THE MENTAL HEALTH OF THE CHILD

Julius Segal, Ph.D., Editor

Authors:

Antoinette Gattozzi
Gay Luce
Maya Pines
Clarissa Wittenberg
Herbert Yahraes

U&SUPT.OFDOC&

Program Analysis and Evaluation Branch
Office of Program Planning and Evaluation
National Institute of Mental Health
5600 Fishers Lane
Rockville, Md. 20852

June 1971

Public Health Service publication No. 2168
Printed 1971

For skle by the Superintendent of Documents, U.S. Government Printing Office

Washington, D C 20402 - Price $5

Stock Number 1724-0132

Page

FOREWORD v

PREFATORY NOTE ix

STUDIES AND DEMONSTRATIONS IN PREVENTION i

Nursery Schools in the Service of Mental Health 3

Reducing the Effects of Cultural Deprivation 25

A Pre-school Program for Disadvantaged Children 44

Baker's Dozen: A Program for Training Young People as

Mental Health Aides 59

Behavioral Consequences of Alienation 72

Alternatives to Violence 86

Child Development Counselors: Lessons From Their Training

and Use 104

Operation Hope: Educating New Leaders 113

FACTORS THAT INFLUENCE THE CHILD'S MENTAL

HEALTH 129

Childhood Influences on Intelligence, Personality, and Mental

Health 131

An Anthropological Investigation of Child-Rearing Practices

and Adult Personality 155

The Effects of Early Experience on a Child's Development . 178

Determinants of Mother-Infant Interaction 193

Studying the Mother-Infant Relationship for Clues to the

Causes of Aberrant Development 200

How the Child Separates From the Mother 211

Early Social Development in Children 222

Family Communication and Child Development 235

Young People of Normal Mental Health 240

The Impact of Visual Media on Personality 247

Parental Behavior and the Origins of Schizophrenia 267

The Causes and Treatment of Childhood Schizophrenia . . , 293
Schizophrenia: New Light from the Life Histories and

Biochemistry of Siblings and Twins 310

in

Page

ADVANCES IN DIAGNOSIS AND TREATMENT 327

Why Adolescents Kill Themselves 329

Studies of Child Abuse and Infant Accidents 343

The Re-Education of Criminals 371

Delinquent Gangs' An Answer to the Needs of the Socially

Disabled 384

Delinquent Girl Gangs 395

Psychiatric Drugs for Children 398

Brief Psychotherapy vs. Drugs: Fitting the Treatment to the

Illness 401

New Approaches to the Treatment of Very Young Schizo-
phrenic Children 412

An Experiment in Foster Care for Seriously Disturbed Boys . 429

The Role of Learning in Relapse of Narcotic Addicts 440

Psychodynamics of Asthmatic Children 450

Infant Stimulation As Part of Well Baby Care in A Dis-

advantaged Area 463

BASIC STUDIES OF CHILD DEVELOPMENT 477

The Development of Intelligence in Babies 479

How Children Learn English 491

Experiments in How We Learn to Coordinate 495

Heredity's Effect on Behavior Under Stress 503

Genetics of Human Behavior 512

Patterns of Sleep Over a Lifetime 523

The Physiological Imprint of Learning 539

Biological Bases of Memory 557

Hormones in the Development of Behavior 567

ACKNOWLEDGMENTS 588

In every age, men have recognized the special importance of the child's
role in society. Whether through ancient tribal initiation rituals or the
pronouncements of contemporaiy psychoanalytic theorists, each succeed-
ing civilization has acknowledged that the child will cany after us the
imprint of the world we create around him. Paradoxically, however, the
child has also been a traditional victim of our neglect and abuse. An
inability and unwillingness to respond to the special needs of the child
haunt us in the ghettos and suburbs of twentieth-century America as they
did in the factory towns of the industrial revolution. Tonight, thousands
of American children will fall into a troubled sleep, bearing the scars of
emotional tiauma and physical pain inflicted by adults no less misguided
and ill than those who nurtuied the cruel children's prisons of eighteenth-
century England. We have not yet matched our best instincts and insights
with comparable action— and nowhere is this more apparent than in the
field of mental health.

Over one half of the U.S. population is now under 25 years of age, and
it is estimated that over 10 percent of this precious resomce— about ten
million of oui youth-require mental health services. Their needs range
from hospitalization to treat and reverse serious psychopathology, to the
early interventions of doctors, teachers, counselors, and parents able to
handle mild, transient problems and thus prevent later developmental
crises.

The child's mental health needs have challenged our best efforts to
date. In the State mental hospitals of our Nation where, for the past
fifteen years, the number of adult patients has been declining steadily,
both the first admission and the resident population rates for children
have increased at an accelerated pace. While many children are receiving
the mental health care they require-some 52,000 in community mental
health centers, 33,000 in public and private mental hospitals, 26,000 in
residential treatment centers, and 526,000 in psychiatric outpatient
clinics-millions more are going without help. Of the Nation's ap-
proximately 2,300 mental health clinics in 1968, somewhat less than one
tenth were child guidance clinics; moreover, only 40 percent of the
268,000 patients under 18 years seen at such clinics were actually treated,
the remaining 60 percent receiving no more than a diagnosis. A large
proportion of all counties in the United States are without mental health
clinics altogether, and most of these also lack agencies that substitute in
some measure for such services. The cost to society— in wasted resources
and human suffering-can only be guessed. But in 1969 nearly one million
children aged 10 to 17 were brought before juvenile courts, and during the
past two decades the suicide rate among adolescents and young adults has
increased by 60 percent.

These data do not minimize the success of efforts to date conducted or
supported by the National Institute of Mental Health. From its very incep-
tion, NIMH has devoted a large part of its program to activities relating to
the mental health of children, many of them reflected in this report.
Contributing to the overall effort today is the entire range of the Insti-
tute's resources.

• Community mental health center programs now not only provide
mental health services directly to troubled children and their families, but
also serve as a rich vehicle for acquainting a variety of community agencies
with mental health principles and practices, thereby enabling them better
to foster the well-being of our young. In nearly two thirds of the com-
munity mental health centers funded to date, specialized services to
children are highlighted.

• Projects designed to improve the Nation's mental hospitals have
focused heavily on upgrading the quality of institutional care provided to
seriously disturbed children.

• Interdisciplinary training programs directed at developing skilled child
therapists have been expanded, many of them emphasizing the prevention
of child behavior disorders. Such programs have led to a significant
increase in the number of professionals and paraprofessionals qualified to
work with children and their families in both preventive and remedial
activities.

• New light on the dynamics of child development— and on our ability
to influence its course— has been shed by scientists devoted to child re-
search. Our understanding of the complex biological, psychological and
social factors involved in mental illness has considerably increased; we are
better able to detect incipient emotional trouble and therefore to take
preventive and curative measures early, when they are most likely to be
effective. A number of promising approaches have been tested and
demonstrated for building in the very young a strong foundation for both
emotional health and intellectual competence.

• And, finally, we now have more hopeful treatments for the whole
range of childhood disorders—from problem behavior to childhood
schizophrenia.

Despite such advances, much more must be done. Activities directed at
improving the mental health of children now carry the highest priority for
NIMH. Our intent is to provide a base of knowledge, techniques, man-
power, and services that will not only reduce significantly the number of
our mentally and emotionally ill children, but also enhance the well-being
and productivity of all our youth.

This volume, issued as the Institute marks the 25th anniversary of the
National Mental Health Act, is intended to display the range of the Insti-
tute's current and past efforts in the child mental health field. Rather than
an exercise in self-celebration, the report serves as a reminder of the
beginnings that have been made-and of what more we can do. It is
intended to stimulate further the creative approaches of research person-
nel, and to provide information of value to clinicians and all those

others who deal directly with the child. A number of studies and demon-
strations reported here may well serve as models for future efforts else-
where.

Highlighting its importance, the area of prevention is the focus of the
first section of this volume. Emphasis here is on the child in his normal
environment— the family, the school, the community—where the task is to
so enhance the child's environment that mental health is maintained and
pathology is aborted at the primary source, before the child must be
separated from the general population. A number of NIMH efforts in this
direction are reported here— including attempts to demonstrate methods
for enriching the intellectual and emotional world of deprived and
minority children and to show how the corrosive effects of prejudice can
be prevented early in the life of the child. Another goal is to help the
older disadvantaged youth, often troubled and embittered, to work for his
dreams and realize his capabilities. One team of investigators has
provided dramatic evidence that worldly wise, angry young people in the
ghetto need not live without hope; properly motivated and placed in a
college setting, they demonstrate untapped depths of wisdom and ability.

The second section of this volume is concerned with a variety of
circumstances that can affect a child's mental health-and thus relates not
only to the etiology of behavior problems, but to their prevention as well.
Reported here are a number of the Institute's efforts to learn precisely
which factors in early life either enhance or reduce the potential for
intellectual growth and emotional well-being. These projects include, for
example, detailed observations of the effects of various relationships
existing between infants and mothers, as well as long-range studies of the
connection between certain characteristics of parents and the psycho-
logical development of their children, One study is concerned with the
influence of visual media-in particular, of television-on childhood
behavior, while another deals with child-rearing practices and outcomes in
societies other than our own. The section ends with three studies of
factors contributing to schizophrenia— the most serious and devastating
form of mental illness, which so often and so cruelly afflicts young
people.

The third section presents an array of projects concerned both with
improving the diagnosis and treatment of disturbed children and
adolescents, and with increasing our understanding of some of the con-
ditions leading to disturbed behavior. The studies reported here deal with
a broad range of problem children-including those diagnosed as
schizophrenic, neurotic, and hyperkinetic, as well as seriously disturbed
boys, members of delinquent gangs, narcotic addicts, young criminals, and
suicidal adolescents. The range of efforts in their behalf is also very broad,
embracing, for example, behavior therapy, brief psychotherapy,
psychiatric drugs, a foster home program, hospitalization and counseling
for potential suicides. The children of poor, unmarried, teenage girls, one
recent study shows, need not be victimized by the nature of their
beginnings-nor are the mothers themselves without the capacity for
psychological growth and maturation. Another report describes attempts

vii

to understand and deal with the problem of child abuse, and still another
provides a demonstration of how young criminals might be reeducated in
a prison setting.

The foundation of all our efforts in child mental health must be basic
research-an effort to understand both normal and abnormal development
and behavior-and the final section of this volume includes reports of
some of the Institute's endeavors in this area. The studies described here
explore such fundamental variables as the biology of learning and
memory, the effects of hormones on behavior, normal and abnoimal
patterns of sleep at various age-levels, the influence of heredity on
behavior under stress and on intelligence and personality, and the growth
of intelligence hi babies. Findings from such studies considerably
strengthen our ability to prevent and treat mental disorders in children.

The NIMH efforts in the child mental health field clearly cannot be
encompassed in a single report, for they involve the varied activities of
research scientists, clinicians, community agencies, and training insti-
tutions. The program includes work in the most basic sciences— for
example, in biochemistry, genetics, and experimental psychology-along
with clinical and community studies. The examples reported here, how-
ever, will help provide the reader with an appreciation not alone of the
scope and complexity of the NIMH program in behalf of children, but also
of its guiding purpose. Underlying all of the Institute's varied efforts-
from basic research to community consultations— is the endeavor to meet
our children's mental health needs.

In pursuing its child mental health program, the Institute seeks every
opportunity to collaborate with programs and agencies whose efforts
complement our own. The factors affecting the mental health of children
are too vast and complex to be dealt with by any one organization alone.
Mental health problems make themselves felt in virtually every aspect of a
child's life— at home, at school, in the world of work— and their solution,
therefore, requires the contributions of agencies with varied missions and
programs, each emphasizing different dimensions of the child's world.

The mental health of our children rests, ultimately, on the health of our
total society-from the smallest unit to the largest. The stability of the
home and the well-being of the family, the compassion of the surrounding
community, the social conscience and social action of our Government
and its citizens-all of these are crucial. The twenty-five years past arc
viewed with satisfaction, but also with an awareness of the formidable
tasks not yet completed. Our pledge is to continue and expand these
efforts, now well begun, to improve the mental health of children and
thus to enhance the quality of their lives-and our own.

0-

Bertram S. Brown, M.D.

Director
National Institute of Mental Health

viu

This volume, intended to provide examples of past and current efforts
by NIMH in the child mental health field, includes primarily reports
published earlier in the Institute's Research Project Summaries and Pro-
gram Reports series. Seven reports, describing more recent advances,
appear heie for the first time. They begin on pages 3, 44, 1 13, 310, 343,
463, and 479. The earlier reports are presented as originally published,
though in many cases the investigators nave done further work on their
projects, and in some instances the locations of the investigators, their
titles, or the official designation of their institutions have changed.

The 42 reports included here are grouped into four major areas,
covering prevention, etiology, diagnosis and treatment, and basic research.
To a considerable extent, however, the groups overlap; work reported in
one group, in a number of cases, could have been included with equal
justification in another. It should be emphasized also that the projects
described here were chosen simply as representative of NIMH activities in
the field of child mental health. They do not deal with more important
subjects and they do not present more useful findings than do scores of
other Institute-supported projects in the area.

IX

Bitter are the tears of a child;

Sweeten them.
Deep are the thoughts of a child;

Quiet them.
Sharp is the grief of a child:

Take it from him.
Soft is the heart of a child:

Do not harden it.

Lady Pamela Wyndham Glenconner

Investigator.

Mary B. Lane, Ed.D.

San Francisco State College

San Francisco, California

Prepared by.

Herbert Yahraes

Introduction and Summary

An Institute-supported program in San Francisco has been quietly
fighting racial prejudice-regarded as a major mental health problem-
through the medium of little children. In an area of the city known as the
Western Addition, where slum neighborhoods are being replaced by urban
renewal projects, to the resentment of the predominantly poor and
predominantly black population, this program has demonstrated that a
special kind of nuisery school can bring together all kinds of people.

Mary B, Lane, professor of education at San Francisco State College,
who conceived the program, calls this kind of nursery school cross-cultural
or multi-cultural. By this she means a school that enrolls children of
different races and socioeconomic levels and uses special curricular and
organizational devices to promote interfamily, interrace relations.

In 1966, Dr, Lane and her associates opened three nursery schools for
60 children, selected in door-to-door canvassing, who had recently passed
or soon would pass their second birthday. The racial and socioeconomic
background of the children reflected those of the community. About 60
percent of the youngsters were black, 30 percent white, and 10 percent
Oriental, meaning Chinese, Japanese, or mixed. About one-third of them
came from families living in low-cost public housing; one-third of the
children came from new, cooperatively owned, apartment development
for people of middle income; and the rest from families in individual
dwellings. This last group-designated for research purposes as the
"random housing" group-ranged from families of unskilled workers to
those of professional men. In each racial group, the families represented
several socioeconomic levels.

The schools were known as Nurseries in Cross-Cultural Education, or
NICE. They were in rooms lent by the Buchanan Y.M.C.A., the Christ
United Presbyterian Church, and-in a public housing unit named the
Westside Courts— by the San Francisco Housing Authority.

Each school was staffed by a professionally trained nursery school
leader, an assistant who had had experience with children, though not as a
teacher, and a part-time aide who was from the community. The staffs

were tacially mixed-white and black. An effort to recruit Oriental
teachers failed. One teacher or aide at each school was a man because Dr.
Lane thought that the many children, about a fourth of the total, without
a male figure in the home should have one at school.

The general aim was to assess the schools as instruments for promoting
mental health in a community subjected to the stresses of redevelopment.
Results were to be appraised in teims of factors related to mental health
such as basic trust, autonomy, initiative, cognitive development, and social
competence, in the case of the children, and, in the case of the families,
social competence, intergroup acceptance, and utilization of community
resources.

The program stressed parent involvement. Mothers gradually took on
the role of aides during the sessions-three houis every morning -and
participated in a numbei of after-school activities, sometimes with fathers.
During the second year, responding to mothers' requests for more in-
formation about techniques of teaching, Dr. Lane gave a 30-hour training,
course, modelling it on one of her courses at San Francisco State for
prospective nursery school teachers. To meet the demand the following
year, two such courses had to be given.

During the second year also, both to promote closer ties between
school and home and to help mothers become better teachers of their
children, the program developed a series of "home tasks"— things to be
made, experiments to be tried, stories to be read, games to be played. One
of these was taken to each home each week by a member of the staff, or
by students from San Francisco State, and explained to the mother. The
following week it was picked up and the experiences of mother and child
recorded.

Befoie the start of the third year, the parents formed the Parent
Advisory Council to help guide NICE activities during the final year and
to decide the program's future when Institute support ended.

A PARENT SPEAKS

Following is a slightly abridged version of an interview with a mother
who had two daughters in one of the cross-cultural nursery schools. One
girl attended the full three years of this NIMH-supported project. The
other was enrolled during the final two years and now attends the project's
independently-financed successor, in which the mother is a teacher's aide.

Mother: Everytime I talk about it, I get sort of shook up inside because
they learned so much, and I have learned so much. I don't mean only
about toys and things like that but the most important-that I have
learned how to raise children. Because I sort of talk and raise my girls
differently from what I did my older boys. And because 1 didn't know -.

Here is an instance. I was a very over-protective mother and I'm a very
firm mother. Well, in the NICE project I was taught how to help the
situation. I used to spank a lot, you know. But I learned you can talk to
children-sit down and talk to them just like you and I talk.

Like when we first went to the nursery school. When the children was
all 2 and 2 and 1/2, they would hit each other a lot-and the teacher would

The NICE schools closed in 1969, after three years, and the children
went off to kindergarten in public schools. Thanks to decisions by the
Parent Advisoiy Council and the commitment of Mary Lane and her
associates, however, the end of NICE was the beginning of CCFC, which
stands for Cross-Cultural Family Center. This is an organization with the
same goal as the original program— the promotion of interfamily, mtei-
race relations— and a membership comprising 33 of the families who were
in NICE and a number of new ones. It operates a nursery school taught
largely by mothers who were tiained on the job in the original cross-
cultural schools, and a kindergarten-supplement program for the original
NICE children. It also offers afternoon and evening programs for other
children and for parents. The Center depends on fund-raising activities by
its members and on tuition fees from the children of nonmembers. It is
pleasantly housed in the new church school building of the Unitarian
Church, in the area served by the original project.

The accomplishments of NICE may be summarized as follows;

1 . The children at 5 seemed to be without racial or class prejudice. As a
matter of fact, on the basis of observations by project and teaching staffs,
the children got along throughout the project at least as well as any
homogenous group of children. Friendships across group lines were com-
mon. During the second year, teachers reported 31 "best friend" pairs. In
more than half the cases, these were inter-racial. Since the children were
2 years old when the project began, they have little or no conscious
memory of associations in groups of children before that time. They will
remember only playing with children of cross-cultural backgrounds.

2. The children made gains in intellectual development, as shown in
Table 1 . On the Stanford-Binet intelligence scale the average I.Q. gain over
three years was nine points-from 102 to 111. On the Peabody Picture
Vocabulary Test, which tests a child's understanding of words, it was 24
points-from 90 to 114. The gains in all cases were significant at the .01
level, and there were no significant differences between housing groups.

walk up and say: "She doesn't want you to hit her. It hurts when you
hit." And I'm just looking and observing. And then when the child would
hit again, the teacher would say: "No, that hurts, and she doesn't want
you to hit and we're not going to let you hit." It was so fascinating to me,
I really wanted to stay every day just to watch how they handled the
children.

You would have acted differently?

Right, I would have wanted to say: "Now listen, you know better than
to hit," and I would have said, "Now you hit her back!" But that's not the
way, I've learned. There was my little girl and another little girl, which is a
little Japanese, Neisha. Every day this little girl, sometime during the day
she would walk over and just hit my little girl. And my little girl would
start crying. You know me-I'd say, "This is just killing ray child!" So one
day I got so mad about it I went to the teacher and I told her, "Now
listen, I'm just sick and tired of my child getting beat up every day," and I
said, "Something should be done about it." And the teacher, very calmly:
"Well, have you any idea why Neisha hits her all the time?" And I said I'd

White children as a group went from 108 on the Binet test to 121 ; black
children, from 97 to 103. On the Peabody, white children went from 93
to 117; black, from 83 to 110. These differences between racial groups
were significant at the .0 1 level , with the white children scoring higher than
the black children at all time-points. The staff points out, however, that it
is impossible to group the children on the basis of a single factor, either
race or type of residence, without the other factor entering in, Oriental
children were omitted from analyses involving racial groups because there
were too few for proper statistical treatment..'

TABLE 1
Average IQ Scores by Type of Residence, First and Last Tests

Public Random St. Francis

Ho using ^ Housing^ Square^

First Last First Last First Last

Stanford-Binet 98 103 100 113 108 117

Peabody 83 115 92 108 94 117

1. Low income; black.

2. All income levels; about half black, half white and mixed.

3. Middle income; about 2/3 white, 1/3 black.

3. The children made significant gains in social competence as
measured on the California Preschool Social Competency Scale. This scale
covers a wide range of behaviors such as response to routine, response to
the unfamiliar, following instructions, making explanations, helping
others, initiating activities, reacting to frustration, and accepting limits.

not even thought about why she was hitting her. And the teacher said,
"She wants to be loved." The teacher went on to tell me. "When they are
that little, they don't know how to talk; this is their way of com-
municating with one another. That's it. And the only thing she is trying to
tell Mary is1 'I want to play with you. I want to be your friend.' " And
within a month and a half those two were playing together every day. But
you know, my concept of it was a completely different thing, but the
wrong thing.

I wish the project had come around sooner, that my older children
wouldn't have had to suffer so much because I din't know. And I think
that the parents in this project got out of it just as much as these children
or even more.

My oldest girl was in the nursery school for three years. Well, I'm just
so proud of her in the kindergarten. She's marvelous in her class, and they
are getting her ready to read. Because all the stuff that the older children
have done, she's had it. You know, too, at times she helps the teacher a
lot-to help the other children. And so I'm very proud of my children,
those who were in the project. Because they have learned education-wise
and they have learned one of the things -.

The national norms for this scale are based on teacher ratings of children
in preschool programs. On the first test, the average score placed the
NICE children at the 38th percentile; on the last test, at the 78th. The
gains were significant for each housing group, and the differences among
groups were also significant. Over the thiee-yeai period, there was no
significant diffeience between white children, who went from the 41st
percentile to the 77th, and black children, who went from the 35th
percentile to the 76th.

TABLE 2
California Preschool Social Competency Scale (Percentiles)

First Test Last Test

Public Housing1 31 72

Random Housing^ 43 82

St. Francis Square3 42 78

1. Low income; black.

2. All income levels; about half black, half white and mixed.

3. Middle income; about 2/3 white, 1/3 black.

4. The children made significant gains, too, on three scales developed
by the project in an effort to guage three qualities it was trying to instill—
basic trust, initiative, and autonomy. Among the housing groups no
significant difference appeared except in autonomy, where the St. Francis
Square children rated higher than the other groups at all times. When first
tested, white children scored significantly higher than black in trust and
initiative but not on autonomy; on the last test, the scores for whites and

Just like my older daughter, she can tell the difference between a
Japanese and a Chinese. And I can't. Because she has been witli these little
Japanese and Chinese children. I'll say; "Oh, that's a Japanese," and she'd
say, "Oh, Mamma, that's not a Japanese-that's a Chinese!" And this really
shocked me because I couldn't tell the difference. But through us having
all the different races Of the children, and them playing with them -.
Sometime you'll kind of look at a child and they'll be looking at a child
and looking at the child-they're really looking at the child, different from
the way we look at it. They look at how the eyes are made, and how the
nose, and how everything, you know.

Anybody tell you that?

No—but I sort of learned from my children. Children look at each other
in a different way than we look at each other. And then, you know, they'll
go up and touch, you know, to make sure that the hair- they want to
know why is her hair different from mine. And they have to touch it to
really feel and see that that is different and everything. So kids is just
amazing. And when I started in the nursery I would go almost just for

black on each of the characteristics were virtually identical, The three
scales are experimental and their validity untested.

5. In their attitudes toward members of other racial and socioeconomic
groups, the parents have become considerably more open-minded and
accepting. This is the impression of the staff and of persons who talk to
representative parents. It is substantiated by an analysis of ratings for each
mother on several experimental scales measuring "intergroup accept-
ance." The ratings were made at the start of the project and again at the
end. Over the three years, both the average score of all the mothers and
the average score of the mothers in each housing group increased sig-
nificantly. The groups had the same relative positions at the end as at the
beginning; first, St. Francis Square; second, Random Housing; third,
Public Housing.

6. The mothers have become more skilled as parents and more
competent as members of society. Ratings on experimental scales to
measure child-rearing practices and general social competence showed
significant increases over the three years. The St. Francis Square group
again rated first and the public housing group third. Four of the mothers
have become assistant teachers in the CCFC nursery; another is in charge
of the Center's kindergarten-supplement program; another supervises the
after-school activities for older children at the Center. One mother has
been employed by the San Francisco Unified School District as a school
aide. A mother who was on welfare when the pioject started has become a
secretary to a Y.W.C.A.

7. Through open meetings, social affaiis, membership of parents in
community organizations, and other ways, the project's influence has
extended beyond the families immediately involved. Staff and parents
have been called on to consult Head Start staff, participate in panels,
appear on TV, and help the State prepare for the 1970 White House
Conference on Children and Youth. The NICE schools were used in the
training of psychiatric social workers, teachers, home economics students,

observation and then the more I went, then the more I wanted to go. I
wanted to learn. Then I started working in the school with some of the
children, which was very rewarding, because you teach them a lot, and you
learn a lot from them.

Has it been easier in kindergarten for your girl than for your boys?

It's a lot easier. It's so much different in boys and girls anyway; girls
seem to be a little bit more advanced than boys when it comes to learning.
But my daughter, she is as much different in her going to the kindergarten
than when my sons went as day and night. Because when she went in there
she didn't have this fear of being away from me, for a time, because she
had learned that, and then she had learned how to play with other children
and she had learned all these different kind of peoples. Just like a teacher
m kindergarten is a Japanese and they have an aide in there that's a Negro.
Well, she goes in and she doesn't even think about race or why is a person
different.

She notices that -?

In the Nice project, even when they was two and a half and three they
noticed. And they wondered and they questioned the teachers to some

health personnel, and members of the Teachers Corps and the Neighbor-
hood Youth Corps.

8. Materials useful to other projects concerned with preschool children
have been developed. These include descriptions of the home task pro-
gram, which has been widely followed by Head Start and other projects
for disadvantaged children, and of suggested processes and materials for a
"multicultural curriculum," and a film, "Swimming in the Nursery
School." Two films are being developed from NICE video tapes-on the
introduction of a child into an ongoing progiam and on the use of
cognitive materials to stimulate thinking. A series of pamphlets on various
aspects of the program is planned.

If money becomes available, the team that directed NICE— Mary Lane
and two research associates, Mary S. Lewis and Freeman F. Elzey-expect
to follow the NICE children through the fifth grade of public school.

Birth of NICE

One day eight years ago, Mary Lane received a telephone call from an
acquaintance of hers, a young mother who had recently moved into St.
Francis Square. This is the cooperative apartment development, for
families of middle income, that in 1962 replaced several blocks of old,
three-story houses occupied mainly by low-income black families. Other
blocks of similar houses had been torn down earlier to make room for
luxurious high-rise apartments. The San Francisco State professor, an
authority on the education of young children, was asked to talk with a
group of "the Square" parents about the troubling experiences they were
having.

Dr. Lane knew that the development had been planned as an integrated
community for people committed to inter-racial living and that, while the
majority of the residents were white, a number of middle-income black

length about why, you know. You know, just like the car toon -they
would rub the black children's skin; they would just nib it to see would it
come off. And then they ask a lot of questions. Just like m my daughter,
she wanted her hair to be long like Enrica's hair, and so I had to press hers
out so hers would be long like 'Rica's hair.

Do they ask questions of you?

Yes, Because they are interested in things. And there she was, sort of
questionning. She'd say: "Am I really a black person or am I a blue
person?" The way I feel about it she is a Negro, but sometimes other
people call the Negro a black. And then she said: "Well, I think I want to
be a Negro." I felt kind of fine. You know. And after the questions, it
wasn't no problem at all. They played with each other. They'd even come
and they'd say: I'm a Negro, or black," and "You're a Caucasian," and
"You're a Japanese," and "You're so and so."

How did the teachers explain?

Well, they'd say at times: "Some of us are one way and some of us the
other, you know, but we're all human beings because we are all boys and

girls."

families lived there, too. She now learned that many, probably most, of
the parents in the Square felt frustrated and rebuffed. They had moved in
considering themselves liberals in racial matters and eager to get along
with the black people in the surrounding neighborhood, mainly in public
housing. But facilities in the development such as laundry rooms and
playgrounds were being vandalized, and children from the Square were
being attacked and chased home fiom school by public housing children.
Further, a group of parents had started a cooperative nursery school in a
nearby Y.W.C.A. and had hoped that some of the black public housing
residents would participate, thus giving children from the Square an
opportunity to associate with other children who would be entering
kindergarten or first grade with them. The public housing people had not
responded. Residents of the Square were beginning to feel themselves on
an island surrounded by hostility.

The Square children, Dr. Lane was told, were doing very well in
elementary school, where almost all of the pupils were black. The tcacheis
liked the newcomers and were pushing them ahead. "And," one mother
reported, "27 of us have been appointed room mothers!" Apparently it had
not occurred to her to wonder about the women who had been the room
mothers before. And apparently it had not occurred to most of the
parents to wonder how public housing people might feel when they
looked out of their stark concrete dwelling place and saw the Square-
with its attractive buildings and the neighborhood's only grass and
shrubbery-occupying the site where they or relatives or friends once had
lived.

Says Dr. Lane, "I thought to myself: Here are all these middle-income
children- 200 preschoolers alone-in the Square, And here are all these
children in public housing. How wonderful if we could get a cross-cultural
thing going, and see if we could break down some of the barriers. "

A family-centered nursery school for 2-year-olds would be an ideal
"cross-cultural thing," she felt, because even people of the most diverse

I was prejudiced before I went to the nursery school. Because when my
older sons would come home and would talk about then friends, I would
say; Is he a Negro or white?" And one day my oldest asked me: "Well
Mamma what difference does it make?" And that gave me something to
think about And then I said: "Well, really what difference docs it really
make?" And then 1 stopped asking him that. X

This NICE project has helped me to overcome my prejudice, because
we often have discussions and one discussion was on p e ud ce and 1 S
hr e or four sessions were on ,t. And at the first meeting, everybody was
sort of, you know, not really saying what they really feel So Mrs L7o
she was there. "Now," she says, "You should be fair-you willy feel one
way or the other about it." And by the time we got to the econd nVeeting
.telly came out, what you really feel. And then we learned about each

10

interests, if properly approached, would surely work together more or less
harmoniously for the welfare of their children. And working together,
they would learn to understand and appreciate one another. Their
children would accept inter-racial activities as a matter of couise. Ad-
ditionally, the children fiom disadvantaged families piobably would enter
the public school system, after three years in the nursery school, much
better prepared intellectually and emotionally than could be expected
otherwise.

With the moral support of a number of community organizations and
financial backing from NIMH, through a grant to the College, Dr. Lane
spent a year developing her ideas, finding sites for the schools, talking to
families with children of the right age (only two such families declined to
enter their children), and selecting and training the teaching staff. Assist-
ing her through the life of the project were the research associates
mentioned earlier, Freeman Elzey and Mary Lewis. The three worked
together as a team, Elzey designing the research projects and analyzing
results, Mrs. Lewis conducting the interviews and creating the home tasks
program, and all three sharing administrative duties. A psychiatric social
worker, Stanley Seidertnan Joined the staff some months after the schools
opened and became a part of the team in a counselor role.

Key Philosophy; Complete Acceptance

Members of the teaching staff faced more problems than usual because,
in addition to being teachers, they were expected to act as counselors to
parents and researchers, a function that obliged them to make daily notes
and write a weekly record of both child and parental behavior. During
much of the first year, too, they sometimes found themselves acting as
social workers, guiding low-income families to medical, housing, and other
community services. Even after a social worker was added to the project

Are you from San Francisco?

No, I'm from the South Houston. And I remember one of the things
that used to hurt me— when my mother would say Yes, ma'm and No'm to
a 12-year-old Caucasian. That used to break my heart, because that isn't
right, you know, ffow'rp you . . . How 're you .... And as a child coming
up, it would break my heart, because we were taught at home that you
respect age. And you know, I could understand that. But a person youngei
than me! And I would ask my mother, but she never could really explain
to me, to my satisfaction. Or she'd say: "Well, you have to," or "That's
the rules." And I have always questioned things like that,

There were things that came out during the group discussions here, that
one race don't really know what the other one really thinks until you
really get down and start talking about things. Because just like some of
the white persons would say they were taught you shouldn't marry out of
your race and everything, and a whole lot of this your paren ts don 't wan t
you to. And they were shocked to know that these Negro parents didn't
want their children to marry out of their race. You know, and different
things like this.

11

staff, some of the families preferred to take their troubles to the teachers.
A summer of intensive training preceded the opening of the schools but
could not altogether prepare the teaching staff for what lay ahead.

Probably the most important feature differentiating these cross-cultural
schools from other nursery schools was unconditional acceptance of the
parents as well as the children. It was also the teachers' heaviest burden,
Unconditional acceptance meant, for one thing, putting up cheerfully
with lackadaisical observance of schedule. Children and their mothers
were welcomed any time they arrived. Mothers who wanted to take
children home before the session was over were permitted to do so.
Mothers who failed to pick up their children on time found them being
cared for.

More disturbing practices also were accepted. One mother spanked her
2-year-old daughter for wetting her pants. One father, finding his boy
reluctant to leave at the end of a session, took off his belt and threatened
him. A woman, watching her son, not yet 3, color a picture, insisted that
he stay within the lines. In each case the teachers felt shivers running up
and down their spines but said nothing.

In staff meetings, where problems associated with unconditional accept-
ance were often raised, Mary Lane and her team took the position that if
parents were ever to be open to change, in either child-rearing practices 01
racial attitudes, they first had to be accepted as they were. Some staff
members found it harder than others to go along. One head teacher left
after a few months.

Before the end of the first year, the staff had evidence that the concept
of complete acceptance was paying off. Families were changing their time
habits to conform with the schedule of the schools. Parents were asking
teachers why they didn't punish the children for wetting themselves,
threaten them for not doing immediately what the teacher wanted, insist
on neat performances. And the parents were beginning to listen to the
answers.

Making the Parents Welcome and Useful

When the three schools opened, parents were invited to stay with their
children as long as they wished or felt it necessary. Each school had a
parents' corner, fitted out with comfortable chairs, coffee-making

I owe a lot to this project. I had a distrust for professional people, you
know-that they were one way, and you were another way. But then in
this project I learned that although they are professional people they are
just human beings, like I am. You know, we call the staff by their first
names, and everybody always looks at me when I say Mrs. Lane. Most of
us call her "Mary," but I don't know-I have so much respect for her. I just
come from old-fashioned teaching that when you really respect a person,
and admire a person-. When I say, "Mrs. Lane," well, this is my way of
saying 1 respect her as a person for what she stands for and the things that
she has done. It's not because there is a difference in race or anything like
that; it just comes out of respect. I suppose there is something in me that I
just can't come to call her "Mary." I just can't.
12

materials, and magazines. Some of the mothers, delighted at the op-
portunity to watch their children and to talk with other adults, lingered
half the morning. After a while the "bag lunch" became an institution. On
a certain day each week, mothers would drop in with a lunch bag during
the morning and stay to observe the rest of the session and to eat with the
staff. These informal get-togethers gradually moved from general talk
about children to guided discussions of such subjects as sibling rivalry, art
activities in the home, discipline, and toy selection.

Mothers became more and more deeply involved in school operations.
After a few months, some of them were asked to volunteer as additional
aides because, even with one or two student teachers in each school, the
ratio of adults to 2-year-olds was not high enough to give the children as
much individual attention as they needed. Later, with the approach of
summer and the departure of the student teachers, all mothers were asked
to serve one day a week during the summer session. Most of those whose
time was not otherwise committed agreed. They were invited to bring
along their children, so the workday became also a family outing day.
The following fall, Dr. Lane explained at a parents' meeting that writing
research reports and attending teachers' meetings required a sizeable
amount of the teaching staffs' time. So one day a week the schools would
close unless the parents could take over. Teams of mothers signed up to
do this, each team obligating itself for one day a week foi six weeks.

Home Tasks

At the end of the first year, project staff members and teachers
wondered what else could be done to bring home and school together. In
about half the families, parents rarely visited the schools except to drop
off or pick up the child. And some of the parents who did stay longer
were either uninterested in or puzzled by what was going on, in spite of
staff efforts to communicate. These were parents who in the main thought
of learning only in terms of reading and writing. Some way of getting into
the homes and talking to the mothers individually seemed necessary if
NICE was to be sure it was reaching at least most of the families. The staff
decided on an educational intervention scheme it called the Home Task
Project.

Home Tasks would help parents see and make use of the learning
potential in the play activities of their young children. Parents then would

1 would just like to see that this family center could be continued,
because I think it has helped my children and so many of the children that
need help. And not only the children, but the parents. There was one
mother in particular. Her children were enrolled in the nursery school, and
the first two years she wasn't really concerned about taking care of them.
You know-their eating right and things like that. And since working with
the NICE project and the teachers in it, I have seen a big change in her.
She's a completely different person. And then you ponder. You see a
parent that don't care, and after three years, this parent has just turned
about. It's the sort of thing that makes you stop and think.

13

have greater undeistanding of the school's task and of the way the school
uses materials to help the child learn. Also, by working with the child a
few minutes each day, the mother would begin to view heiself as the
child's teacher as well as his mother.

Mary Lewis spent the summer developing the home tasks and the
explanatory materials to accompany them. The tasks were designed to
extend knowledge, sharpen peiception, develop motor skills, or expand
concepts. They included seeds to be planted, a scrapbook to be filled, a
plank to be balanced on, picture books to be looked at, puzzles to be
worked, pieces of wood to be joined, drawings to be colored, games to be
played, and a turtle to be cared for. At Halloween every family would get
a pumpkin, along with a recipe for pumpkin cookies, a candle, and a list of
suggestions on how a pumpkin could be used to help the child grasp such
concepts as round, orange, hard, and hollow. Altogether, 60 such tasks
were developed.

Every week, at a regular, mutually satisfactory time, each home was
visited by a member of the staff or a student teacher fringing the week's
task and typewritten instructions for its use. The visitor demonstrated the
task, explained the instructions, and discussed the educational value. A
bean bag game, for example, was intended to develop not only eye-hand
coordination but also such space-relation concepts as inside, outside, to
the right, to the left, over, and in front of. A "feel kit" included the book,
"What Is Your Favorite Tiling to Touch?", a bag for holding small objects
to be identified by touch, and instructions for playing the "comparison
game," in which the child is encouraged to name things that are "as

smooth as — ," "as sticky as — ," "as soft as ," and so on. One task

called for mother and child to take a short walk, talk about what they saw,
and draw a rough map locating some of the things seen. Mothers were
encouraged to use the tasks with their children at least once a day. The
following week the visitor returned with the next task and asked about
the child's experiences with the last one and the mother's ideas for
expanding its usefulness as an educational item. Expensive items such as
books or puzzles were picked up foi delivery to other families.

The staff reports that the home task project, which continued through
the second year and 18 weeks of the third, was not without headaches.
Items were lost, broken, or kept for longer periods than scheduled, and
visits were missed because of forgetfulness either by the visitor or the
family. From the comments of the mothers, however, the staff believes
that the project succeeded both in drawing parents closer to the schools
and in demonstrating the potential of the parents as teachers. Visitors
reported increasing interest on the part of most families. Many mothers
reported that home task day had become a special occasion for the whole
family. Even middle-class mothers were surprised at the variety of ways
open to them to develop their children's abilities. After each visit certain
information was recorded on IBM cards. It is now possible to determine,
among other things, which tasks were most used and how a given task was
regarded by mothers and children.

14

A Training Course for Parents

At the request of mothers who said they would like more information
on how to work with young children, the project staff decided midway
through the second year to offer a course in preschool learning and
education. It reasoned that such a course would make the women both
better mothers and more competent aides at the schools. Mothers were
told the course would lun two hours a week for 15 weeks and that they
would be expected to work one day a week in the nurseiy. Out of 32
mothers who were not working, 15 signed up— 8 of them black, 4 white,
and 3 Oriental.

Asked by Dr. Lane, the instructor, why they wanted training, those
who signed up said they would like to learn, among other things:

To talk to my children so they will want to mind.

To get rid of the feeling that I can't handle my child.

To read books with more feeling.

To stay calm.

Educational spots in the city for children.

What to expect of children.

To understand myself better.

The mothers were asked to take notes; readings were assigned and
reported on; small groups were formed to make a special study of one area
of interest chosen by the members (such as books, trips, toys); and each
mother had a semester project-to study and report upon one child, not
her own.

To set the tone for the course, Dr. Lane wrote each mother a letter in
which she said, in part:

You are your child's most important teacher. Did you ever stop to
think what life-long learnings come from you? First of all, he learns the
feel of mother as you feed him, change him, bathe him. He remembeis
this always. From you he learns how much he can trust the world.

Then he learns his language from you. His voice tone and the way he
pronounces words will be much like yours. What he first talks about
will be what he has heard and seen at your knee.

Very important is what he learns from you about how you feel about
people. If you are friendly and helpful and think people are pretty fine,
he is likely to feel this way, too.

He learns very early from you how you feel about him. If you feel your
child is just great for a two- or three-year-old, he'll feel great about
himself. These attitudes that he "catches" from you when a child, he is
likely to keep for all his life. . . .

The course included sessions on child study, art activities, music and
fantasy, books and story-telling, educational trips and toys, and science,
Dr. Lane or another San Francisco State faculty member would discuss

15

the topic and answer questions. Then a few of the weekly reports written
by the mothers as part of their semester project-to observe a
child—would be read and discussed. During the following week the
mothers were expected to try out in the schools something they had
learned during the session. If the topic had been music, for example, they
were asked to lead a small group in a song or a dance.

"I thought we were just going to learn from Mary Lane," says one
mother, whose views of her experiences during the course were typical.
"And that would have been fine. But we learned also from each other.
One of us would have a problem with children, and we'd lay it in front of
everyone, and people would come up with some very good ideas. You
learn ten times more with people sitting around a table with you. You let
all your feelings out, and everybody gains."

The course was so popular that next year Dr. Lane offered it again.
And, to meet the demand from working mothers, Mrs. Lewis offered a
similar course in the evening, which was attended by a few fathers as well
as by mothers. All in all, about 35 parents took one of the courses and
earned credits from the extension division of San Francisco State for
doing so. The credits were appreciated particularly by half a dozen of the
mothers who wanted to work in day-care centers and had to meet the
requirements for a license.

Developing Se!f-Esteem

In the beginning, NICE made no effort to develop a cross-cultural
curriculum for its children. Since the project was serving a cross-cultural
population and had a cross-cultural staff, it assumed that the curriculum
was bound to be cross-cultural. The staff put its conscious emphasis on
developing skills-perceptual, motor, cognitive-and trust, autonomy,
initiative, and social competence.

Because many of the children had culturally poor backgrounds, the
project did place more than the usual emphasis on building self-
esteem-on helping each child see himself as someone special and as a
member of a special family. The staff followed a number of fairly
common procedures, such as hanging a full-length mirror low enough for
children to see themselves, using the children's names in talking and
singing, and identifying and exhibiting work produced by the children.
Going farther, the teachers also gave special attention to children with
markedly low self-esteem, sometimes even assigning one staff member to
care exclusively for one child. For example, a student teacher was assigned
to Buddy, who was too fearful to talk or even, at snack time, to reach out
for a glass of fruit juice. The teacher took him on walks, read to him,
played with him. Half a year later, Buddy was talking a blue streak but
articulating poorly, so the staff guided him and his mother to the college's
Communication Disorders Clinic. By the end of the year, Buddy's self-
concept had improved dramatically.

16

For deepening and expanding a child's good feelings about himself and
his family, NICE regards photography as its most important activity. Each
school had a simple camera, which the teaching staff was encouiaged to
use for portraits, candid shots, and group activity pictures. Since the
project had access to a dark room at the College, processing was in-
expensive. The schools usually had many pictures on display. In one
school, children's photographs were used to identify lockers; in another,
each child had his own bulletin board, on which weie displayed his
photograph, samples of his work, and group photographs of classmates.
All the schools displayed family pictures. The photographs were often the
subject of discussion among children, teachers, and parents.

Special events such as birthdays, picnics, and trips were photographed
and the prints put on display for use as learning materials. And occasional-
ly pictures taken a year or two earlier would be brought out for comment.
"When a child has a opportunity to see himself in this way through time
and space," Dr. Lane comments, "He views himself as a changing person.
We feel he gains a sense of openness about himself and his potential. These
are important factors in the self-concept."

During the second year, families, teaching staff, and children
cooperated to produce a "Self Book" for each child. This was a collection
of photographs that the child wanted to have in his own book, together
with his own comments about them. Each book began with a photogiaph
of the child and a typed transcript of the story he dictated about himself.
Then came other photographs-usually of members of the child's family,
his teachers, and his favorite toys. The photographs at home had been
taken by someone from the school in accordance with the child's wishes.
With each picture was a story the child had dictated. The Self Books were
kept in school. Often a child would ask a teacher or a parent to read to
him from his book; he would listen delightedly. From this project, Dr.
Lane is sure, the child learned not only that his family was something
special but also that someone cared enough about him to record what he
said.

Later, staff members and parents under the guidance of Mary Lane
produced for each child "My NICE ABC Book," which opened with his
photograph and included, under F, a photograph of his family. Each of
the other pages had a verse about one of the letters of the alphabet,
together with photographs of the children or teachers whose first name
began with that letter. Thus a child could learn the ABC's and at the same
time the names of all those who shared with him three years in the
nursery school.

A Cross-Cultural Program

It was apparent from the start that some of the children noticed dif-
ferences in one another, for they would feel another child's hair as
through to verify its differentness, and touch and comment on each

17

other's skin. Late during the second year, when the children weic ap-
proaching their fourth birthdays, the teachers noticed that they were
becoming more consciously aware of differences, or at least talking moic
often about them. One noon as one of the black mothers was leaving
school with her daughter-after a session in which the music hour had
included a song about black being beautiful— the little girl said : "Mommy,
you and I are white, aren't we?" She was one of seveial black childicn
who identified rather strongly with one or another of the white teachers.
On the othei hand, one of the white boys kept telling his mother he
wished he were "dark, like Michael," because he liked Michael very much.

So the staff began thinking perhaps it should be paying more conscious
attention to the cross-cultural elements of the ciiiriculiun. The feeling was
that children should not only become aware of differences between one
ethnic or cultural group and anothei but also come to accept and ap-
preciate these differences. In addition, a child's image of himself and Ins
family should be tied to that of his group. As Dr. Lane puts it: "We hoped
each child would come to feel that ll am Bobby Lewis. My folks are Uic
Lewis family. And the Lewises are — black, white, Oriental, mixed, or
whatever.1 "

Parents were urged to attend staff meetings to talk about these goals
and what more might be done to attain them The parents were to consti-
tute the Parent Advisoiy Council. Only nine accepted, but these were
outspoken in their comments. For example:

"There are people I don't want him to identify with, no matter what
color they are-people, you know, of lower character."

"How can you explain to your child that you get angry with people
in high office and that you don't like a ceitain person even if he is
President?"

"Yeah, it's hard to build respect, especially in a black home. They're
always talking about bad white people-they do this to us, and they
did that, and we're going to get 'em. Like she knows somebody white
killed Martin Luther King, but she's kept her white friends."

"You know, it's a funny thing, I never even thought of it that way
myself-a white man killed Dr. King, but I guess it's a way of
thinking about it. I think I would say it was an individual not a while
or black man, and it's wiong no matter who did it."

"I think the children need to know blacks do bad things and they do
good things, just like white people and pink people. But I think when
my little girl sees a black do bad things she thinks it too bad."

"Well, I want my kid to feel good about herself and if she does that,
maybe she won't have to hate other people."

18

All of the parents were kept informed of the Council's discussions. The
attending parents, as well as some of the others, contributed ideas for the
third year's program. In the end, the staff had many suggestions for ways
of teaching differences, particularly in matteis of food, dress, games,
customs, and holidays. Through its own research it also had a list of books
and anothei of recorded music and songs, from many cultures, considered
appropriate for prekindergartners.

As the result, the third-year program included:

• Development of a multicultural calendar that marked such dates as
Martin Luther King's birthday, Chinese New Year, Japanese Children's
Day, and Jewish and Christian holy days. The calendar was intended
primarily for home use, but some of the special days were observed in
the schools.

• Visits to the library during Negro History Week to sec an exhibit of
photographs of people around the world.

• Showing of a moving picture about a black boy and a white boy
attempting to span an ocean inlet by building a biidge together. ("Bridge
Tomorrow," written, filmed, and produced by Oscar Williams, San
Francisco State College.)

• Displays of photographs of prominent people-white, black, Oriental.

• Games and songs from different cultures.

• Exchange of recipes among the mothers and the sharing of charac-
teristic dishes-such as chittlings, teriyaki, sweet and sour pork-at the
weekly school lunches.

Services for the Family

The teachers frequently encountered problems that could not be
handled with simply a few minutes of advice. For instance, a mother was
sick, and her family needed homemaker assistance; a marriage was on the
rocks; a family had to move but could find no place to move to; a credit
agency was threatening suit; an older child was in trouble with the police.
Particularly during the first half year teachers often spent part of their
afternoons, which had been set aside for planning and report-writing, and
even some of their free time, consulting with such families and bringing
them together with the appropriate community services.

The addition to the staff of a psychiatric social worker eased the sit-
uation. He acted as a consultant to the teachers and a counselor and
advocate of the families requiring help. With the aid of two graduate
students in social work, he studied the public resources of the area-
health, employment, legal aid, and the like— and told people about them
in a meeting to which the parents in particular were invited but which was
open to all, A number of the poorer families hadn't known what was
available to them almost next door, free.

During the three years of the NICE schools, the project helped 3 5 families
to deal with specific mental health problems through counseling by the
social work staff or action by other community agencies, or both. About

19

two-thirds of the cases were child behavior problems, half of them
involving nursery school children and the other half, oldei brotheis and
sisters.

In other ways, too, the pioject became involved with a family's older
children. For boys whose parents said they needed something to do, it
organized a club led by students in educational sociology at San Francisco
State. One activity was going to baseball games with their fatheis. If a
boy's father couldn't attend, the boy went with another father in the
project. A black boy might be picked up by a white father, a white boy by
a black one. Other activities included trips on foot or by bus to interesting
places in the city. For older girls, the project formed the Girls' Friendship
Group, which met weekly with a student leader. For a dozen older
children who were having trouble in school, it procured tutoring by
student teachers. During the project's last summer, the social work staff
arranged for 19 youngsters to attend good, small camps.

In addition, the schools provided intervention almost continuously and
as a matter of course as the teachers identified special needs. If a child
always hung back from group activities, he was encouraged in a number of
ways to join in. If a child lacked a male model at home, he was chosen
more often than others to go with the man teacher to buy fish for the
acquarium or food for the guinea pigs. If a child just sat and looked blank,
though his hearing was normal, teachers made sure that some smiling,
interested person was near him, often talking to him. Because staff
members knew children and families so well, the project estimates that all
but half a dozen families received special help directly related to menial
health.

The project was concerned with physical health as well, the purpose
being to see that each family knew how to get the medical services it
needed. To this end, the project required annual physical examinations Tor
both children and mothers, and when necessary it guided the families to
facilities where the examinations-and any treatment required-could be
obtained at little cost. It won the cooperation of the Health Dcparlmcnl
in providing vision, hearing, and dental screening services. It arranged for a
comprehensive serology test of all the children, primarily to screen Cor
sickle cell anemia, which has a relatively high incidence among black
people. Several children were found to have the condition and (heir
families were helped to get the necessary medical care. The projccl
arranged also for parents to participate in several discussions of cancer.

Other activities included:

Swimming classes for nursery school children and their older brothers
and sisters.

A dancing class for mothers.

Swimming parties for mothers.

A sewing class for mothers, with aid and instruction from home
economic majors at San Francisco State.

An art class for mothers and fathers.

Moving pictures on family life and education.

20

A monthly meeting for mothers, where the topics discussed included
"Piejudices and How to Discourage Them" and "What Advice Would You
Give to an 18-Year-Old Mother-To-Be?"

School and tri-school suppeis and picnics.

Involvement in the Community

The project was alert for ways to facilitate the involvement of families
in community affairs. One opportunity occurred early when the play yard
of the West Side Courts Nursery School was subjected to vandalism,
including the partial burning of a playhouse. At a meeting of the residents
of this public housing unit, quickly ananged by the staff, people who
were not members of the project had an opportunity to voice hostility
toward the school for serving "white kids" as well as black. And parents
with children in the school had an opportunity to explain the goals of the
project and thus win for it greater acceptance by their neighbors.

Beyond that, the discussion brought out a legitimate grievance: the lack
of recreational facilities, particularly for children from 9 through 13, who
seemed to have done most of the damage. Through the efforts of the head
teacher, a second meeting was attended by the staff of a nearby neighbor-
hood center, and arrangements were made to bring Westside Courts
youngsteis into that center's recreational program. As an outgrowth of
this experience, staff and parents worked with community leaders to
obtain better planning for the use of a small park for the Westside Courts
area and became more involved in the Police Community Relations Pro-
gram. Also, one of the teachers at the nursery school started a wood-
working class for older boys and this developed into a club for young
teenagers, girls as well as boys. The burned play house, incidentally, was
rebuilt and the other damage repaired by a group of the fathers-black,
white, and yellow-in the course of a "work Saturday."

During interviews with staff members before the nursery schools
opened, many of the parents-of all the ethnic and socioeconomic groups
represented— expressed concern about the area's public school facilities
and programs. So a Public Education Committee of parents and staff
members was set up. It arranged for meetings at which school issues were
explained and discussed; organized a program of parent-school visits,
under which parents visited the kindergartens their children would attend;
and sent representatives to meetings of the city's School Board. The
Committee became an area-wide group.

One mother-black, six children, on welfare-represented NICE as a
board member of the Mental Health Consortium, formed to work for
better use of the area's mental health facilities. Another participated in a
charrette, a community-based meeting open to the expression of the
residents' wishes and complaints, organized by the Far West Educational
Laboratory, The social worker represented NICE on a community plan-
ning group, where he helped develop a program for the sick children of
working parents.

21

Breaking Down Boundaries

Several months after the schools opened, a white couple invited all the
other parents from one of the schools to a party in their home in St.
Francis Square. It was the first parents' party. One black couple from
public housing decided not to go because they could not believe they
would be welcome. Another got as far as the host's door. "If anybody
looks funny at me when they open the door and see who's here," said
the man to his wife, "I'm turning around." Today, according to staff
members, the idea of rejection would never occur to any of the parents,
because they have come to trust one another.

The development of trust was a rather slow process. Until the schools
had been in operation a year, there was little mingling by the parents
across either racial 01 income lines. The mixing process speeded up during
the first summer, when more of the mothers were helping at the schools
and getting to know one another better, and whole families were being
thrown together at school picnics. During the final two years, middle-
income parents, who were predominantly white, habitually picked up
low-income parents, who were predominately black, and took them to
parents* meetings and other activities. And some of the women with a car
in the family organized car pools so that working mothers, generally from
low-income families, would be saved the trouble of getting their children
to and from school.

Comments from parents tell this part of the story best.
One mother who had never associated with white people ("oh, we had
little friendships in school but it was always kept in school, you know--
they didn't come to visit me at home and I didn't go to visit them ut
home") found herself liking almost eveiybody she met in the project.
"And, someway, it's not a forced like: I've gotten to know them and it's a
real friend ship- not just having something to talk about, mostly school.
They like me, and I like them." She and her children go to dinner at white
families' homes and the white families and their children go to dinner at
her home.

"We are meeting people on a social level that we wouldn't have met
before, except possibly on a business level," said a professional man of
Chinese ancestry. "This is the most magnificent part of it." He recalled the
first time he and his wife had been invited to a social occasion-tin an-
niversary-at the home of one of the black families. The father managed n
clothing store. "The thing that struck us very profoundly," said the
professional man, "was that these people have as much capability as I
have, yet they are not able to live nearly so well for their efforts as I am.
And there are many black families who live even less well-in one of those
very inhuman monoliths that architects have created to house those
families. When you get them that kind of housing, how can you expect
high-class citizenship performance?"

Said the wife of a lawyer: "We didn't go out of our way to solicit
friends among the parents, just as we don't go out of our way to solicil
friends among our neighbors. It was just a natural happening, and it took

22

time. It's genuine." Her husband noticed a couple of plainclothesmen
looking on at a fund-raising dance given by the new Cross-Cultural Family
Center. "It must seern odd to them," he remarked, "this peculiai mix of
people, having a good time, drinking and dancing and talking and
laughing." His wife reported: "It was really beautiful."

Another mother, asked what she thought of cross-cultuial education,
said: "I think it's wonderful, because I think that's the answer to every-
thing— all the problems that we're having now. The more you are together
the more you really know it's no mystery.... You find out that people are
actually doing the same things and having the same type problems,
regardless. You really find that out."

One woman said that even before NICE she had been acquainted with a
number of people different from her. "But I just didn't have— I'd say
'Yeah, it's a nice day,' and I'd be finished." She mentioned a nursery
school mother who had been born in Japan. "One day I met her in the
wash house, and we started talking. She was the sort of person I could say,
'It's a nice day,' and we were finished— after that— I don't know what it
was. I think it was a change in both of us. The races don't make too much
difference— it depends on how they act. That should be a thing we should
remember. She talks about the classes where people were talking about
how they grew up and what happened to them— their way. And I sort of
look at a lot of people and say, 'Yes, that's how you become you are what
you are,' And that helps. It just helps in knowing a lot of people in
general." This mother said she was joining the Family Center because "I
might learn to know somebody else that I didn't know this year at all."

The New Family Center

The nursery schools closed in the summer of 1969, as scheduled, and
that fall the Cross-Cultural Family Center opened as the successor to
NICE. The Parent Advisory Council and the project staff had spent half a
year planning for it.

The two biggest problems had been where to locate the Center, since
none of the school sites was adequate, and how to finance it, since NIMH
was unable to continue support. The site problem was solved
providentially. Some of the nursery school parents appeared on a
television program, dealing with battered children, and told what they had
learned from NICE about child-rearing and children's education.
Impressed, members of the Unitarian Church, which had recently finished
building a new education center, asked what the project was going to do
next, and offered the use of classrooms, play space, and an auditorium.
The church representatives explained that NICE, more than some other
worthy organizations which had been considered, seemed particularly able
to carry out the church's commitment - to use its new plant for the good
of the community.

Some members of the project staff were not eager to accept. Dr. Lane,
notably, felt that being in a church might put a damper on activities. She

23

also felt that the superioi, almost luxurious, facilities being offered might
intimidate some of the persons who would be using them. She expressed
herself at a meeting of the Parent Advisory Council and suggested that the
search for a location be continued. The Council voted her down. One
member remarked: "It won't take me long to get used to a little luxury.11
Dr. Lane soon felt altogether happy with the decision.

The financial problem was tougher. For the first year it was met largely
through contributions of time and talent by parents and by members of
the original project staff from San Francisco State, and through money
raised by cake and rummage sales and by benefits - a dance, a concert,,!
fashion show. The fund-raising effoits were so extensive and exhausting
that for 1970-71 the Center decided to operate the school in two sessions,
morning and afternoon, and to admit children of non-members and charge
tuition for them. Fund-raising efforts are still necessary but should be less
arduous. Some members of the Center are opposed to this direction but
have been unable to suggest a viable alternative.

The head teacher of the Center's nursery school was an assistant teacher
in the original project, and her assistants are mothers who were trained as
part of that project. The Center's activities include a kindergnrten-
supplement program for the oiiginal NICE children three times a week;
individual tutoring in the afternoon for elementary school children who
need help in order to keep up with their classmates; afternoon classes in
sculpting, dancing, and science for children of 6 and older; and evening
classes in cooking and sewing for children of 10 and older. For boys and
girls from 7 to 10, there are clubs that meet weekly. For adults, there arc
discussion groups, pot-luck suppers, occasional socials.

The Parent Advisory Council for the NICE schools became the Board of
Trustees of the Family Center, which is a non-profit corporation. The
chairman of the board, a man of Chinese ancestry, says: "The three years
of the nursery schools simply started something that we expect to grow
to become increasingly more beautiful." Another member, a black
woman, adds: "The Family Center speaks stronger and louder about the
value of NICE than anything we can say."

Research Grant. MH 14782

Dates of Interviews- November 1 969 and September 1 970

24

Investigators. 1

Msgr. Paul H. Furfey, Ph.D.

Thomas J. Harte, Ph.D.

The Bureau of Social Research

Catholic University of America

Washington, D.C.

Prepared by.

Herbert Yahraes

Introduction and Summary

During the first 1 5 or 1 8 months of life, a number of investigators have
reported, children from one racial or socioeconomic group score about the
same in tests of intellectual functioning as those from another. By school
time, though, different groups have reached different levels; in particular,
children from families of low socioeconomic status have a lowei average
IQ than children from families of higher status. The schools do not change
this difference.

Other studies have shown that intellectual level, as measured by mental
tests, is closely related to verbal ability, as measuied by tests of vocabu-
lary and information, and that disadvantaged groups score lower on these
verbal tests than on others. It has also been shown that the children in
these groups receive less verbal stimulation from parents-through being
talked to, read to, taken on trips, etc. -than children in middle-class
groups, and that the parents are usually not very good examples for
children to follow in learning language.

These findings suggest that:

• Children from low socioeconomic groups develop deficits in intel-
lectual functioning because they lack adequate intellectual, particularly
verbal, stimulation.

• One reason many disadvantaged families remain disadvantaged,
generation after generation, is the lower ability of their children to profit
from our educational system and, therefore, a lower ability as adults to
compete in our economic system.

• If adequate stimulation can be provided early enough, it should
prevent deficits in intellectual functioning and thereby help break the
cycle of poverty and cultural deprivation.

1 This research was planned by Earl S. Schaefer, Ph. D. of NIMH and carried out in
consultation with him. Doctor Furfey served as project director as well as Co-Principal
Investigator.

25

To test one means of providing such stimulation, Earl Schaefer, Ph.D.
of NIMH's Center for Studies of Children and Family Mental Health,
initiated the Infant Education Research project in 1965. This is a pioject
guided and financed by the Institute but carried out by a staff directed by
Msgi. Paul H. Furfey,Ph.D. icsearch associate, Bureau of Social Rcscaich,
Catholic University. Under it individual tutoring-an hour a day, 5 days a
week-was provided to a group of 28 Negro boys from slum aicas of
Washington, D.C.

The tutoring began when the youngsteis were 15 months old and
continued for 21 months-the period in which differences between groups
in their average level of intellectual functioning first appear, and also the
period of early verbal development. The tutors were young women, who
talked to the childien, read to them, played with them, brought them toys
and puzzles and picture books, took them on walks and trips-in short,
tried to provide the kinds of mental stimulation necessary to a child's
development but not commonly found among the most disadvanlaged
families. The tutors dealt with emotional and behavioral problems as well.
When the children were 3, most of them were enrolled in nursery schools,
and monthly discussion groups were set up for the mothers. Serving as
contiols were 30 Negro boys from similar neighborhoods, who received
the same tests as the others but were not tutored.

The investigators hoped to learn whether or not this kind of experience
between the ages of 15 months and 3 years makes a difference in intel-
lectual functioning and, if so, whether or not the difference persists.

The results to date show that the project has indeed favorably affected
the mental development of the tutored youngsters for at least a time.
When the infants were tested at the age of 14 months, a month before
tutoring began, both groups scored above normal on the Bay ley Infant
Scales, which are concerned mainly with sensory and motor development
and with alertness and interest. The children in the experimental group
were found to have an average IQ of 105; those in the control group, of
108.

At 21 months, after half a year of tutoring, the average IQ of the
experimental group was 97; of the control group, 90, a difference sig-
nificant at the .05 level (it would occur by chance only 5 times in 1 00).
The test this time had a number of verbal and cognitive items, which are
included in the Bayley scales for this age level. Scores on such ilcms,
which are closely linked to success in school, are strongly affected by
environment. Presumably the environment had affected both groups
adversely, but 6 months of tutoring had brought the experimental group
close to normal.

At 27 months, the average IQ of the experimental group, as mcnsured
by the Stanford-Binet test, had advanced to 101, and at 36 months, to
106. For the control group, the scores were 90 and 89. The difference
between the groups was significant each time at the .01 level. At 4 years,
on the same test, the children who had been tutored rated exactly normal,
100, as a group; the other group rated 90.

26

In addition to the IQ test at 36 months, the investigators used:

The Johns Hopkins Perceptual Test, in which the child is given a
form— a triangle, say-and asked to match it with one of several other
forms.

The Peabody Picture Vocabulary Test, in which the child is shown
several pictures at a time and asked to point to the one illustrating a given
activity or object ("show me running" *** "show me hoise"
11 * * "show me vegetable").

A preposition test, developed by Schaefer, which asks the child to
place something over something else, behind it, by it, and so on.

The tutored children scored significantly higher on the perceptual and
vocabulary tests than the other children, but not on the preposition test.

After each test, the children were rated for task orientedness, meaning
their attention to the job at hand and their cooperation with the tester. At
three, this factor was found to correlate with the test results, as expected.
More important, both this factor and the test results were found to cor-
lelate highly with the mothers' interest in the children, as rated by the
tutors. Where maternal interest was rated high— as judged, for example, by
the mother's efforts to adequately feed and clothe the child and to
express herself verbally with him— the child's test scores were also likely to
be high. Mothers rated low in this respect were likely to rate high on a
child neglect factor, whose elements included inadequate care, irregular
meals, inadequate clothing, sickness, accidents, and beatings. The children
of such mothers tended not only to do poorly on mental tests but also to
display behavior marked by belligerence, irritability, and negativism.

Nothing in the family—income, parents' education, presence or absence
of the father-correlated with the intellectual functioning of the child
except the mother's interest. At four, maternal interest no longer cor-
related with task orientedness, but it did still correlate with the children's
IQ.

Just why the experimental group dropped from above normal in intel-
lectual functioning at the age of 3 to normal (although still 10 points
above the control group) at age 4 is a matter of conjecture. Doctor
Schaefer thinks it is because the tutoring stopped, except for sessions twice
a month. "It is quite clear," he observes, "that if we tutor a culturally
deprived child for 21 months we increase the rate of his intellectual
growth during that period. But it is not clear that he will stay at the level
he reaches then unless the extra stimulation is continued; the evidence at
this point is that he will not.

"The question is not how to offset a poor genetic potential, The project
demonstrated that these lower-class black children had the genetic
potential for an adequate level of intellectual functioning: with only 5
hours a week of stimulation, they scored above the norms during the
period of stimulation. The question is, how can we foster that genetic
potential over the long run?" Schaefer 's answer, set forth more fully in a
later section, is: by early and continued stimulation-to be achieved by
educating future parents to be good parents.

27

The Infant Education Research Project is continuing. When the children
are in kindeigarten, between the ages of 5 and 6, tutors will visit the
homes of those in the experimental group twice a week, primarily to work
with the mothers in the use of educational materials. The mothers will
also be taken on visits to the school, in the hope of developing better
relations between home and school, and they will continue to meet
monthly. Both the experimental and the controls will be tested during
the first year of elementary school to learn if the benefits of the inter-
vention program continue. Also, younger children in the families are being
tested to learn if there has been a carryover from the tutoring. (At prcscril
there is only anecdotal material. For example, one mother spoke of the
child being tutored as "your" child and of her new baby as "my" child.
When the tutor came to teach the older baby, the mother would sit in the
next room and play with and talk to the new baby. This one became quite
precocious in language, speaking in sentences at the age of 13 months,)

Staff members believe that a program based on the one reported here
but not intended as a research project could be mounted in any com-
munity at no great cost. The tutors could be either mature women or
high school girls, chosen for their ability to relate to children and properly
trained and supervised. Most of the recordkeeping and psychological
testing couid be eliminated.

Howevei, staff members also believe, on the basis of the work so far,
that ways should be found of working with eveiybody in the family— not
just with the youngest child and to some extent his mother. If the whole
family were brought into an educational project, they think, the older
children and the parents would become more interested in and capable of
carrying on measures that stimulate intellectual growth. So the project
should have a broader and longer lasting effect.

Two subsidiary studies likely to be of help in future work with dis-
advantaged groups are under way. One, by Doctor Furfey and Doctor
Mary Elizabeth Walsh, of the Bureau of Social Research, is an analysis of
the children's cultural background, based on the tutors' observations of
daily life in the homes of their charges. The other, by Doctor lidna
O'Kern, of the Department of Sociology, St. Francis College, New York,
is an analysis of the children's language, based on their taped responses to
pictures.

Findings of Earlier Studies

When Schaefer, working with Catholic University's Bureau of Social
Research, proposed the Infant Education Research Project in 1965, con-
siderable evidence had accumulated that the intelligence level of children
in poor environments could be raised. The most dramatic example had
been reported by Harold M. Skeels and Harold B. Dye in 1939.

Skeels and Dye studied 25 children who as babies had been committed
to an orphanage which had few attendants and provided little stimulation.
At ages running from 7 months to 3 years, 1 3 of the children-all of them

28

showing evidence of marked retardation— were transferred to an insti-
tution for the mentally handicapped and placed in wards of the older and
brighter women, one or two babies to a ward These women, with mental
ages of 9 to 12, gave the babies adoring attention. Over the next yeai or
two, the children made extraordinary IQ gains, ranging from 7 to 58
points. Most of the children were later placed in adoptive homes. Of the
12 children in a contrast group, left in the orphanage, only one showed an
IQ gain, of 2 points; the others had losses langing from 8 to 45 points.

In 1 966 Skeels published the results of a followup study made while he
was at NIMH. As adults all 13 persons in the experimental gioup showed
average or better than average achievement "as indicated by education,
occupation, income, family adjustment, intelligence of the childien, and
contribution to the community." Of the 12 persons in the contrast group,
one had died in a home for the mentally retarded, four were in insti-
tutions, one was self-sufficient at a middle-class level, and the other six
had such jobs as dishwasher, part-time cafeteria workei, and assistant to
the gardener in an institution. Most or all of the 12, Skeels believed,
would have achieved within the normal range had they been placed early
in infancy in suitable adoptive homes piovided with equivalent stimula-
tion.

Other investigatois, among the many who influenced the planning of
the Infant Education Research project, had reported.

• No differences between Negroes and whites in mental tests given
during the first 15 months.-Nancy Bayley.

• Negro elementary children in five southeastern states have a stable
average IQ of 81 for the years 8 through 1 l.-W. A. Kennedy, V. Van de
Riet, J. C. White, Jr.

• Infants whose mothers had been instructed to lead and talk to them
began at 17 months to produce more speech sounds than babies in a
control group.— Orvis C. Irwin.

• The intelligence of children reared by foster parents was more like
that of the foster parents than of the less educated, lower socioeconomic
true parents.-Marie Skodak.

• At 18 months, the IQs of 170 London children were virtually
identical, regardless of social class. At 3 years, childien in the upper class
showed a gain of about 22 points; those in the middle class, of about 9
points. Lower-class children dropped about four points.-C. B. Hindley.

• Groups of Negro children of different socioeconomic classes showed
consistent class differences in language skills. — Vera P. John. Later came
the finding that the teaching skills of 160 Negro mothers and the per-
formance of their 4-year-old children differed greatly according to the
families' socioeconomic level. -Robert D. Hess and Virginia Shipman.

• Among 292 children, both lower class and minority-group status were
associated with poorer language functioning.— Martin Deutsch,

• Fifteen children committed to a State institution as feebleminded and
given a special 2-year training program showed an average gain 2 years
later of 10,2 points on the Stanford-Binet IQ test and 10.5 points on the
Vineland Social Maturity Scale. Children in a control group dropped 6.5

29

points on the IQ test and 12 points on the social maturity scale; in other
words they were more retarded at eight than they had been at four.
Follow-ups showed that six of the first group but none of the second had
been permanently paroled from the institution. -S. A. Kirk.

• All gifted children apparently have received intensive early stimula-
tion. This special stimulation may be indispensable to the development of
high abilities. There is no evidence the children suffered ill effects of any
kind. -William Fowler.

• "* * * it appears that the counsel from experts on child-rearing dining
the third and much of the fourth decades of the twentieth century to let
children be while they giow and to avoid excessive stimulation was highly
unfortunate. * * * It is no longer unreasonable to consider that it miglil be
feasible to discover ways to govern the encounters that children have with
their environments, especially during the early years of their development,
to achieve a substantially faster rate of intellectual development and a
substantially highei adult level of intellectual achievement. * * * The fuel
that it is reasonable to hope to find ways of raising the level of intcllcclmil
capacity in a majority of the population makes it a challenge to tlo ilic
necessary research."-!. McV. Hunt.

Training the Tutors

The tutors were chosen because they were intelligent, outgoing women
interested in working with children. All had had experience in jobs-
teaching, social work, nursing-that brought them in touch with childicu,
A few were mothers. With the exception of a woman who was still in
college, all were college graduates. Five were black and four white. They
worked together beautifully, a project official reports, and those of one
color seemed to get along with the children as well as those of the other
(Each child had two tutors -generally one of them black, the oilier
white— who alternated weekly in working with him. This arrangement was
intended to lessen disruptive effects in case a tutor left or was changed, to
reduce any feeling on the part of the mother that the tutor was competing
for the child's affection, and to provide two different observers of the
child and his family.)

During the initial training period, which ran for about 2 months, half-
time, the tutors heard and discussed lectures by Schaefer, Furfey, and
others on child development, the special problems of the disadvantage!
child, and means of overcoming those problems . The need to develop and
maintain a relationship that would foster the child's interest, happiness,
and success in his new experience was emphasized. If this need were to be
met, the tutors were told they would have to accept the child's interests,
praise his achievements, and enthusiastically explore new learning op-
portunities with him. The interaction between tutor and child was to be
playful, spontaneous, and pleasant; formal instruction would be avoided.
A tutor must be aware that a child's early learning involves a great denl of
looking, listening, tasting, smelling, and feeling. She must also be aware

30

that she is a model for the child— and the mother-to imitate. She will
name tilings casually but often— toys, household objects, pictures in
books, events, clothes, the child's body.

"We aie all investigators, researchers, expeiimenteis together," Schaefer
told the tutors. "We are searching for a method of early education which
will raise the intellectual functioning of childien and increase their ability
to do well in our American school system. We think we know some of the
characteristics which help children to succeed in school. We think it helps
a child to be cooperative, outgoing, verbal, friendly, helpful to others,
resourceful, curious, attentive, to have a goal-directed behavior and be
able to concentrate and persevere. It helps if he has developed a feeling of
competence and of human worth, if he is able to asseit himself in a
positive way and make a worthwhile contribution to a gioup. It is of gieat
value that he have good comprehension and a good vocabulary. We hope
to find ways of developing at least some of those characteristics."

Examples of a number of specific activities that might be used during
the tutoring sessions were discussed -among them, blowing bubbles,
playing with beads, making things with pipe cleaneis, playing a guessing
game, making a pull toy, having fun with paper plates, and playing a
sotting game. The tutors were asked to encourage the child to participate
in any activity as much as possible and to talk to him about what was
going on. If tutor and child were making a bean bag, for example, the talk
might go something like this:

We will make a bean bag.

Please hold the cloth for me.

1 will cut it.

Be careful. The scissors are sharp.

Move your hands.

That's it.

Now we can sew it.

Look at this needle.

See the tiny hole.

We have to get the thread in this tiny hole-the eye of the needle.

Look at my thimble.

Try it on your finger.

See how it fits on mine.

Please hold the cloth while I thread the needle.

Watch me thread it.

You can string beads. Some day you will be able to thread a needle.

Now, I will hold the bag open while you put the pebbles in.

Put in one; now one more.

Put a whole bunch in at once if you can.

Now we will sew up the last side.

We must sew it so no pebbles will fall out.

See, it's all done.

Thanks for helping, Johnny. Johnny helped to make a bean bag.

See, isn't it pretty. We're all finished,

31

Now move back, so I can throw it to you, etc.

Now let's try throwing it into that pail. Let's take turns, etc.

Making a bean bag would be fun. It would also help the child build his
vocabulary. And it would help him in learning to count, learning goal-
directed activity, learning that a combination of materials can make
something different from the original materials, and learning that a person
can create something and then have fun with it.

Emphasis throughout was placed on language stimulation. Schaefcr
urged the tutors to get the childien "hooked on books."

The technique of transfer, too, was emphasized. "Children are happier
and more comfortable with the familiar," Schaefer pointed out. "They
can even become frightened by something entirely novel. Begin work with
something familiar to the child. If it is something he likes and enjoys you
will soon be able to transfer to a slightly different use or activity and thus
capitalize on the child's initial inteiest and pleasure.

"For example, if he finds it fun to shake jingle bells because he enjoys
the sound, you may latei be able to transfer to sorting jingle bells ac-
cording to size or color, counting them, making things with them, etc,,
and still maintain his interest and joy, while at the same time stirnulaling
new learning."

As part of their training, the tutors observed the activities of children in
day centers, an orphanage, an institution for delinquent children, and the
homes of middle-class families having children of roughly the same age as
those in the project. The tutors also recruited the families for the research
program, by knocking on almost 200 doors in several of Washington's
worst slum areas, and thus became acquainted with the types of homes
they would be going into. And each worked for two weeks with a pilot
case, a child who was a little too old to be chosen for the project.

There was in-service training, too. Each tutor discussed cases frequently
with the leader of the tutoring team (in the beginning a speech therapist
who had been a teacher of preschool deaf children, and since mid-1967
Mrs. Lillie Davidson, formerly a nursery school teacher, a supervisor of
teachers, and a supervisor of children's counselors in a center for homeless
children). Case conferences involving Schaefer, Furfey, the head of the
tutoring group, the tutors, and sometimes a mother and her child were
held frequently. All the tutors met weekly with other staff members to
discuss problems and to give and receive suggestions about educational
materials and techniques.

The Children and Their Families

The effect of certain environmental factors upon intellectual function-
ing has been found to differ according to sex.2 Since the project could no!

The Hffect of Childhood Influences Upon Intelligence, Personality and Mental
Health in Mental Health Program Reports-3.

32

afford a large enough sample to study both sexes, it limited itself to boys
because lower-class black boys and men seem to have a more difficult time
in school and society than girls and women.

The subjects were chosen from census tracts selected because they had
high rates of crime, delinquency, infant mortality, joblessness, dilapidated
housing, and families on welfare. The neighborhood environment as well
as the typical home environment was deprived. Though the experimentals
lived in a slightly worse neighborhood than the controls, the children from
one area had been scoring much the same as those from the other on the
Metropolitan Readiness Test, given to children on entering school. The
median scores were approximately at the 20th percentile on national
norms, meaning that they fell in the lowest fifth.

Families having a boy of the right age were invited to participate if the
home situation was relatively stable-that is, if it did not look too chaotic
for daily tutoring session— and if the family met two of three other
qualifications: its yearly income was less than $5,000, the mother had not
finished high school, the mother had never been employed as a skilled
worker.

Among the 98 homes the tutors visited to select the experimental
group, 12 were rejected by the staff as unsuitable (because they were
considered unsafe for the tutors or too ciowded to work in, or because
the family's attitude was judged likely to present unduly severe problems).
Five other homes were ruled out because either the income or the
mother's education was too high. Most of the other rejections were made
because the children were not of the right age. Four families refused to
participate.

The control group was selected from 89 homes in a slightly better area.
Ten were rejected by the staff as unsuitable, and 12 were ruled out
because the economic or educational level was too high.

Stipends-$10 for each testing session and $1 for each tutoring
session— were offered primarily in the hope of reducing losses from the
sample, particularly losses of children with poorly motivated mothers.
Since losses were expected to be high in any case, and the project wanted
to retain at least 20 in each group, 31 children were chosen to be tutored
and 33 to serve as controls. A number of the families have broken up and
some have moved, as many as five times. But only three in each group
have been lost. (Among the experimental families, one moved out of
town; one placed the baby in a day center so the mother could take a job,
and the center refused to admit the tutors; and one notified the project
that the tutors would no longer be safe: the wife's mother had moved in
and she liked to throw things.) This unusual record is attributed both to
the interest of most of the mothers in having something done for their
children and to the tutors' persistence in following families and staying on
the job in spite of any unpleasant circumstances. The staff now thinks
that without the financial inducement, no more than three or four of the
families in each group would have been lost.

The homes as a rule had neither toys nor books, and the mothers at
first seemed to think it a joke to suppose that anything could be taught to

33

a child as young as 15 months. Some of the parents, even some of those
on welfare and with half a dozen children, were well organized. The
families always had food; the children were sent to school daily; the
mother knew how to go to second-hand stores and buy clothes. Other
parents seemed overwhelmed by their problems.

The following quick sketches of representative cases are drawn from
tutoi's reports:

W. lived with his parents, a brother, a sister, an uncle, and an aunt-7
persons in all-in a three-room apartment on the third floor of a rapidly
deteriorating house. The rats were so bad, his mother said, that she had to
beat on a pan to frighten them away before she entered the kitchen. They
had chewed the nose off a doll. The father had a steady job as a truck
driver, and the mother woiked part-time as a counter girl in a nearby
drugstore. She kept the apartment tidy, showed her love for her children,
and participated in the tutoring sessions.

At 15 months, J. was a shy, neglected, poorly dressed child who lived
with his baby sister and his 17-year-old mother in a dingy six-room apart-
ment, part of an old house. It was occupied by his grandmother, who
worked as a domestic, an unemployed uncle, and the uncle's 4-year-old
daughter. J.'s father was in jail. The apartment was uncrowded but un-
pleasant. Fumes from an oil stove, which supplied inadequate heat, made
the eyes smart. Chunks of plaster had fallen from the walls. The bedroom
floors were littered with soiled clothing. Cockroaches crawled allaiound.
There were no toys or books, but there were newspapers, a television, a
radio, and three telephones which had been disconnected. The mother, a
high school dropout, was quiet and timid, knew little about raising her
children, and was apparently unable to show affection for them. She was
interested mainly in going to school and getting a job. J, was cared for by
the grandmother, a warm and friendly person and apparently the only
member of the household bringing in money, and a babysitter,

When the tutoring began, V. lived with his parents in a spacious one-
bedroom apartment and slept with them in a double bed, His father had a
steady job, and his mother worked in the afternoon until her husband
asked her to quit so she could spend more time with the boy. This wns
one of the few homes with books and toys. V. was as spoiled as any
middle class only child could be. When the tutor and he went for a walk
and passed a store without going in and buying something, he had a
temper tantrum. One night some men came to the apartment, threatened
V.'s father with a gun, and chased him out, The next day the family
moved to another section of the city. This was near the end of the tutor-
ing program. The mother seemed despondent over the course of events,
and the boy showed hostility toward the tutors.

A. and his mother, brother, and sister lived in a two-bedroom basement
apartment that reeked with the odor of urine because the mother washed
clothes and diapers together and did not properly rinse them. Sometimes
the odor was so bad that the tutor had to take the boy outside. Cock-
roaches added to the discomfort. The mother, strict and brutal, would
punish the children either by striking them or locking them up in a room.

34

E. was a happy, sociable child who icadily sought and gave affection.
He lived with his five brothers and sisters and his parents in a two-room
apartment, which needed repaiis but was kept in leasonable order. His
mother worked nights in a carry-out restaurant. She loved him and did a
good job of caring for him. E.'s parents separated during first year of
tutoring, and the mother moved with all the children to another two-room
apartment. When E. entered nursery school, she began going to night high
school so she could get a better job.

R.'s mother showed the tutor a scar on the boy's upper thigh that had
come from a rat bite. She had four other children but managed to sit in on
many of the tutoring sessions, often went along on trips, and was proud of
R.'s accomplishments. There were books in the house. The family seemed
to pay little attention to R.'s fathei, who did not work regularly. When
the mother got a job as a teacher's aide in a Head Start center, the
children were sent to a baby-sitter's home and the tutoring continued
there.

T.'s mother separated from her husband during the course of the
project and moved with her six children into a one-bedroom apartment.
Three boys and their 8-year-old sister slept in a double bed, another boy in
a crib, the newest baby in a bassinet, and the mother in a single bed in the
same room. A bed in the living room was often used by overnight guests.
The mother, who seemed disturbed as a result of problems first with hei
husband and then with a new man, constantly screamed at the chikhen
and threatened them with belts but got no response. The toilet was
usually broken. Some of the children had ringworm. T. kept complaining
of being hungry. The tutors found it difficult to get him to concentrate.

Working With the Children

The tutois set out to build friendly, easy relationships with the children
by playing with them and talking to them— activities in which some of
the mothers had engaged only rarely. Some children made friends almost
immediately. Others hung back. With one boy, shy almost to the point of
being withdrawn, the tutois had to work almost a year to establish
rapport. In other cases, productive relationships developed only after the
tutors had worked to change maternal attitudes and behavior.

At the start, the children were given manipulative toys-blocks, pop
beads, nesting sets— which1 they learned to put together and to separate.
One toy would lead to a number of activities. Given pop beads, for
example, a child would feel them, swing them in the air, and chew or suck
on them. Perhaps he would learn by accident that they could be pulled
apart. This delighted him. Putting them together was too difficult for
most of the children at 15 months, so the tutor would show how to do it
and then guide the child's hands till he could do it himself. Other activities
with beads included hiding some of them around the room for the child to
find and add to the string, using a long string of them as a pull toy,
counting, naming colors, and matching according to color or shape. Along

35

with the simple manipulative .toys, the child was given opportunities to
play with balls of various sizes, a kiddie-car, a pounding bench, and other
toys that helped develop the larger muscles.

Always the tutor talked to him about what was going on. When ;i new
toy-for instance a ball -was presented, she would name it. The games thai
could be played with the toy gave the child opportunities to hear, imitate,
and learn other new words-for instance, roll, throw, catch, and kick As
the child matured, the tutor would talk about the different sizes, kinds,
colors, and numbei of balls.

If a boy liked a toy car, the tutor would talk about that car and other
cars. Then she would get a book containing pictures of cars and talk with
the child about what he saw on each page. When tutor and child went for
a walk, they would talk about the cars seen. Looking at picutres of cars m
a book helped develop an inteiest in books, too. One tutor was surprised
and pleased to find the children associating one make of automobile with
her. When they saw this auto in a book or on the street, the response
would be, "Tha1 teacher's car!"

Toys were used not only to aid physical development and to teach
language but also to develop two of the characteristics essential to the
mastery of schoolwork-attenriveness and perception. Here puzzles were
considered especially valuable. First came simple, isolated-objecl
puzzles—for examples board containing pictures of three kinds of fruit,
to be taken out and put back in-which were introduced between the ages
of 15 months and 2 years, More difficult puzzles followed. One tutor,
Betty Pair, describing her experiences for the benefit of other persons
working with children, wrote:

"We talk about the puzzle while it is still intact; then dump out the
pieces (this act I leave to the child, because he seems to derive great
pleasure from the dumping); then talk about the side with colors on it and
the dark, rough side; then trace with our fingers around the inside of the
puzzle; then attempt to fit the pieces in the puzzle.

"It is important that the child complete the task, but it is imperative
that he not become so frustrated in his attempt to do so that he sets up a
negative block against the activity. For this reason, I initially put the
pieces back slowly in the puzzle so that the child can observe me. This is
the 'I Can Do It' part. We then see if he can do it, with the assistance he
may require to prevent overt frustration. When there are signals that
assistance is required, 1 put my hand over the child's hand on the puzzle
piece, and I explain, 'Turn it around,1 or 'Turn it over,' or 'try another
space/ as we do what each command directs."

Another tutor, Lucile Banks, offered these suggestions;

1 . Present puzzles as enjoyable games.

2. Demonstrate how pieces are placed.

3. Give each piece a name.

4. Begin by taking only one piece out of the board at a time.

5. Finish one puzzle before starting another.

36

6. If frustration persists, direct the child to a moie relaxing activity-
such as painting 01 a favorite toy.

7. Praise the child.

8. Present puzzles already masteied for relaxation and reinforcement.

Through playtime activities, the tutors also worked to modify un-
desirable behavior. If a child were hyperactive, the tutor would try to get
him interested in one activity— like putting rings on a stick, arranging
animals, building a wagon-and would work with him to see it completed.
"Well, Bobby," she'd say, "let's finish this. I will help you. * * * No, we
aren't going to play with that until we've finished this" or "until we've
picked this up and put it away."

Toys are credited with having aided social development, because they
usually had to be shared with other children in the house, either during
the tutoring sessions or between them. Generally a toy was left in the
home for several weeks. If it became a favorite it might be left longer, or
one like it bought as a birthday or Christmas present.

Tutors and mothers worked together to make toys. Milk cartons-cut
into squares and the squares covered with paper-became blocks; oatmeal
boxes, drums; pierced bottle caps on a string, a tambourine. Bleaching
compound bottles were shaped for use as dolls' cribs. A roll of shelf paper
filled with sketches ("This is Jim * * * he lives in a house * * * he has 3
brothers * * *") became, as it was slowly unwound, a movie. Two cans
and string made a walkie-talkie. Tutors and mothers also spray-painted
cartons for use in storing toys, and they cut down large soap-powder
boxes to make cases for childien's books.

Books were introduced early in the project, though the attention span
of these 15-month-olds was very short. To win the children's interest, the
tutors tried to relate things seen in books to things known in daily life.
Sometimes they would carry books with them on walks and point to a
picture of a dog or a tree, bird, truck, and then to the real thing. Adults
and children in illustrations showing family groups would be named after
the persons in the child's family. Books that a child found especially
appealing would be presented again and again, unless he showed boredom;
books that had no attraction for him would be taken away but might be
introduced again later,

One boy clearly preferred books about horses. So his tutors dug up as
many horse books as they could find and they took him on trips to see
horses and to ride ponies, Whenever a tutor showed him a picture of a
horse, he would tell her about these trips. Sometimes he would sit for an
hour looking at a book with pictures of horses in it. A boy who preferred
books about animals in general was taken not only to the zoo several
times, but also to the Rock Creek Park Nature Center, the natural history
exhibits at the Smithsonian, and the circus. And his mother took him to
the country.

With the child, and sometimes his mother, watching and helping, the
tutors made books by cutting pictures from magazines and pasting them
in scrapbooks. Some of the books were concerned with a single subject
-babies, for example, or automobiles, Others had pictures chosen because

37

they would remind the child of things he and the tutoi had seen at the
zoo, or in a stoie, or during a walk around the block. Scrapbooks were
also made to illustrate such concepts as big and little (an elephant and a
mouse), one and many (automobile tires), old and new (shoes), and ciicle
and square. Qthei scrapbooks were used to teach numbers. A picture of
one sheep would be headed, "This is 1 "; of two donkeys, "This is 2," ami
so on. The numerals were cut out of sandpaper so the child could feel as
well as see them. The homemade books were left in the home for use
whenever the child liked. So were some of the others, particularly the
cloth ones.

The tutors also told stories, read stories, and used the combination of
story book and record player.

During their third year, most of the children went to the library a
number of times and checked out, carried home, and kept for a few days a
book of their own choosing. The children by this time had come to expucl
the presentation of at least one book, old or new, dining each session, mid
books often took up more of the session than toys. "Hey, Teach,*' some
of the children would say as the tutor entered, "I wanna book."

Music, too, was part of the curriculum. The tutors would sing to the
children, play songs and other music on a record player, and use home-
made or inexpensive children's instruments— drum, tambourine, bells,
xylophone— to help the youngsters express rhythm and time.

"When the babies were young (15-18 months)," reports one of the
tutois, Patricia Chernoff, "I held them in my lap or arms, facing me, mid
moved my body or knees to the ihythm of the song, at the same Lime
articulating the words carefully and diawing the child's attention to my
singing by holding him close and using exaggerated facial expression, I
repeated the same songs until eventually the child attempted to sing. At
this point, I simplified the words, concentiating on those which were
repeated most often in the song and therefore easiest to perceive mid
repeat. For example, in presenting "Shoo, Fly, Don't Bother Me/' 1 sting
Shoo Fly with greater emphasis and volume than the rest of the words,
expecting the child to repeat only those two words. Gradually, when he
was able to sing Shoo Fly in the correct places throughout the song. I
encouraged him to add the remainder of the phrase, don 't bother me. The
phrase, for I belong to somebody, because of its length and the rhythm
with which it is sung, comes much later. When the child is unable to
perceive the words from the recorded presentation, I have repeated Ihoni
more slowly, later without the recording.

"Once the child has become interested in the actual singing of the
songs, I have lessened body contact and emphasized the rhythm, con*
centrating only on the words. When the child becomes tired of singing, I
terminate the music session rather than changing the emphasis to clapping,
etc.

"I have found that the length of time required to learn a song has
lessened considerably (in some cases, the child enters in during the first
presentation) as the children become able to focus their attention on tlic

38

words and as I continually reinforce with praise and enthusiasm their
attempts to sing the words."

A child's birthday was always lecognized, usually by a party and a book
or a toy from the tutor. If a party could not be arranged, tutor and child
would walk to a store to pick out a present.

Influence on the Mothers

The tutors believe that they have reached each family to some degree.
One mother related with pride that whenever she took her boy to the
store, he dragged her over to the display of books, and she sometimes
bought him one. Another would sit in on the tutoring sessions and often
ask questions about aims and techniques. Though she herself could not
read, she took out a library card at the tutors' suggestion, and began
boirowing books for her three youngest.

At the start, one mother kept screaming at her 15-month-old, who liked
to get into things: "You are bad --I'm going to beat you." Her attitude was
hostile, almost rejecting. As she watched the boy's progress under the
tutors, though, she came to recognize his inquisitiveness as a mark of
intelligence, and her attitude turned to one of acceptance and even pride.
Another woman frequently made out-of-town visits, leaving her boy with
a relative. During these absences he was obviously upset; even when the
mother was at home he seemed to feel insecure. He became a happy
interested youngster only after the tutors persuaded the mother that he
loved and needed her, and she began staying home. Under the influence of
the tutors, A.'s mother, mentioned earlier, eased her harsh discipline and
became so interested in learning how to help children that she took a
volunteer job in a nursery school. A.'s IQ, which had dropped from 1 1 6 at
21 months to 89 at 36 months, rose to 102 at 4 yeais.

Some of the women would buy a toy or a book like one brought in by
the tutor, or they would ask what they ought to buy. One displayed a
28-piece puzzle she had bought as a Christmas present. Since the boy in
the program was only two and a half, the tutor supposed it was for the
older children. "No, it's for him, " said the mother. She dumped the pieces
on the floor, and the youngster put them together in a few minutes. The
boy's older sister found it difficult at times to keep up with him. The
mother called her silly or dumb, till the tutor explained that the girl was
smart enough-she just had not had her brother's training.

A number of the families began taking more trips with their children.
Some mothers had not known that such resources as libraries, museums,
and a zoo existed; others had not known how to get to them. A few
mothers hadn't even known what bus to take to go downtown, and one
nevei left the house to go anywhere unless one of the children went along:
she didn't think she could find her way back.

The tutors often encountered problems that were outside their province
as tutors but not as human beings. For instance, they would find a family
sitting around in winter coats because the heat had been turned off. Or

39

the mother would be distraught because the family faced eviction. Or t
mother and father had quarrelled, and the mother had gone off to vi
relatives. The children in such cases would be upset-fearful, or apathct
or quarrelsome; one child kept asking his tutor to take him with her.

One woman with half a dozen children had an especially difficult tii
with her money. (Says a tutor: "To budget welfare money-which 1st
low anyway-for a month is really a challenge even to someone who 1
taken a home economics couise in college. When the welfare checks coi
out, food stores raise their prices, and people go around in cars trying
get you to buy nice looking but flimsy clothes that won't last ovei
month.") After the fast two or three weeks of the month, this worn
would run out of both food and money. Several times a tutor found t
children picking crumbs from the floor and eating them. The project h
no funds to feed the family, but the tutors sent the mother to agcnc
that could give her emergency help. They also went shopping with her a
showed her how to choose economical foods, and they brought in snac
for the children.

When the children were three, the tutoring ended, except for twic<
month visits, and the project tried to place them in nursery schools.
succeeded in all but two cases, where the family situation was particula
unstable. (An effort to place children from the control group succeeded
only half a dozen cases, partly because the effort was less intensive a
partly because the families were little motivated to send their children
nursery school.)

At the same time the mothers were invited to meet once a month*
groups of four or five, for a planned program of discussions of subjects
which they were interested as well as subjects about which the proj<
thought they needed to be informed. Most were pleased to accept and o
was delighted: she had always wanted, she said, to belong to something.
All kinds of problems have been discussed-for example, how to t
older children about sex (some of the women said that their own moth
had been too shy to talk to them, so they hadn't known about havinj
baby until they'd had it); what to do about a child who is having troul
in school or refuses to go to school; how to cope with poor health, t
housing, marital problems. The tutors conducting the meetings have f
equipped to answer some of the questions themselves. They have hand
the others by giving out pamphlets on the subject or suggesting that 1
mother talk with her child's counselor at the school. Movies have be
differences between children at different ages, about sibl

others have gone back to school or have entered training programs, in
order to get better jobs. Some have sent their children off to school or a
day care center, traveled an hour or an hour and a half to attend a training
program across town, and hurried home late in the afternoon. After the
children entered nursery school, several of the mothers coupled an all-day
job with a training program at night. They would leave their children at a
sitter's home five days a week, picking them up on Friday night and
retaining them on Sunday.

During the kindergarten year, the homes of working mothers will be
visited on Saturdays, and the children of these mothers will be seen at the
babysitter's during the week.

Implications for Educational Policies

To summarize, the Infant Education Research Project found that:

1 . A tutoring program beginning at the age of 1 5 months for Negro
boys from disadvantaged families and continuing, 1 hour a day, for 21
months significantly raised their level of intelligence until they were 4
years old, at least,

2. The average reached at the end of the third year, 106, dropped six
points during the next year but was still significantly higher than the
control group's average,

3. When maternal interest was high, a child's IQ was likely also to be
high.

How long the benefits from the additional stimulation will continue
remains to be seen.' But the drop in the peak IQ level after the tutoring
stopped, at 3, suggests to Schaefer that short-term programs of early
education are not sufficient to develop and sustain the child's potential
over the long run. Other studies using different programs of stimulation
point to the same conclusion.

The investigator thinks the answer lies in both early and continued
education. "Genetics may determine the potential range of a child's intel-
lectual level," he points out, "but the quality of the environment deter-
mines the actual level. Evidence is accumulating rapidly that because of
physical, social and emotional, and cultural deprivations, many children
are not developing their genetic potential, and therefore do not function
effectively in school and in society."

Physical deprivations include inadequate medical care, insufficient and
low-quality food, poor 'housing, and inadequate clothing. Social and
emotional deprivations stem from the lack of stable and supporting
relationships with the mother and father or their substitutes. Cultural
deprivation occurs when parents fail to provide a stimulating environment.
Children who lack pencils, paper, crayons, books, games, and other educa-
tional materials do not receive enough training in certain qualities and
abilities -such as attention, concentration, perseverence, and perceptual-
motor skills-that make for success in school. Children who are not en-
couraged to talk about their experiences, who do not have models of good

41

language use to follow in their early yeais, and who have little 01
opportunity to make visits beyond their immediate neighborhood
likely to reach school age lacking the language skills, the interest, and
knowledge of more fortunate children.

As Schacfer sees it, the intellectual development of the typical child
be viewed as having four basic stages. In the first, the paient devclo
loving acceptance of the child and a positive involvement with him. In
second, this involvement elicits from the child the development \
positive relationship with the parent. In the third, the parent and the c
engage together in activities, such as piling up blocks, rolling a ball, li
ing at a picture book, and the parent by word and by example teaches
child language, skills, and task-oriented behavior. From this early
perience with the parent, the child reaches the fourth basic stage- he
acquired the interests and the skills that enable him to learn on hist
"Successful achievement of these early developmental stages in
home," says Schaefer, "may be a necessary basis for a successful ccl
tion in the school. But deprivation during their own childhood may I
parents without the personal resources to support the optimal devi
ment of their children."

Do we then need more nursery school? More child-care centers thai
take children at 15 months and even younger? A spread of tutoring
grams, like the one in Washington but lasting longer and made part ol
public education system so that every child will be reached?

"We do need to recognize that education goes on from birth," Scha
answers, "but I think it would be more fruitful in most instance
support parents in their educational role than to set up educational i
tutions to supplant them. Parents, or at least the mother, are there t
and continue to be there. This study found that both tutoring and
quality of maternal care were related to the children's intellectual dcvi
ment at three and four. Tutoring can be thought of as suppleraon
maternal, or parental, care because in the middle class a good mothci,
a good father, does what the tutois have done."

As one hopeful development, Schaefer points to a new Insli1
supported project in Prince Georges County, Maryland, that seeks lo
grate the educational efforts of home and school. When children ur
fants, teachers will go into the home to work with the parents; late
the parents will go into the school. But he thinks we ought to look
farther ahead. "If you assume that almost everyone becomes a parent
that one of the most important jobs for each generation is to rear the
generation," he says, "it follows that we should be giving children, to
ning in kindergarten and running through the twelfth grade, some o
skills needed if they are to be competent as parents. We should
programs for future parents."

The Washington project has demonstrated that lower-class black in
can benefit from additional stimulation. Could other infants also ben
Schaefer answers that we cannot be sure until programs of stimulatio
tested with other ethnic and socioeconomic groups. "Some peopU
that middle-class parents are doing as well as they can with their chil

42

—that nothing else they could do would lead to higher levels of intelli-
gence and achievement. I don't believe that, but it needs to be tested."

"Many parents— and I think they can be found in all social classes-
don't have the skills to be effective in their roles as teachers. We need to
develop methods of improving the education of young children, and we
also need to develop better ways of communicating what we learn— and
what we already know, for that matter-to all parents and future parents.
If the whole culture became aware of the importance of parents as
teachers, I believe it would lead to an educational revolution, and to a
bettei adjusted, more competent, and more intelligent population."

Research Grant. MH 9224

Dates of Interviews. August and September 1969

References:

Bayley, N., and Schaefer, E. S., Correlations of maternal and child behaviors with the
development of mental abilities: data from the Berkeley Growth Study. Mono-
graphs of the Society for Research in Child Development, 29:6, 1964.

Schaefer, E. S. Home tutoring program. Children, 16:2, 1969.

Skeels, H. M. Adult status of children with contrasting early life experiences. Mono-
graphs of the Society for Research in Child Development, 31:3, 1966.

U.S. Office of Education, Department of Health, Education, and Welfare. Infant
Education Research Project. Washington, D.C.,: Superintendent of Documents,
U.S. Government Printing Office.

U.S. Department of Health, Education, and Welfare. PHS Health Services and Mental
Health Administration. The edge of education. HSMHA World, 4(1): 18-23,
Jan-Feb 1969.

43

Director.

Constance N. Swander

Co Director-
Gladys R, Blankenship
The Good Samaritan Center
San Antonio, Texas

Prepared by:

Herbert Yahraes

Introduction and Background

Carmen, David, Pablo, Maria, and the dozen other 3-year-olds playfo;
the coxirtyard of the Good Samaritan Center in San Antonio strike
observer as happy, energetic, and bright. And they are. However, wer
not for a special educational program developed and tested by the Con
with financial support from NIMH, many of them would be destined i
to enter the public school system so poorly prepared as to be consiile
and treated as slow-a few of them, possibly, as retarded. Typically, tl
course through the school system would grow increasingly difficult, im
number would drop out early, prepared only for living at the I
socioeconomic and cultural level into which they were born.

The Good Samaritan Center is a neighborhood center serving abi
20,000 people on San Antonio's west side. This is a low-income ar
more than 90 percent of whose residents are of Mexican descent. Thoi
the neighborhood is clearly poor, it is not a slum. About 650 families 1
in two public housing projects; the others in small, one-story house
some of them little better than shacks-covering block after block. Mi
of the houses, which would sell for perhaps $6,000, are owned by I
families living in them or by a relative. They are usually ncnt, I
crowded; living rooms must often double as bedrooms.

In the four census tracts surrounding the Center, the median nnm
family income in 1960 ranged from $2,830 to $4,190. More than
percent of the families being served in the Good Samaritan's health a
guidance clinics had monthly incomes of less than $200 to support froiv
to 1 5 persons. Among the families of children now being served by t
pre-school project, about a third are on relief. Fathers are absent in abo
a fifth of the homes. The men typically are unskilled or semi-skill'
workers; the women typically do not work outside the home.

Good Samaritan, located in the midst of the area it serves, is housed
half a dozen simply designed, one-story structures built around a com
yard. It is sponsored by the West Texas Diocese of the Episcopal Chun

44

and is supported by the church and by the United Fund of San Antonio
and Bexar County.

Studying the problems common to the people of its neighboihood, the
Center found them to be rooted in lack of education, which in turn was
rooted in lack of preparation for school. Sixty percent of the children in
first grade were considered problem children; 15 percent of all the
children had been held back at least one year. Many of the teen-age
dropouts could read and write English only haltingly; they could not
express themselves adequately even in their first language, Spanish. Half of
the adults in the neighborhood had not completed fifth grade.

Constance N. Swandei, executive director of the Center, decided in
1964 that the long-run solution to the area's poverty and associated
problems lay in preparing the children for success in school. She planned
to do this through a pre-school program that would teach English while
pieserving and reinforcing the children's use of Spanish, and at the same
time would develop the children's ability to learn by guiding them
through planned learning experiences.

The program, which won NIMH support and opened in 1965, had those
two principal objectives because Mrs. Swander recognized that the
children had two principal handicaps-lack of an opportunity to ,learn
English and also lack of the verbal stimulation necessary for a child to de-
velop whatever intellectual capacity he was born with. In the typical dis-
advantaged home of San Antonio's west side, as in many disadvantaged
homes elsewhere, children are likely to be ignored unless they misbehave,
and language is used more for controlling their behavior than for telling
them about objects in their environment or for otherwise instructing
them. So they grow up lacking real facility even in their native language.
And without language facility, points out Shari Nedler, until recently the
project's psychologist, "the child cannot organize his concepts, he cannot
reason at abstract levels, he cannot describe, analyze or synthesize; he
cannot solve any but the simplest problems."

The 1 6 children who enter the program each year are chosen at random
from the neighborhood's eligible 3-year-olds, who are found by a
house-to-house canvass. To be eligible, a child must come from a low-
income,* Spanish-speaking family that has lived at least five years in the
city and two years in the neighborhood. This residential requirement
makes for a stable sample; in four years the project has lost only two
children. Until the 1 969-70 class of 3-year-olds was chosen, there was also
a requirement that the child be able to speak only in Spanish. This
requirement has been dropped, partly because most of the neighborhood's
children do know a little English, picked up from television and from
brothers and sisters who go to school, and partly because the Center
wishes to emphasize that the program is potentially valuable for any

*The Orshansky poverty index is used. A family is considered in the low-income
group if, for example, it numbers seven persons and has an annual income of $4,700 or
less.

45

culturally disadvantaged child, whatever the language he hears at home.
When the eligible children-there were 54 of them out of the 130-3-year-
olds found in the most recent canvass-have been identified, a table of
random numbers is used to choose the 16 whom the program can accept,
During the first two years, the staff had some difficulty persuading
families to let their children participate. Several families, indeed, flatty
lefused. Others had to be visited as many as half a dozen times by tlte
Center's principal emissary, Gladys R. Blankenship, who is of Spanish
descent herself and thoroughly bilingual. Mrs. Blankenship is super-
intendent of the school and co-director of the project with Mrs. Swamter.
The families knew and ti listed the Good Samaritan Center; many of them
just did not think a child of 3 was old enough to be parted from Iris
mother, even if she did have-as she generally did-toc many other
children to pay him much attention. Today, motheis throughout Hie
neighborhood are eager to have their children accepted in the program.
And "Los Ninos," a weekly television show starring Mis. Blankenship and
several children from the Center, undertaken in 1969 at the request of
Station KENS, is popular in San Antonio and neighboring communities.
In 1968, with the grant from NIMH due to expire in two years, (lie
Office of Education made the project an arm of the Southwest Educa-
tional Development Laboratory, in Austin, Texas, one of the Office's 1 6
regional laboratories, and the laboratory named the project's psychologist,
Mrs. Nedler, as its program director for early childhood education. Known
now as the San Antonio Urban Educational Development Center, the
project is serving as a model for other efforts with pre-school children from
disadvantaged groups. As of early 1970, programs to further test the
methods and curricula developed under the NIMH grant were going

forward :

-in San Antonio with 400 Mexican-American children, a program
conducted with Model City funding in the public schools;

-In McAllen, Texas, with 150 children of migrant Mexican- American
farm workers;

-in Dallas, Texas, with 3-year-old Negro children in a recently
established school for disadvantaged children;

-in Bossier City, Louisiana, with 50 Negro children.
Children from disadvantaged Negro families, Mrs. Nedler points out,
have basically the same problems as those from disadvantaged Mexican-
American families: they are not learning good English at home, and they
are not getting the intellectual stimulation necessary for the development
of language skills and cognitive abilities.

School sessions at the San Antonio center run for three hours each
morning, five days a week. Each class of 16 has one young woman as
teacher and another as assistant teacher. The teachers are college women
who have majored in education or in child development and who speak
Spanish fluently. The assistants are high school graduates from the same
neighborhood as the children. Trained on the job, they serve not so much
as aides-if this term connotes someone who helps with coats, serves
snacks, carries messages-hut as second teachers,

46

Building Confidence

Childien in the Good Samaritan area, like children in other disadvan-
taged neighborhoods (and some children in more favored neighborhoods),
typically receive little encouragement to begin a task and can have small
hope of a reward for completing it. As long as a child is doing what he's
supposed to do, his parents say little or nothing to him; they wouldn't
think of praising him. But let him do what he is not supposed to do, and
he is reprimanded. Such conditions stifle initiative and promote passivity.
They may promote misbehavior as well, because misbehavior becomes one
way of getting attention.

Through all its many activities, the school tries to develop a child's
confidence and wholesome self-regard. At the very start, for example,
each child is introduced to the class and given a name tag, which he wears
proudly. Also, he is taken on a tour of the classroom and shown where
things are kept, how to put them away, how to care for books, and how
to carry a chair. Putting this information to use not only promotes good
classroom discipline but also gives the child a feeling of accomplishment.

Classroom performance checklists, along with other observations, are
used to assign the children to three smaller groups based on ability levels.
Such grouping gives each child opportunities to experience success in
work geared to his ability 01 readiness. The groupings vary for diffeient
activities and are adjusted throughout the year. The teacher leads one
group and the assistant another; members of the third group engage in
individual projects-painting, building, cutting out designs, and so on.

Another aid to building self-confidence and a feeling of identity is "tell
time," a penod right after roll-call during which a child may come before
the class and say anything he wants to say. The children don't have to
participate, but most of them usually do. They talk about their mother or
father, or something that happened at home, 01 a cut finger-anything.
Even the 3-year-olds are eager to express themselves.

Cliildren are praised when they answer correctly. They are also praised
for working hard and behaving appropriately. A child who is unusually
shy or passive is given special attention.

Teaching English

In the beginning, the Center thought that the children would learn
English much as they had learned Spanish, by being exposed to it. As they
took part in nursery school activities, the teacher would talk to them in
English, using Spanish to explain. They would repeat her English words
and gradually begin to use them: they would "pick up" the new language.
Tests at the end of the year, however, showed that while the children had
made progress on other fronts, they had advanced hardly at all in their
knowledge of English.

The school then began developing a program for teaching English sys-
tematically. Based on its first-year experience and on word lists used in

47

the public schools, it drew up a 2,800-word vocabulary and organized
by topical units-words dealing with the body, with food, with clothin
with transportation, and so on. Vocabulary building, however, is not*
end in itself. The school's aim is to develop a child's competence in h
first language by expanding his basic fund of information through ne
labels, or words, and new concepts, or ideas, and then to introduce him 1
English by using the same labels and concepts.

During the first 15 weeks of school the 3-year olds hear nothing bi
Spanish, for these children typically have a poor vocabulary even in the
own language. As their teacher expresses it, they lack labels. The
mothers, unlike most middle-class mothers, have not been inclined t
name things— and thus to teach the names— when talking to their childrci
They have been saying, "Get me that," instead of "Get me the apple,
and "Put it here," instead of "Put the apple on the table."

Lessons on body awareness, which is the first topical unit, begin wit
the use of a mirror to aid in self-identification, a pre-requisite of sel
esteem. Many of the children have not used a mirror to any extent— sonn
not at all— and are not fully aware of their own features, The teacher ask;
in Spanish, ""Whom do you see?" Generally the child responds, "Me,"S
the teacher asks, "Well, what is your name?" Then she asks everyone t
say, "Good morning, Juan."

After other members of the group have been similaily introduced, labe
ing begins. The teacher points to her nose, eyes, mouth and names their
then she has the children, one by one, say nose, eyes, mouth— in Spanish-
and point to these parts on their own faces. Next the teacher touches
child's arm. "Este es el brazo," she says. "Este es el brazo, verdad?1
Touching his ears, she says, "Y, estas son sus orejos— orejos, si." When sh
asks a child where his ears are, he may hesitate. At home, undoubtedly h<
has heard the word orejos, but he has not quite connected it with his owi
orejos. The teacher comes back to him every once in a while until it i
clear that the connection has been made.

As part of the body awareness unit, the class makes a life-size piizzto
The teacher traces around a child; then asks, "What is missing-what doe
he need to be able to see? To be able to hear?" and so on. Once tin
missing features are put in, the teacher cuts the figure into parts, and cad
child gets the opportunity to put it together. He is encouraged to tiill
about what he is doing.

Games such as Simon Says are also part of the curriculum. As witl
every new activity, the teacher and the assistant teacher first demonstraK
how the game is played. Then they lead groups in playing it. Eventually
as a means of building self-esteem, each child takes his turn at being tin
leader.

At the conclusion of a unit, a performance checklist is administered, li
the case of body awareness, the teachers ask, "Can this child identify
himself by name? Label parts of the body? Locate them on a doll and or
himself? Describe simple functions?" Lagging children receive special at
tention.

48

From body awareness, the curriculum proceeds to clothing. The teacher
introduces the unit by giving the rule for this class of objects: "If you can
wear it, it's clothing." She then labels articles of clothing. "Who has a
dress on?" she asks. "I have," several girls respond. Says the teacher:
"How do we know the dress is clothing? Because we can wear it." After
she has labeled several pieces of clothing, she points to a chair and asks,
"Is this clothing?" "Why not?" "Can we wear it?"

The clothing lessons introduce the children to the concept of fasteners.
"Can you find buttons on your dress?" asks the teacher. "What do you
suppose would happen if you didn't have buttons on your dress?" "What
other ways can you fasten clothing?"

Sometimes a child will point to a zippei and name it in English. "Yes,"
the teacher will say, "that's very good. That's the way it is in English-
zipper Now, do you know how we say it in Spanish?" They want to
know, so she teaches them: segadura.

The children are also taught the labels of such things in the house as
stove, sink, chah, and table. The 3-year-olds have a model kitchen where
they look at and touch the objects they are labeling.

In many learning situations, action is required because it strengthens
the learning process. Half a dozen 4-year-olds, for example, gather with
their teacher in a screened-off corner of the classroom for an English
lesson. "We are standing," the teacher says, and places each child on his
feet. "We are standing," the teacher repeats. "We are standing," say the
children. The teacher begins jumping up and down, and the group follows.
"We are jumping," the teacher says. "We are jumping," say the children.
"That's right," says the teacher. "We are jumping. Now say it again: 'We
are jumping.' " "We are jumping," say the children. The teacher sits down,
and the group follows, "We are sitting," she says. "We are sitting," the
class repeats. "Good," says the teacher, "very good."

All age groups study the same topical units-food, clothing, house, and
so on-but the older the children, the more deeply each subject is ex-
plored. In the case of fasteners, for instance, the 3-year-olds learn the
labels for some types of them, and the 4-year olds expand this vocabulary.
The 5-year-olds are led to consider the purpose of fasteners. The teacher
holds up a child's shirt and asks first David and then Maria to try it on,
They can't get into it, though, because the front has been sewn up. "What
can we do about this?" the teacher asks. "How can we make this shirt
easier to put on?"

"Cut it," one child suggests.

"Tear it open," says another.

"Fine," the teacher says, and proceeds to rip the shirt down the front.

But now when the children try it on, it doesn't function as a shirt.
Again the teacher asks for suggestions. "What can we do now? How can
we make this shirt stay together?"

"A button," somebody ventures.

Somebody else says: "A zipper."

49

"Good,1* says the teacher. "Let's see what a button will do." She s
one on. The group talks about what a button does and then generalize
othei kinds of fasteners.

Lessons are based on what the child aheady knows. When English
sons begin for the 3-year-olds in January, the introduction of words a
concepts in English is preceded by a short review of the same words fi
concepts in his first language. For instance, starting a unit dealing w
vehicles, teacher and children talk about un carro, tin aemplano, and
bus. Then the teacher announces that everyone will speak in English.

Teacher, holding up a picture of an automobile: "This is a car."

Childien; car.

T: All right. Say it, "This is a car."

C: This is a car.

T: All right. This is a car. Good.

T: holding up a picture of an airplane: This is an airplane.

C: Airplane.

T: Good. Again-

C; This is an airplane.

T: All right, Martin, say "airplane."

Boy: Ah plane.

At each age level the teacher listens to each child in the group as he ti
to reproduce a new word.

Children, as teacher displays a picture of a bus: "This is a bus."

Teacher: A bus. Again-

C: This is a bus.

T: All right. Let's have Martin say it.

Boy: Bus.

T: Bus. Good. Now let's say the whole thing.

C: This is a bus.

T: Very good,

Then the teacher asks questions requiring a "yes" or "no" answer ('
this a car?") and other questions testing whether or not the child li
related the woid to the picture ("All right, Martin, show me the a
plane").

The next year, when the children are 4, the language lesson is entire
in English.

Teacher (displaying pictures of an airplane, a truck, and a ship): "The
are vehicles."

Children: Vehicles. These are vehicles.

T: Good. Why are these vehicles?

Several children: Because they have motors.

T: Because they have motors and because we can ride in them.

T: (showing airplane): Do you ride in this?

C: Yes.

T: Then it is a vehicle. Say it.

C: Then it is a vehicle.

T: (showing ship): Can you ride in this?

C: Yes,

50

T: Say it. Then it is a vehicle.

C: Then it is a vehicle.

T: Listen, if you can ride in it, it's a vehicle.

T: (with picture of a shirt): Can you ride in this?

C:No.

T:So?

C: Not a vehicle.

T: It's not a vehicle. That's veiy good.

T: This vehicle is a truck.

C: This vehicle is a tiuck.

T: Good for you. This vehicle is a truck. Cynthia?

Girl: This vehicle is a truck.

T: Good.

The teacher varies the pace of presentation in order to keep the chil-
dren involved. They must listen carefully if they are to answer correctly:

T: Is this a vehicle: Is this a ship?

C: Yes.

T: Okay. I want Martin to show me a vehicle that is an airplaine ....
Very good. I want Cynthia to show me a vehicle that is a truck ....
Good. I Want Juanita to show me a vehicle that is a banana.

Girl: It's not a vehicle.

T: No, you're right and I couldn't fool you. That's very good .... I
want Olga to show me the vehicles .... Very good. She pointed to all of
them because they are all vehicles. But are they all ships?

C:No.

T: Are they all airplanes?

C:No.

T: Are they all trucks?

C:No.

T; What are they all? What can you call all these things?

C: Vehicles.

T: Vehicles, That's very good.

A year later, when the children are 5, they learn that vehicles includes
still other types and even sub-types. After the group has discussed the
function of a passenger train-it carries people places-the teacher shows
them a freight train.

Teacher: What kind of train do we call this one?

Child: Where arc the people?

T: There are no people on this train. What does this train take?

Children: Food, Gasoline ... .

T: Food, gasoline. What else could this train take?

Child: A refrigerator.

T: A refrigerator. You're right. This is a freight train. Say it .... Say it,
Gonzalo, what kind of train is this?

Boy: Freight train.

T: Say the whole sentence.

Boy: This is a freight train.

T: Very good.

51

The lessons on vehicles-and on the other units-provide opportunities
for taking up language differences that present special problems. One ol
these problems has to do with differences in length; largo and mas largo i*i
Spanish become, when translated literally, long and more long in English
Teacher; This train is short. Say it.
Children: This train is short.

T: (pointing to picture of a long train): Is this train short?
C: No.

T: This is not short. So what could we say about this tram?
C: It is long.
T: It is long. Say it.
C: It is long.

T: (pointing to picture of a longer train): This train is

C: More long. Longer.
T: Longer. Say it, This train is longer.
C: This train is longer.

T: (pointing to longest train): What can we say about this train?
C: More longer.

T: No, we're going to say this train is ....
Child: Longest.

T: Good. Longest. This train is longest. Say it.

Language lessons for the 5-year-olds also include a review of prcposi
tions (the children learn, for example, that in English an airplane is on tin
ground but in the sky) and a sequence of questions enabling the child ti
relate present, past, a,nd futme tenses (the airplane is in the sky; bcfor>
that it was on the ground; when the airplane on the ground takes off, i
will be in the sky).

Expanded Language Program

Complementing the language lessons in all three years is what th

Center and the Southwest Educational Development Laboratory call "th

expanded language program." This is so planned as to arouse the chil

dren's interest in talking about the topics they have encountered in th

language lesson, and about related topics. It gives them practice in the nci

words-and concepts and structures-and at the same time requires ther

to use their memories, reasoning powers, and imaginations. An EngHs'

lesson dealing with fruits, for example, is followed by a discussion of thes

and other foods, The teacher gets the talk started by asking questions

"ruits do you peel before eating? Does an orange have more juic

lanana? What would happen if you squeezed a banana? And UK

i language activity involves the cooking of rice and corn, in ortk

wt-and talk about-how cooking affects their volume. Food a Is

0 help teach the concepts-and the labels-of hot and cold £oa
cornflakes) and big and little (apples).

rt of this program, books pertaining to the unit being studied ai

1 the classroom for the children's use, and at story time each da

the teacher reads from one of them. If the lesson unit deals with houses,
for example, the teacher reads a story about houses. Then the children are
encouraged to talk about their own home, or a friend's home, or anything
that the story has brought to mind. Five-year-olds may spend three or
four periods on the same story. The first day they may just look at the
pictures and talk about what may be happening. The second day, the
teacher reads the story and asks the children to compare their predictions
with what actually happened. The third day the children draw a picture—
the scene or event they liked best-and the next day they talk about it to
the group.

For the 3-year-olds, the expanded language program is in Spanish
throughout the year; for the 4-year-olds, much of it is in English, but the
story period remains in Spanish. For the 5-year-olds, practically eveiy-
thing is in English. Daily music and art periods also serve the language
program.

Developing Perceptual Motor Abilities

Many disadvantaged children reach first grade deficient not only in
language skills but also in perceptual skills, which are even more basic to
intellectual growth. The Good Samaritan school seeks to develop these
abilities through planned daily exercises.

Training in attentiveness and auditory skills begins during the first week
of school, when the 3-year-olds learn a rule about listening— that when
they hear the bell, which is used to signal a change in activities, they will
stop what they are doing and listen. Each child rings the bell and repeats
the rule. Another time the teacher demonstrates two bells having different
tonal qualities; then the children close their eyes, the teacher rings one of
the bells, and the children tell her which one it is. In another exercise the
teacher plays one of four instruments-bell, drum, triangle, or cymbals-
and a child who has been blindfolded is asked to walk to it. Other lessons
deal with sounds in the home -a door closing, dishes rattling, a window
being opened, water running. Such exercises get the children accustomed
to paying attention to the differences between sounds and to locating the
source of a sound. Many of the children, from crowded, noisy homes, may
have learned to tune out; now they are learning to tune in. Later on, more
advanced auditory discrimination lessons will sharpen their ability to de-
tect and recognize differences in the sounds of words,

The visual training program includes practice in discriminating among
objects on the basis of size, color, shape, or function; paying attention to
the boundaries of objects, as in cutting out or coloring a picture; and
noticing and reproducing patterns, as with blocks, pegs, and beads. Thus,
the program strengthens visual skills and reasoning ability and at the same
time helps prepare the child for an important task that he will face when
learning to read-distinguishing one letter from another. Under visual
training, too, come many exercises to strengthen attentiveness and memo-
ry. In one, for example, the teacher displays three pictures, then asks the

53

children to close their eyes as she turns one over. Can they remember
which one is missing? When there are four pictures, can they remember^
Five? Seven? Another exeicise uses letters instead of pictures. Another
calls on the children to reproduce from memory patterns they have been
shown-simple ones, like circles and triangles, at first; later on, numbers
and letters.

While the typical disadvantaged child is proficient in such gross motor
activities as running, jumping, and climbing, he lags in the development of
the fine motor skills needed for classroom success, particularly in writing.
The Center's daily schedule, therefore, includes such activities as lacing
shoes, tieing ribbons, manipulating small building blocks, dropping bill-
tons through a narrow opening, using a pair of scissois, tracing, coloring,
and pasting.

Evaluating the Project

By the time the children have completed two years at the Center, most
of them-whether at play in the couityard or at work with another child
on some project in the classroom-are using both Spanish and English.
There is no conscious choice, the teachers think; the children use the
words that come to mind first.

Will these children, after the third pre-school year, be able to compete
in first grade with children who have learned English at home?

The staff feels that they will. In substantiation, the teacher of the
5-year-olds points out that in first grade, where she taught for four years,
teachers are supposed to begin the reading readiness program immediately,
a very difficult matter with children who know little English. The 5-ycar-
olds at the Center, she thinks, are ready for such a program even before
they have finished half of their last year. They have been trained to listen,
which is one of the goals of the public schools* reading readiness program.
They have learned enough English to be able to follow instructions; for
example, they know when the teacher is talking about the top of the page
and when she is talking about the bottom. They have learned other con-
cepts and the English words for them. Give some first-graders from dis-
advantaged homes two or three objects and ask whether they are the same
or different, and the children will just look at the teacher or say "Maybe."
But the children at the Center, she says, "can tell you right off." The
Center's children have also had experience in a common reading-readiness
exercise: "Let's look at this picture and you tell me what you see." The
Center's children have done this many times; they know the labels i»
English- "boy," "girl," "father," "mother," "house," "car"-for what
the pictures show, and they have had practice in expressing themselves.

Approximately 30 of the Center's graduates are now in first or second
grade, in half a dozen different schools. Two teachers have spoken highly
of the few they have encountered, and at least a dozen mothers have
dropped by the Center to report that the Good Samaritan graduates are
doing considerably better in school than their older brothers and sisters.

54

For lack of funds, the Center notes, a systematic follow-up of these
children has not yet been undertaken. However, under a ptogram financed
by the Southwest Educational Development Laboratory, an effort has
been made to evaluate the San Antonio project on the basis of (a) changes
in the children attending its classes during the 1968-69 school year, and
(b) a comparison of these changes with those experienced by children in
two other groups. One of these groups came from three day-care centers
in San Antonio funded as Head Start programs undei the Office of Eco-
nomic Opportunity and concerned with children from Mexican-American
families of low-economic status. These centers offered some of the ele-
ments of the Good Samaritan program but were concerned in the main
with providing all-day care and supervised play. The second comparison
group comprised 16 3-year-olds who were eligible for the Good Samari-
tan's regular program but could not be admitted. The parents of these
children were encouraged to participate in a parent-involvement program;
through it, the Center hoped to learn whether or not the children could be
affected indirectly by increasing the parents' interest in child develop-
ment. As it turned out, the fathers and mothers involved in this program
attended the scheduled semi-monthly meetings quite irregularly, and the
discussions— though they included such topics as hygiene, mental health,
and story-telling techniques-were not so specifically concerned with child
development as had been planned. Essentially, the children in this group
may be considered controls— that is, as having experienced no significant
intervention.

To try to determine the intellectual development of the children in the
three groups during the nine months between September 1968 and May
1969, the staff used:

1 . The Leiter International Performance Scale, a non-language test rely-
ing heavily on visual discrimination. Some items call for matching one
object with another; others, for grouping objects that belong together. The
examiner demonstrates what is to be done—for example, he takes a red
block and puts it with a red square, Then he gives the child other items in
the same category to do by himself.

2. The Peabody Picture Vocabulary Test, which provides an estimate
of the child's verbal intelligence although the child himself is not required
to talk. The examiner names an object (such as "dog") and then asks the
child to point to it in one of several pictures presented to him. In addition
to the standard English version of this test, the Center developed and used
a Spanish version.

The results are shown in Table 1 .

In the first examination, as expected, since all the children came from
educationally deprived, non-English-speaking homes, all three groups
scored well below the national average in a test -the Peabody-requiring
the use of English in its administration. Each group fell below the average
by at least 40 points. Even in the Spanish version of this test, the groups
scored from 12 to 25 points below standard. On the instrument not
requiring language in its administration-the Leiter— all three groups fell
within the normal range. This result, too, had been expected.

55

Table I

AVERAGE SCORES BEFORE AND AFTER NINE MONTH TRIAL PERIOD
Test Good Samaritan Day Care "Controls"*

Sept. May Sept. May Sept. Moy

Uiter 107 121" 99 101 97 96

Peabody-
English

59 68

60

63

58

58

Peabody-
Spanish

88 102*"

76

77

75

80

"Children in Parent-Involvement program. See text,

"Significant at less than the 01 level, meaning that the results would have occurred by chnncolt
than 1 time in TOO.

Nine months later, only the children who had attended the Good Samar

itan Program showed marked gains. On the English version of the Peabodj

Test, this gain was not-by a slight margin -statistically significant; on the

Spanish version and on the non-verbal test, it was. The scores on these

tests were also significantly greater than those made by the other groups.

In sum, the Center's program for 3-year-olds significantly increased

their intellectual performance as compared with what it had been and as

compared with the performance of children in a traditional nursery school

program and with that of children not involved in a program.

The staff emphasizes that these findings are for a single year only and
for small samples. It points out, too, that the evaluation tests used so far
are not ideal by any means. Though the Leiter eludes the language barrier,
the test is time-consuming and those who administer it may unwittingly
provide clues to answers unless specially trained. The experience at San
Antonio suggests that the vocabulary of the Peabody Picture Vocabulary
Test is too advanced for disadvantaged pre-school children and contains
too few items at each age level for accurate measurement of differences
between groups,

Improved evaluation measures. In conjunction with child development
authorities in Texas universities and the public school system* the South-
west Educational Development Laboratory is developing and standard-
izing several tests intended to measure a pre-school child's proficiency in
language more adequately than tests now availnhlfl. Onp. snnh effort is

Working with Parents

The Center keeps in touch with parents through monthly meetings to
which all fathers and mothers are invited and through conferences with
individuals when desnable. The meetings are quite informal, almost like
family gatherings; mothers even bring their new babies to be admired. At
one recent session, the talk got around to how children ought to behave
when they got to public school. One mother expressed the opinion that the
teacher was always right. "Keep quiet," she said she told her fiist-grader;
"then the teacher is going to like you, and you'll pass into second grade."
The other parents seemed to agree. Mrs. Blankenship suggested, though,
that it was healthy for a child to ask questions, that he had a right to ask
them, and that, in fact, a teacher wouid like him to ask them.

When a problem arises that the Center alone cannot meet, the people
who know the child and his family— generally at least one teacher, hei
assistant, and Mrs. Blankenship-talk it over and decide how to advise the
parents. Usually the mother will be asked to drop in; sometimes a staff
member will visit the home. The mother of an extremely timid girl was
advised to scold her less, praise her more, and give her some opportunity
to talk. The parents of a boy who had no motivation because at home his
older brothers and sisters did everything for him, were advised to let him
have more responsibility and to praise him whenever he undertook a new
task.

Parent Education Program. This program, a revision of the parental
involvement project noted earlier, is for 16 mothers of 3-year-olds who are
not included in the regular pre-school course. At the start, each mother
was video-taped as she taught her child simple tasks, such as sorting
blocks, and as she read to him, The staff wanted to learn how well the
mothers explained a task to their children, how they organized informa-
tion, to what extent they used praise, and how much affection there was
between mother and child. Now the Center is trying to train the mothers
to work with their children more effectively. This involves showing them
how they can present educational activities, such as comparing fruits of
different color and size and labeling and counting common household
objects. In many cases it also involves changing certain behavior patterns,
the most common of which is to ignore good behavior and punish bad
behavior, so that the only way a child gets attention from his mother is to
misbehave,

Mrs. Blankenship, working with mothers at the Center, demonstrates a
different activity each week. Once a week another staff member visits the
homes and observes how the mother works with her child and whether or
not the child has learned the activity for the week. The Center plans to
continue this program for three years so that it can compaie the effects of
working directly with children in the classroom and working indirectly
with them through their mothers. Evidence that the children of the
mothers in the Parent Education Program are substantially benefitted
would be welcome, would point toward a way of reaching educationally

57

The program is double-barreled in that it helps a group of deprived, social-
ly disadvantaged, poverty-inhibited adolescents who then, each in turn,
provide a significant mental health service to the community. Help is given
at less cost and more effectively than it could be given by a progiam
limited to traditional use of professionals

The area has an acute lack of clinical facilities for children and adoles-
cents. However, it is obvious that the resources of such facilities, primarily
oriented as they are to the middle class, would be inadequate to the job
anyway. The usual agency is not oriented to the real problems of the
poor, and the poor genei ally do not seek its services. Treatment personnel,
drawn from the middle class, are not in time with the clients and find it
difficult to work with them Theie is a special need for integration of any
treatment program into the local institutions, i.e., the schools and courts,
and this generally is not done.

Another problem is that in this type of neighborhood the impact of
treatment may be compromised by the stigma of being a patient. At the
same time, the undeistanding and knowledge of the mental health practi-
tioner is valuable, and it is primarily a new approach that is needed. One
theory (held also by the staff at Baker's Dozen) is that the best kinds of
prevention and therapy aie those which can be done in the normal social
context of the youth-that is, within the schools, the work, or the natural
groups of the community.

In addition, the services must be realistically accessible. For instance, at
Baker's Dozen a variety of services are offered because there are so few
facilities in the neighborhood. The services are free, as the clients have no
money. Baker's Dozen has no authority in the sense that the police or
schools have authority, and so it must attract and hold boys and girls by
the opportunity for change, for friendships, for the activities that it offers.
It must be open the hours that it is needed, not just 9-5 on weekdays.
Therefore its schedule is extended to evenings and weekends.

Aides, the term used for the young indigenous workers, were recruited
and selected through a process of "screening in" rather than "screening
out/1 Signs, encouraging people to apply, were put up in a wide variety of
places including bars and laundromats. Radio and TV news announce-
ments were made. Applicants were considered with characteristics that
would ordinarily bar them from employment. Only a fifth-grade educa-
tion was required and this only so that reading would be possible. No
previous work experience was necessary. Most jobs call for a clean police
record, but this was not required here. It was essential, however, that no
court action be pending which would interrupt the training. As to
personal characteristics, the only requirements were that applicants be
free of serious physical or mental problems and communicable diseases.
Psychological testing was used to identify gaps and problems and was
referred to again later when an evaluation of the training was made.

Applicants were encouraged from low socioeconomic groups and from
families with low incomes. Five men and four women were selected, all of
whom had lived within five blocks of the Center. They ranged in age from
17 to 21. Several had dropped out of school around the seventh grade;

60

one had completed high school. The reading level ranged from a minimum
of fifth grade to a maximum level of llth grade. One young man was
diopped because of legal problems.

The eight trainees were subdivided into high- and low-risk groups, four
in each. High-risk youths were described as deprived youths who had had
a series of police and criminal involvements, some emotional or delin-
quency problems, and those who may have spent time in an institution for
an offense. They read at a minimal fifth-grade level, dropped out of school
early, worked only at odd jobs, and never worked longer than 3 months at
any given job. The low-risk youths were defined as deprived youths who
had no police record, who continued in school until family and poverty
circumstances forced them out. They worked at menial jobs but for longer
periods than did the high-risk group. The average number of siblings for
the aides as a group was five. All the aides chosen had multiple social
problems and were so accustomed to rejection, failure, and defeat that
they had to be convinced that all this was true. They suspected that
"there must be a catch to it somewhere."

Recruits were numerous, and the rate of completion throughout this
project and otheis run by the Institute for Youth Studies was very high.
The staff reports that over 1 50 multiproblem youths have been trained in
Institute for Youth Studies projects and that the dropout rate has been
less than 1 percent.3 One aide selected for the Baker's Dozen project was
dropped as he had to serve a prison term, and at this writing he had still
not returned to the project. This points up how strong the holding power
of the project has been as many of these youths would have undoubtedly
been in police trouble and perhaps given stiffer sentences due to their
histories of past offenses. In this and other ways they might have been
prevented from continuing in the project had they not exercised extreme
caution in order to be able to participate.

Applicants were seen in groups during the selection process. Every
effort was made not to impose unnecessary barriers that would cause
people to drop out. Even so, one aide said at a later date that he had
arrived at the Center and found instructions to go to Freedman's Hospital
for a physical, As he didn't have busfare, he almost quit until a staff
member organized rides for the women and encouraged the men to walk
there together, thus solving the problem.

The initial training period lasted 3 months. When it was completed, a
graduation was held that was the fiist such ceremony most of them had
ever attended and marked for many the longest period of employment to
date. The aides were given a stipend of $20 a week during the training
period; this jumped to $75 a week at the end of that time. Presently they
are earning $80 a week, and increases are written into the budget. They
are considered GS-2 level ($4,108), and provision has been made that they
will move up to Government Service levels 3 and 4. The District of
Columbia Health Department has written job descriptions and positions

2These aides are working in schools, settlement houses, and children's institution
recreational centers throughout the city.

61

into their budget, and this new type of position will be continued after
the initial demonstration phase has ended. An exciting part of the pro-
gram has been the development of new caieers and the additional resource
for manpower in mental health.

Training Program for the Aides

Major training goals were:

1 . Development in these youths of the necessary motivation, identity,
values, and capabilities for maximally utilizing the offered training.

2. Leaining the basic personal, social, and interpersonal skills, attitudes,
and knowledge which would help them successfully cope with and
solve group, client, and personal problems.

3 . Learning specialized skills for their roles in mental health.

4. Developing flexibility of attitude, role, and viewpoint.

Training was designed in three parts which ran concurrently -core group,
specialty workshop, and on-the-job training. These were continued after
the initial 3-month training period and comprise the vehicle for inservice
training which continues throughout the program.

Core Group

This is the basic group which meets several hours a week and is led by
the same professional staff person. It provides a place where a variety of
issues can be raised and day-to-day problems of work can be discussed.
The group itself provides a laboratory for group interaction and manage-
ment techniques. Work throughout the training program is done primarily
with the entire group. It has been found that this makes possible greater
participation on the part of the aides as they have the support of their
peers.

The level of anxiety among the aides was high due to the new demands
made upon them, and the group was found to be effective in dealing with
this in a way that allowed for maximum change. It gave each aide a group
to move with in his transition from an unemployed school dropout to a
semi-professional. This is very important because these young people,
drawn away from familiar patterns and attitudes, often feel very isolated.
They have helped each other to correct the distorted views they have held
as a result of their background. They have learned new ways of solving
problems, often by applying the new techniques to problems that come
up in the core group. The leader is a visible example of how to lead a
group; what he does is more likely to be influential than anything said in a
lecture.

The trainees and the staff have had some difficulty in establishing the
role of the aides. For instance, such questions arise as to how much the
aides should have to say in policy decisions. How much responsibility
should they have? The core group was used to define an appropriate job

62

identification. Initially, the aides wondered, "Should I be like the Man or
myself," Although they were encouraged to absorb attitudes and tech-
niques, they were discouraged from becoming carbon copies of the staff as
it was hoped that a new type of professional, comfortable with lower class
behavioi and at the same time able to meet middle-class demands, would
emerge. Since it was highly desirable that they remain in touch with and
empathic to the lower-class group from which they came, this was en-
couraged. Their observations about the shortcomings of professional tech-
niques were encouraged. The staff had to learn to stand the criticism and
separate the useful information from the aggressive attack.

The aides were also encouraged to examine their own backgrounds.
With the help of their leader they developed a very perceptive paper,
"What It Means To Be Really Poor." This was an outgrowth of their
examination of the process of adolescence and the difficulties of living in
a slum. They began to acquire some perspective and to sort out what
could be done. In their paper they comment that when you are poor you
know only your own troubles and your own neighborhood.3 If you are
failing in school you do not have the time to ponder what it is about
American education that makes it difficult for the lower income student
to achieve scholastically. Nor do you have the energy to do something
about the situation.

In the core gioup, methods of social action are discussed, and the aides
have been encouraged to participate in community groups. They have
attended schoolboard meetings and Senate hearings. They have been en-
couraged by the staff to speak up regardless of whether or not other
groups in the community find their opinions embarrassing. A prominent
issue for these trainees have been that of discrimination. All are expected
to have opinions and to act accordingly. They are expected to act
responsibly and knowledgeably. Early in the IYS training programs there
was discussion as to whether trainees should attend community meetings
as observers. This was found to be confusing and tended to weaken the
piogram and was abandoned. A major problem of these young people has
been their attitude of defeated, resigned acceptance of an unsatisfactory
status quo, and group action protest methods have been helpful in showing
them ways to achieve improvement.

Tiie issues discussed are varied. At one core meeting questions were
raised about vacation hours, a secretary's attitude, and what to do about a
letter received from an invalid old lady who wanted a Christmas basket.
(Although this is not a routine agency service, the group had already
called her and taken up a collection.) They also discussed whether or not
an aide, despite provocation, should ever hit one of the children in his
group. Other issues, such as the use of the credit union, are raised by the
leader. The staff reports that a great deal of information taken for granted
by middle-class workers has to be discussed with the aides in order to be
sure that they clearly understand both their rights and their obligations.

3 "What It Means To Be Really Poor." Baker's Dozen aides and Lonnie Mitchell,
Ph.D, Baker's Dozen Youth Center, Washington, D.C., unpublished paper.

63

The job of the leader of the coie group is a difficult one. The young
people aie angry, suspicious, distrustful, vulnerable, hopeful, sensitive,
brutally realistic, and terribly unrealistic all at the same time. There is a
gap that must be spanned, Neither side wants to be compromised, T!\c
leader must be comfortable with authority and know when to turn
responsibility over to the group, He must act as a liaison with the staff. He
must interpret the staff to aides and vice versa. He must be a good and
knowledgeable teacher. The group will scrutinize him and be critical as he
represents the people trying to remake them. He is the personification of
the professional. All staff working with the aides must be able to handle
passivity without becoming dictatorial. If the leader or other staff mem-
bers become anxious about group inactivity or lack of response, tlic
group's confidence in their own ability will be undermined. A good ex*
ample of this occurred in an interview with the group. After an initial
explanation of the broad questions to be discussed, the group was utterly
unresponsive. The leader then restated the issues. The group remained
unresponsive. Then upon questioning they said that they found the
second explanation insulting because it made them look stupid. Upon
hearing that a response is looked for as evidence of understanding, they
said that that was "middle class" and asked who wants to "react and
commit themselves." The staff reported that complaints about the pro-
gram and lack of response were common manifestations of anxiety and
feelings of noninvolvement in the program.

The staff reports also that they have found a rather strict approach as
opposed to a permissive one most effective. Rules are strict, but anything
is open to discussion. Aides are docked for being late even if it is a small
amount. Hours are checked. A great deal is asked of the aides in terms of
meeting professional standards. The aides by and large have been able to
do this. On the other hand, the project has not discharged aides for some
issues, angry episodes, minor police violations, etc., that would be suf-
ficient grounds for dismissal in some agencies. The fact that the finances
are administered through a university has made its employment standards
binding on the aides and has imposed stricter rules than the staff would
have wished. These, however, have been subject to appeal.

The aides are interested in the process of becoming a professional. Most
of them have now grasped the concept of the steps leading to a profession
and have either gotten tutoring or resumed special classes in school. They
now see that it is not magic which makes a person a doctor, a psy-
chologist, or a social worker, but a process pursued step by step. The
program in general and the staff personally have given support to any
interest in further education on the part of the aides. One boy audits
classes in psychology at Howard. They are also encouraged to continually
relate the more academic work in their training to their own lives. The
aides have commented on this issue in their paper:4 "The learning en-
vironment of the slum child is dismal. He is often emotionally disturbed.
It is a mistake to urge such a child to get an education because it will help

"What It Means To Be Really Poor."
64

him to get a good job and allow him to leave the slums. The most
promising motivation for a child in the culture of the poor is the acquisi-
tion of knowledge as an end in itself and for its own sake." The irony is
that, having found a good job, the aides have found their own educations
too limited to allow them sure access to better jobs. One young woman,
married and with two children, would like to be a social worker. She is
young but would have to complete high school and go on to college, and
remedial work takes more time than original schooling did as it must be
done around the demands of other responsibilities. This, coupled with the
lack of money, poses almost insurmountable problems despite her intel-
ligence and abilities.

Specialty Workshops

The aides were given didactic work on such subjects as interviewing,
history taking, record keeping, group observations, psychological testing,
etc. The classes resemble college survey courses in style, They sought to
provide an overview. Needless to say, the aides needed a great many
background issues filled in. The staff opeiated by giving talks and lectures
and utilizing extensive questioning and give-and-take to be certain that the
information was being understood and absorbed. The aides often brought
up questions about such things as the effects of LSD, alcoholism, etc. The
staff talked about issues reported in the newspaper and tried to keep
stimulating the group but tried at the same time not to ovei whelm the
group. The investigators report that it was necessary to keep a current
tone to the work because the group as a whole had such a negative feeling
about schoolwork that initially they found it almost intolerable, The
problem of even sitting stilf was difficult for some to manage. They had
felt so out of touch with previous teachers, so attacked and so unsuccess-
ful, that it posed a difficult teaching task to the staff to create an
atmosphere which could foster their learning. Efforts were made also to
pace the work so that the aides would learn answers to problems that
arose in their groups so the practical significance of the theory would be
visible.

On-the-Job Training

Many visits to community agencies were also part of the training. The
group visited Congress, the juvenile court, Junior Village (a children's
residential home), St. Elizabeths Hospital (a psychiatric facility), and
other social institutions. They saw many films on mental health and child
development. They interviewed members of the neighborhood for ideas
about community needs, and they learned the rudiments of research
methods.

Throughout the training it was intended that the aides would learn to
become more sensitive to interpersonal feelings. They report an awareness

65

of a considerable change in how much they perceive of one another's
feelings. They cite this as a problem in the training as sometimes they feel
awkward and ill-equipped to deal with this. This problem is familiar to
anyone who has experienced similar training. Although any staff member
would talk to any aide who sought his help, there has also been a group
psychotherapy program for the aides. This makes it possible for issues
inappropriate to other meetings to be referred back to that group. The
same psychiatrist has met with the group since their entrance into the
program. He feels, as does the rest of the staff, that the aides have over-
come some, but not all, of their initial difficulty in talking about their
feelings-an idea quite alien to their way of life. They were, as a group*
much more comfortable with activity and movement than with verbal
modes of expression.

There is a fear that you can be manipulated when people know your
feelings. The aides talk about "gritting," which means maintaining silence
as a way of controlling a situation and still remaining technically non-
obstructive. The staff has learned to call the aides on such tactics and lias
earned the respect of the aides to the extent that they have done this. The
psychiatrist who meets with the group makes no administrative decisions
about them because it was hoped that this would further a therapeutic
atmosphere and provide one place where the aides could speak without
fear of losing their jobs. At times of acute personal stress the aides have
been referred to other psychiatrists in the community, The group meets at
Baker's Dozen, and it is considered part of their program and is com
pulsory.

Basically the aides were trained to carry out specific duties and to fill
certain roles. These included;

1 . Be group leader, helper, and planner for 1 0 children in each of two
groups.

2. Participate with the psychologist, psychiatrist, or social worker in
developing structured therapeutic programs foi their groups.

3. Observe and record individual and group behavior.

4. Conduct interviews with group members and provide information to
the professional staff for feedback and quality-control purposes.

5. Escort groups on trips and tours,

6- Participate in individual and group supervision.

7. Attend staff conferences.

8. Write progress reports and keep records of daily observations on the
children with whom they work.

The Groups Led by the Aides

Each aide is responsible for two activity groups. Activities are struc-
tured to provide ego-strengthening and therapeutic benefits and include
recreation and cultural, social, and community activities. A major purpose
is to raise the behavioral standards of the children involved. A strong
emphasis is placed on the concept of the aide as a good figure for the

group to identify with and on the use of the aide's management to estab-
lish beneficial controls over the childien. The aides seek to reduce
symptoms, lessen police contacts, and improve the social functioning of
the children. The children, who are from the area around the Center, had
school problems, difficulties with the police, defiant attitudes toward
authority, and many symptoms of social and personal disorganization.
Referrals were accepted from the juvenile courts, public schools, the
Urban League, and other agencies. The age range was 12—16. Both boys
and girls were accepted. Youths excluded were those who were being
committed to an institution as a result of court action, those in need of
immediate hospitalization, and mental defectives whose difficulties posed
certain management problems in an outpatient setting. In addition, the
aides ran dances and open houses to bring children into the Center and
interest them in the program. This further tended to bring in natural
groupings of children. Although these children were not unusual in the
community and not technically referred, they had many of the same
characteristics as those who were. This points up the degree of social
disorganization in the community when so many multiproblem children
can be found by simply dipping into the community. Speaking of one
such child an aide said, "When I say he gave the secretary a hard time, I
mean he pulled a gun on her."

The emphasis is on work with the teenager rather than the parent. The
aides know from their own experiences that work with the parents would
be less fruitful than work with the children. They feel a positive non-
parent-connected relationship is helpful to these adolescents. They are
only too well aware of the hostility and rejection in the home situations
of many of their group members. They point out that these young people
would be on the streets if they were not at the Center. The importance of
"someplace where nobody yells at you" is underestimated in the aides'
opinion. Neither the aides nor the group members see this as a patient-
therapist relationship. They see it as working together toward getting
along, and planning activities. The aides are rather permissive but have
certain taboos, such as not allowing the boys to play the "dozens."5 They
usually try to control fighting by manifesting their disapproval. They also
have learned that withholding privileges helps, but by and large they feel
that as the boys stay with the groups these problems diminish by them-
selves. They have more difficulty when they themselves are attacked be-
cause the old patterns of self-defense come into conflict with their newly
found professional approach.

The groups are designed to improve the coping skills of the teenagers
and to help them towards more positive attitudes. At times the group
seems to repeat past experiences of the youngsters-for instance, when

5 "Playing the Dozens"-the act of talking about another person's parent with the
intent of hurting the person's feelings. Foul and abusive language is often used. "The
dozens"-a term used for the act of "playing the dozens." Definition from "A Diction-
ary of Local Terms and Expressions" edited by Mitchell, Lonniej Ph.D. ("Definitions
primarily contributed by Baker's Dozen aides.)

67

they visit Washington museums, etc- However, one finds on closer exami-
nation that although all of these children have been herded through oji
educational tours, raiely has someone discussed it all with them in terms
they could understand and helped them with it. When necessary the aides
intercede with other agencies such as the schools. Contact is made and
maintained with the young people even when they leave the group. The
adolescents sometimes seek intense contact with the aides and the aides
have neither discouraged nor encouraged this, but accept it. One little girl
used to show up at an aide's apartment on Sunday just to say hello and
then leave, The aides lend money if asked, and all have been asked. They
try a variety of approaches and are not bound by tradition.

The aides comment that the predominant motif in the boys' groups is
that of aggression and sex. Sex is the predominant concern for the adoles-
cent girls' groups, All the aides agreed that the girls in this neighborhood
"get wise" too young. By ages 11 to 12 they are too seductive, loo
stimulated. Many of the girls will become illegitimately pregnant, The
aides stress that this tends to fmther trap the girl who wishes to escape the
ghetto. "Theft, murder, fornication, desertion are so much a part of their
lives that they become indifferent to what would shock other people."6
The aides see in their group members the patterns of impulsive living for
the moment, and they try to help their group members find other ways of
living. They talk to their groups. One aide, herself pregnant, has been
asked "how it feels to be pregnant," All of the women aides find such
issues under discussion, and they meet them openly and honestly and Iry
to give guidance. They are in close contact with the context of the ques-
tions and can answer them more appropriately than someone from
another background. Although the planned programs are developed by the
teenagers and include such activities as movies, parties, cooking, makeup,
etc., depending on the sex of the group, doubtless the informal activity
and discussions are also valuable.

The aides try to foster a feeling of concern among their group members
and to combat the feelings of helplessness, isolation, and indifference so
common among their group members.

Since Baker's Dozen also serves as a training placement for social group
and caseworkers and psychologists, the aides' intimate knowledge of tins
culture is passed on to professionals in many ways. In their training pro-
gram, the aides have developed a book of slang vocabulary phrases com-
mon to the area. They have included street talk, homosexual jargon, drug-
addict talk, and prison terms. Publication is being considered at this time,
In addition, a series of radio programs was written by the aides and
presented over a period of months. They discussed their training program,
their perspective on the community, their ideas on why children mis-
behave, and other topics.7

6 "What It Means To Be Really Poor."

Radio program series; "The Nonprofessional Youth in the Community." Station
WOL, Washington, D.C., November 1 965-February 1966

68

Research

Research data are being collected utilizing the self-reports of the aides
and projective measures and observations by others of the aides in both
experimental and natural situations. The program is being studied in terms
of such issues as staff roles and decision-making proccduies. Crises are
being noted and followed by the research staff. Records are being kept on
attendance of both aides and group members. Job performance lalings on
the aides, background data on the social situation of the aides, their
families, etc., are being kept. Periodic evaluations are added to the initial
comprehensive personal evaluation. The aides aie asked foi indications of
their self-image and their self-esteem. Their patterns of behavior (as seen
in such things as impulse control), their levels of aspiration, values,
anxieties, and other issues are being noted and measured.

Other Issues

The entire issue undei discussion— of the effectiveness of the indigenous
worker due to his close understanding of his group's culture— leads to a
discussion as to whether middle-class people, the professionals, can them-
selves be effective with the aides. The staff report that this is entirely
possible but that certain conditions are essential. It is essential that the
staff understand the realities of life in the slum. They must undcistand the
obstacles that have been presented the aides. They must understand the
ways in which the aides have not been included in the mainstream of
opportunity. They must understand the emphasis placed upon money. At
Baker's Dozen the aides are all Negro; the staff is primarily Negro. The
staff and aides see the bonds and differences not primarily in terms of
color, however, but in terms of class background. All staff, regardless of
race, must be able to tolerate hostile remarks about the white community.
In this area of the city the problems faced by the residents arc brought
about both by being poor and being Negro, and both factors have to be
considered. On the other hand, the middle-class professional provides
these aides with a glimpse of a life they hadn't really seen firsthand
before. One aide said he couldn't wait to leave "this lousy area" and that
his ambition was to be successful enough to buy a "house next to
Dr. . . .and put a little cast iron black boy out front and paint it
white . . .," the best thing of all being he would know it would still be
there in the morning. This also highlights one of the problems of the
program. The job requires that the aides not move out of the area for the
duration of the program, and they are chafed by this restriction. They
have gotten a good job and they want to move out. They do accept it as a
realistic requirement but vow to move as soon as possible. Therefore it is
reasonable to assume that they will leave this area. It has been an interest-
ing aspect, and one now being studied, that these aides had lived in the
area and were known by many people. They have provided a visible model
for other young people as long as they have lived in the area. The effect
on those who knew them and on the area is being assessed.

69

Although many issues have to be dealt with sensitively, the aides hi
been found to have a far higher tolerance for frustration than predict
With guidance, the number of severe disruptions has been few. The nc
verbal nature of the group makes it difficult for staff. The aides tend
gieet each new topic with silence. They also tend to cast the staff in 11
role of boss. Then, too, the very process of helping people to a new vie
of themselves makes the leader vulnerable to their new strength, ft
instance, the aides have learned the power of group activity through tlie,
study of the civil rights movement. Now they are aware that if the
become seriously dissatisfied they can quit en masse and jeopardize th<
program and the leader.

The aides cite money as a major asset of the program, The importnnct
to them of the opportunity for a decent job has been tremendous. They
are deeply conceined about future work. They question whether or not
they will actually be able to get jobs in other similar settings. They still
feel as though the system is closed to them. Many of them are not deeply
committed to human relations work, although they would probably con-
tinue in it if a future existed for them there. All feel the program has
changed their lives. They are strongly attached to the staff in the program
and are loyal to the agency.

Results

The most impressive finding is that young people such as these can be
trained as aides and can do the work successfully. Although levels of
efficiency vary, all have been working at an acceptable level. An effective
training piogram has been developed and has been found to motivate and
hold the young people. Despite their youth, and many are in need of
employment at an age far younger than most professionals reach a job,
they are able to handle the responsibility. The aides have served as a
bridge between the professionals and the people being served and have
served themselves as well.

The staff reports that major changes seen in the aides can be accounted
for by having steady, meaningful employment which has enabled them to
support themselves and to stabilize their lives. Marked personality change
has not occurred, but social adjustment has improved markedly. Both the
high- and low-risk groups performed well and, with the exception of one
boy who dropped out early owing to trouble with the police, there were
no essential differences. However, many of these people have histories of
difficulty that follow them. For instance, the one dropout would have
returned, but because he was in prison he fell behind in payments set by a
previous paternity suit and was put back in jail.

Despite the fact that the backgrounds of the aides were similar to those
of their clients, they seemed to see the problems clearly and to want to
help the kids get out of patterns that would lead to trouble. The staff
found that the aides could cope with many difficult situations and that,
with the supervision provided them, they could perform many functions.

70

In a program with a strong rehabilitative design like this, one has to
search out the needs of the aides and provide opportunities to fill these
needs during the period of training. Staff must be tolerant and capable.
The aides say that their neighborhood is slowly getting better. It is ob-
vious that it still needs improving. The mental health of anyone living in
such an area is inevitably impaired. The awareness that much needs to be
done has not always sharpened the understanding of how to do it. The
staff at Baker's Dozen and the Institute for Youth Studies ate finding a
way.

Research Grant- MH 14837

Dates of Interviews: Jan. 31 and Feb. 3,7,8, 10, and 17, 1967

References.

Denham, W., Felsenfeld, Naomi, and Walker, W. The neighborhood worker, a new
resource for community change. A monograph on training and utilization. Institute
for Youth Studies, Howard University, Washington, D.C., May 1966.

Klein, W,, Denham, W., MacLennan, Beryce, and Fishman, J. Training nonprofessiortal
workers for human services, A manual of organization and process. Institute for
Youth Studies, Howard University, Washington, D.C., May 1966.

Klein, W., Walker, W,, Levine, Myrna, MacLennan, Beryce, and Fishman, J. Leadership
in (he training of human service aides: First report on the counseling intern pro-
gram. Institute for Youth Studies, Howaid University, Washington, D.C., 1965-66.

Mitchell, L. Psychotherapy with the cultiually and economically deprived youth.
Paper read at annual convention of the Ameiican Psychological Association, New
York, 1966.

Mitchell, L. Training for community mental health aides as leaders of child and adoles-
cent therapeutic activity groups. Institute for Youth Studies, Ilowaid Univeisity,
Washington, D.C., May, 1966.

Mitchell, L. (Ed.) A dictionary of local terms and expressions. The Baker's Dozen
Community Mental Health Center for Adolescents; Institute for Youth Studies,
Howard University, Washington, D.C., 1967.

Mitchell, L., and Fishman, J. Mental health for the poor-the use of trained problem
youth in a neighborhood treatment program for children and adolescents. Paper
read at 122d annual meeting of the American Psychiatric Association, Atlantic
City, May 1966.

Training for new careers. President's Committee on Juvenile Delinquency and Youth
Crime. June 1965.

"What It Means To Be Really Poor." Bakei's Dozen Community Mental Health Center
aides and Mitchell, L. Mimeographed paper, 1966.

71

Investigator.

Metvin Seeman, Ph.D.
University of California
Los Angeles, Calif.

Prepared by

Antoinette Gattozzi

Although it is probably true that there never has been a human com-
munity without its critics, the case against modern industrial society may
be unique in history for the intellectual sophistication and emotional
appeal of its arguments. Distinguished scholars such as Arendt, Marcusc,
Fromnij and Mills have formulated the radical critique for our own time,
The critique has an illustrious history, moreover, with main roots in the
19th-century writings of Durkheim, Weber, and Marx. The works in this
genre now form a more or less cohesive body of literature known as mass
society theory, The mass society has been criticized in many particulars,
but the common judgment spanning the decades is that a mass society is,
on the whole, an unhealthy one. Indeedj mass society analysts have been
aptly called the pathologists of contemporary industrial societies.

Alienation is the dominant theme in mass society analysis. Alienation is
regarded as the psychological effect generated by the structural forms that
define a mass society and, in turn, it is considered to be the primary cause
of a multitude of personal ills and social evils. The pivotal role assigned to
the alienation concept can be made clear by outlining the central thesis of
the mass society literature. The thesis is essentially composed of three
elements-a historically oriented view of contemporary social forms, the
concept of alienation, and judgments about the quality of contemporary
life.

A mass society is defined by it? structural features, which are the forms
inherent in the historical developments of democratization, urbanization,
and industrialization. It is a society in which most major institutions are
designed to deal with people in the aggregate without distinguishing
among individuals or small groups of individuals. "Mass" refers not to
large size or huge numbers per se, but to the enlargement of the scale of
social forms. Thus national governments, nationwide corporations, trans-
portation and communication systems regularly make decisions that affect
whole populations, Mass culture, mass production, mass consumption sup-
plant the natural heterogeneity of the people, Pressures for technical
efficiency and rational control lead to the bureaucratization of organiza-
tions: community is lost. Geographical mobility displaces stability. Kin-
ship ties are attenuated: anonymity and impersonality come to character-
ize relations among people,

72

Mass society theorists have argued that alienation is fostered among the
members of a society with these structural features. The alienated man
suffers incalculable losses of many personal satisfactions as an individual
human being. He is, as a consequence, especially vulnerable to mental and
emotional disturbances. Further, a society in which there are large num-
bers of alienated men has little cohesiveness and organic integrity. A mass
society may be beset, for example, by such social evils as widespread
political passivity, ethnic and racial prejudices, and wildcat strikes of in
dustrial workers.

This capsule statement of the mass society theory does not do justice to
the persuasiveness of the formulations, but it does suggest the seriousness
of the charges made. There is no doubt that these ideas and assertions,
particularly the concept of alienation, have persisted because they are
intellectually and emotionally seductive to many thoughtful people. Yet
the literature of mass society theory tends to be discursive and rhetorical,
and the use of the crucial alienation concept rather more exhortatory than
analytical. As a result Prof, Melvin Seeman has noted, "the debate con-
cerning alienation has often remained sterile, however valid the critique of
society and however proper and humane the values involved." Professor
Seeman has been trying to clarify the terms of that debate and to derive
some specific propositions about alienation that would lend themselves to
systematic testing.

The task began in 1959 with the publication of a conceptual paper in
which Professor Seeman suggested that the various connotations of aliena-
tion could be distilled into five related but separate ideas.1 In defining
these ideas, he drew upon the social learning theory of Julian B. Rotter, a
theory that Professor Seeman thinks has much in common with the aliena-
tive notion of powerlessness and its consequences. These five varieties of
alienation, then, were defined in terms of an individual's expectations and
his values.

• Powerlessness. The person who experiences a sense of powerlessness
expects that forces outside himself control his personal and social rewards.
He has little expectancy that his own behavior can be efficacious in gain-
ing these rewards.

• Meaninglessness. The individual regards social affairs as incompre-
hensible. Therefore, he has little expectancy that he can predict the out-
come of social events.

• Normlessness, The individual believes that he is not bound by con-
ventional standards of conduct in the pursuit of his goals, Normlessness,
on the contrary, implies a high expectancy that socially unapproved
means must be used to achieve these goals.

• Value isolation. The individual rejects the values of society. He
assigns low value to the goals and behaviors that are highly valued by
most other members of his society.

1 Since that time, Professor Seeman has added a sixth form of alienation, social
isolation, which he defines as an individual's low expectancy for social acceptance, as
reflected, for example, in the feelings of loneliness and exclusion experienced by
members of minority groups.

73

a Self-estrangement. While this idea has taken a number of definitional
forms, the one in which it is perhaps most easily grasped is this: To be
self-estranged is to be engaged in activities that are not rewarding in them-
selves. This is self-estrangement in the classic Marxian sense of alienated
labor. In social learning terms, the self-estranged person is continuously
engaged in activities he does not value highly.

Professor Seeman has noted that the obverse of these alienative forms-
order and trust, consensus and commitment, integrity and engagement-
represent humanistic values that are highly esteemed by democratic soci-
eties. Thus, to the extent alienation is engendered in a society such as outs
in the United States, it exerts profoundly detrimental effects on individual
lives and threatens to make a mockery of our most cherished values, The
crucial questions, of course, are whether or not the social forms of our
modern industrial society really do spawn alienation among individual
Americans and, if so, whether or not alienation has the behavioral conse-
quences that mass society theory would predict.

It is the second question that has chiefly interested Professor Seeman.
In a series of empirical studies carried out over the last 10 years, lie has
looked at two forms of alienation, powerlessness and self-estrangement,
and has sought to discover whether an individual's level of alienation is
related to his behavior in certain circumscribed areas.

Powerlessness and Learning

As noted above, Professor Seeman utilized certain constructs from a
theory of social learning in his explication of alienation . The similarities
between this theory and the mass society view of the behavioral conse-
quences of an individual's powerlessness are quite striking.

The social learning theory formulated by J. B. Rotter holds that a
person's behavior is significantly influenced by two factors and by the
situation in which they occur or are embedded, It depends on his expect-
ancy that the behavior will lead to a successful outcome and on the value
he places on that outcome. Moreover, the theory distinguishes between
internal and external control of rewards. Dr. Rotter and his associates
have postulated that, in laboratory experiments in learning, subjects will
not do as well when they perceive their success as being dependent on
chance or luck or on the experimenter's control of the situation as when
they believe that their own skill can decide the outcome, (A number of
investigations have since shown that this is indeed true.) Similarly, mass
society analysts have contended that the individual living in a contemp-
orary industrial society, believing that he is powerless to determine the
successful outcome of his social behaviors, turns away from political activ-
ities; his sense of powerlessness makes him indifferent to and a poor
learner of political information and of other knowledge relevant to his
social functioning.

74

"Thus," Professor Seeman has commented, "the idea of powerlessness
extends downward, as it were, in its potential for reorganizing the rela-
tively 'microscopic' studies of laboratory learning . . . But the idea of
powerlessness also extends upward in its significance, being an integral
element in sociological descriptions of 'macroscopic' concerns: The occur-
rence of mass movements, the conditions of political democracy, and the
like."

The first of a series of controlled studies testing the relation between
powerlessness and learning was conducted by Professor Seeman and a
colleague, John W. Evans, among the patients in 10 tuberculosis sanitari-
ums in Ohio. Their hypothesis was that a patient's sense of powerlessness
influenced what he had learned about tubeiculosis. They predicted that
high poweilessness would be associated with poor learning.

A total of 86 white male subjects were selected from a much larger
sample and grouped into 43 pairs closely matched forage, income, educa-
tion, length of hospitalization, estimated discharge time and, most impor-
tant, hospital experience— each individual in a pair lived on the same ward
of a hospital and was exposed to the same routines and staff care. They
differed, of course, in their positions on the powerlessness scale.

The scale consisted of forced-choice items designed to reveal an indi-
vidual's expectancies for personal control.2 Most items referred to socio-
political expectancies. For example, the patients were asked to choose, on
the basis of personal belief, between these two statements: 'The average
citizen can have an influence on the way the government is run," and
"This world is run by the few people in power and there is not much the
little guy can do about it." The scores of the total sample were divided at
the mean of the distribution into "low" and "high" alienation. Each pair
of matched patients, then, included an "unalienated" and an "alienated"
person.

The investigators also needed a measure of each patient's knowledge
about tuberculosis. This they obtained by a standard true-false informa-
tion test based on one used by the National Tuberculosis Association.
Then, to get an idea of the extent to which a patient's objective knowl-
edge was manifested in his ward behavior, the staff was asked a number of
questions about the behavior of individual patients. Two questions were
pertinent: How good is this patient's understanding of his illness, and of
the disease generally? This gave the investigators a measure of what they
called reputed knowledge. Finally, they obtained from each patient a
measure of his subjective knowledge; that is, an indication of how satisfied
he felt with the knowledge about tuberculosis that he possessed. These
latter two measures were taken solely to help shed light on the re-
searcher's central inquiry, which concerned the postulated relationship
between an individual's expectancies for control and his objective knowl-
edge about an event— tuberculosis— that strongly affected his life career.

2 The alienation measure used in this and the other studies reported here was
adapted from the I— E Scale (internal-external control) developed by Professor J. B.
Rotter and the late Professor Shephard Liverant at the Ohio State University. The I~E
Scale has been extensively tested and refined and shows satisfactory reliability.

75

The results confirmed the prediction: Alienated patients scored lower
on the information test than did unalienated patients. The difference
between the two groups was small but statistically significant. The investi-
gators showed that the difference in knowledge about tuberculosis did not
reflect any consistent differences in intelligence among individuals in Uic
two groups.

Staff evaluation of a subject's knowledge, so-called reputed knowledge,
was in line with the main finding. Patients high in alienation were judged
by the staff to be less well informed about tuberculosis than palicrtls
relatively lower in alienation. No significant coirelation was found be-
tween subjective knowledge and alienation, but an inteiesting interaction
was discovered between ward stratification and alienation. The iclalivcly
more controlled environments (controlled by the physician in clungo)
drew fewer responses indicating dissatisfaction fiom the unalienated than
from the alienated patients. The researchers had predicted that the oppo-
site would be the case-that those who felt a greater mastery would resent
the tight control exerted in the highly stratified ward. They offered one
possible interpretation of this finding, A highly stratified environment is
not congenial to the transmission of knowledge and it could be, (liey
suggested, that on such wards the alienated patients actively sought m\
gained more knowledge than alienated patients were inclined to do, In a
less controlled ward setting, knowledge may be acquired more passively,
and thus be equally available to those willing to seek it (the unalienated)
and those for whom knowledge presumably has little value and, therefore,
is not worth any effort to attain (the alienated). This interpretation was
modestly substantiated when the investigators compared objective knowl-
edge scores of the unalienated and alienated groups on the two different
kinds of wards. There was less difference between the group scores if
patients resided on a low-stratification ward than if they lived on a highly
stratified ward.

The results of this study, then, provided a reasonably satisfactory
demonstration of an association between powerlessness and learning. At
least two questions were left open, however. Was the relatively poor learn-
ing shown by the more alienated patients the product of their greater
sense of powerlessness or did their powerlessness come from possessing
little knowledge? Second, information from only one domain of informa-
tion was tested-the control-relevant domain of tuberculosis in forma (ion;
would the more alienated patients have made an equally poor showing in
any other area of knowledge, which might suggest a general withdrawn! of
their interest in learning? The design of Professor Seeman's next study
provided ways to look into these questions.

The next study was carried out among the young male inmates of an
Ohio reformatory. The choice of setting was particularly apt, "It is pos-
sible to conceive of the reformatory and its associated training apparatus
as a vast learning mechanism," Professor Seeman pointed out, "but one to
which the essential features of powerlessness dominate institutional life-
where, for example, paroles are denied and inmates left in ignorance of

76

the reasons, and where the inmate culture is a response to the more or less
total threat to personal control."

The alienation scale used in this study was like the one used with the
tuberculosis patients. Once again the scores were divided at the mean of
distribution so that the subjects could be described as high or low in
alienation. The tests of learning weie quite different. This time Professor
Seeman tested for new learning rather than measure prior knowledge. To
do so, the group of 85 men (none having less than an IQ of 100 or a
ninth-grade education) was presented with 24 items of information about
correction. A third of these items dealt with the reformatory , a third were
concerned with parole matters, and a third dealt with long-iange oppor-
tunities. The items were compiled from documents not readily available to
the men and presumably represented new information.

The men were presented with this material twice in one session. In
order to insure that each item would be read, the investigator first asked
them to mark off how interested they were in each item, moments after
this task was completed, the men were given the items arranged in a
multiple-choice-test format which they were asked to complete. The latter
constituted their learning scores, for it tested how much of the material
they had retained.

The investigator found that men low in alienation achieved significantly
better recall of the parole items than did men high in alienation. There
was no statistically significant difference between the alienation groups in
the recall of the two other kinds of information.

Professor Seeman also made an estimate of the value each man placed
on the conventional norms of the reformatory by compiling the number
of merit commendations each had earned from prison authorities. Dif-
ferences in powerlessness were found to be unrelated to learning among
inmates who had earned no merits, but high or low powerlessness made a
significant difference in learning among men who had earned one or more
merits. The highest correlation between alienation and the learning of
parole items appeared when low powerlessness was accompanied by a
degree of commitment to the values of rehabilitation (as symbolized by
the earning of merits). Thus, as Rotter's social learning theory would
suggest, the combination of high expectation for personal control and
high evaluation of the goals in question was most revealing of the associa-
tion between alienation and learning.

Merit commendations were an indication of the men's behavior in the
reformatory; Professor Seeman also obtained some indication of their
behavior outside the reformatory, prior to confinement. Although age and
IQ were found to be unrelated to learning scores, continuation beyond the
ninth grade and achievement relative to capacity were relevant. A man's
willingness to stay in school and to achieve in accordance with his capac-
ity correlated well with both low powerlessness and good learning of
parole material. "The most interesting feature of these results," Professor
Seeman commented, "lies in the fact that the inmate's learning of
corrections-relevant material (the parole information) is related not only

77

to his generalized expectancies for control but to his behavior-both out-
side the prison and inside it- which presumably reflects such expectan-
cies."

This investigation went a long way towards claiifying the relations be-
tween behavior and alienation in the powerlessness sense of low expect-
ancy for control. It yielded a clear demonstration of an association be-
tween learning and alienation. More, the findings suggested some limits of
this association-it appeals only in the learning of control-relevant
information-and revealed the fact that some behavioral concomitants of
alienation may be found outside the sphere of learning. These two impor-
tant additions to the body of evidence about alienation were explored in
depth in the third study of this series. The study was made in Sweden;
Professor Seeman carried out a number of related investigations in
Sweden, which gave him an opportunity to examine the cross-cultural
validity of alienation effects on learning and other behaviors. The study to
be described next was conducted among some 300 students .it Lund
University; the other studies wil! be discussed below in different contexts.

The basic design was similar to that of the other investigations. Stu-
dents were judged to be high or low in alienation on the basis of tlicir
scores on the powerlessness scale, and each student's knowledge in one or
the other of two domains of information was tested. One was the domain
of cultural knowledge and the otliei was that of nuclear weapons (for ii
small subsample, political information was also tested). All tests weic
equal ia difficulty. Professor Seeman predicted that high alienation
would be associated with poor scores on the nuclear (or political) test, but
alienation would be irrelevant to scores on the cultural test. In other
words, he hypothesized that alienation would affect learning differential-
ly, as was suggested by the results of the reformatory study, and that the
acquisition of control -relevant information is the specific learning most
affected by one's relative sense of mastery or powerlessness.

Another alienation proposition was tested in this work. Prior to the
initiation of the study, each student had completed a short veision of tlie
alienation scale, and these scores were in hand when the study proper got
under way. For the study proper, then, most students received a cultural
or a nuclear test and a long version of the alienation scale by mail, along
with a covering letter and return envelope. Professor Seeman was inter-
ested in those who delayed returning the tests or failed to do so alto-
gether, that is, those who showed avoidance behavior. He predicted that
high alienation would be associated with late or nonreturns among the
group that got the nuclear test but that no such association would appear
among those getting the cultural test. This, he suggested, would allow for
a "mierodemonstration" of the proposition that powerlessness leads to
avoidance behavior™ "micro-" because what was involved was "the small
world of everyday tasks, like responding to an inquiry from the univer-
sity."

The results concerning alienation and control -relevant learning con-
firmed the prediction: High alienation, especially for women, correlated

78

with poor knowledge of nuclear weapons or political affairs, but alien-
ation, high or low, showed no associations with cultural knowledge, poor
or good.

The data did not leveal so clearcut an answer to the question about
avoidance behavior and alienation. The results tended to confirm the
prediction— that is, alienated students were slower to return and more
often failed to return the packet of tests if they had gotten a nuclear test
than if they had received a cultural test— but did not reach statistical
significance. There was, in fact, an astonishingly high rate of return (85
percent) before followup letters were sent out, which might be attribut-
able to the fact that in Sweden theie are strong, though informal,
pressures to cooperate in social research.

This study showed that the proposition concerning the effects of
powerlessness on differential learning holds in the domain of sociopolitical
information and that it is valid for college-age students of another culture.
Taken together, the three studies of alienation and powerlessness convinc-
ingly demonstrated the applicability of this proposition to a range of
information domains and to different populations of people. In so doing,
Professor Seeman's work greatly strengthened the empirical basis of the
proposition's credibility and its usefulness in other social research. For
example, its applicability to a sensitive and vital area of American life, the
education of young children, was recently demonstrated by James S. Cole-
man and co-authors in their report, Equality of Educational Opportun-
ity,* Professor Seeman has cited the relevant passage: "* * * a pupil
factor which appears to have a stronger relationship to achievement than
do all the 'school' factors taken together is the extent to which an individ-
ual feels that he has some control over his destiny." To the degree the
proposition linking high alienation and poor learning is valid in other areas
as well, it bears important implications for those trying to increase public
understanding of international political issues, say, or those responsible
for public health information campaigns about alcoholism, for instance, or
the value of prenatal care.

The Role of Organizational Ties

Mass society theorists have made an important recommendation in their
writings. They have repeatedly asserted that organizational ties must be
established by the individual in order for him to have an effective media-
tor vis-a-vis the mass-scale institutions that surround him. This theme is
sounded throughout the mass society literature, from Durkheiin to Mills;
although Mills, in The Power Elite, expressed doubts about the efficacy of
mediating organizations in our time because they, too, as he saw them,
had begun to assume mass-scale qualities.

'U.S. Government Printing Office. Washington, D.C. 1966.

79

Perhaps the most familiar example of a mediating organizatioi
labor union, and mass society theory would predict that union mci
ship mitigates a worker's sense of alienation. To test this, Professor
man and a colleague, Arthui G. Neal, formulated a concrete hypot
that could be examined in an empirical manner. The hypothesis:
expected members of a work-based formal organization to exhibit
powerlessness than individuals without an organization to speak for t
in the crucial area of occupation,"

The setting for the study designed to examine this hypothesis
Columbus, Ohio, where the investigators assembled a random samph
adult male subjects from the city directory. Mail questionnaires were u
to gather necessary information about each man's union membership i
his sense of powerlessness. In addition, each was asked to answer qi
tions about his experience of occupational mobility and to complex
scale designed to determine his attitudes towards mobility. These lal
measures were needed, the researchers noted, because several studies 1;
shown mobility and mastery (the obverse of powerlessness) to tie relate
Finally, subjects also received the well-known anomie scale, developed i
Leo Srole, which can be interpreted as measuring generalized despair, T!
was included because the investigators wanted to find out if nonmciubt
ship was associated with other forms of alienation in addition to powc
lessness.

Through the use of personal interviews done after the returns were if
the investigators checked the possibility that more alienated than uiutHcr
ated men would have delayed or not bothered at all to comply with tli
mailed requests. They found this was not the case: Neither late return no
nonreturn correlated with high alienation. The investigators received re
plies from slightly more than 600 men (57 percent of the original sample),
The occupational statuses of the respondents ranged from high-level exec-
utives to unemployed manual workers. Analyses of the data yielded (he
following results.

First, the basic prediction was confirmed. Compared to the group of
organized workers, the group of unorganized workers felt themselves to be
more powerless. This finding held when the variables of occupational
status and income were controlled. When the mobility variables were
taken into account, on the other hand, the investigators did find some
interesting interactions.

When the groups of organized and unorganized men were separated into
manual and nonmanual workers, the investigators found the same strong
association between high powerlessness and nonmembership among (he
manual workers, regardless of mobility history or attitude. The picture
was different among non-manual workers, For this group, a mobility atti-
tude of nonstriving reversed the effect. Thus, among the mobility-oriented
white-collar workers, the unorganized were higher in powerlessness titan
the organized (as expected), but it was the organized workers who were
higher in powerlessness than the unorganized in the group of nonstriving
white-collar workers. These data enabled the researchers to add a useful

80

lefinement to the thesis about organizational ties: It is particularly appli-
cable to woikers committed to mobility.

Second, no consistent relation emerged between membership or non-
membership and anomie, although unorganized workers tended to be only
slightly higher in anomie than organized workers. By implication, anomie
and powerlessness were not found to be strongly and systematically re-
lated either. This negative evidence was welcome on two counts. First, it
supported the investigators' assumptions that their hypothesis was a rela-
tively specific one-i.e., membership or nonmembership was linked to the
powerlessness form of alienation— and second, it was suggested that pow-
erlessness, as they defined and measured it, was a satisfactorily specific
form of alienation that excluded more generalized feelings of hopeless-
ness,

Although low powerlessness was clearly shown to be associated with
organizational membership, the fact by itself does not support the mass
society thesis regarding mediating organizations. One may wonder, as the
investigators did, whether membership leads to low alienation or vice
versa. The researchers were able to make a start toward the answer by
obtaining data on the union situation (open or closed shop, maintenance-
of-membership contracts, and the like) in the firms where their respond-
ents worked. While conceding that their findings had to be regarded as
only tentative, they concluded that both options were valid. In other
words, the mass society argument about membership ameliorating the
alienative effects of social structures was not inconsistent with the evi-
dence; alternately, the data also provided support for the interpretation
that workers low in alienation may be motivated, by virtue of their expec-
tancy for personal control, to join organizations which are vehicles for the
exercise of control in the work sphere.

The proposition concerning the ameliorative effects of organizational
tics was directly tested in one other population. This was a large random
sample, some 550 men, of the work force of Malmo, Sweden. Professor
Seeman worked intensively with this group; in addition to the data re-
quired for the mediation thesis, he gathered many other kinds of informa-
tion pertaining to alienation and its putative effects on behavior and atti-
tudes, Much of what he discovered will be described below in the section
on alienated labor. Two aspects of this omnibus Swedish work are relevant
at this point of the report, the evidence concerning the mediation thesis
and the data demonstrating anew, in the context of the mediation thesis,
the relation between powerlessness and learning.

In Malmo', as in Columbus, high powerlessness and poor knowledge of
political affairs were found to be associated in the group of unorganized
workers, manual and nonmanual; there was no such association to be seen
among the organized workers. The same pattern held when such pertinent
variables as education, income, and occupational prestige were controlled.
Moreover, there was a modest connection between degree of involvement
and both powerlessness and learning: Workers who were little involved in
their organizations scored higher in powerlessness and lower in political
knowledge than did those who were more involved.

81

Thus, in two dispaiate cultural settings, men who were members of a
work organization experienced less personal powerlessness In the socio-
political arena than did their unorganized fellow workers. The evidence
gathered in the two studies was consistent with the mass society thesis
assigning a meliorative role to such oiganizations. More investigations arc
needed, however, to determine how much a worker's sense of powerless-
ness is minimized by his membership in a woik organization as opposed to
how much his sense of his ability to influence conditions that affect him
leads him to join the organization in the first place.

Alienation in Work

The world of work occupies a prominent place in mass society theory,
which devotes much attention to alienated labor. An alienated worker is
defined as one who does work that is not intrinsically rewarding to liim.
He does it because he feels he must, not because he gains personal satisfac-
tion in the activity. To engage in such work regularly, then, is to experi-
ence a variety of self-estrangement. Mass society and Marxian theorists
have contended that the existence of an alienated labor force is itself
responsible for a plethora of social problems. The deleterious effects arc
seen to be so pervasive and profound, in fact, that Marx and others have
argued that many of the social ills of modern industrial societies could be
made to disappear if only the alienation of laboi were ended.

As expressed in the mass society literature, the consequences of alien-
ated labor include attitudinal effects such as powerlessness and norniless-
ness as well as behavioral effects such as minimal political participation,
racial and ethnic hostility, and the substitution of extrinsic goals (job
status, for example) for unattainable intrinsic satisfactions. Professor See-
man has characterized this theme of the mass society literatuie as tlic
generalization thesis. He has begun to test its validity in separate investiga-
tions carried out in Sweden, France, and the United States. The Swedish
study has been completed and it can be reported in full.

A random sample of the male work force of Malmti constituted tlie
study population. Data were obtained through personal interviews witli
the workers. The basic measure-whether or not a worker felt alienated
from his work-was obtained by reference to a work alienation index,
which was developed by factor analysis of the responses to pertinent
questions. The scale finally developed by this method was composed of
seven items, for example, "Is your job too simple to bring out your best
abilities, or not?" The investigator also obtained measures of powerless-
ness, ethnic and racical prejudice, anomie, political knowledge, orienta-
tion to experts, and mobility orientation. According to the mass society
generalization thesis, a man who is alienated from his work feels power-
less, hostile to "others," and anomie; he is rather ignorant of political
affairs, tending to leave them in the hands of experts, and strives for the
extrinsic rewards that status confers.

82

Reasonable though this thesis may sound, the analyses of the data
yielded very little support for it. Work alienation was not found to be
related to any of the theoretical outcomes in a statistically significant
manner, although some of the outcomes did relate to one anothei-for
example, high powerlessness was associated with scanty political knowl-
edge and high expert orientation and with racial hostility. Even when
variables such as age, income, occupation, and the like were controlled, no
clear-cut associations with work alienation emerged. Further, when con-
sidering solely the question of how much control a worker felt he excited
in his work process-supposedly the quintessential condition of alienated
labor— the associations to the postulated effects were similarly minimal.
The conclusion to be drawn from these results, then, was that the attitudi-
nal and behavioral consequences repeatedly attiibuted to alienation in
work simply do not reflect the reality as it exists in Sweden.

An obvious question that arises is whether Sweden, after all, should not
be considered an exceptional case because of its homogeneous population,
its long history of peace and of social and economic stability. To answer
this question, Professor Seeman undertook to perform essentially similai
studies of alienation in work in France and the United States. These, too,
are highly industrialized societies, but their political and social histories
differ markedly from Sweden's and are different from one another as well.

Results from these two studies are not yet available. Preliminary analy-
ses suggest, however, that neither the French noi the American situation
differs from the Swedish in terms of the consequences of alienated labor.
On the other hand, it appears that the relation between high powerlessness
and low political knowledge will be found in both the French and Ameri-
can populations, as will additional validation of the mediation thesis. (The
study in Los Angeles should also help illuminate the uniquely American
racial situation, In addition to interviewing some 500 adult male white
subjects, the investigator assembled a subsample of some 270 Negro
workers. It should be instructive to see how the data collected from the
white workers compare to those gathered in the Negro subgioup.) When
the results from all three countries are available, the generalization thesis
of mass society theory will have received a substantial test.

Even if alienated labor does not produce the adveise personal and social
outcomes that have been imputed to it, it remains a source of concern.
"For one thing," Professor Seeman has noted, "work life absorbs a major
portion of the day, and an ethical stance concerning it must come to
terms with that fact, regardless of any other consequences of alienated
labor. For another, it is reasonable to argue either that the outcomes we
have treated constitute an insufficient list (the data say little about the
quality of family life, for example) or that the consequences will reveal
themselves in a longer-term, cumulative way-for example, in revolutions
or in the irregular outbursts of a wildcat strike."

Nor can it be assumed that men alienated from their daily work are
indifferent to the fact. The sort of data required to examine the general-
ization thesis did not touch on such matters. It cannot be said, therefore,
whether or not the worker detests his work and is angry at himself for

83

doing it. All that the data do allow us to say, Professor Seenian has
pointed out, is that "people can work out fairly effective adjustments to
varied kinds of work, if by 'effective' we simply mean leading a work life
that has little generalized effect on the standard forms of hating, striving,
withdrawing, and complaining reviewed here." Thus, although the
predictions implicit in the mass society generalization thesis may
ultimately prove to be invalid, the ideological indictment of industrial
forms that breed alienation in work is a sepaiate issue and one still most
worthy of attention.

The concept of alienation bears on some of the most urgent problems
confronting our society. Yet the resounding rhetoric in which discussions
of alienation are often couched makes it too easy for many people lo
dismiss the ideas along with the words as baseless pessimism. Professor
Seeman's concern as a scientist has been to help turn "the parable of
alienation into a proposition," or rather a series of propositions tlint can
be accepted, rejected, or amended on the basis of deductions drawn from
evidence methodically gathered in controlled b'ut real situations, As a
result of his work, we begin to see that the "pessimism" of the mass
society radical critique is not, in fact, groundless.

Much more research on problems involving alienation needs to be
carried out before we know just how valid the critique is and, more
constructively, how much insight into man's social behaviors can be
gained through explorations of the alienation concept. Professor Seeman
recently outlined a general program of research to be done in this area and
enumerated several studies he thought would be useful. In that paper,
which will be a chapter in a book to be published by the Russell Sage
Foundation, the investigator also described the unique challenges Hint
such research makes on social scientists.

"The fact remains," Professor Seeman concluded, "that there were con-
centration camps and genocide, that there are now widely disrespected
qualities in American life-including a widely disrespected war, jit the
moment-and that there will be further urban violence born of frustrated
hopes. That kind of past, present, and future create a special tension for
the sociologist interested in the problem of alienation. It is the tension
between keeping a craft that is worthy of the name, and at the same time
making sociological investigation practically and morally relevant. The
danger on the craft side is that the work deteriorates into a kind of
alienation in itself-bound by technical rules, limited in vision, devoid oi
personal involvement, and largely oriented to careers. The danger on the
side of relevance is that this deteriorates, too-into a subtle anli-
intellectualism that is impatient with any thing but the immediate; or into
a kind of self-indulgence that emphasizes stance over analysis, so that
what becomes crucial is one's identification (as radical, as realist, as
humanist, or whatever). My hope is that the secularization of work on
alienation can continue to be achieved while avoiding both these
dangers-whieh is to say that clarity and demonstration can be success-
fully wedded to scope and human concern."

84

Research Grant: MH 10460
Date of Interview. October 1968

References,

Neal, A. G.; and Seeman, S. Organizations and powertessness: A test of the mediation

hypothesis. American Sociological Review, 29 216-226, 1964.
Seeman, M. On the meaning of alienation. American Sociological Review,
24:783-791,1959.
Alienation and social learning in a reformatory. American Journal of Sociology,

69:270-284, 1963.
Alienation, membership, and political knowledge: A comparative study. Public

Opinion Quarterly, 30:353-367, 1966
Status and identity: The problem of inauthenticity. Pacific Sociological Review,

9:67-73, 1966.
Powerlessness and knowledge: A comparative study of alienation and learning.

Sodometry, 30:105-123; 1967.
On the personal consequences of alienation in work. American Sociological

Review, 32:273-285, 1967.

Alienation and engagement. Chapter for a forthcoming publication to be

published under the auspices of the Russell Sage Foundation (edited by A.
Campbell and P. E. Conveise).
Seeman, M,; and Evans, J. W. Alienation and learning in a hospital setting. American

Sociological Review, 27:772-782, 1962.

Note. -In addition to NIMH grant support, the investigator's research has been
aided by grants from the University of California and The Swedish Social Science
Research Council,

85

Investigator:

Saul Bernstein, Ph.D.
Boston University
Boston, Mass.

Prepared by

Clarissa Wittenberg

The unruly issues of riots, slums, and racial problems are now of major
importance. Our state of information is extremely limited, and major
decisions are often made on the basis of scant knowledge. Saul Bernstein,
a professor of social group work at Boston University, has completed two
extensive interview studies to explore these and related issues. His focus
has been on our most alienated young people, those who are minority
group members and who are living in slums. Initially interested in delin-
quency and its contemporary forms, he then turned to the impact of the
explosive events of the 1 960's.

Twice in the mid-1 960's Professor Bernstein traveled to nine major
American cities to talk to people working closely with teenagers in the
ghettos, as well as with some of the youngsters themselves. In the first
study, completed in 1963, l he found that many of the young people In
this country live in terrible housing in slums, are blocked educationally,
cannot get jobs, are undermined by family problems, and are caught up in
destructive cycles with new babies being born into new one-parcnl
families. Many of these young people are bitter and intensely cynical, and
feel hostile and destructive towards this country. Many engage in delin-
quent acts and are in trouble with the police. These young people do not
feel encouraged by the new legislative landmarks which are designed Eo
secure their rights. If anything, many felt more impatient and intolerant
of existing inequities than ever before. Many felt that this country had let
them down. Mr. Bernstein concluded that many agencies working in the
slums were doing good work but against great obstacles, and he especially
singles out the street workers as of great importance in reaching these
alienated young people.

In'the mid-1960's this country was torn by riots and stunned particular-
ly by the Watts riot. Many people were puzzled and shocked because
progress had been made in civil rights legislation and the antipoverty
program had begun. The civil rights movement at that time was also strong
and tasting the fruits of success. It was at this time that the second study
was planned to determine if any of the socially approved forces had
touched the alienated youth with the same force as had the riots, An

'Supported by grants from the Duncan Russell Memorial Delinquency Committee
of the United Community Services of Metropolitan Boston and the Permanent Charity
Fund of Boston.

86

exploration of the role of these young people in the riots was also
planned.

Dining both studies the investigator visited Boston, Chicago, Cleveland,
Detioit, Los Angeles, New York, Philadelphia, and Washington, D.C. The
second study done in 1965—66 substituted Rochestci, N.Y., for San
Francisco. At that time Rochester had had riots and San Francisco had
not. The design in both studies called for the location of major social
agencies working with young people in the slums, and to inteiview staff in
each of these agencies. In addition to interviews with staff, the young
people themselves were intei viewed whenever possible. In some cases,
staff membcis were themselves indigenous workers and very representative
of the ghetto population. In both studies, the cities were selected because
they had large numbeis of hostile, alienated young people and because
agencies in these cities had long expeiicnce in work with young people.
The focus was on the poor, including members of minority groups;
Negroes, Mexican-Americans, Puerto Ricans, and poor whites.

In the study done in 1965—66, the investigator visited nine cities. He
held inteiviews with 289 people. Most of those interviewed wcie staff
members at various levels of responsibility in agencies active in slum areas.
In most cities there weie interviews with officials of the Human Relations
Commission or its local equivalent. Police, particularly those dealing with
juveniles, weie included. Educators were seen. Staff members of various
poverty piograms and of community planning councils were included.
Repiesentatives of such Federal agencies as the Office of Juvenile Delin-
quency and Youth Development, and various research groups, were inter-
viewed. Experimental agencies, such as Mobilization for Youth and
Haryou-Act in New York, and the Cleveland Community Action for
Youth, were included, as were new training programs such as Manpower in
Rochester. Limited time of the second study and the inaccessibility of the
young people themselves prevented more than a small number of inter-
views with really alienated young people themselves. A number of
representatives of militant groups, such as Saul Alinsky's Industrial Areas
Foundation, were interviewed. AH interviews took place between
December 1 965 and June 1 966. In many cases group interviews were held.

This technique resulted in a broad and detailed series of observations of
experts and young people about some of the major events of the
mid-1 960's, While events have been occurring at rapid speed in this area,
this study provides a valuable cross section of opinion at that time. The
investigator acknowledges that the research is limited by several factors.
Obviously, not all alienated youth come into contact with such agencies
and the experts interviewed may have some biases. The study is treated as
documentary rather than as "hard" research.

The major question of the study was to find if we as a society had
found any way other than riots to touch the alienated young person and
help him change his status, There is also great concern about the role of
the ghetto youth in these massive riots. Although many things have oc-
curred to correct injustices and open up opportunities, it is still question-
able if they have made sufficient impact. The study was designed to

87

increase our information about how these young people feel about ric
about legislative landmarks, protest marches, desegregated schoc
"black" schools, etc.

The first study (in 1963) showed that there are many people caught
our slums in very destructive patterns. Some changes in patterns of del
quent gangs have occurred. The highly structured large gangs appear
have diminished, although not to have disappeared. Smaller grou
referred to as clusters tend to piedominate now. Although it may
reassuring to see the disappearance of the highly visible signs of gitii]
such as matching leather jackets or gang giaffiti on walls, this may be cl
to the accumulated pressure brought to bear on visible gangs, rather (h
to their lack of strength. The gangs may have "learned the game
Although rumbles on a large scale have decreased, "snagging"
"Japping" was commonly mentioned. In this form of fighting, a looko
watches the regular movements of one or two members of an antagonist
gang. When it is established that a suitable place will be passed by t!
victim at a predictable time, about six of the aggressors go homo, was
shave, put on good suits, and then saunter towards that spot. They a
careful to go in pairs and not to show in any way by their behavior what
planned. When the victim comes along, they give him a bad beating an
then go casually home to change clothes. In some places the ritual \v«
different so as not to risk their good clothes. Clothing is very impoiUnl i
the gang psychology. Another form of gang fighting mentioned is the "fa
one." Heie a repiesentative of each group fights without weapons t
assistance. This, however, calls for a high level of discipline which is rare!
achieved. This is relatively infrequent as the fighters are in a very Icnst
public, and vulneiable position as well.

Most of the aggressive incidents occur between groups with the sail)
ethnic or racial background, although in some places even the nios
confined ghetto population is quite mobile and there are some incident
where groups have traveled to other areas for conflict.

The study revealed that although these bitter, hostile young pcoplt
would seem ripe recruits for militant 01 protest groups that relatively few
had joined. The Black Muslims, for example, had little to offer in terms oJ
meeting concrete needs, such as jobs. The NAACP methods often seen
too slow and removed to be attractive to these young people. However,
some young people, generally the less hostile ones, have joined NAACP. Id
Boston it was reported that a small number of white gang members were
recruited by the American Nazi Party.

The first study also examined the role of the "street worker." Agencies
in slum areas have found that "detached" workers can move into the
environment of the young people and reach them in a way no office-
bound worker could do. They provide a link between agencies and some
young people who have lost contact with all legitimate agencies or social
groups. Street workers vary; some have master's degrees, some have not
yet finished high school. Most have undergraduate degrees. Some have
delinquent or prison backgrounds and some are from very stable middle-
class backgrounds. As a group they tend to be very active and nol

attracted by a 9 to 5 life. Most are deeply involved in helping people find
themselves and achieve some success. They tend to be not well accepted
into any one professional group. They are "lonely" professionals.
Agencies have a hard time retaining street workers, and the demands of
the job are very grueling, so most workers tend to be in their twenties or
thirties. Job opportunity tends to be limited in terms of opportunity for
advancement and increased income. These issues are important as more
and more innovative agencies designed to serve the ghetto population
depend heavily upon the services of the street worker. Social work is
considered the nearest profession. Mr. Bernstein comments that street
work, which has been estimated to cost about $200 to $600 annually foi
each youngster, is little enough to pay for work which genuinely reaches
them and often their families.

After examining patterns and services and the state of some veiy

alienated young people in the first study, the investigator turned to what
was happening and what alternatives theie might be to the violence oc-
cuning in our cities. An early task was the division of the various riots and
incidents into broad categories. In the mid-1 960's riots occurred that were
the result of spontaneous events, e.g., the riot in Boston following a dance
which excluded large numbers of young people who could not be ac-
commodated. Other riots were between racial or ethnic groups, such as
the incident sometimes called Watts II on March 15, 1966, between
Mexican-Americans and Negroes. However, violence within the ghetto
itself was the most serious and disturbing type. At that time Negroes were
primarily involved, although some episodes were thought to have had their
beginnings in the tensions of Puerto Ricans or Mexican-Americans.
Despite much "get whitey" talk and the destruction of white-owned
businesses, the aggression at the time of the study was largely confined to
slum areas. White men were injured primarily as they came into the riot
area in an official capacity, such as happened with police or firemen.
Threats to burn and loot white areas did not materialize.

To realize the magnitude of the riots of the mid-1 960's a review of the
McCone Commission figures is valuable. Although Watts was the most
memorable riot of that period, many others occurred. Many, many people

McCone Commission Report summary of riots occurring in 1964:

City

Date

Killed

Injured

Arrests

Stores
Damaged

New York

July 1 8-23

1

144

519

541

July 24-25

4

350

976

204

Jersey City

Aug. 2-4
Aug. 11-13

0
0

46
8

52
65

71
20

Aug. 11-13

o

6

18

17

Aug. 16-17

0

57

80

2

Dixmoor (Phila.) . ,

Aug. 28-30

0

341

774

225

89

were injured, many arrested, and some killed. To make the figures imm
dramatic, it must be remembered that only a small fraction of Iliok
involved were arrested or noted in official statistics. It is speculated, too
that many more who never took part still tacitly supported the rioting.

All of these occurred between July 18 and August 30, a 6-weck purm?
Five dead, a total of 952 injured, 2,484 arested, and 1,080 slorr-
damaged. Despite all this, the riot in Watts overshadowed the icst. Ir
began on August 11, 1965, a Wednesday, and ended the following
Tuesday, August 17. It was estimated that at times as many as tcr
thousand Negroes participated. This, however, was still only 2 percent of
the population living in the riot aiea. The size of the riot area was ar.
incredible 46.5 squaie miles. The dead totaled 34, the injured 1,032
arrests 3,952, and damage was estimated at about $40 million. Only ois
public building was destroyed and only 14 damaged or burned. Tlwiun
interesting issue, as these buildings, while not attractive to looters, dv
symbolize the dominant "establishment." Although there is always
present the explanation of riots being a spontaneous mass protest wlicw
people get out of control while attempting to express a protest agaimi
degrading living conditions, this explanation fails to account for (lie con-
trol exhibited in some areas and the relative immunity given, for instance,
to schools, normally a target for vandals. Rioters in Watts were Jilso ob-
served stopping at traffic lights and driving with caution. Examples were
given during the study interviews of looters apologizing when bumping
into each other. There was considerable evidence as well that riots nmylio
provoked or related to specific incidents or situations rather than emerg-
ing spontaneously from the blue.

Some situations were cited as definite causes of riots. For example,
Garfield Park in Chicago, a predominantly Negro area, had a fire station
with an all-white staff until after the riot when it was integrated. The
California vote defeating Proposition 14, an open-housing ordinance, was
considered an initant in Watts. The inadequate public transportation in
Los Angeles was also cited. A man looking for work in Los Angeles who
lived in Watts might have to spend several hotus on a bus and pay it lound
trip cost of almost a dollar. Budd Schulberg, who later established a
Writer's Workshop in Watts, told of seeing a group on a street in Walts. .A
6-month-old baby had died. The mother's grief was intensified by (lie
bitter knowledge that the prompt arrival of an ambulance and a hospitJf
closer than the County General might have saved her child. In April of
1 966 there was still no public hospital in Watts, and Los Angeles voleis
later rejected a bond issue to construct one.

Others interviewed told of comments by law enforcement officers that
were irritants, Particularly cited were "abrasive" remarks by Los Angeles
Police Commissioner Parker. A comment he had made in 1958, that
Negroes committed 1 1 times the major crimes as other races, was siiii
remembered with bitterness by Negroes in that area.

The rigidity of the Boston School Committee about de facto segrega-
tion was considered an outrage. Even the withholding of State funds for
education, because of the unwillingness of the Boston School Committee

90

to prepare and put into effect adequate measuies for the desegregation of
public schools, did not produce significant change in 1966. Elections for
posts on the School Committee, which produced large votes for the most
intransigent members and defeat for the ones in favor of desegregation,
added to the affront to the Negroes.

Heat, "the long hot summer," adds to the tensions and problems of
crowded living and may bring a state of irritability that is explosive.

Accidents can trigger incidents. A fire truck on an emeigency call hit a
Negro woman and killed her. Since the fire station involved was all-white
and in a Negro area and already the source of tension, the rumor spread
that the killing was on purpose.

The presence or absence of social controls is an important element. In
Philadelphia, at the "Wall" of the Girard College, a crisis was described
which was headed off on at least one occasion by the strong activity of
the Human Relations Commission which marshalled clergymen, probation
officeis, police, street workers, and others to help keep the gang members
who were gathering under control, Other types of social control are
punitive and may work temporarily but appear to be self-defeating.

For instance the mayor of a riot-torn area in Illinois said he would meet
with Negro ministers but not with "violators or demonstrators." Participa-
tion in riots was justification for being sent eviction notices if participants
Jived in public housing. In other places parents in public housing were very
concerned about involvement of their children in any kind of a protest for
fear it would lead to the family being evicted. Officials of some cities had
considered the legality of cutting people off public welfare or un-
employment compensation. The assumption is that those who are arrested
are guilty and were more active in rioting than those who were not
arrested, which may not necessarily be true. These measures, such as
eviction, are punishment beyond that established for criminal acts and
make life worse for those affected. The accessibility of public officials to
those in the ghettos is very important and can reduce the impulse to riot.

Role of the Alienated Young Person in the Riots

Those interviewed in this study described considerable activity by the
young people in the ghettos. They looted, burned, and fought the police.
Some were members of gangs, although a more frequent pattern was for
small cliques to riot together. These were units of three to five youngsters.
The older teenagers rather than the younger, and boys rather than girls,
predominated. It was the general finding that these young people </W not
plan the riots. Many of the young people who took part were in their
twenties or thirties. Strong feelings were expressed by ghetto youth
against the middle-class Negroes who attempted to stop the riots. One
group of teenagers in Rochester articulated this antagonism, claiming that
middle-class Negroes thought that the white reaction to the riot would
hurt them. These same young people were bitter that many better-off
Negros moved out of the bad areas and then abandoned all responsibility

91

for them. The investigator notes that, despite the concern for the ghc
problems shown by middle-class Negroes interviewed in this study, then
great resentment towards them by those still in the ghetto.

A wide range of feelings was expressed about the riots. Some you
people admitted that they enjoyed them. Others said they were dangero
and foolish. Some felt that they are essential to attract attention to t
ghettos. One comment was, "It is better to spill blood in Watts than
Viet Nam." Theie is little doubt that the liotsgive people in the ghetto
sense of community and power. In Los Angeles it was said that particlp
tion in the riots became a status symbol, and many felt good to have bet
a part of it. Many spoke of the dangers of being shot, hurt, or arrcstd
Young people in Boston spoke of envying Watts and yet did not speak <
riots in their own area. Detroit and Washington seemed calmer than othi
cities, yet both had riots later. At the time of the study, the riots scemc
very remote and unrelated to the Pueito Ricans and Mexican-American
interviewed.

There is tremendous complexity in the reactions to riots. The impact o
the experience itself and its realities, such as food shortage, jail, etc.
bring about additional factors that become added to those present prior U
the riots. The mass news media and many civic leaders add interpretation1
and give reasons, and these all become incorporated into the discussion
The reasons given after a riot may be quite different from the motivations
of the riot period itself. Still, an overwhelming certainty was that for tltc
people in this study the reality of the unbearable living conditions and (lie
profound feeling of helplessness was a major factor in every riot.

Agency Activity During Riots

Street workers who had established influential relationships with young
people on the streets were able to make major contributions. In Gar field
Park,. Chicago, the workers were able to persuade the young people they
knew to leave the riot area. In other cases the street workeis were able to
enlist gang leaders in the effort to keep their friends out of the riot
activity. Several agencies took young people away from the areas to lessen
the opportunity for involvement, During the Watts riot the staff of the
Special Service for Groups took food to desperate areas. Later, when it
was considered unsafe for whites to enter the area, the Negro staff carried
on.

All of the cities need more street workers in the rough areas. The street
worker is seen by the militant Negro as a threat because of his commit-
ment to nonviolence. Mr. Bernstein quotes a letter sent in plain envelope
to three street workers at Lawndale Neighborhood Services of Chicago
Youth Centers:

"Grettings Brother:

"Have you ever considered WHY the GREAT WHITE FATHER is
continuing to spend MORE money on the various street work pro

92

giams throughout the entire country and especially in the BLACK
GHETTOS. Also why the GREAT WHITE FATHER is expanding
these piograms to include community organizations If you are aware
of the current struggle being waged by Black Americans to be recog-
nized as human beings in the so-called 'land of the free and home of
the brave,' it is not difficult to arrive at the most logical answer as to
why WHITEY is expanding the program.

"The white man wants to contain our people, to keep them from
rebelling against living in slum housing, receiving poor education, and
thus the worst jobs, if any at all. The people who run America do not
want the Afro-American to know who he is, to have an identity, to
know of his glorious past, and the prospects of an even richer futiue
if they can get Charlie off their backs.

"Today there is a fierce battle being waged for the minds of the
youth of America, especially Black Youth. For young people are
invariably more idealistic than older people, and thus are not likely
to go for as much bull-as older people. Thus, it is principally the
Black Youth who fought the police during the ziots of last summer,
age range from 14 to 19 years old.

"Because Mister Charlie pays your salary the same as he pays me,
ask yourself, if the real-breaks out here, what will be my ROLE? If
your wife and children are in the streets during a riot, the BIG
WHITE COP will crack their heads just like any Negro's head. MY
BROTHER, what will you do with your group? Will it be a, 'cool it,
baby, they ain't done no thin' to us,1 or will it be a 'defend your-
selves, brothers'? Things are bound to get worse all over. The War on
Poverty is a farce, a throwing of ciumbs to the poor, both black and
white, to stem the tide of rebellion.

"Every Afro-American street worker has a very important role to
play in the coming days ahead. In the language of the street, you can
teach the young brothers and sisters to be proud to be BLACK. You
can help destroy the often subconscious inferiority complex that
exists in many of our ghetto youth. BROTHER, it is a question of
whether or not you will truly be a BLACK MAN who cares above all
else about the hopes and aspirations of his people. The choice is
yours to make and I have confidence that you will not be a SELL-
OUT. The fight is ours and we cannot afford to not be victorious.

Your Soul Brother"

In addition to the efforts of the street workers, some agencies were able
to make and keep active telephone contact with clients in the affected
areas. This is an especially good technique for verifying or defusing
rumors.

93

A common feeling among Negioes, even those who would work to
prevent a riot, is that no Negro can totally regret that riots have occurred-
Although many people suffer and there is a trauma involved, there are
definite gains in terms of community awakening. There is heightened
sensitivity on the part of police and politicians to their handling. Whites
have a range of reactions from dedication, to eradicating the basic wrong*,
to buying guns. Riots in some situations increased communication and in
others intensified the polarization.

Participation in the Civil Rights Movement

The civil rights movement in this context is described as a broad
combination of legislation, oiganizations, public opinion, and activities
aimed to extend all civil rights to those who have been deprived of them.
The means are many, including marches, demonstrations, sit-ins, boycotts,
political pressure, etc. In the early 1960's many white people were
involved as well as members of minority groups. At the time of the study
some militant groups of Negroes had given up on the civil rights move-
ment as an effective force. However, they must also be included here
because of their actual and potential impact on alienated youth. The civil
rights movement has been concerned with human rights as well as strictly
civil rights and often includes a strong appeal to decency as well as to law
enforcement.

Little participation in civil rights activity on the part of alienated youth
was discoveied by this study. One group of boys in Chicago had helped to
promote a large meeting to collect food and money to be sent to Negroes
who had been cut off from welfare. This had occurred when they had
registered to vote in a Southern community. Some other participation
took place largely through agencies. For instance, Mobilization for Youth
in New York took a large number of ghetto residents to Washington for
the March on Washington in 1963. By and large, participation was limited
and almost accidental. If a protest march went through their neighbor-
hood, they might join in. The young people in the civil rights movement
have been those whose motivations and aspirations have been more clearly
defined than those of the young people here under study. By and large,
the young people who are most alienated are also bitter and less likely to
participate in a civil rights movement. They are quicker to strike out,
more hostile to the police, and less disciplined. Some groups would even
discourage their participation so as not to jeopardize their cause or
inflame a sensitive situation. The more quick-tempered prefer to work at
things in a more direct way. The staff at Haryou-Act in Harlem described
a group that was concerned about the lack of toilets at a playground,
They then got publicity for the problem by urinating in cups at the
playground. They did get the toilets. Many of the most alienated young
people did not even really understand the civil rights movement. Many in

94

the North thought it was only fo: 'colored people in the South." The
delayed results of the civil rights movement are not attiactive to them.
This group has had little experience with altruism and cannot believe that
anyone does anything for a reason other than his own gain.

It was clear from the interviews that the nonviolent theme is not
compatible with the impulses of the angry young men in the ghetto. They
feel a strong need to retaliate. Although many situations have been
corrected legally, the problems still exist in an extremely frustrating form.
For instance, these young men know that legally many restaurants are
open to them and that stores cannot refuse to serve them. They also know
that they will be ill-received or self-conscious if they venture out of home
territory. Their clothing or their hairdos will mark them as different. They
see the stores and do not have the money to patronize them. The jobs that
are available to them pay so little that often they do not materially
advance their ability to have the things they so desire. They see the quick
financial rewards that come from illicit operations. They resent having to
"break their backs" to earn a legitimate salary and yet know that "their
backs" are all they have to offer due to their poor educations.

This study showed that the civil rights movement has to a certain
extent increased the awareness of the average minority group member of
the discrimination that he suffers. The need for an increase in black pride
and a self-respecting racial identity is crucial. There is still the question as
to whether any attempt to revive dignity can be totally successful without
significant changes being made in the slums and in employment, and in
general acceptance and respect on the part of the dominant community.

At the time of the study, the heroes of the civil lights movement were
little known to ghetto youth. The Reverend Martin Luther King, Jr.,
known to everyone since his death, was relatively little known by ghetto
youth in most cities, although he had been widely honored in this country
and abroad. Malcolm X was an authentic object of hero worship. During
the riot in Harlem in 1964, youth were quoted as chanting: "We want
Malcolm X." Many knew Cassius Clay whom they admired for his boxing
skill. Feelings were mixed regarding his lack of humility. In Roxbury, the
Boston Celtics were highly regarded by the Negro youngsters.

The special emphasis on the concept of "black" or "Afro-American,"
with its repudiation of slavery and its labels, is congenial to the most
alienated persons in the ghettos. The concept of Black superiority is wel-
come to them. However, this study showed that, despite the compatibility
of the concept, few of the young people joined the movements; most
remained as isolated as before from the larger society.

Puerto Ricans and Mexican-Americans have been much less involved
than Negroes in the civil rights activities. The reality of Puerto Rico as a
place and the recognizable and respected culture helps them tolerate the
problems here. This is true of Mexicans as well. The emergence of the new
Black African countries is important but very remote to most American

95

Negroes. While the Puerto Rican can and does return to Puerto Rico, the
average American Negro does not go to Africa, and, if he does, he may
feel more alienated and foreign than in the United States. Although the
militants have revived or made ties to Africa, by and large the American
Negro feels that, good or bad, this is his country.

Antipoverty Programs

The War on Poverty generated a great many programs to attempt to
attack serious problems. Of those specifically designed to help adoles-
cents, perhaps Job Corps and Neighborhood Youth Corps arc most
pertinent.

Job Corps Camps were set up away from slum areas. They offeied
healthy environments, good food, remedial education, and job training.
This might have been beneficial for these particularly unequipped and
hostile young people, except that a few things were wrong. The publicity
was attractive, but the delays in processing applications caused people to
become discouraged. The location of the camps away from home sur-
roundings made some uneasy. The similarity for some youth of being sent
away to training school was too marked. Agencies which did send young
men had a hard time finding out what happened to them. Information was
rarely sent to the referring agency when a young man dropped out or was
asked to leave. So, for a variety of reasons, the Job Corps did not make a
noticeable mark on alienated youth in the cities studied.

The Neighborhood Youth Corps included programs of jobs for school
dropouts, special intensive summer job programs, and activities for in-
school youth. Social agencies, government offices, and industry par-
ticipated as work stations. Youth whose families had incomes al the
poverty level were acceptable. Usually assignments were for about 20 to
24 hours a week for those who were out of school and less for (hose in
school. The pay rate was about $1 .25. A common complaint was thai (his
rate was too low. Many participants regarded the assignment as work and
not training and felt that it was not enough money. Delays in issuing
checks and other such problems lowered motivation to stay on in some
cases. Even the "poverty" label was offensive to some. After 6 months
those out of school had to resume some type of educational program or
be dropped.

The Neighborhood Youth Corps was most successful where the policy
was clear and counseling was an integral part of the program. The naUurc
of the work assigned and its potential benefit to the young person were
also important. Most members valued the opportunity to use business
machines and other skilled types of jobs much more than the menial jobs
usually available to them.

Agencies which employed these young people found that their early
employment phase was often difficult. A combination of patience, under-
standing, and firmness was needed to deal with absences, unexplained
lateness, or early leaving. Some young people were in agencies whose staff

96

understood the hair styles or dress of the youth and the identity issues
involved, but also felt these posed a barrier to later employment.
Language and attitudes about authority also caused problems. The way
that each issue was handled was important in each case. The youth had to
be accepted as they were at the beginning, and later the realities of the
employment market were introduced.

Local programs were developed in various cities with the approval of
the local antipoverty organization. Mr. Bernstein reports on a few that
seemed successful or interesting.

C/wa£o-STREETS (Socialization, Training, Education, and Employment
Technical Services)

This was a cooperative venture by the Chicago Boys' Clubs, the Chicago
Youth Centers, Chicago Federation of Settlements and Neighborhood
Centers, and the YMCA. The purposes in general were to enable dis-
advantaged youth to function better in relation to requiiements of our
society. Other goals were to make fuller use of social welfare and other
systems, to help the youth move toward adequate education, satisfying
jobs, healthy maniages, and so on. The neighborhoods selected for service
were among those designated by the Chicago Committee on Urban Oppor-
tunity as having the greatest poverty and related problems.

The main feature of STREETS was the hiring of neighborhood adults
and youth to help in the work of the agency. The youth worked primarily
with younger children. Those with leadership quality were sought. The
adults, called neighborhood aides, shared, with the professional staff, in
the supervision of the youth, called program aides. The aides also
benefited from the agencies' programs as well as promoting them for
others. The program was set up in units designed to serve 1 ,500 youths.
These were eight units and a staff of 488 to serve 12,000. Those inter-
viewed were enthusiastic about the program although some aides did not
achieve the level originally set for them. The respondents felt that great
success had been achieved in some areas. They felt that some aides learned
to give up antisocial acts, assumed great responsibility for children, and
returned to school. They also carried over to their own peer group what
they had learned.

Central to this success was the familiarity of the agencies involved with
the neighborhoods and clientele. They already had skill in working with
alienated youth and this made their commitment to the most difficult of
the young people moie durable. They seem to have a good sense of the
types of jobs aides can perform, and, in general, the program incorporates
the advantages of using indigenous workers. Aides received salaries of
about $45 for a 30-hour week.

Detroit— SWEEP (Summer Weekend Evening Emergency Program)

This is a summer weekend program conducted by the Neighborhood
Service Organization in high delinquency and poverty areas, in coopera-
tion with the Youth Bureau of the Police Department, with schools,
universities, museums, and others. It was for girls and boys, ages 12 to 17,

97

referred by schools and other agencies or people because they were drop-
outs or were having difficulty. There was also concern that these young
people might present problems dining the "long hot summer."

The active program was well attended. In addition to group meetings
and home visits, the program took children horseback riding, to Detroit
Lions' practice, to the Tigers' games, and to many other interesting places
in the area. Street workers and teachers made up most of the program
staff. The youngsteis were asked to write their evaluation of the program,
too, and were positive about the experience. Again with this program,as
with the one just described, the experience of the parent agencies was
important, and the woiker's skills and orientation were of vital impor-
tance to the success. The weakest part of the program is its temporary ami
limited nature. When summer ends, the stresses and deprivations continue,
and young people live with and react to them all year. The choice hcicis
between a siiccessful fragment or nothing at all.

Los Xflge/as-ESCAPE STRING (Education in Service Careers and Hm-
ployment-and Service Toward Redirection of Impressionable
Neighborhood Groups)

A major objective is to help predelinquent youth, and another is to help
low-income Negro and Mexican-American youth of academic promise to
continue their education with the help of salaries paid for work with this
agency. A team of professionals contacted gangs and their families, and
then the college students from similar low-income families were assigned
to work with them. Professionals supervised the college students, A very
satisfactory aspect was the involvement of the students in the agency. As
the program went on, if any of the college students were interested In
social work, everything possible was done to assist them in attending a
graduate school of social work. The staff felt that in some cases tlie
experience gave some college students the motivation to stay in college.
The young people helped also benefited. Again, the long experience in this
field of the parent agency was important.

Los Angeles had many other programs:

The South Central Volunteer Bureau recruited volunteers for work in
ghetto areas. It was very active in helping get food and supplies into the
Watts area during the riot. It was felt that many volunteers benefited from
this experience.

The Neighborhood Adult Participation Aides worked with a wide
variety of agencies to allow them to extend their services into areas thai
normally would not be offered. Homemaking services or foster father
activities were two examples.

The Los Angeles Human Relations Commission had a number of
projects. They hired some militant young Negroes to be a link between
them and the community. It was felt that this helped in preventing
trouble in their areas.

The Los Angeles County Civil Service instituted a program to employ
poverty youth in 30 types of civil service jobs as aides. They also searched

98

for other types of employment. Testing and orientation were given prior
to employment. Role playing and movies were used to help educate and
prepare the young people. Psychiatric consultation was used when neces-
sary and monthly employment followup was done where necessary.

Travelers Aid gave special help to newcomeis to the Watts area. Money
was available to help with interim crises. Public assistance was not
available here until residency requirements were met. Help was given with
school problems, medical care, legal difficulties, mental illness, etc. Aides
from the previously mentioned Neighborhood Adult Participation pro-
gram weie extensively used.

Rochester, New Yoik State Division of Employment: Special Youth
Project; Manpower; and City-County Youth Board: Youth and Work
Project.

The focus for these three related programs was employment of
poverty-level youth and appropriate training and job experience for them.
The State Employment Service had "outreach" programs to recruit youth
in streets, bars, pool halls, etc. Some workcis could speak Spanish so that
Cubans and Puerto Ricans could be reached. Contacts were followed up if
office appointments were not kept.

The Manpower Section was a training piogram run by the Board of
Education and the Employment Service. It offered remedial programs,
counseling, and primarily vocational training.

The Youth and Work Program was for girls and boys 16-18 years old.
Those who were least likely to find satisfactory jobs and most likely to get
into trouble were selected. Odd jobs in the program office were used for
diagnostic purposes. Then some of the young people were given individual
assignments. Others were given group assignments with group discussion
first, and then were interviewed for jobs or received more training.

Many other programs were examined during this study. One conclusion
was that the experience of local groups well grounded in work with this
population was very important. The locally shaped programs were charac-
terized by flexibility and imaginative use of existing agencies and facilities.
It was a general consensus that moiale was high on these programs and the
most discouraging feature was their time-limited existence in some cases.
These programs were, by and large, well able to attract the number of
young people they were designed to serve. Young people did want jobs
and training. The participants felt that the jobs they obtained had the
value of being hopeful examples for others in their neighborhoods.

However, these piograms, it was stressed by those interviewed, did not
basically change the ghetto. The stresses that unemployment cause, and
the increasing problems caused by automation, would take far larger and
more potent projects to combat.

Racial Feelings

This study discloses the feelings that have become more pronounced as
time has passed. At the time of this study the issue of Black Identity and

99

Black Pride was well in evidence in the ghetto. The study showed that
despite the riots of that period the anti-white feeling was not as high as
might have been expected. The Black Militants influenced directly only a
small part of the ghetto population, although the climate of hate thai tlicy
preached was moving in that direction, An awakening of many young
people to the racial inequities was described. Even the most successful
young people in our city slums are finding that they cannot love anyone
who helps to keep them down. The anger is high and the chances that it
will erupt against a policeman or some other symbol of their oppression is
very great.

The study showed that a key problem is in their own identity. The
tremendous necessity to reach back and find a cultural history that can
bolster respect has led many to study Africa. Others are searching for
Negro achievements in more recent slavery and post-slavery periods. A
difficulty here is that many Negroes who are tremendously accomplished
have achieved this by "doing it Whitey's way" which causes many Negroes
to classify them as "Uncle Toms." The solid achievement of even ac-
complishing that is often Iqst in this issue. Despite the fact that this stud/
showed that many "successful" Negroes who have achieved middle-class
status are very interested and concerned about what happens in Die
ghetto, this concern is not believed or recognized for the most part. Un-
doubtedly this is one of the reasons that the civil rights groups are unable
to enlist more basic support. The Urban League and the NAACP are often
characterized as "Uncle Tom" associations,

Mexican-Americans and Puerto Ricans were beginning in 1 965 to show
signs that they, too, would organize and ferment to achieve more oppor-
tunity. In some parts of the country their conditions are the worst
possible. A particular problem is the language issue. Schools have been
unable or unwilling to fully meet the complication of having Spanish be a
primary language for this group and still helping them to learn the neces-
sary English to compete successfully here. This is a major problem. In
1966 there were about eight hundred thousand Mexican-Americans in Los
Angeles County alone. The struggle of the National Farm Workers' Asso-
ciation to get higher wages and better conditions for grape workers was
beginning at the time of this study. It has subsequently become more of a
political issue. In 1 968 one California congressman passed out grapes with
a sign that indicated one should enjoy the "forbidden fruit," and fie
received many packages back from congressmen in other States who did
not wish to anger constituents or labor groups by being seen to accept
them.

One Mexican-American group of about 50 members walked out of a
Federal Equal Opportunities Conference because they were not repre-
sented on the commission that planned the conference. A large group held
a banquet to celebrate this act,

Though in some ways the problems resemble those of the Negroes,
there is a more marked resemblance to the patterns experienced by other
immigrant groups. This study and the one in 1963 showed that the youth
seem to be between two cultures. It was the impression at this time that

100

they felt superior to the Mexicans and inferior to the Anglos. This causes
the classic pattern of family tension and lack of respect for the parents
and the old ways. Gang formation is a common way to deal with this and
was attractive to those unfortunate enough to lemain in the slums.

The Puerto Ricans are in a similar but different situation. In New York
in 1 964, figures show that Puerto Ricans were even poorer as a group than
Negroes. They share with Negroes high rates of public assistance cases,
large families, high unemployment, excessive interest lates for loans, etc.
They share language and cultural tension with the Mexican-Americans.
They, however, seem to have considerable pride in being from Puerto Rico
and tills is a definite plus, just as the island is a retreat and a refuge for
them from life on the mainland.

The Puerto Rican Community Development Project is a comprehensive
plan and program for meeting a wide range of needs. It was financed by
antipoverty funds and was built upon existing programs.

Conclusions

By 1965 the apathy had coalesced into anger. The conditions in the
slums and the state of day-to-day life proved too provocative and unyield-
ing. The methods of protest were too tame and unsatisfying to the needs
of the ghetto resident. The alienated young person living in the slums and
feeling he had so little to lose was immersed in this atmosphere. The
release felt by many who had been pent up was stronger than the prohi-
bitions. Although mixed feelings about the riots are common, the pride at
turning their image from "shiftless" to dangerous was welcome for many
Negroes. It is definitely more manly to be dangerous in this country than
it is to be "shiftless and lazy." Whereas the civil rights movement was able
to achieve change in the South and the new laws were making changes, the
latter were not of sufficient measure to make a difference in the ghettos
of the North. The riots undoubtedly hurt the Negro communities
involved-they have suffered real loss, food scarcity, and loss of sustaining
businesses. But the gains are real also and unfortunately are often more
visible than the gains of slower, more moderate methods. Politicians and
policemen, for instance, have become more conscious of ghetto problems
and their explosive potential.

Whites, too, have suffered. All have suffered. The investigator found
that during this study it was virtually impossible to limit the time spent on
this subject with any respondent as the feelings flowed in such an intense
and meaningful way.

A warning has come out of this study not to lump minority groups
together. Specificity and tailoring to meet conditions in individual areas
have been successful. Other suggestions are that intergroup contact should
be fostered. Conference, tutoring programs, training programs are all
valuable if they contain meaningful contact between ethnic groups. Goals
which are yielding of fairly quick success win alienated young people into
organized campaigns faster than more abstract goals.

101

The study suggests that the formation of overall city poverty agencies
may be unwise. It was found that they tend to generate problems that get
in the way of their own work. The investigator suggests that perhaps it is
better foi local agencies to deal directly with Federal officials foi anti-
poverty funds. Another point is that funding is important to agencies and
must be given a firm enough and long term enough basis to make it viable,
This study revealed that many agencies are forced to cm tail and limit tncii
involvements due to fear of cutbacks or loss of funds. There is the concern
that programs begun and dropped can arouse more antagonism than those
never stalled. Many agencies have to devote themselves mainly to keeping
alive.

Further, the investigator urges that the real serious nature of the
problem of the ghettos be faced. He recognizes that a realistic appniisuf of
economic and social network that causes slums and its tremendous human
toll be faced. Varied programs are needed. This study clearly shows that
no one type of program will reach all. Even if at times the cost may be
high, such programs must still be supported. The difficulty and, :it the
same time, the ease of predicting riots is thought-provoking. In 1 969 some
politicians and civil rights leaders are predicting that our major riots are
over, but this is hardly assured, and, if it were, the human cost of our
racial problems and the loss of any of our young people would still be loo
high. The commitment has been made in this country many times and in
many ways and needs now to be made a reality.

The investigator suggests that many legislative and other steps arc
needed to combat poverty. The guaranteed annual income is suggested as
one idea that might be successful. Cooperative housing might overcome
the need for large investments on the pait of poor people and give them a
chance at ownership lather than at being renters. Street work and social
agencies have a strong chance at helping touch alienated youth. The pride
in racial identity needs to be supported, but not the hate messages tlial
sometimes accompany it. Indigenous woikers reciuited from an area to
serve an area are also particularly successful, and these positions provide
employment for those who might otherwise be unemployed, which is also
valuable. Birth control information is often sought in the slums anil
should be made available. The cities need to improve on services to Die
deprived areas and so on. Primarily, tolerance must be found for Ik
forward and then backward and forward again patterns of work in this
area.

This study suggests that education in the slums is a complex issue and an
important one. While the need for technical education exists and is impor-
tant in preparing young people for our mechanized society, the as-
sumption should not be made that this type of education is all that is
needed in the slums. Many students need assistance, educationally and
financially, to achieve their potential. Many more could attend college.
Programs with liberal admission policies, even pteferentia! admissions,
etc., need to be developed.

102

The investigator also postulates that, if each large firm undertook to
hire and train some of these depiived young people, the effect would be
striking.

As has been previously mentioned, community service positions
utilizing indigenous workers can fill some important needs in oui society.
Almost all large community institutions, such as hospitals, schools, and
social agencies need help. The presently alienated, almost wasted young
peison could provide the manpower needed for better operation in all
these community institutions. Training is the key in both private and
government programs for effectively teaching and employing these young
people. '

A major mistake would be to underestimate the forces that keep slums
as slums and keep minority groups in them. The solutions to these prob-
lems are extremely complex. One conviction of the investigator is that the
poor should be consulted and involved with progiams, even though this
may at times make progress slower and more complicated.

Research Grant. Mil 11396

Dates of Intel view: September 1968

References:

Dems tern, S. Youth on the Streets, Work with Alienated Youth Gtoups. New York:

Association Press, 1964. 160pp.
Alternatives to Violence, Alienated Youth and Riots, Race and Poverty. New York;

Association Press, 1967. 192 pp.
Schulberg, B. Fiom the Ashes. New Yoik: New American Library, 1967. 275 pp.

103

Project Director!

Reginalds. Lourie, M.D,
Children's Hospital
Washington, D.C.

Prepared by:

Herbert Yahraes

A pilot project for training a new type of mental health worker, ji child
development counselor who would make a frontline effort to intcrcep!
emotional disability before it could take hold, is now complete and the
results are being evaluated. Because the lessons being drawn should be
useful in other programs to develop unconventional manpower soutces.

The reasoning behind the project ran like this:

In the typical well-baby clinic, the pediatrician and the nurse can give
only a short time to each mother and child. Tins has generally proved
sufficient for noting signs of physical illness and of irregularities in
physical development and for being concerned with preventive approaches
in general. But it is hardly sufficient in most clinics for noting and
counseling about signs of emotional maladjustment in the child or of
maternal attitudes likely to lead to such maladjustment. If the clinic staff
could include a person whose main concern was the normal emotional and
psychological development of the child, the clinic might be better able (o
head off crippling problems. Well-baby clinics, in short, might better fulfill
their mission of bolstering the mental as well as the physical well-being of
the children passing through them.

Since the traditional mental health disciplines-including medicine,
psychology, nursing, and social work-cannot meet even the existing
demands on them, however, where are these new workers to come from?

An earlier pilot project conceived by Dr. Margaret Rioch, a psycho-
logist, had found a new manpower supply in mature, intelligent mothers
with a keen interest in the community and a desire to serve. The piojcct
had demonstrated that such women could be trained for counseling
adolescents and adults in mental health clinics and.other centers, and tlint
a variety of agencies and institutions were eager to employ them.

Wouldn't it be possible to take another group of such women and tiain
them-again over a 2-year period, half time-for a new profession, child
development counseling? And couldn't they then contribute significantly
to a highly important but relatively neglected part of the mental health
campaign-preventive approaches in the earliest years of life?

104

To answei such questions, Dr. Reginald S. Loune, chief of the depart-
ment of psychiatry of the Children's Hospital, Washington, D.C, and
medical director of the affiliated Hillcrest Children's Center, undertook to
direct an Institute-financed tiaining program for child development
counselors. The project was sponsored both by the research foundation of
the hospital and by the Bureau of Maternal and Child Health of the
District of Columbia's Department of Public Health. Dr. Samuel Schwartz,
the Bureau's chief, was codirecloi with Dr. Lourie. Through most of the
project, Dr. Rioch, who had diiected the earlier program and stimulated
the thinking in this one, served as training director, and Margaiet Stolzen-
bach, a graduate of that program, was her executive assistant and
coordinator. The evaluation phase is being conducted by Dr. Stuart E.
Golann, of the University of Maryland and the American Psychological
Association.

The tiaining centers were, in the main, the well-baby clinics of the
District of Columbia Department of Public Health and the one at the
hospital. The project directors felt, however, that child development
counselors probably could be highly useful at nurseiy schools, day-care
centers, and almost anywhere else that mothers and young children are
found, so the trainees were made acquainted with a variety of institutions.
Most of the instruction was carried on in the pediatric and psychiatric
facilities of the hospital. The teachers included two dozen authorities-
many from the staffs of the sponsoring institutions but a number from
outside— in such fields as psychiatry, psychology, child development,
family and child therapy, social anthropology, and psychiatric social
work.

The Trainees and the Program

To recruit trainees, staff members discussed the project's hopes and
plans with each of about 50 persons who were leaders in the community
or represented such organizations as PTA's, women's clubs, community
service agencies, and college alumnae groups. This was done by telephone.
Descriptions of the projects were then mailed to interested organizations
so they could be posted, read at meetings, or published in newsletters.
Stipends of $1,000 per year were offered trainees. Women requesting
application blanks were given a preliminary telephone interview that
served, among othei purposes, to correct misunderstandings about the
project and to emphasize the uncertain future of applicants selected for
training.

Out of the 101 women who completed their applications, eight were
finally selected. All eight -on the basis of autobiographies, interviews, and
group discussions with members of the training staff-rated high in intel-
ligence, perceptiveness, self-awareness, integrity, and emotional maturity.
Says Dr. Lourie; "We ended with bright, warm, flexible, empathic, verbal
women who were interested in others and intellectually curious."

The women chosen ranged in age from about 35 to about 50. All but
one were white, though it had been hoped that about half would be

105

Negro, because in Washington the well-baby clinics serve a primarily I
population. (A second Negro woman had been recruited, but she
drew before the training began.) All were mothers, though their chi.
were either grown or well along in school, and all were paiticipath
community activities. Five were college graduates, including one wo
with a master's degree, and the otheis had had at least 2 years of col
Most of the husbands were professional people.

Preceding the start of training, in February 1964, the stuff n
arrangements with American University to grant credits toward citli-
bachelor's or a master's degree. It was reasoned that the counselors wo
always be woiking alongside professional people and that (hi
presumably, would be interested in their academic backgrounds. Also
was hoped, academic recognition for a new type of health-field prof
sional might eventually be developed.

The training piogram was divided into four semesters which wi
spread, college-style, over a 2-year period. The trainees worked half tttu
They had the following weekly schedule during the first semester:

• Three hours of lectures on the physiological, sociological, an
psychological aspects of child development.

• A 2-hour seminar on personality development.

• Four hours of practical classioom work dealing with clinical tfls
histories-from well-baby and other types of clinics-in order to learn Hit
kinds of problems that arise and how to handle them.

t Half a day at a well-baby clinic, generally one of the 12 opera ted by the
District of Columbia Department of Public Health and known officially as
Maternal and Child Health Clinics, For two weeks the trainees observed
clinical routine by accompanying an assigned family through all the clinic
procedures, including a home visit by a public health nurse. Then for (wo
weeks they took routine case histories of mothers selected by the clinic
staff. After that, clinic staffs were asked to refer to trainees those mothers
who expressed interest in discussing problems at greater length than stuff-
time permitted. The trainees were urged to invite additional interviews
with these mothers and to spend unscheduled time in informal waiting
room contacts.

• An hom of counseling, at a well-baby clinic, under the Jmmcdinic
supervision of a member of the training program's staff.

• A conference with staff members, primarily to give the trainees nn
opportunity to discuss their work and to ask questions and to give the
staff an opportunity to gauge progress.

For the second and third semesters, the fieldwork was expanded to
include one full day a week at a well-baby clinic and half a day In a
suburban nursery school, where the trainees could observe the behavior of
children-and the attitudes of their mothers-who were growing up to a
privileged section of the metropolitan area. The trainees also visited a
variety of other agencies to learn the kinds of helping resources available
and the problems facing them. Lectures and observational work during
the second semester dealt with the following topics, among others:
psychological and psychiatric examinations, pediatric consultation, the

306

child's early development, mental retardation, speech and hearing,
nutrition, nursery school orientation, child rearing, and planned parent-
hood. Classroom work during the third semester included 4 hours of
lecture-discussion seminais covering child-rearing practices, family inter-
action, psychodynamics, the psychosocial effects of illness, and tech-
niques of parent-teacher consultation.

During the fourth and last semester, six of the trainees continued to
spend a day a week at a well-baby clinic, while the other two worked in a
maternity clinic, supervised by an obstetrician. The trainees also spent one
day a week working with one of a number of other organizations-
neighborhood centers, a mental health clinic, the Family and Child Service
Agency, the District of Columbia Day Care Association, the Jewish Social
Service Agency, and the Prince Geoiges County, Md., Maternal and Child
Health Clinic,

Course work during this final term included 3 hours a week on "The
Process of Child Development." The first half of the couise was given in
the maternity wards at D.C. General Hospital, where trainees observed
newborn babies, talked with the mothers, and participated in a seminar
with nurses assigned to the maternity section. Later, at the Children's
Hospital, children up to 6 months of age were observed intensively. Then
children between the ages of I and 2 years were studied, Ward observa-
tions were supplemented by lectures and films.

Among their other activities, the tiainees attended a course on adoles-
cence given at Howard University School of Social Work; sat in on
diagnostic sessions at the Department of Psychiatry of the Children's
Hospital; attended lectures in the pediatric-psychiatric training program of
the same hospital; and observed a number of interviews between therapists
and patients at the Clinical Center, National Institutes of Health.

Major Problems Encountered

In the disadvantagcd areas served by most of the Washington well-baby
stations, many of the mothers were uninterested in talking with the
trainees. They wanted mainly to get Ihiough with the pediatrician and the
nurse and to get away from the usually crowded clinic. The many who did
want to talk generally presented urgent problems of their own-the need
for a place to live, for clothes, for legal aid in marital problems, for help
with a delinquent older child.

The trainees felt obliged to offer this pressingly needed assistance,
usually with the aid of community agencies the mothers had not known
about or had been afraid to try, because it was the humane thing to do. In
addition, particularly during the early part of the training, when the
women often wondered if they were being really useful, it gave them a
sense of accomplishment. Sometimes the help with the urgent practical
problems opened the way to do something about the basic principles of
preventive mental health work centered on the child-to try to get across
the idea, notably, that the way a youngster is tended, talked to, and
played with will affect his future emotional well-being and his ability to

107

learn. Generally such efforts had to be neglected or abandoned, howei
because the mothers visited the clinics infrequently. A few did maker
keep special appointments with the trainees.

During the latter part of the training program, the counselors rcpori
greater success in directing the interviews to the field of child devel<
ment and in dealing with problems directly concerned with the chilrtn
Included were problems of nutrition, shyness, destructive bchavh
nervous habits, and slowness. Information that the mothers would n
otherwise have had on such matters was made available. A report on (
project's fourth semester notes that "there seemed to be more counseli
and less social work,"

During this semester, in the couise of their work at the well-baby aj
maternity clinics, the trainees talked with 450 mothers fora total of 6(
interviews. This was an average of about 83 interviews per trainee, and <
about six per working day at the clinic. Most of the interviews Instc
between 15 and 40 minutes, but some ran more than an hour, Aboi
three-fourths of the mothers interviewed had been referred by the cliiu
doctor; most of the otheis had been approached directly by the trainee i
the clinic waiting room. Some 300 of the mothers were seen only once.

Dr. Lourie points out that even when the main concern had to be til
compelling problems of everyday life, the trainees were actually doin;
preventive work with the child. For if they could help take care of lfn
mother's needs, she in turn would take better care of the baby's. The goa
of the training program, though, had been different. In short, the well
baby clinics, with only sporadic family contact available, proved less than
ideal for providing the counselors-in-training with experience in the kind
of preventive counseling it had been envisaged they would do and for
which their other work was preparing them,

A second important problem had to do with the attitudes of the other
people working in the clinics. The Department of Public Health had joined
with the project's training staff to explain to the clinic personnel lite
concept of preventive work by a child-development counselor. The results
varied considerably, the project reports, ranging from eager acceptance al
one end to open resistance at the other, with indifference in the middle.
Resistance and noncooperation tend to be met whenener a new group of
"manpower multipliers" is proposed for the health field, Dr. Lourie notes.
The fundamental cause is the tendency of professional workers (o view
the new group as a threat to the standards of their profession-die
academic degree, the length of training, the other requisites they them-
selves have met.

All the trainees reported they could work effectively only when [lie
clinic pediatrician-and preferably the nurse also -understood and
accepted the program's aims, was willing to make referrals to tin
counselor, and could give her consultation time. These requirements were
not always met, particularly in the early part of the program.

As trainees and clinic staff got to know one another, the resistance
generally gave way to cooperation. Three pediatricians became especially
interested in the program, so the well-baby clinics they directed were tlie

108

ones used during the fourth semester. At the end of this term they and the
director of the maternity clinic were asked to rate the counseling service
from the standpoint of usefulness; all rated it as significant or potentially
significant. They were also asked to rate the individual counselors on such
items as ability to observe child with discretion, understanding of child,
ability to function usefully with patients as individuals, and behavior as a
professional person. Out of a total of 48 ratings for the eight trainees,
only three ratings fell as low as satisfactory; all the others were good or
excellent

Perhaps significantly, it was not until this last semester that the
pediatricians still in the program felt able to score all the trainees on all
the individual items.

The Post-Training Experience

Contrary to expectations, the District's well-baby clinics had no jobs
for the counselors at the end of the training period, in January 1966.
Project officials explain that in spite of improved relationships, there were
people at administrative levels not yet ready to accept the concept of
counselors trained as these had been. A defensive attitude by the
traditional professions may be only part of the explanation, these officials
think; another part may be a conviction that the preventive approach to
mental illness is impractical— as it often seemed to be, in the atmosphere
of the clinics, even to the trainees themselves.

Seven of the graduates took half-time positions with another Health
Department activity, a community mental health program, and the eighth
went to work for the well-baby clinic at the Children's Hospital. Several of
the half-time workers took on another part-time job-with a mental health
clinic, the Jewish Social Service Agency, or the National Capital Day Care
Association.

The seven counselors in the community mental health program had
been hired to do preventive work, they thought, but the pressure to help
the children who were already emotionally ill, and their parents, was so
great that preventive action had to be postponed. "It's like fire fighting,"
Dr. Lourie remarks, "If you fireproofed the buildings, you wouldn't have
so many fires. But you can't stop fighting fires in order to do the fire-
proofing. Though we did not expect our trainees to have to fight fires,
they did apparently make good emergency firemen."

As part of the project's evaluation process, which is continuing, the
counselors' supervisors and fellow workers were interviewed about 3
months after the graduates had been on the job. Those interviewed
included pediatricians, psychiatrists, public health doctors, public health
nurses, social workers, and teachers. Virtually all said that they were
pleased with the ability and the performance of the new workers, and that
they considered the women to be examples of a potentially significant
manpower source. The counselor employed by the well-baby clinic was
reported by the clinic's director to have been trained "better than the

109

pediatrician to help with the hundreds of patients who don't
psychiatrist." This supei visor also icported: "She does bcttei than you
doctors. I'm continually delighted she's here."

The women themselves had a different story. Intei views and job din"
showed them discouraged and dissatisfied, uncertain both about wh
they could do and what they were expected to do. The fiist fcw wtvkS'
months on a job aie almost sure to be an upsetting period for any u'tvnti
trained person, Di. Golann remarks, even if the tiainee has a tradition
degree and is woiking in a chosen, and traditional, field. But Ihesi1 cftrf-
development counselors had been trained in a new field, they did ncl
possess univeisally recognized credentials, and, worst, moat of t fit-in WLVV
working in fields diffeient from the one foi which they had been lianicJ

Salaries, too, were a disappointment. In the eyes of graduates ;m.l o!
the project staff as well, the Civil Service ranked the new counsolois Jew
low. Further, it distinguished between those with a college degree, wliotf
it placed in GS-7 classification, and those without a degree, whom fl
placed in GS-5, though the training for child development coumt'loi li;id
been the same fot everyone. This meant that a counselor going to wwt
for a District of Columbia agency could expect to start at an annual salary
ranging from about $6,500 if she had a degree to about $5,500 if slu-diJ
not.

The child development counselois have dealt with their clissiitisfauta
in various ways:

• After some months, two of the eight left their positions and yiuolk-d
at Howard Univeisity to earn a master's degree in social work, one of
them studying full time and the other part time. They knew (hoy wouM
have to repeat much of the woik already covered, but, says Dr. Louru1,
"They felt the need for a label that eveiybody understands. In (mining
new categoiies of manpower the identity problem is a prime one."

• The counselor at the well-baby clinic has stopped woiking in tliehopf
of finding a position in which she feels more useful. In spite of the praise
from the clinic's director, she didn't feel needed.

• One woman moved to New York City with her husband and lias token
a job with an adoption agency. She counsels both the mothers who lime
adopted a child and the natural mothers.

• Three counselors are still with the community mental health program
but, having received on-the-job training, work mainly with troubled
adults.

• The eighth counselor left her part-time job with the mental health
program and is employed full-time by the National Capital Day Care
Association, where she works with mothers, teachers, and children. Of all
*lm" " training program, she comes closest to carrying out (lie

-liat the counselors try to head off potential problems

~)r, Laurie is hopeful that through the work of

k> develop additional centers for babies and

placements for some of the other

Summing Up

The project has demonstrated, members of the tiaining staff conclude,
that it is possible to train mature, selected women in 2 years, half time, to
do a satisfactory job as mental health counselois having a specialty in
child development. At the end of the progiam the eight women could
counsel effectively with mothers of young children.

The project has been less successful in fitting the counselois into the
existing system of health agencies. By and large, the trainees are doing
useful work in the general field of mental health, but it is not the kind of
woik for which they had been trained. Dr. Lourie thinks this situation
may well change as the new community health programs develop and new
centers for parents and children evolve into developmental centers.

Some of the Lessons Drawn by Project Officials

Very early in the planning of any program looking to the development
of a new type of mental health personnel, the professional people-
including personnel officers— of the agencies in which the trainees will
work should be drawn into it. They can help ease the annoying everyday
problems, such as shortage of space, that arise when an additional service
is injected into established routines. More important, they can be exposed
to the concepts of mental health counseling and to the potential value of
the trainees in putting them into practice. The understanding and
acceptance of these concepts, even by professionally trained health and
welfare personnel, reports one member of the project's staff, and the
willingness to cooperate with new types of manpower cannot be taken for
granted. Says another; "Even more time must be spent educating the
existing professionals than the trainees themselves." Included in the
educational program, he reports, should be the professional associations
that set the standaids for the fields in which the new counselors will be
working in part.

Dr. Lourie adds that if agencies nnd institutions experiencing manpower
shortages—shortages especially of intelligent, sensitive, mature people-
would analyze their needs, they would find almost certainly that a large
part of them could be met by women like those selected for training in
the pilot project. "The question is whether existing institutions are
flexible enough to use people like this once they have been trained. The
women in our program are eager to be useful; but they also want to learn,
and they will not stay long in positions which do not allow them to use
their capacities."

To insure the best use of people trained in nontraditional ways, plans
for their employment should be carefully made well in advance,
preferably at the same time that the training program itself is planned.

It is speculated that most of the Negro women who would have been
eligible for training were already working and could not afford to under-
take the program. If more Negro women are to be recruited for programs

such as this one (and it may very well be that more of the mothers at tlic
clinics would have sought the counsel of Negro trainees), stipends Luge
enough to support them probably will be necessary, along with a guaran-
tee of employment at the end of training.

Except on a superficial level and with limited objectives, preventive
mental health woik probably cannot be carried on in well-baby clinics
such as the ones, serving a disadvantaged population, used for the field-
training program in Washington. The attitudes and problems of the
mothers, along with crowded conditions and the lack of cooperation of
some regular clinic staff members, worked against the Washington project.
"This is not to say that the child development counselors couldn't and
didn't make a useful contribution there," Dr. Lourie observes. "They were
trained hopefully to make a better one."

If he were starting over again, the director of this pioject would stilf
want to train child development counselors, but he would envisage putting
them to work in a different setting-day-care centers that accepted
children from infancy onward. He hopes such centers will become part of
the public school system. If differences in learning abilities, beyond (hose
set by heredity, are determined to a large extent when a child is vciy
young, perhaps mainly during the first 18 months, says Dr. Lourie, why
not get the educators interested in the child far earlier than at present?
Educators, welfare workers, and health people should collaborate in our
lowest socioeconomic neighborhoods, he believes, to provide continuing
services from birth on up into the school years as we now know them, in
such a program the child development counselor could do her most
effective work.

Research Grant: MH 8322
Date of Interview: May 18, 1967

References:

Golann, S. E. Initial findings of the follow-up study of child development counselors
1966. Mimeographed.

Louire, R. S., Rioch, Margaret J., and Schwartz, S. The concept of a training program
for child development counselors. 1 966, Manuscript.

Rioch, Margaret J., Elkes, Charmian, and Flint, A. A. Pilot project In training menial
health counselors. Public Health Service Publication No. 1254. Washington, Super-
intendent of Documents, U.S. Government Printing Office.

112

Director

A.Paul Parks

Operation Hope

Los Angeles, California

Prepared by:

Gay Luce

At a time when the voices of despair are registering with ominous
clarity, a small project in Los Angeles is offering an antidote to the ugly
waste of young people in urban ghettos. Operation Hope is an experiment-
al training program in which 20 young men and women from Central and
East Los Angeles aie being helped through college, and encouraged to
acquire the skills and credentials that may make them the new leaders in
mental health professions such as social work. The program provides a
monthly stipend of $200, the sine qua non, which combined with part-
time job income, permits the participants to study. Moreover, the three
staff social workers, Director of Research, Administrative Assistant, and
outside consultants of Operation Hope provide special courses, legal and
psychological counseling, and assistance during financial and family crises.
These young people from the ghettos are likely to revitalize the mental
health professions, at a time when welfare and social workers are often
seen as middle-class visitors, as outsiders who are ineffectual and hated by
ghetto residents, and who seem incapable of devising programs that will
attack the consequences of poverty and crowding.

No neat formulae can be offered for selecting and educating ghetto
residents, but the program evolved by Operation Hope offers a guideline
that could be adapted to any city location. The impact of this program
cannot be measured by statistics, since only a few participants have been
funded, but the school performance and careers of its trainees will be
noticed in the next decade as these people become poverty lawyers,
teachers, and mental health workers in the neighborhoods they know well.

Background

It is shocking to discover that close to one out of every six Americans
lives in a poverty pocket, an island of hopelessness surrounded by af-
fluence. Innercity ghettos do not communicate with the "outside world";

113

however, the young, who have grown up in an era of television, have seen
the wideJy advertised wealth and "opportunity" that lies only a few miles
away. For them it is as distant as Hawaii. These street-wise young people,
who are old by 21, know all the discrepancies between promise unj
actuality. They know that the promise of free education, of job oppor-
tunity, of legal equality differs from the actuality they experience. Thcs
know that the ostensibly free clinic is a place wheie they are rudely
treated, and wheie a sick man may suffer for many hours in a waiting line
If they are black or brown, they know they may be sent to jail while a
white Anglo is set fiee for the identical offense. They may know of jobs,
yet in a city like Los Angeles they cannot get to them because they do nw
own cars and theie is virtually no public transportation. Many of these
students are blight and do well in grammar school, but by high school
they diop out as they begin to be inundated with financial and family
problems. If they are spunky they are likely to be pushed out.

Traditionally, the people who have tried to help have been sincere and
hardworking social workeis, people who still come from comfortable
middle-class backgrounds and merely visit the ghetto, escaping to aclenn
quiet home each night. Despite their good intentions, social workers arc
inevitably seen as transient visitois, and hated by the people they aim to
help. They are viewed as spies and purse wardens who enforce moialistk
rules which have no relevance. At an early age many children conclude
that the casewoikei is merely snooping when he asks questions, ferreting
out some misdemeanor 01 technicality that means the family will noted
money. The familiar stipulation that welfare recipients would receive no
money if an able-bodied man were in the house has created deceptions
and family disruptions that have warped the lives of entire generations.
The tragedy is typified by a girl in the program who fought and hated her
mother throughout her life, and only later discoveied that her fathered
been driven from the house by the welfare act, not by her mother, By thai
time she and her mothei were permanently estranged.

Under pressure and without outlet, despair, passivity, and withdrawal
into multisubstance use, such as alcohol and dangerous drugs, are (lie
inevitable concomitants of a life with ramified health problems, and
perpeptual family crises. Hundreds of thousands of potentially bright am!
creative youngsters are being warped and wasted. Ghetto existence cannot
be patched. Today, ghetto youth live in an ironic vortex of history.
maiJced by the threat of population growth, the exodus of middle-class
families to the subuibs, inflation, and war. Ironically, the pressure is on
the young, themselves, to rebuild or create the institutions that will
readmit the poor into the mainstieam of American life, a life of decency
and opportunity. Nobody else can do the job.

Considerable institutional and social changes must be constructed
before this portion of the American people can participate in its own
future. At present theie are no institutions that can fully supply
city ghettos with the most basic necessities - good housing, health
care, and education. The need is for change in institutions, policies or

114

social organizations as they now exist. In a dynamic society institutions
should change to meet changing needs. The question that faces our society
is how to revitalize our institutions so that they maintain health and
well-being. It seems improbable that these institutions can be fashioned by
people who live far from the ghettos, and who know the problems only by
reading. The hope for constiuctive institutional changes emanates from
within. Opeiation Hope was evolved to select potential leaders, and give
these young people the instiumcnts for institutional change, supplying
them with the education that is the key to social participation. It is hoped
that this program may lead to changes in the selection of peisonnel for the
mental health fields.

History

Mi. A. Paul Parks, the originator and guiding spirit of Operation Hope,
came to Los Angeles with the conviction that his profession of social work
needed to be revitalized. An Easternei with a varied work experience, he
had extensive experience in the field of drug addiction. He had conducted
educational seminars foi community physicians on the social implications
of drug abuse. Like many of his associates, lie began to see that drug
addiction and other major social problems could not be treated as if they
were local "infections" that could be diagnosed and cuied on an indi-
vidual basis. The traditional medical approach is rarely lelevant in the
attempt to counter drug abuse. Drugs of all varieties have been the escape
of people whose lives are intolerable throughout the centuries and in all
countries of the world. Life is indeed unendurable for the vciy poor and
no amount of patching can alter the fundamental trap or the isolation in
which the poor, and paiticularly the black and brown minorities, now live.
Neither preachment nor punishment can eradicate the alcoholism and
drug use and other consequences of a degrading mode of living in a dis-
honoring environment.

The helping professions have not always given credence to the fact that
human behavior is adaptive and learned. This implies that each commun-
ity paiticipates in shaping the lives of its citizens and their behavior. In
actuality, social work like many other helping professions has often used
its power and position to label behavior — an approach which does not
produce the techniques that contribute to healthy reactions or change. In
gaining status as a profession, social work in some settings has also moved
further away from the people it originally set out to work with.

In the early 1960's, social work groups were seeking new solutions to
ferment in the ghetto, by recruiting and training local young people for
social work, Mr. Parks moved West to join an agency known as Special
Service for Groups, Inc., at a lime when it was training gang Icadeis from
Watts and East Los Angeles as social workers' aides. He arrived in Los
Angeles on a hot August day in 1965, one day after general discontent

115

had combined with heat of sumrnei in the destructive insurrection that
came to be known as the Watts riots.

Watts is a remnant of a real estate tract in South Central Los Angeles,
but the action that America remembers occurred west of Watts in an area
housing some 400,000 people, 300,000 of them black. These people have
no access to the resources of the larger city. Pooi, they lack transporta-
tion, and are unable to work outside their neighborhood. Thus, in their
out-of-the-way enclave, they remain invisible to the affluent majority.
Here, on a hot night a small incident with the police generated rumois and
crowds that slowly swelled into a populace-police battle culminating in
the deaths of 34 persons and the destruction of millions of dollars in
property. Watts became the predictive symbol, reminding Americans that
poverty and racial oppression were hidden in the affluence, and thill the
problems of huge gioups of American people were being neglected. 'I he
"Watts riot" announced that this double state could not exist any longer.
Two years later Boyle Heights became the center for the development of
new solutions, the headquarters for what is now Operation Hope.

Location and Staff

Roughly two miles fiom the modern Civic Center of downtown Los
Angeles is an old barrio - Boyle Heights. With the adjacent neighborhoods
of East Los Angeles and City Terrace, it houses about 180,000 people of
whom 135,000 have Spanish surnames. Boyle Heights is bounded by free-
ways, an area of 150 blocks through which many waves of immignilioii
have passed. Like a port of entry to the city, it has received successive
waves of Russians, Serbians, Italians, Jews, Mexicans, and othcis. At
present Spanish is the major language and many of the older Mexican-
Americans carry on the traditions of their mother country. In the ghettos
and barrios of South, Central, and East Los Angeles few new homes have
been built for 20 years, but as middle-income people fled to the suburbs,
black and brown people took their place. It is in the center of this inulU-
lingual ghetto that Operation Hope was established in 1968, setting up
offices in a few rented rooms.

The four staff offices, the large anteroom and classroom were carefully
chosen, for these are not only the working headquarters of Operation
Hope, but the center of all activities, seminars, and consultations, the one
meeting ground for its many disparate participants. The rooms arc
spacious and friendly, containing the bulletin boards of events, duplicating
equipment, and an incipient reference library for students,

Mr. Parks drew together a staff as varied in completion and background
as the students he would seek out It had to be a group of exception-
ally experienced and committed people, because the entire staff would
include only five people, and they could anticipate working around the
clock, in every conceivable exigency. The staff all had extensive social
work experience, and each person came from a different part of the
country as well as different ethnic backgrounds, They are supplemented
by consultants and instructors.

116

In philosophy the staff agreed. They had all personally seen the failuie
of the medical model of social work. They had, themselves, been trained
to act as professional social physicians who would diagnose a needy
client's problems and try to repair them in the manner of setting a broken
bone. Since the ills almost inevitably lie in the complex social environ-
ment, not the individual, the medical approach to social work is extremely
frustrating. The staff, therefore, felt that a new approach was essential
that the people should be selected for the helping professions, out of
ghetto areas, not the middle class, with the hope that these indigenous
people would invoke needed social and institutional changes. Foitunately,
one member of the staff has long been a resident in the neighboihood, and
is known among many people in Boyle Heights. It is rare for a project to
have on tap a person who is familiar with the history and problems of the
people just outside the door.

Many Los Angeles people from wealthy neighborhoods have visited
Mexico, but have not heard of Boyle Heights only a few miles away. They
would feel as out of place there, as in a foreign country. It is a distance
and contrast that cannot be overemphasized, for Los Angeles sprawls over
an area of about 400 square miles. Among its eight million people, some
epitomize the wealthiest in suburban living, while close to a million repre-
sent the most hopeless isolation of poverty. The facade of poverty in Los
Angeles is prettier than that of Eastern cities because there are houses and
trees, but the isolation is more exaggerated because people who do not own
cars cannot leave then neighborhoods. Beaches, museums, plays, and con-
certs, taken for granted by middle-class youth, cannot be reached without
some form of transportation. It is worth repeating that jobs and schools
are similarly out of reach for those who cannot afford cars. Thus, the
misleading facade of small white houses and palm trees belies a greater
isolation and deeper abandonment than the visual filth of a New York
City "slum."

Most of the institutions that might change the viscious cycle of mental
illness and misery that takes place in the ghetto are also too far away. The
people who make the decisions allocating money and setting up basic
requirements for housing, health, and education are far removed from the
realities of the poor, particularly the poor minorities, the blacks, Chicanos
(the current local term for socially active Mexican-Americans) of the West
and Indians throughout the country. Middle-income Americans hear sta-
tistics about Government budgets and visualize free, tax-supported educa-
tion that is available to all, as well as free clinics in hospitals, that make
medical care accessible to everyone. The services are available on paper. If
they are to be made a reality for the people who need them, it will happen
when ghetto people themselves, have the education and credentials to
attain positions of authority. As all the great voices from the depths of
ghetto life have tried to explain, the forces of jungle survival bruise and
also strengthen a person, thus creating an individual for whom the games
of middle-class education are particularly difficult. The very people who
must lead ghetto improvements are therefore not easy to mould; they
have needs that are not easily met.

117

Recruiting the Model Trainee

In discussions with a community psychiatrist, a psychologist, and social
workers, Mr. Parks and his staff had evolved a profile of desnable traits -a
checklist that might be used in inteiviewing prospective trainees. They
wanted youngsters who were dissatisfied, but not passively disgi untied,
students who wished to bring about change, yet who had attiactive per-
sonalities They needed young people with fire who wanted to learn rather
than destroy; people with a sense of justice and motivation. How did one
find a young man in the ghetto with fire and talent who had not been loo
embittered, beaten, legally entangled, mentally warped, or addicted lo
drugs? The young people had to have developed the strength to survive in
the ghetto. They had to be people who knew how to fight, were wise to
every con, and canny in the brutal realities of their own neighborhoods.
At the same time, they were going to be asked to incorporate a new style
of living, adopt a genteel mannei, and survive in another world of middle-
class college life, learning the subtleties of bureaucracy, and the language
of the affluent woild. They had to be people who could survive 1.1
both life styles simultaneously. These qualities should strike the leader as
remarkable, given the environment from which the trainees were sought
Ft is astonishing that people with such qualifications were actually found,
Most of them had been in trouble with the law. Some had been involved
with drugs. Any youngster with the spirit and independence of mind to
become a leader was automatically too spirited to be considered desirable
by teachers and educational authorities. Yet, it was precisely the school
"troublemakers," the questioners, fighters, activists, who were the objects
of the search.

The Opeiation Hope staff began their quest by asking for refenals fiom
social workers and teachers. After the Watts riot, Central Los Angeles was
inundated with community programs. Nonetheless, when the staff looked
for applicants from the Educational Clearinghouses, from schools, social
agencies, or ethnic organizations, these agencies could not pioducc a single
person to fit the criteria. In general, schools and agencies were antago-
nistic to the aggiessiveness and spirit that would qualify a student, and in
the end they did not help at all.

Not many social workers or teachers actually live in Boyle Heights or
other ghetto areas. The ghetto's daily rhythm begins with an immigration
and ends with an exodus. From the arrival of milkmen in the morning, the
police, the teachers, storekeepers, bankers, and social workers flow into
the area. In mid-afternoon, when the teachers begin driving to their homes
20 and 30 miles away, the exodus begins. In this ebb and flow, the teacher
or social worker is, in fact, a stranger, a visitor, who does not know the
families of the children he teaches or people he tries to help.

After six months of searching the network of social agencies and the
neighborhoods, the staff of Operation Hope had to change its initial image
of a potential leader and the means of finding him. Many of the young
people with leadership qualities had been so emotionally damaged and
brutalized during childhood that they could not remain sensitive to other

118

people, Often they were withdrawn and suspicious. The staff began to
look at the habits of potential candidates, seeking people who had spent
time helping others. Ultimately they did find anomalous people, a man
from Watts who was teaching youngsters by coaching them in -sports, u girl
who had worked with Head 'Start. After interviewing some 60 people, fom
prospective students were found, and through them a gtiivepine was begun
that finally attracted some 200 others. The kinds of students who had
unbroken spirits, who wished to help others, and could use their anger at
social injustice for constructive change were to be found on street corners,
in picket lines or community demonstrations, not in welfaie agencies.

After months of recruiting, the staff found 33 qualified applicants, bill
they had funds for only 20. They selected the people who seemed to have
the rarest combination of qualities, leadership ability, and sensitivity lo
other people. Now they were asking these students to hold their auger and
their action in abeyance, to take on a middle-class life style, and to posl-
pone their effective action until - perhaps seven or eight ycms later- (hey
had their credentials. It meant a long-term commitment to mi educational
program, postponing rewards, and suppressing many of their normal feel-
ings. At a time when many of the disgruntled youth of the United States
had given up hope of working through institutions, these students were
asked to gamble on the hope that they could later constructively change
conditions through the instruments of the establishment. This is some
measure of the remarkable men and women who have been chosen to
participate. To them, indeed, the program seemed unreal at (list. As one
man said,

"At fust, I thought you people were from Mars, then I decided
you were the 'fuzz' trying to find out what minority people were
doing. Right now I do not know what your bag is because I know
nobody gives a person money to go to school, but it's a groovy
program, and I am making B's for the first time in my life."

The Col leges

Originally, the staff had planned to pick 20 trainees and supply them
with money enough to graduate from a State or city college, while offer-
ing them special courses in social sciences and remedial work. It was soon
apparent that the tax-supported schools were not amenable to these stu-
dents despite their facilities. Theoretically, California offers virtually free
education for all qualified residents. Actually, the State colleges charge
about $150 a year and some fees are over $200, an amount that is formid-
able to these students. In addition, books cost as much as $125 a year
State universities charge $775 per year plus special fees and books.

Most of the ghetto youngsters have missed classes In order to work or
have dropped put of high school in discouragement; one girl dropped oui
of high school two weeks before graduation, and nobody in her school
acted to get her a diploma. Many others have been flunked out for ques-
tioning the teacher or expressing opinions. Since the State universities

admit only from those students in the top 12 percent of the high school
giaduates, while the State colleges accept students from the top Nurd,
most ghetto students could not qualify, however innately intelligent they
were. State colleges may admit 2 percent of incoming classes as members
of minority groups and another 2 percent who show promise despite low
test scores. On the other hand community colleges take anyone with a
high school diploma. A few of the trainees had already entered local
two-year colleges on theii own, but the administrative bureaucracy was so
frustiating that after two-and-a-half years one student had never hncl tfie
opportunity to discuss his curriculum with an advisor. He had taken end-
less courses and yet had not fulfilled the requirements to giiidtciti:.
Another student had piled up 30 course credits without completing any of
the basic couises to graduate. At the largei schools, the State colleges, Hie
coldness and rigidity of the administiation made the students fed they
were not wanted. The unfamiliar routine of application forms and inter-
views lesemble social welfare agencies. As one student had said about
bureaucracy, "The place is designed to frustrate you. By the time you
explain what your problem is to a number of people, answer endless
questions regarding why you had the problem, fill out endless papers -iind
the waiting is awful - you will be told you will be seen at such and suit
time by Mr. so and so, and you wait some more."

It was finally necessary for the Operation Hope staff to negotiate
scholarships for their trainees at small private institutions, and the s(n-
dents were finally placed at Occidental College, Pepperdinc, Wittier,
Marymount, USC, and other schools around the metropolitan area. Tlic
students had to be helped through high school equivalency exams, the
red-tape of application foims and transcripts, and enrollment. They
needed additional funds for books, and continuous emotional support,
One college administiative official told Operation Hope staff: "We will
consider applicants from your program as long as you are available to give
support. From our experience minority students will not use school
counselors within the institution."

The process of educating young men and women may seem a straight-
forward matter of funding and organization but it is a subtle and ramified
undertaking. Unless helped with all the exigencies of survival, with Jiving
needs, books, and emergencies, the student would have to be absorbed by
the all-demanding hustle to exist. Although studying may ultimately liber-
ate the student, the first impact is one of extreme tension. Students liavt
to survive not only at home, but in a new environment that is hostile, on a
campus whose very philosophy is antagonistic to them. For instance,
Chicane students are affronted to find that their hero, Cesar Chavez, is
considered a communist agitator by many students, and the black stu-
dents are amazed to learn that the charismatic Malcolm X is considered
merely a criminal by middle-class whites. This is tantamount to sending a
devout Christian among people who call Jesus a communist provocateur.
On campus the values, dress, language, and social life are all unfamiliar and
somewhat threatening. Moreover, the average college student is still partly
a child, involved and dependent upon his family, while the ghetto student

120

has been a street-wise adult for some years, depending upon his wits, often
supporting or helping to support a family. For the white middle-class
student, college may be a slight strain, a first venture away from the
comfortable family enclaves. For the ghetto student it is another country,
one in which he is scorned because he does not know the school game, a
woild in which he feels he hostility and cruelty of people who have been
comfortable all their lives.

One gifted student in Operation Hope commented that money to facil-
itate college was the most important contiibution any project could have
made to his life. No other Federal piogram operates precisely this way.
Still, in attempting to help these students through college, theOpeiation
Hope experience emphasizes how wide is the breach between rich and
poor in America today.

Operation Hope

The staff and directors' offices at Operation Hope are open at all times
for students, and a "hot line" telephone is maintained at night and on
weekends. At least one staff member is always available to help with
psychiatric emergencies, health problems, draft exigencies, or legal pro-
blems. If a staff member vanishes for several hours, or remains at his office
until late at night, it usually means that a student is in a crisis, These
students acquired adult problems before they entered college. One
21-year-old, for example, needed medical and legal counseling concerning
divorce and the death of his 2-year-old child. Draft counseling is another
important need. Since most of the men have jobs and families, they can-
not finance the 30 hours of couise time that would qualify them for
student deferments. Thus, consultations and appeals to local draft boards
are necessary staff functions. These students also need advice about
courses, professors, and examinations. They require coaching in the tech-
niques of studying, paper writing, and exam taking. All of these areas are
routinely part of staff guidance.

These students had been so intellectually and emotionally starved that
they devoured attention. None of them had enjoyed the kinds of family
or social contacts that would develop intellectual curiosity. Indeed, most
of them had such battered images of themselves they were afraid to ask
questions in college classes. Many of them needed someone with whom
they could discuss the Viet Nam War, the tax system, the meaning of the
moon shot, or racial discrimination. One student who began the program
with a poor self-image, put it this way: "Now I can question, but I lack
power, I've learned how to conform enough to get by ... ." The staff,
who are black, white, and Chicano, are by turns the surrogate parents,
doctors, lawyers, philosophers, and companions to whom the students can
turn.

Since students live in perpetual crisis, staff members know there will
never be a weekend without its emergency, no night without a call from a
panicked student. The services of Operation Hope span the gamut from

121

counseling and financial assistance to aid in getting things done. This may
mean calling a medical doctor, getting a family welfare check expedited,
or getting a family member admitted into a hospital. Any middle-class
person takes for granted that he can lift a telephone to accomplish such
things, but these needs can become major crises for the ghetto person. The
students needed these reality services, as well as step-by-step advice on
how to dress, how to behave at an interview, and lessons in the social
amenities, the things white students take for granted. They needed lessons
in speech, training in group speaking. Many of them needed to learn how
to write, to organize papers and express their ideas cleaily. Because the
campus atmosphere is cold and unfamiliar, the one comfortable phice
where all the participants could meet has been at classes offered by Opera-
tion Hope.

Campus lectures, particularly in the social sciences, have frustrated and
angered these students. Often the courses seemed inelevant or instructors
offered opinions that sounded incredibly naive to the ears of these ex-
perienced, old-young men and women.

One of the important needs of the students was a sense of ethnic
identification and history, and a forum where they could freely raise
questions about universal social ideas. Courses in black or Chicano history
are often badly taught in colleges Indeed, these courses often are taught
with a condescending point of view that insults the student who is
attempting to learn about his own heritage. Consequently, such courses
have been taught at the Operation Hope headquarters. During the fall
semester of 1 970, June Moore of the UCLA School of Social Work, taught
a course on the history and philosophy of social work institutions in the
United States. The informality of such classes permits the students lo
interact with each other and with teachers in informal discussion and
comradery that is lacking in most of their lives.

Involuntarily most of these students are loners. During the seven-eiglil
years when they are holding jobs and also studying, they are under un-
usual strain. They need the staff help in homemaking, in finding an apart-
ment, buying a car or filing insurance. Many of them have debts, often
from the illegal claims of door-to-door salesmen.

The small "Hope" offices, with their modest library and open doors,
provide a kind of home base for these students where they can "rap" with
the staff, and admit their real feelings, or ask for help. It is the only such
place in their lives. Not at home, at friends' houses, nor in school or clinics
can they find support and assistance.

By contrast with the impersonal, bureaucratic schools, and institutions,

the staff has an open-door policy which means nobody is ever too busy to

HiHpnt wnen he drops in. There is never a wait. Since the students

r *he program, staff members are always on hand to help

+o provide coffee and a snack. Students rely on tin's

jring exam periods, or when a term paper is due.
wvmg db mey do, in two very separate life styles, the students are
under extreme tension. As one Chicano put it, "This project is great but,
man, am I having problems. For the first time I am learning to think, I

122

listen and try to figure out what's being said. I'm so excited about hearing
people, but I'm bored with my family. I'm finding my girlfiiend a drag,
we can't talk anymore. And the guys I know, well I haven't seen them in
weeks. Man, is this lonely."

For ghetto students in the middle-class colleges the strain of a double
life is continuous. Theie is an internal price they pay foi living in two
worlds. Part of the price may be in terms of physical tensions, anxieties,
or general loss of self-confidence. Realistically, life is a daily struggle. In
Older to study, the ghetto student leads a tired and lonely life. As one
man from Watts described his schedule, he attends school in the morning,
works afternoons, sleeps in the early evenings, and studies all night, after
his family goes to bed. He has no time or energy for close peisonal
relationships The stress is such that he must keep his feelings undci
control at all times. It is not surprising that he lias a stomach ulcer. Many
of the students in the project suffer from psychosomatic ailments al-
though they are very young. Some have hives, or colitis, ulcers, or hyper-
tension. A few suffer from insomnia, panic, and nightmares. When a stu-
dent leaves the personal and familiar world of the barrio for the college
environment, he leaves his status and sense of acceptance, and must adapt
to a highly impersonal structure. He must be careful in his dress, guarded
in his manner and speech. When he leaves campus he again returns to a
wife (01 a woman may leturn to a husband) whose standards and outlook
are the antithesis of all that the campus represents.

To the average student on campus, the ghetto student seems of a dif-
ferent breed. Tough, street-wise, analytical, many of them have experi-
enced jails, and brutalities that the other students have only lead about.
They have suffered hunger, exhaustion, and have seen their friends turn to
alcohol or drugs. They have had to use cunning and muscle to survive. An
affluent youngster arrives in college certain that theft and alcoholism arc
signs of weakness, inferiority. Neither students nor their teachcis realize
that the crimes they deplore may be tlie vehicles of survival in the ghetto.
The kid who doesn't suivivc that jungle docs not get to enter a project
that will help him through college. A person who has not used his cunning
to suivive will ultimately despair, entering the no-mans-lancl of inertia and
drugs. A ghetto student who seems angry, suppressed, cynical, and hard
can only be understood if his armor of hostility and suspicion are also
understood; young people who have lived in the ghetto or barrio foi 20
years have developed defenses that allow them to cope with drug users,
police harrassment, with rejections by schools, social agencies and hos-
pitals. Most of them have needed to devise a hard surface, never revealing
their feelings to anyone, never trusting anyone, always analyzing a situa-
tion for possible traps. Since they have had to hustle for money all then
lives, they may be busy trying to beat the welfaie system, or avoiding
someone else's con game. Typically, as children, some of the students have
been shifted from one foster home to another, in the manner of one girl
whose mother was declared mentally incompetent. At age 5 she recalled
being put to bed in a relative's house where, as she hugged the pillow, she
found a gun. For her, school was & relief from washing, cooking, cleaning,

123

and being beaten with a belt buckle arbitrarily. At 14, she and a friend
weie put in jail for stealing Anothci girl was abandoned at the age of 14
and left alone in an apartment. In ordei not to be evicted she rented
looms, but when she tued to go on welfare to raise the rent money, the
casework ei told hei to thiow a rent patty and charge admission which
could be used to pay the month's rent- Two weeks before high school
graduation the girl quit school and worked at odd jobs foi six years before
anyone suggested that she had a good mind and might go to college. To
the son of a migiant worker, an avid readei in a large family, the possibil-
ity of a college education seemed like the wildest of dreams as he worked
in the fields year after yeat , missing months of school.

To classmates or outsiders these students might seem cool and brusque,
with an appearance of self-confidence. Actually, they of ten live in a stole
of conflict that is bordering on panic. They know that they must be many
times as strong, intelligent, and controlled as their peers in school, and
that they are expected to endure and prevail in a manner that would be
impossible for many adults, Raised by television, these students havea
sophistication that is underestimated by their peeis. They know a good
deal about the way they look to the other side, and they are familial with
the comforts of most Americans, with the social work jargon in which
ghetto disadvantages are discussed, and they realize that they arc put
down for qualities that are really the product of their environment, They
must survive in two worlds, and ate expected to be idealistic when they
are surrounded by despaii and futility and to be willing to give some part
of their lives to change a society that would not ordinarily give them a
chance.

By adolescence most ghetto children have given up. They see no alter-
natives and no future. In the southern section of Central Los Angeles they
are likely to turn to alcohol, while the eastern section they would more
likely use drugs. Many youngsters between 15 and 25 try or actually do
commit suicide. Theie is nobody to help them. Their families are dis-
rupted. Teachers who have taught in these neighborhoods for 20 years,
but who live elsewhere, do not even know the parents of their students.
As one black student explained, high school students are never told thai
they have possibilities of further education. Typically, one student stated,
"Nobody evci told me 1 was college material." One very able athlete, wilh
honor roll marks, wasted two years after high school not knowing that he
could have attended a tax-supported college without paying thousands of
dollars. Scholarship offers had been made through his high school because
of his athletic ability, but he had never been informed by his advisors.
Another girl wasted six years in dead-end jobs because she similarly did
not know college was even a remote possibility. Isolated from the infor-
mation that every middle-income person takes for granted, and misled by
their own schools, the best of these youngsters have a bitter tale to tell.
By age 24 many of them say they feel 50, On a college campus this is no
advantage.

124

Impact

The first most important impact of Operation Hope was the fact that
the students were sticking out their difficult academic programs. Pre-
viously, when faced with crises they had withdrawn from college. In this
project, bolstered by the staff, they weatheied continual crises and man-
aged to go on studying. Even success was not without conflict. One girl
who had, with considerable help made a C in English, remarked, "Well, I
guess this means I've given up. I'm conforming, the fight is over, I've
entered the system - what can I possibly do to make anything better for
myself or anybody else. I did what the instiuctoi wanted and I passed, he
wasn't the least interested in what I am, what I feel, or what I want"

The students' marks have steadily improved and several begun to be
eloquent spokesmen, writing reasoned essays on the issues close to theii
hearts. Drug abuse is a topic on which these students have more than usual
understanding while the illegality of the drugs generates the crime for
which they aie punished, it seems clear to them that there is little concern
foi actually helping addicts; there aie no adequate medical or community-
based programs available. One of the students wiote:

"How rational would it be to appoint a doctor as a judge in a
court of law? It would seem that this wouldn't be very rational at all,
for although he might know medicine he would be quite unfamiliar
with the mechanics of law. Taken in this light the question is laised:
Why are administratois and law enforcement officials considered to
be more knowledgeable in the aiea of a medical and social pioblem
such as drug addiction?"

The participants are impressive people. Selected for qualities of leader-
ship, they also have displayed considerable aesthetic talents. Many of
them write excellent poetry, sing, and paint. Candid and cynical, they are
at the same time deep and more idealistic. Although the men and women
supported by the project have come from different homes, their stories
are painfully similar. Most of them come from families on relief. In grade
school they were demeaned by their teachers. They weie often absent
from school, and were arrested at an early age for some theft such as
gouging nickles from a parking meter or for gang fighting, By 17 the
"average" young man had been in a house of correction, may have gotten
a girlfriend pregnant and have had to marry, spending a year or so washing
cars or doing clean-up work in a factory. By age 20, they could see
themselves at a dead end, with the odds stacked against them. Yet they
had intense intellectual curiosity, and some inner stature that was not
altered by the meanness of life around them. As one girl remarked:

"A long time ago I made up my mind I would rather die than to
treat people as I had been treated, I knew what it meant to be
disliked for nothing and always used, I suppose having compassion
has been the only thing that has saved me from hating the world."

One Chicano student recalls being "busted" on a marijuana charge.
When he appeared in court he was sent to jail, while a white kid on the
identical charge was released, "Jail is a school where you learn many

125

things," His experience in the army taught him yet more about discrimina-
tion as he watched some episodes of mistreatment of the Vietnamese.
Now, on his leturn, he had a mission, to create changes, and as a beginning
he has been tutoring kids in an East Los Angeles Paiole Centei. His own
image of his past life was succinctly expressed in a newspaper essay he
wrote:

"I cannot help but bring up a thought that would enter my mind
every once in a while in jail. I would picture myself as a peison trying
to leain how to swim (cope with every clay life problems) in a deep,
dark, cold sea (the Barrio), and finding myself submerging because of
my inability to swim. Along comes a lifeguard (institution: rehabili-
tation) and pulls me up on the platform (institution) and shows me
how to swim (rehabilitates me) and simply throws me back into the
deep, dark, cold sea (Barrio) with little knowledge as to whether I
had developed the ability to swim. . . ."

Although many of the students in the program came from disrupted or
nonexistent families, a few have shown the stability that comes from tacit
family support. A student from Watts, who has watched his friends de-
teriorate along the route of drugs and apathy, somehow manages lo
study while his family sleeps. He is lucky enough to have a family thai is
together. He also manages to continue athletics, coaching younger chil-
dren in school subjects and basketball. He manages to live without close
friends, without a confidante or wife, without time for himself. He works
in the afternoon, sleeps until midnight, studies until dawn, and goes to
school in the morning. When he finally has his credentials he wants to
become a poverty lawyer. He has every reason a man could have to play
an escape from the poor, enclosed, drugged, hopeless world in which his
life has been lived -but he is not content to do as many others have, to gel
his skiils, "go make his pile" and separate himself. Another student wlio
intends to be a lawyer, commented, "I don't look for big hopes anymore
but I've learned I can do something and I'm not going to stop fighting,"

These are the people who can provide community mental health serv-
ices, legal aid and instruction from within their communities. These are
the new people who are needed in the mental health professions. The
attitude of these students offers some measure of the urgent need for
change. One girl in the program, who had been quite upset, was asked
whether she would like to see a campus psychiatrist.

"For what? I have had two of those, a psychiatrist and a social
worker. One told me I had problems with authority when I was late
for an appointment, so I came early the next time and was told I was
too anxious. When I got the social worker, I tried to be exactly on
time and you know what she told me? I was compulsive. How do
you win with that kind of closed system? The only thing I could do
is drop dead. They had no awareness that I had no car, had to beg a
ride most of the time, and had no money all of the time."

Another student told a conference of mental health professions ami
social workers, "You mean well, but you really can't help with the kind of
problems we have."

126

Colleges and piofessional schools could altei this impasse by training
low-income people and by developing techniques for working with the
pool, instead of sending emissaries by day to help the pooi adjust to being
poor. By drawing professionals from within poor communities, social
work and the mental health professions might promote community activ-
ity, patticipation in city planning, in education, and in the establishment
of agencies for health caie and housing. Such participation would enhance
a constitutive process of change that is a matter of ethics as much as
money.

It has been estimated that some 200,000 youngsters who are in the top
of the U.S. population (judged by ability) will never acquire the means to
go to college. These are the potential leaders who could rage the real war
against poverty. Operation Hope began with the idea that the selection
and training of young leaders from the ghetto would offer a model for
i emitting and educating a new kind of personnel in social work profes-
sions. These would be people who could inject the energy of personal
involvement and leality understanding into helping professions such as
social work.

In its short existence, Operation Hope has already shown that there are
potential leaders within the ghetto, whose ideals and intellect could make
them the agents of peaceful social change. The education of such people is
difficult because it must compensate for so much: the students' lack of
family, confidence, gamesmanship, academic skills, medical, psycho-
logical, and legal aid - in addition to money. No program to aid and
educate potential young talents from the ghetto can realistically meet
these needs, nor leave them to the universities and colleges. The ghetto
environment so rapidly destroys trust and normal human emotions that
the Operation Hope staff has begun to see that its program should begin
with high school students. No person can cope with life and death crises in
his family day after day and yet devote himself to learning, yet this is
what is expected of the ghetto student. As early as possible, students
should be relieved of some of the reality conflicts. In addition, paucity of
funds leaves huge gaps in their education; unlike many of their peers they
never had the opportunity to relate to man's universal problems through
visual art, drama, or music. Most of these students have seen little of the
world outside their neighborhoods. Yet, as periodic analyses show, they
are aware of the culture outside, and as their education has made them
more perceptive, self-confident, and articulate, they are creating art of
their own.

This program while only a miniscule experiment in the overall context
of America today, has shown that a remarkable transformation can be
made. It has helped to liberate a new kind of student. Because they have
experienced more of life than most men of 50 when they arrive in col-
lege, these "youths" are not passive intellectuals. They approach the arts
and social sciences with a personal sense of involvement and judgment.
They do not merely accept; they question. If they are old in outlook,
their minds are capacious, Their brains are young. They have vitality,
compassion, motivation. The result is a kind of genius, a combination of

127

wisdom and brightness, a depth that is not often found. Trivin will no(
distract them, nor will minor hardships present obstacles. They have a
highly developed sense of justice, and an eagerness for history, and indeed,
at 23 and 24, they resemble the idealistic young men of the American
Revolution who, at the same age, were writing the Declaration of Inde-
pendence and attempting to fashion a constitution. The qualities that liavc
been elicited in these men and women include a kind of idealistic willing-
ness to gamble on the future. Many young people have given up, and fee!
that the establishment will not permit changes for the benefit of the
left-out Americans. The participants of Operation Hope have no ceitaiuly
that change is possible, but they have committed themselves to tiy, With
their nascent social genius they may indeed tiansform the helping profes-
sions from ineffectual stopgap procedures, to a realistic confrontation of
the social problems behind individual ills. They may infuse new creativity
and strength into the mental health piofessions by enteiing as a new
kind of professional - one who is fearless rather than cowed, analytical
rather than sentimental, one who is undeludecl about the detailed nature
of the ghetto and incapable of forgetting its manner of shaping people.
Unlike many social workers in today's piofessional schools, these people
will be able to function in the poorest levels of society as well as among
middle-class professionals. They will be able to offer the kind of innova-
tion that stems from understanding where one started and what one's ends
are. If their affluent classmates are currently unsure of where they aio
going, these students are seizing their intellectual honing as a liberation,
for this is the only way out of the real hopelessness in whicJi they have
lived. The template offered by Operation Hope is there to be extended
and copied throughout the countiy,

Training Grant. MH 11513

Dates of Interviews. September and October 1970

128

You are the bows from which your children as
living arrows are sent forth.

-Kahlil Gibran

Investigators:

Jean W. MacFarlane, Ph.D.

John A. Clausen, Ph.D.

Institute of Human Development

University of California

Berkeley, California

Prepared by:

Herbert Yahraes

Introduction and Summary

The usual way of studying how the circumstances of childhood affect
the characteristics of adulthood is to start with the grown person and try
to work back. So-called longitudinal studies, though, begin with the child.
Three of these, directed by the Institute of Human Development of the
University of California, Berkeley, and supported recently with NIMH
help, are now approximately 40 years old and probably offer the richest
collection of data ever assembled on human beings over a long period.

The projects have attempted to answer such questions as:

• Do personality and intelligence change during the years or remain
constant? To what extent are they related to a person's very early experi-
ences?

• How is the mental health of an adult related to his life at home and
to other influences during childhood and adolescence?

• "What factors contribute to an adult's attitudes, achievement, psycho-
logical health?

In the beginning each study had its own set of objectives:
The Guidance Study was primarily interested in personality develop-
ment. It began studying its subjects as infants. There were 252 of them -
every third child bom in Berkeley over an 18-month period beginning
January 1928. The children were weighed, measured, tested, interviewed,
and observed at various times through their eighteenth year. Special atten-
tion was given to their life at home during the preschool years. Informa-
tion about them was obtained also from their parents, brothers and sisters,
teachers, and classmates, At 30, when they were rearing children of their
own, 167 of them were studied again. (The project got its name from one
of the original objectives: to learn whether or not psychological guidance
offered to parents would lead to better mental health for their children as

131

adults.) The project's director until recently was Jean Walker MacFailuiw,
Ph.D., who is now Professor of Psychology and Research Psychologist,
emeritus, at the Institute of Human Development and is still working on
the study. The present director is Marjorie P. Honzik, Ph. D,, Research
Psychologist and Lecturer. in Psychology.

The Berkeley Growth Study, which also began m 1 928, has been pai lie-
ularly interested in physical and mental growth. Its original sarnpla com-
prised 61 healthy hospital-bom babies, who were studied while they were
still in the hospital and then every month until they were 15 months old.
After that, they were studied every 3 months until they were 3 years old,
then eveiy 6 months until they were 18. They were examined and inter-
viewed again when they were 21, 26, and 36 years old. The sample now
numbers 54. The project's dhector is Nancy Bayley, Ph.D., research
psychologist.

The Oakland Growth Study has been concerned with the effect of
adolesence-the physical and psychological changes occurring then, ;ui;l
the accompanying attitudes and behavior-upon later life. The study be-
gan in 1 93 1 with the fifth grade pupils of five Oakland, California, schools
who would be entering the same junior high school. There were 200 of
these. They were studied intensively-through measurements, tests, obser-
vation, self-reports, ratings by classmates and teachers, and other means-
through the six years of junior and senior high school. At graduation, 165
were still in the group. Follow-up studies made 15, 20, and 26 years ta
have reached as many as 123. The study began under the direction of
Harold E. Jones, Ph.D., and Herbert R. Stolz, M.D. Its director since 1%&
has been John A. Clausen, Ph.D., professor of sociology and lescarcli
sociologist.

Almost all of the subjects were white. In the Guidance Study, though, 3
percent were Negro, a proportion representative of the community's
Negro population in 1928. Though all socioeconomic levels were repre-
sented, the families of the subjects were predominantly middle class.

In some respects-in a ntliro pome tries, intelligence tests, certain person-
ality measures-the three studies overlapped, making it possible for one set
of findings to be compared to another. In some of the analyses now being
made, data from more than one study are used. AH the subjects, it is
planned, will be followed through life, with the principal research interest
fmm nmw nn i^g factors connected with the aging process.

particular, much of that obtained during the most recent follow-ups— is
still being analyzed.

This report is concerned with the most recent findings (roughly, those
of the past five years), which have a number of implications for mental
health. For example, findings of the Berkeley Growth Study indicate that
the level of intelligence continues to rise until at least the age of 36. This
was true for persons with relatively low IQ's as well as for those with
relatively high IQ's, and for persons from all socioeconomic levels. The re-
sults suggest that, on the average, adults reach their peak of mental ability
at a later age than has often been supposed. This age for the group under
study will be determined in future follow-ups. But the work already bears
out what individuals often discover for themselves: people from whatever
level retain for many years the basic brainpower to learn new ways, new
skills. The study also points to a number of environmental factors in very
early life-parental attitudes among them—that seem to influence a per-
son's IQ for years afterward and perhaps permanently. When knowledge
of these factors is widely held and applied, the mental competency of our
population can be expected to rise.

By studying the same people over such long periods, the investigators
have also been able to demonstrate that the mental well-being of adults-
those in their thirties, at any rate—is related to certain childhood charac-
teristics and even to events during infancy. Those people who came down
with psychosomatic ailments in adulthood, or became mentally ill, or
showed other evidence of psychological trouble had been reporting worse
than average adjustments to life even as 1 1 -year-olds. And when these
individuals were less than 2 years old, their families had shown more than
average amount of disturbance. One of the investigators reports that a
major factor in the onset of psychosomatic illness, (which may include to
some extent almost all illness) seems to be a tendency towaid depression
reaching far back into childhood.

Looking into the records of their subjects as children, the investigators
have also identified certain characteristics held in common by those who
later took up smoking and continued to smoke. And the person most
likely to become a problem drinker, preliminary evidence suggests, could
have been spotted while he was still in high school.

Most of the recent findings should be useful in detecting children who
need special help if they are able to realize their intellectual potential and
be able to lead a satisfying life. But the Guidance Study staff warns that in
predicting at 18 a person's psychological health at 30 it is easy to over-
emphasize the effect of traumatic experiences and of the youngster's
response to them. Many of the most stable young adults were those for
whom the staff had forecast a poor outcome. Disturbing experiences and
behavior during adolescence had apparently been, in these cases, maturing.
Dr. MacFarlane concludes that pediatricians and other doctors should look
for strengths in children and not just pathology.

On the other hand, many of the Guidance Study youngsters who had
been highly popular in high school failed to live up to their potential-by
the age of 30, anyway-and were puzzled and dissatisfied, perhaps because

133

life had been too easy for them or because they had poured their adoles-
cent eneigies into maintaining an image of success.

Some of the other findings covered in the following sections are
summed up here:

• Certain personality traits become established early and persist, The
child who at five was either reserved and shy or expressive and gay tended
to show the same characteiistics at 16. The child who was either reaclne
and explosive or calm and phlegmatic at five was likely to be the same at
16.

• Personality is correlated with intelligence, at least in the case of boys.
Adolescent boys with high IQ's were generally described as friendly ,
social, and independent-as they had been since the age of four. There was
a positive relationship between these characteristics and the IQ in thease
of girls, too, but it was not statistically significant. Men of high inlclli-
gence were less likely than men of low intelligence to be hostile, self-
pitying, or impatient.

• Frequently the effects of a given environmental circumstance art
quite different for one sex than for the other (as they are for subjects witJi
different tempeiaments). The differences are found in the devolopmeii!
of both intelligence and personality.

• The speed of development during childhood, which presumably is
determined by both genetic and environmental factors, seems to influence
peisonality characteristics into adulthood. The early talkers (generally
those who had received more than the usual amount of parental attention
during infancy) were more introspective as adults, perhaps because
language rather than action had always been for them the favored re-
sponse pattern. Boys who matured early had an easier time during adoles-
cence, a highei income at 30, and a strong tendency to conform.

How the Early Environment Affects Intellectual Development

A child's experiences during his very early years seem to affect tlie lowl
of his intelligence for at least many years afterward. The main conclusions
on this subject, summarized here, stem from analyses either by Uoiirik
and her associates in the Guidance Study or by Bayley and her associates
in the Berkeley Growth Study.

Socioeconomic conditions. The socioeconomic level of the parents was
unrelated to the intelligence of their children during the first IS or 24
months. But after that, and particularly after the age of five years, the
relationship became fairly strong: in general, the lower the parents' status
when the child was young, the lower the children's intelligence as mew-
ured by mental tests. This was true through at least the age of 18, which is
as far as this part of the analysis has gone.

Family income during the early preschool years was significantly re-
lated to the children's scores on mental tests at most points up through
the age of 15. At the age of 30, it was significantly related to the sons'
scores but not the daughters'. Superior play facilities during early cfirW-
hood were also related to the IQ scores later on.

Parental attitudes toward education. Both boys and girls whose parents
had expressed concern when the children weie babies that they get a good
education were likely to make higher IQ scores at all ages through 30 than
the children of parents who had shown little or no concern. Boys were
motivated by their mothers' concern; girls, by their fathers'.

Parents' marital adjustment. When the home atmosphere very early in
life had been one of parental harmony, or at least of lack of conflict, a
girl's mental test performance— but not a boy's—was likely to be signifi-
cantly bettei right on up into adulthood.

Maternal characteristics. Mothers who appeared worrisome, tense,
highly active, and energetic had childien who were more likely than other
children to score high on mental tests through the age of 30. The child's
need for tactual, auditory, and visual stimulation, Honzik notes, aie best
met in the family where the mother is icsponsivc and actively concerned
with the infant's welfare, even to the point of being worrisome about him.
A genetic factor may be at work, too, for the worrisome mother tended
also to be the better educated mother, and educational level in this
sample, the investigators think, was evidence of native ability as well as of
stimulation during childhood.

For the boys, the one best predictor of test peiformance during the
period between 8 and 1 8 was the closeness of the mother-son relationship
as rated at 21 months. Apparently, it is verbal competence that is fostered
by a close lelationship. For at 1 8, a boy's verbal IQ was much more likely
to be high if the early relationship had been close; his performance IQ was
unaffected. At 30, when the test was for performance, not verbal ability,
there was no correlation.

Boys whose mothers were anxious, irritable, strict, and punitive toward
them during their first few years tended to have IQ's below average during
the school years and even as adults. Maternal love or lack of it seemed to
have no effect on the mental test scores of girls. But mothers judged to be
instrusive -forever meddling in the child's activities-had daughters whose
IQ ratings through adolescence tended to be low.

Paternal characteristics. A girl's intellectual development was likely to
be increased when, during babyhood, her father had a close, warm rela-
tionship with both his wife and with her. It was the father's interest in his
daughter, rather than his expression of affection for her, that seemed to
count. The father-daughter relationship apparently influenced IQ scores
from the age of seven through adolescence. During the preschool years,
the relationship between mother and daughter was the more influential.

Childhood IQ and adult success. In the case of men the IQ through
childhood and adolescence is roughly a good indication of their success-
conventionally judged-as an adult. On the basis of the number of years
they had gone to school and on the kind of jobs they held at the age of
30, the men seemed to be achieving "pretty much in accord with their
mental abilities." This was not true of the women. But then, Bayley
points out, women in general have different educational and occupational

135

goals than men. As a group the women who classified themselves as house-
wives had had, all their lives, higher IQ's than the women who were
working for pay.

The influence of sex. As noted in the findings above, certain factor*
that influenced the IQ's of boys had little or no influence on the IQ's of
girls, and vice versa. Bayley suggests there is a genetic difference between
the sexes in the ability to resist certain environmental influences or to
recover from them. "Boys," she adds, "appear to be less able than girls to
recover from hostile, rejecting treatment; but they may also profit mow,
in the long run, from understanding loving acceptance." There also scenii
to be, Honzik reports, a sex difference in the rate of mental growth, Tfic
parents' education showed an increasing correlation with the children's IQ's
between 21 months and 1 5 years, but it became significant for girls by tit
age of three; for boys, not until five. When there were boys and girls in Hie
same family, this difference still held. The lelationship between socio-
economic status and IQ also became significant for girls earlier than For
boys. Girls' abilities, Honzik suggests, develop earlier.

The home environment in later years. The family situation was assessed
each year until the children were 1 6, In general, the relationships between
a given factor, such as parental warmth, on the one hand, and menial
development on the other were less-if they were present at all-lliai
they had been during early childhood. In other words, the family situation
when the child is very young influences mental development mow
strongly than the situation later on, when out-of-family influences in-
crease.

The Course of the IQ as People Grow Older

After the 1964 follow-up of members of the Berkeley Growth Study,
Bayley began analyzing the IQ's of the subjects at that time, when they
were 36 years old, in relation to some of the earlier IQ's-those shown at
the ages of 16, 18, 21, and 26 years. She finds that until the ageof36,al
least, the intelligence level tends to rise.

This is particularly true with the verbal scale of the test, which meas-
ures the extent of a person's vocabulary and information, his ability to
comprehend, and other factors heavily dependent on his capacity to
understand and use words. On this scale, all the subjects, show an increase
in score with age, between 16 and 36, though the rate of increase ii
slowing down.

On the performance scale of the intelligence test, there is a slight
decline after 26. The men score the same at 36 as at 26, but the women
score lower. This scale deals with the ability to perceive patterns, to
visualize the whole from some of its parts, and to manipulate objects in a
logical way.

As a group, Bayley points out, the subjects in the study have had a
better socioeconomic background than average and have scored a bow

136

average in the mental tests. However, the group does include several per-
sons whose scores during childhood were quite low, and these persons,
too, have continued to grow in intelligence.

The person with the lowest IQ was a man whose ratings were in the low
60's from the time he was 5 until 16. After that they went up, and at 36
reached 80. His performance IQ then was 92; his verbal IQ, 72. He had
not learned to read until after he was 21 .

Bayley concludes that the intellectual potential of the people in her
study remain unimpaired through 36 years and that "in the attainment of
information and word knowledge their intelligence is continuing to in-
crease." She finds some evidence— the lowered speed with which the
women in general completed some of the tests on the performance scale-
consistent with the findings of other investigators that advancing age is
accompanied by loss of speed in learning. However, she points out that
other investigatois have found evidence also that loss of speed is often
compensated for by an increase in knowledge and skill.

As people grow older, this investigator believes, they probably show
increasing resistance to learning new techniques and new ways of organ-
izing knowledge, but the extent to which these resistances are overcome
may be matters of motivation and opportunity, rather than of intelli-
gence.

She points to two men in the study who went back to school in their
thirties, after a decade in other employment, one to become a lawyer and
the other a doctor. Both have started their new careers "with bright
prospects of success." She concludes that motivation and drive and ample
time, rather than a small variation in intelligence, seem to be "the im-
portant determiners for much of learning in adults."

Do smart babies become smart children and smart adults?

Not necessarily,

Bayiey and some of her associates have been analyzing the correlations
between mental test scores in infancy (the average of several scores at 10
to 12 months) with test scores at 24 ages, running from 1 month to 36
years. For the first 2 or 3 years the relationships are close; then they fall
off rapidly. From 4 years on, there is very little relationship, particularly
in the case of the boys.

The investigators have also developed "precocity scores" in order to
observe how the precocious infants fared later on. These scores are based
on the ages at which a child first put a block together, first noticed a
pellet on the table, first said a syllable that had meaning for adults, and so
on. There are 1 1 5 items in all.

So far the analysis has been completed only for the vocabulary or
vocalization factor, comprising seven items or steps that the normal child
generally completes between the ages of 8 and 15 months. These include
expressing emotions with distinctive sounds and, later on, saying words.
During the first 3 years of life, the higher the score on this factor, the
higher the IQ, for both sexes. Then the boys' correlations drop sharply
and in most cases become negative, meaning that the higher the score on
the vocalization factor, the lower the IQ. For the girls, however, the

137

correlations remain liigh: in their case vocal precocity as a baby is a
prediction of high IQ as children and as adults.

Intelligence as Related to Behavior and Personality

The follow-up of the people in the Berkeley Growth Study when they
were 36 included a detailed personality assessment made on the basis of
the 100-item Block Q-sort. (In a Q-sort, characteristics noted during an
interview are given numerical weight by scoring each item on a scale. The
interviewer thus can say that the given quality was absent, present loa
veiy high degree, or present to one of the several degrees in between, In
this study the transcribed interviews were Q-sor£ed by the interviewer and
by two clinical psychologists.) Bayley has now analyzed the findings and
compared them with the scores made on IQ tests during the same follow
up.

The men who had been scored high on such items as impatient, negatfr
istic, self-pitying, and hostile were found to be those who in general tori
the lowest IQY The men described as critical, introspective, socially per-
ceptive, and having wide interests were those in general with the highest
IQ's. Little or no relationship was found between IQ scores on the one
hand, and on the other hand, characteristics described as either distant
and avoiding or warm, calm, and gregarious.

"The men m this sample with high intelligence," Bayley reports, "arc
best characterized as intiospective, thoughtful, and concerned with prob-
lems, meanings and values; they are men who are perceptive and have a
wide lange of interests. The least intelligent are most often found to be
impatient, prone to vent their hostilities and to project them onto
others."

The correlations between the women's [Q's and various personality
attributes are much weaker but similar in pattern. Again it is the thought-
ful, insightful person with wide interests who is more likely to score high
on the IQ tests. Women described as bland, conventional, or anxious arc
much less likely to rate high on the tests; so are women described :is
cheerful, poised, and gregarious.

Another measuie of personality styles and psychological attitudes
administered during this follow-up was the California Psychological Inven-
tory, a questionnaire designed to measure such characteristics as sociabil-
ity, self-acceptance, sense of well-being, tolerance, and responsibility. II
has 17 scales. Bayley has compared the scores made on each of these at
the age of 36 with the IQ scores at that age and also at 1 6, 1 8, 21 , and 26
years.

Certain of these characteristics appear to be significantly associated
with a high order of intelligence at all the ages studied. For men, the
clearest and most consistent associations with IQ's are socialization (refer-
ring to social maturity, integrity, rectitude); the ability to make a good
impression; potential for achievement, whether by conforming with the
group or acting independently; and intellectual efficiency. For women, the

138

cleaiest and most consistent associations aie with tolerance, potential for
achievement by acting independently, and flexibility.

Ratings on self-acceptance and self-control, qualities usually associated
with mental health, were not significantly related to intelligence in cither
men or women.

When Bayley used the scores on the subscales of the intelligence test,
she found some other provocative patterns. Little or no sex difference
appeared in the correlations between scores on the verbal-academic scales
of the IQ tests and the ratings for achievement potential, intellectual
efficiency, and inteiest in intellectual pursuits. But the other scales
pointed to marked differences. With men but not with women, for ex-
ample, the score on the picture-completion test correlated strongly with
the score on socialization; with women but not with men it correlated
with flexibility. Scores on the object-assembly test correlated with flexi-
bility, achievement potential, and intellectual efficiency with the women
but not with the men. The highest scores in arithmetic were made by the
women rating highest in feminine qualities and by the men rating lowest
in them. The highest scores in the digit-span test were made by the men
ranking high in sociability, well-being, and interest in making a good im-
pression and by the women ranking low in these characteristics.

In short, for this small sample at least, there seems to be a relationship
between intellectual processes and personality as manifested in various
social attitudes, interests, and motivations. The lelationship remains fairly
stable over the years between 16 and 36, but differs both with the nature
of the intellectual piocess and with the personality characteristic being
considered. It often differs widely between the sexes as well.

In the case of males, the investigators have also found some relations
between mental test scores throughout the 36-year period and the be-
havior and personality characteristics of the subjects dining their first
three years. Boys who were calm, responding, and happy, and who were
active after 15 months rather than before, weie more likely than the
others to have high IQ's (determined in this case only from the verbal
scale). Girls showed no clear pattern.

With females through the years, considerably fewer significant correla-
tions between IQ scores and personality latings were noted than foi males.
Bayley suggests that a girl's intellectual potential is less affected than a
man's by social and emotional factors. A girl comes into life physio-
logically tougher, it has been shown; perhaps she is by nature psycho-
logically tougher as well.

Psychological Mechanisms and the IQ

After the subjects in the Oakland Growth Study had been interviewed
at 37, Norma Haan, a psychologist, rated them for the presence of coping
and defense mechanisms. Among the coping mechanisms she includes
objectivity, logical analysis, empathy, sublimation, and tolerance of
ambiguity. Among the defense mechanisms are repression, doubt and

139

indecision, and denial of facts and feelings that would be unpleasant
self-threatening to acknowledge. The coping and defense mechanisms
counterpaits. For example, the coping partner of denial, says Haan.
concentration-the ability to set aside disturbing feelings or thoughts
order to get on with necessary tasks at hand,

In general, the adults who tended to make use of coping rather th
defense mechanisms had the highest iQ's. Further, they were the pcrso
whose IQ's between adolescence and adulthood were most likely to ha
risen. Coping, the investigator suggests, leads to the development of om
intelligence; defensiveness interfeies with one's intelligence as well as one
effectiveness.

Persistence of Personality Traits

Dr. Wanda C. Bronson, a psychologist, has begun to analyze the alt
tudes and characteristics of subjects in the Guidance Study to learn i
these become set very early or change with the years,

Between the ages of 5 and 1 6, the period covered by the analysis so fa;
she finds two persistent "behavioral dimensions." One is behavior ctac
terized at one end of the dimension as reserved, somber, shy, and at th
other end as expressive, gay, socially easy. The second dimension is a
contrast between reactive, explosive, resistive behavior at one end-calm.
phlegmatic, compliant behavior at the other end.

If a child was either reserved (somber, shy) or expressive (gay, social!)
easy) at the age of five, he was likely to be the same at 1 6. If he was cttht
reactive (explosive, resistive) or calm (phlegmatic, compliant) at five, IK
was likely to be the same at 1 6.

The reserved individual tended also to be introspective and, to a lesser
extent, anxious and socially withdrawn. In early childhood, he was likely
to be inactive and a poor eater; at 16, uncertain and uncompetitive, Tk
expressive boy was the extrovert. Girls showed only one marked dif-
ference from the boys on this behavior measure. The reserved girl leixled
to be cautious and imadventurous at all ages whereas the reserved boy wis
cautious and imadventurous only between the ages of 8 and 1 0.

In the case of the other behavior grouping, the reactive and explosive
boy tended also to be emotionally unstable, quarrelsome, and complain-
ing. In adolescence but not earlier he was also likely to be rated active an!
adventurous. Girls showed some differences. The correlation between
reactiveness and emotional instability was not significantly strong exwpl
during early childhood; the correlations between reactiveness and the
activity level, strong between the ages of 8 and 13. The reactive giil>
tended to be finicky about their food at all ages and to be exhibitionist
at 16.

Both of these attitude patterns -reserve v, expressiveness and rettctivily
v, placidity^describe characteristics that the individual brings with htoilo
every situation and that affect the environment's impact upon him. An
expressive child, for example, would be more ready to initiate or be dram

140

into an intensive relation with his mothei than a withdrawn child. A reac-
tive child, more than a placid one, would be affected by an anxious,
intrusive mother.

To what extent and through the mediation of what mechanisms these
petsistent personality traits are inherited, affected by the environment,
and developed in the interaction between heredity and environment is not
cleai. Bronson does find that the children tend to take after the parent of
the same sex and to reject or be unaffected by the characteristics of the
other parent. Expressive boys, for example, tended to have fathers of
expressive and even aggressive temperament; expressive girls, mothers of
the same type. More information on this question is expected to come
from an analysis of the children of the subjects in the study.

In another of the Berkeley studies, the boys with the loving mothers—
these were the boys most likely to have high IQ's later on-tended as
babies to be happy, inactive, and slow, Beginning about the age of 4,
though, these boys were consistently rated as independent, social, and
friendly.

Personality as Related to Speed of Development

Some of the personality traits noted in preceding sections seem to be
related to the late at which the people developed— began talking, began
walking, reached adolescence. The early talkers, Dr. MacFarlane reports,
tended to be the late walkers. And those who matured late, as indicated
by the age they reached pubescence, differed considerably in some re-
spects during adolescence and young adulthood from those who matured
early. This was true of boys in particular.

As an example of how the rate of physical development in youth can
influence a person's life for years afterward, this investigator tells of two
boys who differed mainly in speed of maturation. The early maturer (who
reached adult sexual status before he was 13) excelled in athletics and
enjoyed the accompanying rewards. He showed interest in girls at 13 and
they in him. The late maturer (who reached adult sexual status after he
was 17) avoided girls, and they him, till he was 20. The first boy got a
summer job at 14; the second went to Boy Scout camp. After college, the
early maturer joined a firm in another city, married, and by 30 had
reached a responsible position that takes him and his family to all parts of
the world. The late maturer married a girl he had known since grade
school, got a promising job with her father's help, and established a home
in the neighborhood where they had been born and raised. At 30 he had
yet to reach the administrative level in his firm. "I'm too young, they tell
me," he reported. "I've always been too young."

Though the details differ from person to person, the early maturing
male has been found by the Guidance Study to have an easier time during
adolescence and to show more confidence both then and later. Tliis was to
have been expected, MacFarlane thinks, because the boy who matured
early was also likely to have begun early getting into things and exploring
the environment. His interests were outward because there's where the

141

excitement lay. On the other hand, the late maturcr wus
early talker, an intioveited fellow who was appaicntly IIKHO
fiom the start by thinking pioccsses than by action.

At 30, the early matuier was likely to have advanced I'mllkT i>
than the late maturei, more likely to be married, and likely (o I
children.

Findings by Dr. Mary Covei Jones, Piofessor of liduejiltoii,
from her study of a different sample-early and late muUiiinn b(
Oakland Giowth Study-confirm those results, add scum1 inlet1
tails, and cany the comparison a little farther along. A| ,13. llu"
had matured early rated significantly higher in both souiihiliC
sponsibility. They were also more conventional in then ,i[(j(
thinking. Five years later the diffeienccs were less marked, but
who had matured early still appeared to be moic assured ami s
less fearful, and also less insightful and independent. Mine ot
maturing men have attained executive, status-conferring', vocii1ci»

The boy who most rapidly approaches physical mtmhoml, J
gests, is the one who is first recognized by the adult community
therefoie is most likely to take on-if he doesn't have I hem .tin
peisonalily traits likely to be most valued by that eomimirnly. J)i
Peskin, a psychologist, thinks the difference found in holli sin
well have a deep psychological basis. The early matuier, IK- MI^II^
prepared for the changes of adolescence. So he may expi-iieiKv
less tolerable and theiefore less acceptable. He flees, lheiotnK\ in
hood and makes an early and rewarding commitment to [\w \.ili
cultme. So he is "naturally" more sociable and conformhii'. On t
hand, the late maturer, not having to deal with the hoiiiuiiul
duves till later, has more time to look around, expand his st,
develop a vaiiety of psychological mechanisms for ivfuiLiluip i
when puberty comes, he tolerates it bcttei and has less mv»l ol
supports and rewards. This would explain his greater msii'liUulm-*-

Here again an apparent sex difference lias been I'oniul, <i
leached maturity early, MacFailane lepoits, were usually hv» onii
adults. In school they seem to have fell out of thini's, Tltvv,-
described by Mary Jones, on the basis of obscivational uiliu^. .^ '
disadvantaged," However, their responses on the Thematic API<,
Test and their self-report scores indicate adequate scll'-coiuvpl s

Dr. Louis Stewart adds this finding: Among males, :it Ir.^i, l
bom tends to mature earlier than an only child or a hisllmm Ju
psychologist thinks the earlier maturation is somehow assuci.ikd i
events attendant on the arrival of a new baby, for when ,i moi
pregnant with her second child, and for a year or two alU'r iK lv
fhstboin showed an unusual spurt in growth. Numerous
animals and some studies of people show that stimulation in
painful stimulation, makes for growth, and separation lioiti
appears to be an important form of such stimulation. The
nancy and the arrival of a new child, the investigator suun
painful, development-spurring stimulation for the firsthnin.

142

Incidentally, othei work on buth oicler by Dr William T. Smelser, also
a psychologist, throws new light on the recuncnt finding that firstborn
children get moic education than those born last. In two-child families
where both children aie of the same sex, Sinelsci finds, there is no signifi-
cant diffeience in the number of years they go to school. But where one
child is a boy and the other is a gill, the fiistborn, of whichever sex, goes
to school significantly longer And theic is a greater pioportion of these
cross-sex (girl-boy or boy-giil) families. In studying the effects of birth
order on yeais of schooling, then, it is important to ascertain not only a
person's birth position but also the sex of the child next to him.

Early talkers v. late. Analyzing the records of men in the Guidance
Study, Dr, Kenwood Bartelme finds that the early talkers and the late
show significant personality difference both as children and as adults. The
same findings seem to apply to women, too, but in a less clear-cut fashion.

The eaily talkers were taken to be those who had said at least five
words befoie they were 12 months old; the late talkers, those who had
not done so until after 15 months. The boys in this second group talked
well enough once they got started, but they had a different peisonality
style.

Thiough adolescence, at least, the early talkers were on the restrained
and somber side. Their IQ's were consistently higher than those of the
later talkers but largely because of the difference in scores on the verbal
factor tests. In high school, the early talkers were known as eggheads; at
30, they still valued intellectual matters and were inclined to intellectual-
ize about a subject, even to the point of splitting hairs, They were also at
30 more practical, prudent, and conservative. The late talkers were active,
lelativcly uninhibited, and even rebellious. "The late-talker," Bartelme
remarks, "is the social nonconformist."

The middle gioup, who began talking between 12 and 15 months,
turned out to be the most conventional.

Why one person starts talking exceptionally early and another ex-
ceptionally late, even in the same family, is not known, but Bartelme has
found one environmental difference. The early talker had received more
than the usual amount of attention from his parents, particularly from Iiis
mother; the mother, in fact, had seemed to be more involved with him
during infancy than with her husband.

Predicting Adult Psychological Health

When the subjects of one of the longitudinal studies (Oakland Growth)
were 36, Stewart divided a sample of them into three groups:

1. Those with psychosomatic disordeis. Of the 20 afflicted persons in
this group, most had either stomach ulcer or hypertension. The others
suffered from migraine headaches, spastic colitis, asthma, or arthritis.

2. Those with behavioral maladjustments. Two of the 21 persons here
were alcoholic, six had had repeated divorces, another six had failed to
make a satisfactory social adjustment, and seven had been treated for
mental illness.

143

lay. On tlie other hand, the late maturer was typie.ifJy
early talker, an introverted fellow who was apparently more fiiscm
from the start by thinking processes than by action.

At 30, the early maturei was likely to have advanced farther in iiis u
than the late matuier, more likely to be manied, and likely to \M\\C
children.

Findings by Dr, Maiy Cover Jones, Piofessor of Education,
from her study of a different sample-early and late matming boys rii tie
Oakland Growth Study -con firm those icsults, add some intcicstiitfidfr
tails, and cairy the comparison a little fmther along. At 33, the men wlw
had matured early lated significantly higher in both sociability ami re-
sponsibility. They weie also more conventional in theii attitudes and
thinking. Five years later the differences were less marked, but rlic man
who had matured eaily still appealed to be more assured and somewtal
less feaiful, and also less insightful and independent. More of the c;irl>
maturing men have attained executive, status-conferring vocational gojfc
Tlie boy who most rapidly approaches physical manhood, Jones sug-
gests, is the one who is fust recognized by the adult community iind wlio
therefore is most likely to take on-if he doesn't have them alrcndy-lfa
peisonality tiaits hkely to be most valued by that community. Dr. IliirH'S
Peskin, a psychologist, thinks the diffeience found in both studies ma>
well have a deep psychological basis. The early maturer, he suggests* is!c»
prepaied for the changes of adolescence. So he may experience them a*
less tolerable and theiefore less acceptable. He flees, thetefore, into adult
hood and makes an early and rewarding commitment to the values of Im
culture. So lie is "natuially" more sociable and conforming. On I he oilier
hand, the late maturer, not having to deal with the hormonal-inspired
drives till later, has more time to look around, expand his skills, and
develop a variety of psychological mechanisms for regulating discs. Su
when pubeiry comes, he tolerates it better and has less need of outside
supports and rewards. This would explain his greater insightfuiness.

Here again an apparent sex difference has been found. Girls wlio
reached matmity early, MacFailane reports, were usually less confident as
adults. In school they seem to have felt out of things. These gj'ils arc
described by Mai y Jones, on the basis of observational ratings, as "socially
disadvantagcd." However, their responses on the Thematic Apperception
Test and their self-ieport scores indicate adequate self -concepts.

Dr. Louis Stewart adds tms finding. Among males, at least, the first-
born tends to mature earlier than an only child or a lastborn child. This
psychologist thinks the earlier maturation is somehow associated with the
events attendant on the arrival of a new baby, for when a mother was
pregnant with her second child, and for a year or two after its birth, the
firstborn showed an unusual spurt in growth. Numerous experiments with
animals and some studies of people show that stimulation in infancy, even
painful stimulation, makes for growth, and separation from the mother
appears to be an important form of such stimulation. The mother's preg-
nancy and the arrival of a new child, the investigator suggests, constitute
painful, development-spurring stimulation for the firstborn,

142

Incidentally, othei work on birth order by Dr. William T. Smelser, also
a psychologist, throws new light on the recurrent finding that firstborn
childien get more education than those born last. In two-child families
wheie both children are of the same sex, Smelser finds, there is no signifi-
cant difference in the number of years they go to school. But where one
child is a boy and the other is a girl, the firstborn, of whichever sex, goes
to school significantly longer. And there is a greater pioportion of these
cross-sex (girl-boy or boy-girl) families. In studying the effects of birth
order on years of schooling, then, it is important to ascertain not only a
person's birth position but also the sex of the child next to him.

Early talkers v. late. Analyzing the records of men in the Guidance
Study, Di. Kenwood Baitelme finds that the early talkers and the late
show significant personality difference both as childien and as adults. The
lame findings seem to apply to women, too, but in a less clear-cut fashion.
The early talkers were taken to be those who had said at least five
vords before they were 12 months old; the late talkers, those who had
lot done so until after 15 months. The boys in this second group talked
veil enough once they got started, but they had a different personality
tyle.

Through adolescence, at least, the early talkers were on the restrained
nd somber side. Their IQ's were consistently higher than those of the
iter talkers but largely because of the diffeience in scores on the verbal
ictor tests. In high school, the early talkers were known as eggheads; at
0, they still valued intellectual matters and were inclined to intellectual-
;e about a subject, even to the point of splitting hairs, They were also at
0 more piactical, piudent, and conservative. The late talkers were active,
slativcjy uninhibited, and even rebellious. "The late-talker," Bartelme
'marks, "is the social nonconformist."

The middle group, who began talking between 12 and 15 months,
irned out to be the most conventional.

Why one person starts talking exceptionally early and another ex-
'Ptionally late, even in the same family, is not known, but Bartelme has
und one environmental difference. The early talker had received moie
an the usual amount of attention from his parents, particularly from his
other; the mother, in fact, had seemed to be more involved with him
inng infancy than with her husband.

adicting Adult Psychological Health

When the subjects of one of the longitudinal studies (Oakland Growth)
re 36, Stewart divided a sample of them into three groups:

1 . Those with psychosomatic disorders. Of the 20 afflicted persons in
s group, most had either stomach ulcer or hypertension. The others
Tered from migraine headaches, spastic colitis, asthma, or arthritis.

2. Those with behavioral maladjustments. Two of the 21 persons here
re alcoholic, six had had repeated divorces, another six had failed to
ke a satisfactory social adjustment, and seven had been treated for
ntal illness.

143

3. Those who weie symptom-free: 25.

Then he went back to the data showing the social and emotional adjust
ments of these individuals when they had been adolescents. (The ttoh
came from the University of California Social and Adjustment Invcntoiy.
which had been administered each year between 11 and 17.) He found
some important differences.

The people with psychosomatic ailments as adults weie reporting a
poorer-than-average adjustment to life when they were only 1 1 years oil
which was from 15 to 20 years before the diagnosis of the illness. Oi
scales measuring such characteristics as attitudes toward family, feeling
about their own woith, and ability to get along with other people, tlws<
persons had rated themselves toward the low end. So had the individuals
with behavioral maladjustments as adults. During late adolescence, how-
ever, the scores on family and social adjustment had improved among Hi;
psychosomatic group but not among the behavioral maladjustment group
During the same period, those who were later to be afflicted with a
psychosomatic disorder also expressed a number of vague physical com-
plaints.

Members of the psychosomatic group had been marked, too, by UP.
underlying tendency toward depression, as indicated by feelings of worth-
lessness, lack of energy, sleep disturbance, and the loss-actual or feiircd
of parental attention and love. On all such traits there had been timing
adolescence highly significant differences between the psycliosomalic
group and the normal. The group with behavioral maladjustments InJ
fallen in between. The results suggest to Stewart that a basic clcpicssiw
tendency is an important factor in the onset of psychosomatic disorder.
And he is inclined to agree with some other investigators that virtually a'l
illness is caused by psychic as well as somatic factors,

To try to find the basis of the maladjustments noted during udolev
cence, Stewart is now analyzing the records of another study (GuklaiiH)
which go back to infancy. His preliminary findings confirm tluil lululli
with psychosomatic ailments or with psychological problems worse \\m
usual had been poorly adjusted adolescents. Further, the new fimliii£i
indicate that (1) members of both groups came from families whcic.vti)
early in the children's life, there had been more disturbance, and IM
satisfaction and security, than usual, (2) members of both groups tended
to be those who had matured either very early or very late,

This second finding does not imply that early and late ma Hirers aie
inevitably bound for trouble. The processes associated with either cxlrem*
of the maturation rate do seem to produce not only differences in person-
ality, as noted earlier, but also a higher than average potential for illiiM
and psychological difficulties. Stewart is now trying to find curly famib
and childhood patterns that distinguish the two groups-thc ill iuid ifo
maladjusted. He is also looking for childhood factors that distinpMi
people with one type of ailment from those with another.

In related work, Drs. Norman Livson and Harvey Peskin have ten
trying to determine which, if any, specific characteristics of a child, dis-
played at which particular age, can be used to predict his psychological

144

health as an adult. The subjects were 64 young adults, from the Guidance
Study, who had been rated for psychological health at 30 by comparing
their scores on a personality appraisal with theoretically ideal scores. (The
ideal scores were a composite of those made by foui clinical psychologists
as they attempted to define a fully healthy peison.) The adults' ratings
were then compared with their ratings as children, from the ages of five
through 16, on numerous behavior and personality scales.

High scores on certain characteristics during the years from 11 to 13,
but only during those years, were found to be significantly related to
adult psychological health. The healthiest men were those who as boys of
11-13 had been lelatively extroverted, cheerful, telaxed, and expressive,
and relatively immune to irritability. The women had been relatively inde-
pendent, confident, and inquiring— and had shown a hearty attitude to-
ward food.

The 11-13 age period proved significant, the investigators speculate,
because it encompassed the transitional period from elementary school to
junior high. Now once again, as when he had left the family to enter
school, the child had to take an important step toward maturity. "The
demands and opportunities of the junior high school, for both boys and
girls," the investigators suggest, "may represent so profound a difference
from elementary school as to constitute a qualitatively new experience.
The manner in which the child responds to the transition-actively inviting
or passively withdrawing from the new experience-tells us something
about how healthy an adult he will be."

The psychological health of these subjects will be assessed again during
the 40-year follow-up. The investigators plan also to look for factors in
the family environment during childhood that may portend good or poor
psychological health in latei life.

Children Who Turned Out Better or Worse Than Expected

Because of the work reported in the preceding section and of the re-
search under way at Berkeley and other centers, the investigators think we
shall be better able to predict while a person is still in school the probable
state of his mental health as an adult -and, if the outlook is poor, to take
steps to alter it. They emphasize, however, that the relationships reported
are based on group averages and that in every group studied there were
individuals who did not conform.

More than 20 years ago, when the children in the Guidance Study were
IS, MacFarlane and her associates made predictions about them as adults
—their personalities, their success in marriage and work, their ability to
cope with the problems of life: in short, their mental health. Though the
investigators had had few scientific guides, they were surprised by the
results of the analysis after the subjects were followed up at the age of 30.
In many cases, the predictions turned out to have been wrong. The rea-
sons ought to be helpful for parents, teachers, doctors, and everyone else
associated with children.

145

many of the most matuie adults-integrated, competent, clear about
their values, and accepting of themselves and others-were found tolumr
been those who as youngsteis had been faced with difficult situations and
whose characteristic lesponses had seemed to compound their problems.
They included chronic rebels who had been expelled from school, Mac-
Failane reports, highly intelligent students who were nevertheless academic
failures, children filled with hostility, and unhappy, withdrawn schizoids.
But the behavior regaided by the investigatois as disiuptive to growth and
matuiity seemed in these cases to have led directly or indirectly to adult
strength. One of the foinier rebels recalled that he had desperately needed
approval "even if it was from kids as maladjusted as I was." To maintain his
rebel status, he said, he had had to commit all of his intelligence and
stamina, a circumstance he believed had contributed to his adult strength
in tackling difficult problems. "I hope my children find less wasteful ways
to mature," he remaiked, "-but who knows?"

Close to halt of the subjects fell into the gioup for whom crippled or
inadequate personalities had been predicted. But as adults almost all of
them were bettei than had been expected, and some of them far better.

One man, for example, held back three times in elementary school, had
not graduated from liigh school until he was 21 . His IQ over the years had
aveiaged less than 100. He had shown little inteicst in studies, school
activities, or people. The school had not recommended that he go to
college. The staff thought he'd always be a misfit, a sidelines. But 1 2 years
later he was a talented environmental designer, a good father, and an
active worker in community affairs. "Obviously," says MacFarlane, "his
tested IQ's were no measuie of his true ability."

One girl, who early was suspicious of and even hostile to members of
the study staff, lived with a rejecting mothei and a poorly adjusted aunt.
She hated home and she hated school, paitly because of her poor clothes.
To escape, she married while still in high school a boy as erratic mid
immature as she. They soon separated. At 30, with the investigators
dreading the impending interview, in came a personable, well-groomed,
gracious woman with two buoyant but well-mannered childien. She had
married again and was living a stable, contented life.
Why were the predictions wrong in such cases?

For one thing, MacFarlane answers, the investigatois gave too much
weight to the troublesome and pathogenic elements in a child's life-quite
naturally, in view of the studies that have traced neuroses and psychoses
to such elements-and too little weight to the healthful, maturity-inducing
elements. (The lattei were present even in the case of the girl who sought
escape through marriage at 17. She always remembered that another aunt
had given her affection and happiness-had heloed her wlant seeds that

end. As an example, MacFarlane cites hurt feelings, m both boys and girls,
as "a veiy successful parent-manipulation tool." In changed situations,
such early useful devices lost their effectiveness, and the big majority of
the young people then dropped or modified them, sometimes not without
difficulty. With a number of girls, the game of getting their feelings hurt
was can led over into marriage.

Sometimes the undesirable but long-continued patterns were converted,
to the investigators' surprise, into almost the opposite characteristics. For
example, it was predicted that overdependent boys with energetic and
dominant mothers would pick wives like the mothers and continue the
pattern of overdependencc. Instead, nearly all such boys chose girls who
were lacking in confidence. The boys thus won themselves a role as the
proud male protector and giver of support, and in this role, says the
investigator, they thrived.

Along the same line, a number of those in the study who were socially
inept and insecure as children and adolescents became, again to everyone's
surprise, highly successful salesmen. Looking back, MacFarlane sees this as
a quite natural transformation. The boys did not have easy, intimate
relationships growing up because they did not have them at home. As
adults they still fear intimate relationships but have an unconscious desire
for social intercourse. Selling gives them the needed contacts without the
feared intimacy. (She could be wrong, she adds. The director of sales
training for a large firm told her he deliberately picked shy people, be-
cause they would concentrate on selling the product instead of them-
selves.)

One man, who is remembered with special pride, is the highly successful
manager of a large business concern. Years ago he had been a shy little
boy without friends. Though' he had dropped in from time to time to see
members of the Guidance Study Staff, his communications had often
been limited to hello and goodbye. After high school he enlisted and,
since he had taken some shop courses, he was asked to help with the
building and repair work at his Army post. First he was flattered that
anyone should think he could do any tiling; then he was proud that he
could actually do it. After his service, he went to business school, where
he got all A's, as compared to C's in high school. Now he says the most
interesting part of his job is to give people "something to do that is a little
harder than what they have done or think they can do-but not something
they would fail at-and then to watch them expand. Nothing is more
exciting to me than to see people get confidence"-which he himself had
lacked for so long.

A number of other subjects had had similar experiences. They did not
achieve "ego identity"-did not find themselves-until they had been
forced into or been given an opportunity to take on a responsible role that
gave them the sense of worth they had missed at home. Often these
people did not find this new and satisfying role until they had left both
their childhood homes and their home towns.

147

"Don't give up on our present generation of adolescents," MacFarlane
urges. "Many of ours came through bad times and developed into mature,
stable adults in spite of our fears."

In a speech before The American Academy of Pediatrics recently, (lie
investigator quoted comments made spontaneously by a number of the
Guidance Study subjects at 30. Some examples:

"When I was confused and worried, the Institute was the only place I could talk ant
loud to myself and find out what I thought and felt."

"I sensed your respect for me, even when I knew I wasn't acting very sensibly and
knew you wouldn't have had respect for me if there wasn't something there lo respect
because, beheve me, I can tell a phony in a split second-because at times I'm n plumy
myself."

"You asked questions, you listened, but you were the only grown-ups who didn't
give advice. You helped me to ferret things out for myself, to make my own decisions.
I try to carry this on in the raising of my children."

The investigator then pled with the doctors to take a similar role wilh
their patients. "You pediatricians are the only professional group," she
said, "that can furnish continuity and interest over the long age-span of
growth from babyhood to maturity, provided, of course, you have Ik
temperament to be sympathetically interested in the vagaries of Hie
human struggle for competence and maturity. Provided, too, you can
accept the fact that you can't play God or believe you know nil the
answers, because one tiling we have learned is how little we know. Pro-
vided, too, that you can train yourselves to look for strengths in indi-
viduals and their situations and not just for pathology. If you don't fur-
nish this function, what professional will? If teaching departments don't
incite interest, who will?"

About 20 percent of the cases turned out worse than expected. These
included many of the persons who as children and adolescents had had
easy, confidence-inducing lives, free of severe strains and marked by
academic, social, or athletic success. Prominent among these were a high
proportion of the men who had been outstanding athletes in high school
and of the women who as girls had been pretty and exceptionally popular
At 30, many of these people had failed to live up to their potentialities
and were puzzled and discontented.

MacFarlane gives several possible explanations. Early success may have
led to unreal expectations and to a draining of energies into maintaining •
an image. It may have sidetracked the development of patterns ami atti-
tudes that would have made adult life more rewarding. Perhaps there
't enough stress in these youngsters' lives to foster development. She
some of the people in this group will work free from their tails-
's as young adults and will yet live up to their predictions for them; '
«>" go through live wondering what happened and where they

people turned out as predicted . Among Ihe
rcontrolled, who had built a psycho-

-~eiw.. «^iW agauui me uangers 10 be found in other people and in llfcia -
general. As young adults they still had the shield. It seems to have pro-
tected them, MacFarlane reports, but it also-by denying them access to ;

148

many kinds of learning experience-has impoverished them. The second
main group includes the youngsters who had been subjected to marked
variability in family treatment, being handled indulgently one day and
slapped down the next. Neither as adolescents nor young adults had they
developed stable patterns of behavior. Of the nine adults in this study who
were found to be compulsive drinkers at the age of 30, all but one came
from this group. Perhaps significantly, the compulsive drinkers had mani-
fested physical vulnerability, too-acute allergies, beginning in infancy.

The Effects of Guidance

The patents of half of the children in the Guidance Study were en-
couraged to turn to the staff psychologists for discussion of problems
whenever they wished. Many did so. Through these discussions, they were
helped to a better understanding not only of their children's behaviors and
attitudes but also of their own and their spouses'. The control families
were not intensively interviewed and discussions were avoided or kept to a
minimum. The subjects in the guidance and control groups-the children
who have grown up and become parents themselves— are now being com-
pared in order to help explain why they grew up to be the kind of parents
they are and have the kind of children they have.

So far, only hard fact has emerged: The group whose parents received
little or no opportunity for discussion of interpersonal relations and chil-
dren's problems has had about four times as many divorces as the guid-
ance group. Dr. Ann Stout, the psychologist who is handling the parent-
child study, finds other evidence pointing to the conclusion that when
parents are encouraged to discuss family problems with professional
workers, their children tend as adults to have more flexible qualities and a
better ability to cope with situations. As she puts it, the guidance seems to
have tempered some of the negative factors influencing the children's
development. For example, there are persons in both groups who were
ebullient as youngsters but who for some reason, at some point, took on a
ratlier depressive attitude. They are not getting the satisfaction out of life
that they should. However, the depressive subjects in the guidance group
show more resilience than those in the control group, and a tendency to
find more satisfaction.

Antecedents of Smoking

In 1 964 the report to the United States Surgeon General on "Smoking
and Health" stated that "While rebellion may play a role in the initiation
of smoking, perhaps an important one, there is not much evidence for it.
Claims in the literature are at best based on circumstantial suggestive
evidence, linked to conclusions by a chain of questionable assumptions."
Since then, Stewart and Livson, using data from the Guidance and Oak-
land Growth Studies, have turned up evidence that cigarette smokers
actually are more rebellious than nonsmokers and that this greater degree
of rebelliousness is part of the smoker's personality long before he starts

149

smoking. It appears even during the earliest years of school. In a separate
study using a different methodology, Clausen finds that the youngsters
who became smokers tended to be less controlled and more aggressive
than the otheis and to differ also in other important traits and in their
backgrounds. The nonsmoking boys had a stronger drive to get ahead, and
they still have it as adults.

To measure rebelliousness, Stewart and Livson used the teachers'
ratings on the children's behavior in school and on their attitude toward
school and, where the ratings were not available, grades on conduct. A
"resistance to authority" measure, taken during four years of high school,
was also available for those subjects from the Oakland Growth Study. This
was the average of independent ratings made by three staff membeis after
observing the students in a variety of social situations.

From ages five to 15, the persons who later became smokers were
found to show more evidence of rebellious attitudes. For all the subjects
studied, the diffeicnce was clearest during the sixth and seventh grades,
when it was statistically significant for each of the two groups and for
both girls and boys. For those subjects related on "resistance to author-
ity," the difference continued to be significant through high school.

As adults, too, the smokeis— who comprised about 55 percent of the
160 subjects studied during their early 30's~were rated more rebellious.
The measure this time was a scale intended to show the degree to which a
person conforms to the mores of our society. (Bankers have been found to
rate at the top of this measure, which is socialization scale of the Cali-
fornia Psychological Inventory; juvenile delinquents and criminals rate at
the bottom.) The nonsmokers scored significantly highei than the
smokers: the finding could have occurred by chance one time in 100 in
the case of women, and one time in 1 000 in the case of the men.

So what? The investigators respond that smoking and adolescent rebel-
lion against authority have long been linked in popular thought and that
this study provides evidence they actually are linked. If this is so, the
investigators continue, an antismoking campaign based upon authoritative
pronouncements has little chance of success: a more subtle strategy is
required.

Stewart and Livson refer to a handful of recent studies that have used a
measure (the Psychopathic Deviate Scale of the Minnesota Multiphasic
Personality Inventory) very similar to the one employed in this investiga-
tion to rate rebelliousness in adults. These studies have found significantly
higher scores among (a) people addicted either to alcohol or narcotics, (b)
smokers in a psychiatric outpatient population, and (c) smokers in two
samples of male college students. The Berkeley investigators suggest that
smoking is an addiction, sharing with the other addictions a common
origin in some underlying resentment of authority and being, like the
other addictions, not an isolated habit but an expression of "pervasive

nts of smoking, Clausen sent a
subjects in the Oakland Growth
received answers from 1 23, and

then searched for personality and background factors-asrecoided during
adolescence— that might differentiate between those who took up smoking
and those who did not, and between heavy and light smokers.

Perhaps the clearest finding, Clausen reports, is that the adolescents
who did not become smokers were more controlled in high school, more
oriented to adult values, relatively unaggressive, and modest in tncir views
of themselves. The boys were seen as well-adjusted and creative and as
having a strong desire to achieve; in senior high they enjoyed piestige and
popularity. The girls who did not take up smoking were rated as conven-
tional, unassuming, calm, serious, and self-sufficient. Though well-
adjusted, they were less popular than the girls who became smokers.

As a partial, quite incomplete, explanation of these differences, Clausen
finds that the boys and girls who matured later than the others were more
likely to start smoking and continue. Though there is no direct link be-
tween biology and smoking, the investigator hypothesizes, the person who
finds himself lagging behind the gang in physical maturity may turn to
smoking as a sign that he, too, is growing up.

Clausen also finds differences in the characteristics of the mothers of
smokers and nonsmokcrs. The motheis of the boys who remained non-
smokers had been rated as effective, nonneurotic women— significantly
less talkative and more clearheaded and cheerful than the mothers of the
other boys. The mothers of the girls who remained nonsmokeis showed an
accepting rather than a critical attitude.

The actual smoking habits of the parents strongly influenced those of
the girls, but not of the boys. Among those girls with at least one parent
who smoked, a fourth eventually became very heavy smokers and only a
fifth did not smoke at all. When neither parent smoked, scarcely any of
the girls became very heavy smokers and nearly half never smoked.

Lack of poise would be an important antecedent of smoking, Clausen
had thought, but this turned out to be incorrect, especially among the
girls. The girls who remained nonsmokers were rated less poised and
socially skilled than the others and at the same time more unaffected and
more composed.

On the basis of intensive interviews, averaging 12 hours in length, the
adults show many of the same personality traits that characterized them
as adolescents, The nonsmoking men still have a stronger drive for achieve-
ment and are more effective. They are in tighter control of themselves,
less self-indulgent, and more self-satisfied. Though the nonsmokers were
rated significantly less assertive in adolescence, as adults they arc slightly
more assertive than the smokers, Probably this change has occurred,
Clausen thinks, because most of the nonsmokers have experienced a con-
siderable measure of occupational success. Moderate smokers have done
nearly as well, but heavy smokers much less well. Among the women,
nonsmokers remain more conforming. They are also rated more fearful
and more likely to manifest guilt than are women who become smokers.

The questionnaire relating to current smoking has now been sent to
subjects of all these studies. A monograph reporting findings will deal not

151

only with antecedents and con elates of smoking but also with compari-
sons between addicted smokeis and those able to give up or cut down on
tobacco use.

Antecedents of Drinking

As anothei example of using a longitudinal study to answer questions
almost impossible to answer in any other way, Mary Jones has been
studying the personalities of drinkers and nondrinkers. Her question was:
Do personality characteristics associated with a given drinking pattern
show up early in life, before the pattein has been established? The tenta-
tive answer is that they do.

In the work to date, 68 men and 70 women in the Oakland Growth
Study were classified according to their drinking habits and theii reasons
for drinking. The subjects were in thcii middle forties. Then their person-
alities weie assessed, through the use of California Q Set, for three age
levels: junior high school, senior high school, and adulthood.

More than half of the behavioral items that differentiated problem
drinkeis from modciatc di inkers (typically, a drink or two before dinner,
three or four at a party) and abstainers in adulthood were found to have
differentiated them also in junioi high school. For example, the men
problem drinkers, compaied with the other men studied, were found to be
rebellious, self-indulgent, gregarious, unpredictable, and disorganized.
They were less dependable, less considerate, less fastidious, and less
moralistic. As junior high students, these men had been marked by the
same characteristics. And they had been more concerned than the others
to demonstrate their masculinity. Behavior duiing the junior high school
years proved to be a better predictor of adult drinking patterns Hum
behavior during later adolescence.

On the rating that distinguished the three groups both as adults and
junior high students, the men who weic problem drinkers as adults usually
stood at one end of the scale, the abstainers at the other, and the
moderate drinkers in between.

Preliminary findings indicate that women problem drinkers resemble
their male counterparts in respect to instability, unprcdictableness anil
impulsiveness. However, they tend to be introversivc and more marked by
feelings of depression, self-doubt, and distrust than the men.

Looking Ahead

When the longitudinal studies began at Berkeley, there was little docu-
mented knowledge about factors that helped shape intelligence, person-
ality, and mental health from childhood onward. Now there is a good
deal, thanks not only to these pioneering investigations but also to numer-
ous other studies undertaken more recently .

Because of man's complexity and the great variety of the influences
pressing upon him, much remains to be learned -a statement that may
always be true. But the body of our knowledge is being steadily increased.

152

Among the continuing investigations at Berkeley, two seem especially
important: the attempt to identify those factors in infancy and childhood
that predispose to psychosomatic illness and psychological maladjustment,
and the attempt to ferret out the influence of heredity on certain abilities,
and characteristics.

Other important work under way includes research on

o The relationship of physical factors, such as body build, to specific
types of intellectual function, such as mathematical ability

o The ethical, religious, political, and other values held by a poison's
family as he grew up; the values he look with him into maniagc; the
values that now dominate his home.

« What difference it makes if one paicnt is the disciplinarian rather
than the othei, or both.

« The value of the Roischaeh test, which was administered to Guidance
Study members for 7 years during adolescence and again at 30, in predict-
ing psychological health.

» The relationship between a person's interest and satisfactions during
childhood to his personality and psychological health as an adult.

Staff members are also busy fitting together and interpieting the find-
ings so that these can be of the widest use. Important books coming up in-
clude (1 ) "Ways of Personality Development: Continuity and Change from
Adolescence to Adulthood," by Jack Block and Norma Haan, which uses
data from both the Oakland Growth and the Guidance studies; (2) "Chil-
dren of the Depression," an analysis by Dr. Glen H. Elder, Jr., of the
immediate and long-term effects of the depression on the subjects in the
Oakland Growth Study; and (3) "The Course of Human Development," a
collection of major papers from, and new essays about three studies,
edited by Drs. Mary Jones, Bayley, Honzik, and MacFarlane. Also Dr.
Clausen is working on a major "life careers" monogiaph based on informa-
tion from the Oakland Growth Study. It seeks to answer: What are the
major influences upon a person's performance in the most salient loles of
adulthood-thosc of worker, spouse, parent and community participant?

In the near future, Berkeley's Institute of Human Development liopes
to establish a program of Intel-generational Studies of Development and
Aging that will use the people who have been participating in the present
sliidics-thc original subjects, their surviving parents, and the subjects'
children. The questions to be investigated include the patterns and
processes of aging, the hcritability of traits and abilities, and the similar-
ities and differences in family patterns and styles of life from one gen-
eration to another.

Research Grants: Mil 6238, 8135, 5300
Date of Interviews. April 1967

153

References:

Bayley, N. The life span as a frame of reference in psychological research. Vita

Humana, 6 125, 1963.
Bayley, N Consistency of maternal and child behaviors in the Berkeley Growth Study.

Vita Humana, 7.73, 1964.
Bayley, N. Research in child development' a longitudinal perspective Merrill Palmer

Quarterly of Behavior and Development, 11'3, 1965.
Bayley, N. Age-trends in mental scores: ages 16 to 36 yeais. Paper for American

Psychological Association, 1966.
Bayley, N. Learning in adulthood: the role of intelligence. In. Analyses of Concept

Learning. New York1 Academic Press, 1966.
Bayley, N, Cognition. Paper for University of West Virginia Conference on Theory ami

Methods of Research and Aging, 1967.
Bayley, N. Behavioral correlates of mental growth: birth to 36 years. Paper foi

American Psychological Association, 1967.
Bayley, N. and Schaefer, E. S. Correlations of mateinal and child behaviors with the

development of mental abilities: data from the Berkeley Growth Study. Mono-
graphs of the Society for Research in Child Development, 29(6). 1-80, 1 964,
Branson, W. C. Early antecedents of emotional expressiveness and reactivity control.

Child Development, 37:793-810, 1966.
Branson, W, C. Central orientations: a study of behavior organization from childhood

to adolescence. Child Development, 37' 1, 1966.
Cameron, J.; Livson, N.; and Bayley, N. Infant vocalizations and then relationship to

mature intelligence. Science, 157'3786, 1967
Clausen, J. A. Adolescent antecedents of cigarette smoking data from Oakland

Growth Study. Social Science and Medicine, 1:4, 1968.
Haan, N. Proposed model of ego functioning: coping and defense mechanisms in

relationship to IQ change. Psychological Monographs General and Applied, 77 '8,

1963.
Honzik, M. P. A. sex difference in the age of onset of the parent-child resemblance in

intelligence. Journal of Educational Psychology, 54:5, 1963.
Honzik, M. P. The environment and mental growth from 21 months to 30 yeais. Paper

for International Congress of Psychology, 1966,
Honxik, M. P. Environmental correlate's of mental growth: prediction from the family

settingat 21 months. Child Development, 38:337-363, 1967.
Jones, M. C. Psychological correlates of somatic development. Child Development,

36:4, 1965.
Jones, M. C. Personality correlates and antecedents of drinking patterns in adult

males. Journal of Consulting and Clinical Psychology, 32:1, 1968
Livson, N., and Peskm, H. The prediction of adult psychological health in a longi-
tudinal study. Journal of Abnormal Psychology, 72 '509-5 1 8, 1967.
MacFarlane, J. From infancy to adulthood. Childhood Education, 39:336-342, 1963.
MacFarlane, J. Perspectives on personality consistency and change ftom the Guidance

Study. VitaHumana, 7-115, 1964.
MacFarlane, J. The dilemmas of adolescents. Paper for American Academy of

Pediatrics, 1967.
Peskm, H. Puberral onset and ego functioning: a psychoanalytic approach. Journal of

Abnormal Psychology, 72:1-15, 1967.
Stewart, L., and Livson, N. Smoking and rebelliousness: a longitudinal study fiom

childhood to maturity. Journal of Consulting Psychology, 30:325-329, 1966.

154

Investigator:

John W.M. Whiting, Ph.D.

Harvard University
Cambridge, Mass,

Co -contributor:

Shulamlth M. Gunders

Prepared by

Herbert Yahraes

John W. M. Whiting and follow workeis at Harvard and elsewhere have
searched the world over for moie information on some of the forces that
turn a child-born with the same potentialities, on the average, as other
children-into an adult typical of the society in which he has been reared.

The research is a study of certain child-rearing practices and their effect
upon adult personalities and behavior. But it differs from most other
studies having the same objective. These other investigations generally
assume that parents, in bringing up children, have theories as to what is
good and bad for a child's development and that these theories are in-
fluenced by unconscious motives and anxieties. The work of Dr. Whiting
and his group, on the other hand, is based on the idea that differences in
child-rearing practices are imposed by differences in certain aspects of the
physical and social environment.

A good deal of evidence has been collected in support of the Whiting
group's idea. Much of it comes from analyses of data collected by other
anthropologists in earlier studies, but the results are supported by more
recent field investigations, of which the most important is the Six-Culture
Study directed by Dr. Whiting, Irvin L. Child of Yale, and William W.
Lambert of Cornell. For this work, six pairs of anthropologists spent a
year observing child-rearing practices in six widely different cultures: a
Gusii community in Kenya, Africa; a Mextcc Indian community in the
state of Oaxaca, Mexico; a community of Tarongans in the Philippines; a
village of Okinawans; a neighborhood of families of the Rajput caste in

155

the province of Uttar Pradesh, India; and agioupof New England families
in a village identified as "Orchard Town " The fieldwork, done in the
mid-1950's, was financed by the Ford Foundation; the analysis of the
results, partly completed, is being suppoited by Dr. Whiting's grant from
the institute.

John Whiting is professor of social anthropology at Harvard; his co-
principal investigator, who is also his wife, Dr. Beatiice B. Whiting, is a
lecturer in the same department. They plan to spend a year in Africa
directing field studies to add to and verify the most important of the
observations and inlerpietations made by themselves and their fellow

workeis.

"Our aim," John Whiting explains, "is to investigate the process by
which a child learns the moral rules of his society in such a way that he
will not deviate from them even when the piobabilities of his getting
caught are minimal-in other words, the inculcation of self-contiol."

He believes a high degree of self-contiol may have a variety of causes,
among them (a) an exaggerated and paranoid fear of others, (/;) belief in
an all-seeing and all-powerful God who is concerned with the moral be-
havior of mankind, and (c) effective training for the parental role. The
investigators are particularly interested in this third cause and have given it
the bulk of their attention. By "training" they do not mean a series of
instiuctions and examples specifically intended to help fit a girl to cairy
out a woman's iole, or a boy a man's; they mean, most of all, the process
by which a child identifies with an adult and, therefore, consciously and
unconsciously, tries to be like him, and they also mean the characteristics
of a society that affect this process.

The work suggests how diffeiences in ceitain child-rearing customs lead
to differences in certain adult characteristics. It throws light on the
process of learning, and in doing so points to what may be a fundamental
cause of much juvenile delinquency, "Many of the casualties in our mental
hospitals and jails," says Whiting, "piobably would have been spared if we
had had a better understanding of the complex process of socialization,"
That's what he is trying to provide.

The investigators' main findings and conjectures are summarized below
and presented in more detail in later sections of this report.

A Summary of the Results

1 . Cross-cultural surveys offer a support for John Whiting's basic hypo-
thesis-that status envy is a prime force in the development for personal-
ity. A child most envies the person who can withhold the resources he
values most highly, the hypothesis says, and therefore, he tries to identify
with this person. In monogamous societies where the father is home, a
male child generally identifies with the father. But where the father is not
home, and particularly in societies where mother and child sleep together
-and the father elsewhere-during a long post partum sex taboo, a boy's
primary identification may well be feminine. The evidence for this is

156

found in (a) the results of various tests for cross-sex identity; (Z?) a display
of "hypermasculinity"— a bending ovei backward to prove themselves
men— among certain gioups of boys from father-absent 01 father-weak
homes; (c) the practice of strenuous male puberty rites-considered a type
of brainwashing-in societies whose household structure and other factors
would lead to the prediction that a boy's identification would be with
women. The investigatois believe that these findings may explain much of
the abeirant behavior in our own society.

2. Most monogamous societies rated high on an index of guilt; most
polygynous societies, low. John Whiting believes that in a monogamous
household, a boy is moie likely to develop a conscience based on his
father's, and that a man's conscience is stronger and more unforgiving
than a woman's.

3. Training against aggression tends to be most severe in families where
the household comprises not only parents and children but also a number
of theii kinfolk- The investigators believe that severe Liaining stems from
the desire of the adults to steer clear of unpleasantness likely to arise from
the children's fights.

4. Tiaining for independence, too, appears to be related to household
structure. As measured by early weaning and by reduced contact between
mother and child, it is found most often in nuclear households— those
comprised of the father, the mother, and the children. Such training,
Whiting thinks, probably (a} leads a child to grow up with the urge to be
successful and (b} is necessary to enable a couple to establish their home.

5. Evidence is offered that the type of task assigned a young child-
under 10-affects his peisonality. The most responsible children in the
six-culture study were those who had to take care of younger children,
and this babysitting job seemed to be related to the amount of outside
work the mother had to do.

6. Eveiy where in the six-culture study, boys were rougher than girls
and wandered farther from home, and girls rated higher on a responsibility
measure. Girls were more responsible than boys (and women in general
show more sensitivity than men), Beatrice Whiting surmises, because they
were more often assigned the care of younger children.

7. A probable link is reported, too, between the kind of tasks assigned
a child and the kind of supernatural power the culture believes in.
Societies where a child's work is to care for children or cattle, with the
results of failure being immediate and dramatic and punishment severe,
tend also to be societies where the supurnaturals punish a person here and
now. Where the main tasks arc household chores and schoolwoik, there is
a tendency to believe in supernatural who punish only in an afterlife,

8. Some evidence is found that polygyny arises from a long post
partum sex taboo, and that the taboo in turn is related to a diet so poor
that the mother fears to conceive until the nursing child has been weaned,
which may not be for several years,

9. Certain kinds of stress during in fancy -including inoculation,
circumcision, and periodic separation from the mother during the first
days of life-appear to be related to adult stature. In societies practicing

157

these kinds of stress, the average adult is significantly taller than in other
societies. The investigators hope to learn whether or not psychological
differences also occur— as they do in rats.

The investigators emphasize that the findings summarized here are
based almost entirely on studies of cultures, many of them primitive,
othei than our own, but theie is evidence that some of the conclusions—
especially concerning the effect of the fathers absence-apply here, too.
Study of other cultures is an essential means of discovering universal
tiuths about personality and behavior.

Status Envy as the Unconscious Motivation for Behavior

The most irnpoitant mechanism shaping the development of a child's
conscience-or, in psychoanalytic terms, superego-and an important
factor in his behavior throughout life, John Whiting believes, is status
envy. According to this hypothesis, the child envies the person who with-
holds resources from him 01 deprives him of them. The child thereupon
seeks to identify himself with this person; that is, tries to learn to act like
this person so that he, too, can become a controller of resources.

Resources include food, water, love, power, freedom from pain-
anything somebody wants. Every society has a status system under which
peisons at a certain level have privileged access to resomces while persons
at another level do not. The conditions for status envy arise even during
infancy, for no matter how everloving a parent may wish to be, there arc
times when he or she must withhold something the baby wants.

As Whiting views it, the oedipal situation, so prominent in psycho-
analytic theory, is merely a special case of the status envy hypothesis.
Under the psychoanalytic interpretation, the boy child vies with his father
for the mother's love; under Whiting's interpretation, the boy child simply
envies his mother— and therefore, identifies with her for a time—because
she can dispense to other persons a resource, love, he wants for himself.

The more a child envies the status of another with respect to the
control of a given resource, Whiting believes, the more he will practice
that role-openly perhaps, but certainly covertly, meaning that he will
indulge in a fantasy in which he sees himself as the envied person control-
ling and consuming the valued resources of which he has been deprived. It
is this fantasy of being someone other than himself that Whiting defines as
identification. As a simple example, when a child wants to stay up late,
but his parents make him go to bed, he may say to himself, "I wish I were
grown up. Perhaps if I acted as they do, I would be." And he thinks about
grownup behavior as he goes to sleep.

Further, the child will tend to manipulate resources as his parents have
manipulated them with him; for instance, if he has been given resources
when he had a special need for thern-such as solace when he was hurt and
assurance when lie was frightened-he will respond the same way to his
fellows when they are hurt or frightened. The child also will tend to
respond to the naughty behavior of others as his parents have responded

158

to his; and, having taken on the parents' role through status envy, he will
punish himself as well as others.

In some cultures the investigatois suspect a strong tendency for a child
to identify with his mother during at least the early years. Depending on
who controls which resources, however, he may well identify with both
parents. If, for example, the mother has primary control of food and love
whereas the father controls the power to administer or to withhold physi-
cal punishment, the theory predicts that the child will identify with both
parents but with respect to different lesources. He will take after his
mother with respect to love and affection; after his father with respect to
power and authority.

All cultures have rules concerning status that differ considerably from
the desires of a growing child. For instance, a child is supposed to act
neither like a baby nor a grownup; a boy is not supposed to be a sissy, nor
a girl a tomboy. Hence the desire to play certain roles of envied statuses
may lie latent for years. "It has often been noted," Whiting observes,
"that a mother will frequently respond to her fiist child exactly as her
own mother had treated her, even though she is not aware of practicing
such behavior, and, in fact, may even have vociferously sworn that she was
going to bring up her children differently. This suggests that covert
practice of envied roles may often be disguised and unconscious."

The idea that children identify with the persons close to them who
control valued resources is not new, but John Whiting has done more than
anyone else to seek confirmatory evidence and has supplied the name by
which the hypothesis is now generally known.

The Problem of a Conflict Over Sex Identity

Development of the status-envy hypothesis has pushed Whiting into
fascinating speculations for which he has tried to find a solid basis in the
child-training arrangements and the behavior characteristics of numerous
cultures, including some in the United States.

The speculations so far have centered on cross-sex identity. If a mother
has had control of all the resources a young boy values, Whiting's theory
says he will envy her status and try to perform her role. Part of the
behavior that he seeks to emulate, though, is sex-typed and forbidden.
"Such a boy finds himself in dire conflict," the investigator says. "Ho may
practice feminine-role behavior despite the sanctions against it, or, if the
sanctions are too compelling, inhibit his impulses to perform such be-
havior. The theory predicts, however, that he would continue to piactice
covertly and would thus have a feminine self-image,"

As a test of these views, the investigators analyzed sleeping arrangement
in a sample of 64 societies. If a boy sleeps alone with his mother during
his first years, covering the nursing period, the reasoning went, he will
come to consider her as all-important and, because she sometimes with-
holds resources, as the person to be envied. So his primary identification,
or the type acquired during infancy, will be feminine. If he then, after

159

weaning, enters a world in which men are obviously the important per-
sons, able to bestow 01 to withhold the most valued lesources, the boy's
secondary identification, acquired during childhood, will be masculine.

In such a case, the reasoning continued, the boy's conflict over sex
identity should be especially severe, so severe, in fact, that society would
step in to resolve it and assure him beyond any doubt of his masculinity.
This would be done through elaborate male initiation rites at puberty,
including circumcision and tests of strength and endurance— all serving, in
the investigators' words, "to brainwash the primary feminine identity and
to establish firmly the secondary male identity."

Native theoiy is offered in support of this interpretation. Most societies
having male initiation rites, the investigators report, have one word refer-
ring to all women and uninitiated boys and another word referring only to
initiated males. In these societies a boy is born twice; first into woman-
child status and then, at puberty, into the status of manhood.

Fiom the sample studied, 13 societies were found to have such rites, In
all 13, mother and infant slept together and alone. (For the most pait,
these were polygynous societies, with a long post partum sex taboo. When
a wife had a nursing baby, the father slept in another wife's house or in
the men's quarters.) And 12 of the 13 were patrilocal societies, in which
the domestic unit comprises a group of closely related male and a group of
unrelated females brought in from other villages as wives; in such societies
prestige and power are clearly vested in the men.

There are numerous cases, however, where the secondary as well as the
primary identification should be feminine because the child not only
sleeps alone with his mother during infancy but also grows up in a matri-
local world, controlled by his mother, his aunts, and his maternal grand-
mother. Such a society, the investigators reasoned, should give a man
some means to act out, symbolically at least, the female lolc, and they
found such a means in the custom known as the couvade. Under this
custom, during the period just preceding or following the biith of a child,
the husband takes to bed, fasts or limits himself to ccitain foods, under-
goes a puiificalion ceremony, and in general accepts the same attention as
his wife.

Childbirth epitomizes the uniquely feminine part of a woman's role.
"When a man attempts to participate In the birth of his child by closely
imitating the behavior of his wife,1' say the researchers, "this should be a
good index of his wish to act out the feminine role and thus symbolically
to be in part a woman."

The Harvard anthropologists found the couvade a common piactice in
12 societies in their sample. In 10 of these, mother and infant slept
together, and in nine, because of the matrilocal residence pattern, women
even after the nursing period controlled the resources most valued by a
growing child,

"Masculine Protest" Behavior and the Absent Father

In another test of the theory, a husband-and-wifc team from Harvard,
Robert L, and Ruth H. Munroe, recently studied the Black Carib of

160

British Honduras. This group springs from escaped Negro slaves who
assimilated the culture of the natives, the Island Carib, of St. Vincent, in
the Caribbean, and then spread over the coastline of the Gulf of Hon-
duras, in Central America.

As practiced by the Black Carib, the couvade involves a. number of
restrictions on the father's oidinary behavior, among them taboos on
fishing and on extramarital intercourse. If the restrictions are violated, the
Carib believes that the new bom infant will get sick. But some fathers, the
intensive-couvade group, observe many taboos; others, the weak-couvade
group, only a few. The approximately 50 fathers studied fell about evenly
into the two groups.

The investigators hypothesized that the gioups had been brought up
differently; specifically, that the intensive-couvade males had been under
stronger female influence. This turned out to be correct; during the first 3
years of life the intensive-couvade group had spent more than twice as
much time as the others in a household whose only adults were women.

But were these men really more feminine, psychologically, than the
otheis? Cross-sex identity is a tricky thing to measure, but Whiting and
some fellow workers had attempted to do it earlier— and successfully, they
think-with the so-called magic man test. In this test, the subject is told to
suppose that he can be anybody he wishes and is then asked who he
would most like to be-a father, a mother, a son, a daughter, and so on.
Then he is asked to make further choices. American males brought up in
the usual household, with both father and mother attending to them from
infancy, tend to choose all the male statuses fiist; American females, all
the female statuses. But the earliei study had found that among young-
sters with good reason to envy the status of the opposite sex, a number
did make some cross-sex choices. Among the Black Carib, the Munroes
found, males who practiced an intensive couvade chose to be a mother, a
daughter, or a baby girl significantly more often than males who observed
the couvade only weakly.

The Munroes also used the Semantic Differential Test, in which the
subject is given a number of scales having contrasting adjectives at each
end, like good and bad, fast and slow, and strong and weak, and is asked
to use the scales to describe various persons. The men who observed the
couvade most strongly tended to rate themselves on these scales the same
way they rated women.

There were still other discriminating factors. Black Carib men and
women show certain linguistic differences; men, for example, tend to use
a different word than women for "yesterday." Asked to translate from
English to Carib, the intensive couvade men used female words signifi-
cantly more often than the other men. Further, when their wives were
pregnant, the intensive couvade men experienced many more pregnancy
symptoms than the others. Male pregnancy symptoms, which have been
found in all societies where they have been asked about, are believed by
these investigators to have the same psychological implications as the
couvade itself.

161

If all this seems a little beside the point, it may be recalled that numer-
ous investigatois have found links between family conditions and juvenile
delinquency. The Harvard gioup is now offering evidence that the specific
problem is the mother-present, father-absent or father-weak household
structure during the first years of life, because in such a home a boy
originally identifies with women and then later on may go overboard to
prove his masculinity. This cross-cultural reseaich may, then, explain an
underlying cause of much juvenile delinquency, alcoholism, and other
aberrant behavior.

Beatrice Whiting puts the matter this way;

If during the first 2 or 3 years of life a child is constantly with his mother and
infrequently sees, and is handled by, his father, he will identify strongly with his
mother and not with his father; in short, if he is a boy he will have a cross-sex
identification. If, later, in life, he is involved in a world in which men are perceived to
be more prestigeful and powerful than women, he will be thrown into conflict. He will
develop a strong need to reject his underlying female identity. This may lead to an
overdeter mined attempt to prove his masculinity, manifested by a preoccupation with
physical strength and athletic prowess, or attempts to demonstrate daring and valor, or
behavior that is violent and aggressive,

A boy's attempt to prove his masculinity has been used to explain the
high rate of juvenile delinquency among Negroes in lower-class neighbor-
hoods. John Whiting points in particularly to a study concluding that the
Negro juvenile gang member rejects femininity in every form-"and he
sees it in women and in effeminate men, in laws and morals and religion, in
schools and occupational striving."1 And a study of the correlates of
crime in 48 societies found that lack or limitation of opportunity for
young boys to form an identification with their fathers was associated
with a higher frequency of theft, assault, rape, murder, and other crimes.2

The new developments by the Harvard group are the elaboration of the
status-envy hypothesis as an explanation of masculine-protest behavior
and the finding of such behavior, linked with father-missing households,
dining field studies, As one important example, the Munroes found it
among the Black Carib. The men who practiced the couvade most
intensively were described by the people of the town as braver men,
heavier drinkers, and more frequent cursors than the others. "In the day-
to-day situation," the investigators report, "the intcnsive-couvade men
were prototypes of the rugged male."

Carrying their work to the United States, the Munroes then examined
the exhaustive records of 200 men whose wives had given birth at a
Boston hospital. The records were exhaustive because they were part of
the comprehensive, long-time perinatal study -an inquiry into early condi-
tions associated with later defects-sponsored by the National Institute of
Neurological Diseases and Blindness. The men were white and from a
variety of racial stocks and socio economic levels. Forty percent of the

1 Rohrer, J, H., and Edmonson, M. S. "The Eighth Generation." New York: Harper
1960.

"'Bacon, Margaret K., Child, I. K., and Barry, II., HI. A cioss-cultural study of
correlates of crime./. Abnorm, Soc. Psycho!,, 1963,66.

162

\

husbands reported that they had experienced common pregnancy
symptoms— foi example, nausea, vomiting, toothache, and food cravings.
About 30 of these men and their wives were studied, as were a group
similar in size and general characteristics except that the husbands had had
no pregnancy symptoms.

The men who had had symptoms, the team reports, tended to respond
more like females, as compared with the other men, on a number of
measures— on the way they completed a drawing, foi example, on their
preferences in television piograms, and on their attitudes toward children.
The men with symptoms had been happier about the coining event, and
they usually spent moie time feeding, bathing, and caring for their
children than the other men. But on a scale dealing with typically female
activities, such as cooking, washing dishes, and setting the table, the
symptomatology group reported significantly less participation than the
otheis. This indicated, the investigators believe, a shift towaul hypei-
masculinity in the case of activities clearly associated with the other sex.

But the payoff is this. The men were asked if, during their early years,
their fathers had been gone from home, permanently or foi long periods
of time. In many cases they had been, because of divorce, death, or jobs
that had taken them out of town. The men with pregnancy symptoms had
experienced a significantly greater loss in this respect than the others.

In sum, male pregnancy symptoms have been linked by the Munroes to
cross-sex identity in two widely different cultures. This suggests to the
investigators and their Harvard associates that these symptoms may
eventually prove to be a widely usable measure of a man's identification
with the opposite sex. If so, the Harvard group's attempt to prove or
disprove the idea that a peison's primary identification is with the adult
who controls the resources he most wants in early childhood would be
considerably cased. This is because the results of tests to uncover a
person's primary identification are hard to verify. The presence or absence
of pregnancy symptoms in adult males would be a welcome check on the
other tests and hopefully could even serve alone,

Several investigators elscwhcie have compared the families of
Norwegian sailors, gone from home, often for 2 years or more, with
families in which the fathcis were present. The mothers in the sailor
families were found to lay greater stress on obcdicnce-instead of
happiness and self-realixation—than the other mothers. And the boys
tended to be infantile and dependent and to behave in overly masculine
ways. John Whiting hopes to learn someday what happens to these boys.
He would not be surprised if, during adolescence, they become sailors
themselves. Sailing is an occupation suitable for a man who places a high
value on obedience, he says; further, on an extended voyage, it permits a
man to engage in certain work-such as cleaning one's quarters and
sewing— associated with the female role. And Whiting suspects that during
their first voyage these boys may have to undergo a rather severe initiation
ceremony— analogous to the male puberty rites in many primitive
societies, particularly those where a young child has unusual opportunity
to identify with females.

163

The idea that the pubeity ntes serve to stiengthen a boy's masculine
identification wilt he checked in the field during the Whitings' African
yeai. The rites-held in what anthropologists call bush schools, remote
fiom the settlement -continue for days and even weeks and can be
extremely exhausting physically and psychologically. In one reported
case, the young men returning to their village were at fust unable to
lecogm/e its inhabitants. Whiting pioposes to administer a number of
psychological tests to candidates shortly befoie the initiation ceiemonies,
and then again afterward. He expects to find a numbei of cross-sex
identifications the fiist time;veiy few, if any, the second.

In addition to household stuicture during infancy, this reseaich group
emphasizes, prominent elements affecting adult personality aie the
relative importance of the sexes as a child grows up and the personal
attributes a culture values most highly Among people in the lower
socioeconomic levels in the United States, the Whitings note, thegiowing
boy is likely to peiceive men as clearly the more powerful and the more to
be envied; so if his primary identification is feminine, his conflict should
be unusually severe. But among the middle and upper classes, male
dominance is less clear, so the conflict for a boy who has made an initial
feminine identification may be small. Beatrice Whiting refers to a study of
college boys who spent the first 2 years of their lives with their mothers,
their fathers being overseas in World War II.3 These boys were found to
have feminine attributes but to be neither anxious nor defensive about
them. The suggested reason is that academic and intellectual circles place a
high value on sensitivity, aesthetic interests, verbal ability, and certain
othei chanicteiistics usually considered mote typical of the female than
the male.

Answers must still be sought to the following questions, as listed by
John Whiting and a former associate, Roger V. Burton, now at NIMH:

Are theie times when the absence of a father is more critical than at
other times?

How long docs it take for a child to establish identity?

How do the effects of a weak father compare with those of an absent
father?

What is the effect of an absent father on the development of a girl?

Violence and Aggression in the Six Cultures

Further support for the status-envy idea comes fiom Beatrice Whiting's
recent analysis of violence among the societies covered in the six-culture
study. Two of these societics-the Gusii tribe of Kenya, Africa, and
members of the Rajput caste in Uttar Pradesh, India-were found to have,
as compated with the other four, an unusual amount of violence. The men
in the Rajput sample of a few dozen families had been involved in 59

3CaTlsmith, Karolyn Kucfcenberg. Effect of early father absence on scholastic apti-
tude. Ph.D. dissertation, 1963, Harvard.

164

court cases, 1 7 of them concerning incidents that had taken u violent turn,
leading in thice instances to homicide. The Gusii sample -most of whose
men had been involved in at least one court case— remembered seven
instances of violence between men they could name and numcious others
involving men they could not name. The other societies, in conliast,
reported very few if any such instances. Among the Okinawans, the
study's fieldworkcis commented upon "the absence of ciime and the low
incidence of quarrels, disputes, and brawls." The Mixtecans seemed to fear
aggression; children were told that if they became angry and then ate,
they would die. In Luzon, the Philippines, there was a good deal of
quaireling but scarcely any physical violence. In "Orchard Town," New
England, as in these other cultures, assault and homicide cases involving
the people studied 01 anyone they knew by name were exceedingly rare.

Besides the impression that the Rajputs and the Gusii were unusually
aggressive, something else differentiated them from the people in the
other four cultures. In neither of the aggressive gioups was it customaty
for husband and wife to work, play, eat, 01 even-in the sense of oc-
cupying the same room-sleep together.

Most of the Gusii in the sample studied are polygynous. The huts they
build foi each of their wives have a special section, with its own entrance,
for the exclusive use of the husband. The husband either rotates among
these huts or sleeps in a special house, which he may share with his
unmarried sons who have undergone the puberty riles. He seldom, if evei,
visits a wife with a nursing baby.

Among the Rajputs, the men of a household sleep in a separate struc-
ture. When a husband visits his wife, they go to an unused room, or, more
likely, a deserted part of the courtyard, within whose walls the wife
spends most of her young married life, along with her female in-laws.
Aftei a child is born, the husband is not supposed to visit his wife for 2 or
3 years, a taboo which is by no means umvcisally observed.

In both of these cultures, then, the world of Ihe infant is largely a
women's world. For his first several years, the young child sees his father
only infrequently. But when he is able to move beyond the immediate
circle of his mother, the child begins to perceive that the really important
people are the men. "The Rajput 3-y car-old must notice that the women
get down on the floor and cover their heads every time a male enters the
courtyard," notes Beatrice Whiting. "The hungry 3-year-old in Nyansongo
(the Gusii community) must have looked with longing «t the basket of
food prepared for his father. He must have learned that his father has a
private world in his special room or house. He knows that when he is older
he will sleep with his brothers and his father, and will be initiated with
other boys in a ceremony from which all women are excluded," Hence it
is in the Rajput and Gusii cultures that the status-envy theory would
predict protest masculinity, and it is in these two cultures lliat it seems to
be most evident, as indicated by the extent of physical violence, "For the
little boys brought up in the other four societies/' the investigator
observes, "the problem of male control in the world of the 3- to
6-year-olds in not theoretically relevant, as they have already had a chance

to identify with males They have seen men and women mteiact in
intimate settings since biith. Moieover, even if they have made a sttong
feminine identification, there will be less conflict later in life since the
importance of men and women will be more nearly equal and the contrast
between the behavior and personal profiles of the sexes will be less.1'

The Mormons, the Texans, and the Zuni

Status envy and cross-sex identity presumably apply to girls as well as
boys but have been little tested with them. In one of the very few studies
including tests of botli sexes, John Whiting and a group of associates
observed members of three different cultuies living close to one another in
western New Mexico-Mormons, homesteaders from Texas, and Zuni
Indians. The field work, part of a larger investigation supported by the
Rockefcllei Foundation, was clone in 1950 but the analysis was completed
only recently.

The children given the magic-man test weie in grades thiec to six.
Among the Mormon gills, 23 percent chose to be males-reflecting, the
investigators report, the lelatively high status of men among Mormons.
Among the Zuni boys, 10 peicent chose to be girls, reflecting the relative-
ly high status of women among the Zuni. There were no eross-scx choices
by the Texas children, which came as no surprise because Texan men and
women were regarded by the investigators as having relatively equal status.

The effect of cross-sex identification among some of the Monnon
girls-assuming of course that the magic-man test is a valid indicator of
it— is not known. Bui there is some evidence from studies of lower-class
families by other investigatois that girls brought up in mother-child house-
holds become more dominant and aggressive than gills reared in house-
holds where both the father and the mother are present. Perhaps this is
the result of a secondary identification, Beatrice Whiting suggests, arising
from the perception that men in lower-class neighborhoods clearly have
more power than women, Or perhaps it is the result of a primaty identifi-
cation with a mother who has some masculine characteristics because,
without a man, she has to fend foi herself.

In any event, the New Mexico study offers interesting examples of the
relationship between the structure of a family and the way children are
brought up.

The Zuni, as one main instance, emphasized rigid training for the
control of aggression. Was this because of some innate love of harmony?
Not at all, the investigators say, after a look into history. The Zuni
originally lived in single-family houses but by 1300 these had been
replaced by the great pueblos, perhaps because of invading Apaches.
Crowded living conditions and, especially, the requirement that several
women share in the running of the household, led to an emphasis on
harmony. In a worldwide sample of 30 societies, 92 percent of those with
extended families-this is, households comprising not only parents and
children but also a variety of relatives-rated above the median in the

166

severity with which aggressive manisfestations by the children were
punished. Only 22 percent of the nucleai family households— parents and
childien— were equally severe. Punishment for aggression, John Whiting
believes, is directly tied in with the deshe of adult kinfolk, by blood or
marriage, to steer fiee of squabbles likely to be engendered by their
children's fights.

This view gains strength from a recent analysis (by Leigh Minturn,
"University of Illinois, and William Lambert, Cornell) of the data
concerning the mothers in the six-culture study. All the mothers show
some concern for the quarrels of their children, but the Mixtec mothers
are the most concerned, and the New England mothers the least.

The adults in the Mixtec sample, the investigators observe, are highly
interdependent, with many close kinship ties. Brothers and their families
generally live around a common courtyard, where the children play;
relatives look to one another for help with their work and for financial
assistance. Consequently, the Mixtec mothers teach their children to be
unaggressive. Physical punishment is used to punish aggression moie
commonly than it is used for any other type of behavior. Sometimes, to
prevent fights, children are even kept home from school. One result is that
Mixtec adults are unaggiessive and unusually slow to take offense.

The New England families, on the other hand, are the only group whose
membeis are not living next door to relatives and whose livelihood does
not depend upon the support of their neighbors. This means that they can
ignore their neighbors if they cannot get along with them; it also means
they have no claims of kinship to cement relationships in case they do
want to be friends with their neighbors. Under these circumstances, a
mother is likely to tell her child to go play with somebody else if he can't
get along with the children next door; she also seems reluctant to
complain to her neighbois on behalf of hei children.

The investigators emphasize that New England mothcis punish children
if they instigate aggression or if they attack younger childien even though
provoked. More than any other group of mothers in the study, they arc
concerned that children learn the rules of "fail fighting" and the occasions
that justify retaliation.

As another example of the influence of household structure upon child
training practices, the homesteaders from Texas, in the earlier study,
exerted early and strong pressure for self-reliance and independence-
pressure, say the investigators, that does not spring from an innate desire
for their children to be successful but is related, like the Zimi and Mixtec
training against aggression, to living conditions. In the Tcxans' case, events
had brought a swing away from the extended family, a feature of Gieat
Britain in Elizabethan times. The male head of this extended family had
been in full control, and child-rearing piactices had included relatively late
weaning, at two yeais, and swaddling, which hampered movement.
"Dependence was more valued than independence," the investigators
observe, "and obedience was strictly demanded,"

But this type of family fell upon hard times in America, where, because
of the problems of the new environment, achievement came to be valued

167

and patriarchal authority was challenged. The major concern of the Texan
parent, in contiast to that of the Elizabethan, is that the child may be
excessively dependent.

This difference between the Elizabethan and the Texan families js
apparently not unique. In a sample of 30 cultmes, the median age for the
beginning of independence haining-as judged by the tune when thcie is
reduced contact between mothei and child-was found in nuclear house-
holds to be 18 months; for all other households it is 30 months. One of
the factors believed to figure most importantly in independence training is
time of weaning. A survey of the infoimation about 52 societies the world
over found that weaning began at less than a year in only two, the
Chamarros and the Marquesans, of the South Pacific. Among the Texan
mothers studied, the average age was less than 9 months (The Zuni
mothers weaned at the age of 2 or 3; the Mormon mothers, earlier than
the Zuni but later than the Texan.) With respect to child-ieaiing practices
that promote a strong drive for success, the invcstigatois report, these
Texan homesteaders are extreme.

John Whiting and his co-workers consider the Mormons an especially
interesting case. Historically these people had retained the patriarchal
features of the Elizabethan family but adopted polygyny and the mother-
child household. Closely correlated with polygyny the world over is the
post patrum sex taboo, and this was so with the Mormons. Although
definite evidence is lacking* the anthropologists think the Moimons
probably adopted another featuie usually found in polygynous
mother-child households, the practice of mother and infant sleeping in the
same bed.

Under such conditions, Whiting and other authorities have theorized in
recent years, the mother unconsciously redirects her sexual interest
toward her child during the period of the post parlum taboo. Then, in
compensation, and in line with the universal taboo against incest, she joins
with the father in a stiong effort to control the boy's sexual impulses
during childhood and adolescence.

If this is so, then sex training should be most severe in societies where
mother and child sleep together. To test this hypothesis, 18 cultures weic
surveyed. Ten of these were found to be above the median in the severity
with which sex behavior is punished in later childhood, and eight of these
ten have the exclusive mother-child sleeping arrangement. In the eight
cases where the father and mother sleep together and the infant sleeps
elsewhere, sex training was found to be severe in only two,

The Mormons, of course, when this study was made, had long since

id the standard American independent family structure and nuclear

told. But some of the older practices still held: notably, the research

renorts, the mothers had a warm, seductive telationship with their

r worked to severely control sexual impulses and behavior in

"nd adolescence). The dominant value in this culture was

Harmony in the Zuni culture, and success in the Texan.

howed up when children were asked what they most

Texas children most wanted to be successful, and the

Mormon to be good, kind, or happy. The most popular Zuni choice was to
be a man or woman, or, simply, a Zuni.

However, the investigators thought that the Mormons* change in social
stiucture should be accompanied by a shift from virtue to success as the
dominant value and along with this, more emphasis on independence and
less on the control of sex. Considerable evidence was found that the
Mormons studied in New Mexico were indeed moving in this direction:
one-third of the sample could not be distinguished from the Texans in
their child -i earing practices. The investigators predict that "in another
generation the Moimon and Texan family structure and value system will
be indistinguishable."

Monogamy, Economic Progress, Sin, and Guilt

In John Whiting's view, as the preceding section may have made clear,
the association between monogamous societies and material progress is
not happenstance but rooted in chilcl-rcaiing practice. A new baby is
taken into bed with his parents, and soonei 01 later the father says: "Why
don't you wean that baby and get him started?" So there is an early
pressure upon the child toward growing up, Whiting remarks, and this
leads to a strong prcssuie toward achievement. In the polygynous society,
on the othci hand, there is no reason for the child to be weaned early.
And, being in a happy state, why should he struggle for achievement?
Contrary to a common notion, the relative lack of economic progress in
polygynous societies—which include the great majority of those in Africa
and many of those in the Middle East— has nothing to do with the cost of
multiple wives, cither in the bride price paid to acquire them or in the
time, energy, or substance spent to keep them. "In polygynous societies,
by and large," says Whiting, "wives are an economic asset, and a man can
hardly afford not to have more than one."

Children in monogamous, nuclear-household societies, as compaied
with those in polygynous societies, tend to grow up not only with
different ideas about the value of material progress but also, Whiting
believes, with a different type of superego, or conscience. As the investi-
gator explains his idea, a man's role calls for him to be strict and un-
forgiving; a man says: "A rule is a rule." But a woman's rule calls for her
to be aware of contingencies and to exercise forgiveness; a woman says:
"It all depends." So the male conscience is more strict than the female,
and the child who acquires his conscience primarily through identification
with a male finds it less easy to forgive himself than the child who
acquires it primarily through identification with a female. The child whose
primary identification has been with his father should, therefore, have
more guilt feelings than the other child.

If this is so, Whiting goes on, people in monogamous societies, where
the father plays a much stronger role early in a child's life, should have
stronger guilt feelings than those in polygynous societies. To test this idea,
the investigator grouped 28 societies by household structure and rated
each one on an index of guilt, which was the extent to which a sick person

169

blames himself for his sickness. Whiting and Child had developed the
index years before because self-iecrimination as a response to illness
seemed "a probably useful index of the degree to which guilt feelings aic
strong and widely generalized."

The results were as predicted: Most monogamous societies rated high
on this index of guilt; most polygynous societies rated low. Further, (he
proportion of monogamous societies having high guilt levels was great ei
among those with nuclear household stiuctuie than among those with
extended household structure, as might be expected because in the latter a
child is likely to have a greater opportunity to identify with women. And
among polygynous societies, the proportion was higher where the fathei
lived with his wives and childien in the same household than wheie
mother and children lived by themselves.

In monogamous societies, then, as Whiting sees it, social contiol is
achieved largely through the development of a strong-or male-
conscience, which gives one a sense of guilt and a readiness to accept
responsibility for one's actions. But throughout the world he finds two
other independent systems at work to keep behavior within bounds. One
is sorcery, whose believers ate convinced that antisocial behavior will be
punished by the magical power of otheis. Another is a sense of sin, or the
belief that badness will be punished by a supernatural power.

As foi belief in sorcery, the investigator finds that it tends to occur in
societies where mother and child sleep logethei, and the father elsewhere,
and where there is severe punishment for sexual behavior in childhood.
This combination of circumstances, he believes, produces anxiety about
sexual impulses, and the anxiety leads to a paranoid fear-very apparent in
societies having a strong belief in sorcery -of retaliation from other
humans.

As for a sense of sin, Whiting says that "the gods seem to reflect the
parental treatment of childien," He points to cross-cultural surveys by
himself and others indicating that where children arc relatively neglected
during infancy-by not being fed as soon as hungry, for example-and
punished seveiely for aggression in later yems, the society typically dreads
punishment by gods or ghosts; where the parents are more benevolent, the
gods tend to be also.

Whiting is speaking in general terms. He believes that more than one of
the systems-paranoid fear of retaliation by humans, dread of punishment
by the gods, and a sense of guilt and personal responsibility-can act on
one individual at the same time. But he also believes that one or another
system will tend to be emphasized in a given culture, depending upon the
child-rearing practices of that culture.

An Inquiry Into the Origin of Polygyny

Some of the basic determinants of personality appear to be related to
the household structure, and the Whitings think that the household struc-
ture in turn-as suggested in the examples of the Zuni, the Mormons, and

170

the Texan homesteaders— is related to the environment. Recently John
Whiting found himself trying to demonstrate this in the matter of
polygyny, an inquiry beginning with the chance observation at Harvard's
Laboratory of Human Development that societies practicing circumcision,
generally as a part of male initiation rites, aie not evenly distributed. Most
of them are found in the tropics, but the records show none for the
tropical regions of South America.

As noted earlier, male initiation rites occur most frequently in societies
where mother and child sleep together and where a boy, therefore, accord-
ing to the status-envy theory, may acquire a primary feminine identifica-
tion. In turn, the mother-child sleeping arrangement occurs most often in
societies with a long post partum sex taboo, and this taboo tends to be
associated with polygyny. It leads to polygyny, Whiting suspects, rather
than the other way around. But what leads to the taboo?

The tribesmen themselves generally explain that if a lactating woman
has sexual intercourse, her milk will become sour or thin and her baby will
get sick. The real explanation, Whiting conjectures, lies in the musing
mother's conscious or unconscious dread that she will conceive again and
thereby endanger the life of the child she is nursing. There would be good
reason for this dread if the mother was on such a poor diet that pregnancy
would reduce the already low protein value of her milk below the danger
point.

Combing through the records of diets the world over, Whiting found
that the long sex taboo was found most frequently among societies with a
diet low in protein, and that these low-protein societies were found most
frequently in the rainy tropics, where the climate is conducive to the
growing of low-protein root and fruit crops. The investigate! assumes that
a diet based largely upon such crops probably leads to a high incidence of
a protein deficiency disease, kwashiorkor. Aware of this, at some level of
consciousness, a mother avoids getting pregnant while she is lactating. And
when she avoids it by abstinence, the husband is led to seek another wife.
Thus Whiting— while emphasizing that his assessments of climate,
nutrition, and health in many cases had to be crude-traces a connection
between polygyny and climate. As to the polygynous societies in the
South American tropics, he suggests that many of them practice abortion
rather than abstinence as a means of child-spacing.

The Effect of Chores Upon Character

Analyzing the reports of the fieldworkers in the six-culture study,
Beatrice Whiting finds fiuther evidence to support her husband's thesis
that child-rearing practices, and therefore personality, depend to an
important extent upon certain fundamental characteristics of a society.

One of the important variables, the analysis suggests, is the kind of
tasks a child is expected to do. The most responsible, from the standpoint
of recognizing that something needed to be done and doing something,
were those who were expected to take care of their younger brothers and

171

sisters or of cousins. This babysitting choic seemed to be related, in turn,
to the amount of woik a woman was supposed to do outside the house.
The highest ratings for responsibility went to the Gusii children of
Nyansongo, in Kenya, and the Mixtecan children of Juxtlahuaca, Mexico.
And in both these cultures the women had to be away from the house
much of the day.

The children given the lowest tatings for responsibility were those of
Khalapur, India, where baby tending was a common chore but the
mothers were always home, and Orchard Town in New England, where the
commonest chores were household tasks and the mothers did not ordinari-
ly work outside the home while the childien were young.

No matter what the culture, ghls were scored higher on responsibility
and muturance than boys. If a young child was poking a knife at another
youngstei, for example, or needed help in ciossmga road, it was an oldci
sister who was moie likely than an older brother to notice what was going
on and take action. Such responsibility is a matter of training, Beatrice
Whiting thinks, and has a lifelong influence, Specifically, in the six-culture
study, it was found that children who weie assigned the caie of younger
brothers and sisters at an early age were more nurturant and icsponsiblc
about all younger children. Since girls are assigned this task moic
frequently than boys, it is not surprising that they score higher in both
iiurturance and responsibility.

This investigator finds other behavioral differences between girls and
boys that hold true for all six cultures and can probably be considered
univeisal sex differences. Boys are more aggressive; in spite of radical
differences in the way they are brought up, the boys of all the cultures
strike other persons more frequently and engage in moie rough play than
the girls. They are also more mobile. In the six cultures, little boys wander
farther from home than little girls, and older boys stay away fioin home
much more dining the day than older girls. Mrs. Whiting is not sine
whether or not the higher mobility of boys may be ascribed to differences
in the way paients treat the two sexes. In all six cultures, the girls are
assigned more household chores, so the girls must spend more time at
home. Still, little boys may just naturally be more difficult to keep boxed
in.

Differences in the chores assigned to the children may result not only in
some of the peisonality diffeiences between the sexes, Beatrice Whiting
believes, but also in some of the differences in how societies view the
supernatural. Her reasoning goes like this, If the chores involve for the
most part the care of children or animals, the consequences of failure will
be immediate and obvious. If a baby is left untended, for example, he may
put noxious things in his mouth or fall and hurt himself; if cattle are left
untended, they may get into the corn and eat an important part of the
coming year's food. In contrast, the importance of getting the dishes
washed or the house cleaned is less clear because results of failure are less
dramatic. Household tasks must seem arbitrary as compared with baby-
sitting or cattle tending, and schoolwork-whose importance must be
taken largely on faith-must seem even more arbitrary. Mrs. Whiting

J72

believes that a child's view of the tasks and of the rewards and punishment
associated with it will be generalized to his view of the nature of the world
order, and this world view will be leflected in religious beliefs. Con-
sequently, if children perceive their tasks as arbitrary, with major leward
or punishment absent 01 remote, as in the case of schoolwork, the
membeis of the society will tend to believe in a future life wheie rewaids
and punishments are handed out for the deeds clone 01 not during this
one. But if children know that failure to perform a task will be punished
at once, the society is likely to believe that punishment by supernatural
beings is meted out now rather than in a future life.

Two surveys of cultures believing that supernatural beings are
concerned with the moral behavior of man give results viewed by the
investigator as supporting these ideas. The fust suivey dealt with the
effects of schools. Five of the 26 societies in this sample have schools, and
all five believe in a future life where rewards and punishments aie handed
out for behavior in the present life. Of the 21 societies without schools,
14 do not believe in punishment in the futuio life. The second survey
dealt with the effect of cattle raising. Thirteen of the 36 cultures in the
sample are cattle-iaising societies, and 1 2 of these believe that punishment
by the supernatural is immediate. Half of the 22 societies that do not raise
cattle believe that punishment by the supernatural comes now; the other
half, in a future life. "But ceitainly when we aie dealing with such
complex subjects," John Whiting remarks, "moie factors are involved than
the ones we've been chiefly concerned with."

Beatiice Whiting thinks there may well be a critical age -from about 3
to about 8-for the influence of task assignment. Her own research has
dealt with children under 10. Older children have been inculcated with
such values as cleanliness and orderliness, she observes, and they are better
able to understand future goals; so household tasks and schoolwork will
seem less arbitrary.

If certain facets of personality arc indeed related to task assignment,
Mis. Whiting points out, one should expect to find wide personality
differences within a society, particularly one like ours where the assign-
ments show wide variations. Children who grow up in large families and
have to help care for younger children; children whose tasks are obviously
related to the economic welfare of the family, as in families with farms or
grocery stores; and children in households where the mother works and
where the children's failure to do the household chores results in chaos-
all these should be diffeicnt from children whose chores seem arbitiary
and are frequently left undone and where such failure often goes un-
punished, as was the case with the youngsteis of Orchard Town.

The Relation Between Stress During Infancy and Adult Characteristics

Laboratory studies have shown that animals, particularly rats, subjected
to an unusual amount of stimulation during infancy grow up better able
to cope with stress, whether physiological or psychological, and to be

173

significantly largei , too. The animals develop faster and they learn fastei.
John Whiting and several members of his group-in particular, Thomas
K. Landauer, now at Stanford, and Shulamith M. Guilders, now at Bar-
Han Univeisity, Israel-have been much interested in learning whether or
not the same thing applies to human beings, and why. If so, the implica-
tions might be tiemendous. The findings might even point to a way of
strengthening psychological health for a lifetime through a simple
procedure during infancy.

The inquiry began as the narrowest of sidelines to the Harvard group's
main efforts. It was undertaken because the chance to compare laboratory
findings with the findings of "natural experiments," as contained in the
reports on a number of cultuies known to expose infants to certain stress-
ful practices, seemed too good to pass up.

The stimulation found effective in animals has included exposure to
cold and to electric shock, painful manipulations, and supposedly gentle
handling—lifting the infant from the cage, stroking it foi a few minutes,
and then returning it. Apparently all such stimulation has a lasting effect
on the endocrine gland system. Landauer and Whiting, leafing through the
material on scores of societies, noted many practices that might be
considered stressful to the human infant-among them, exposure to
extreme heat or cold, the administration of emetics or enemas, scraping
the skin with a shell or other sharp object, and tight swaddling. Then they
compared the average height of adults in the societies engaging in these
practices with that of adults in other societies.

Significant differences were found when children under the age of 2
were exposed to either one of two main classes of stress: (1) Piercing,
which includes circumcision, inoculation, and piercing the nose, lips, or
ears to receive an ornament, and (2) molding, which includes shaping the
head and stretching the arms or legs, usually for cosmetic purposes. Out of
66 societies, the men in those that practiced either molding or piercing
were more than two inches taller on the average than the men in the other
societies. This relationship between infant stress and adult statute was
found in every major geographical legion and apparently was not
influenced by either diet or climate. It held true for both women and
men.

Further evidence comes from comparing two groups of individuals in

the Pels Growth Study, which was begun in the late 1920's by the Pels

Research Institute, Yellow Springs, Ohio, and is continuing. The aim is to

learn what factors have helped shape the physiological and psychological

characteristics of the 150 persons being studied. When the children were

being selected for this work, John Whiting notes, the practice of early

inoculation-before the age of 2-was just being introduced. So, as it

ha^ru^ed, some of the children in the Pels group were inoculated eaily;

^ot till they were much closer to school age. The Harvard team

Uyzed the records of these two groups and found that the

vho had been inoculated before the age of 2 are now, as

icantly taller than would have been predicted from informa-

tature of their parents.

1/4

Shulamith Guilders then took another look at the reports of the labora-
tory experiments with animals and noted thai in every case, in ordei for
the stress, or stimulation, to be applied, the infant animal had been
separated from its mother. Perhaps, she reasoned, the crucial factor was to
be found in this separation. For each of 75 societies scattered over the
world she worked out a separation score on the basis of information about
the baby and its mother during the first 2 weeks after birth-whether or
not, for example, the baby was nursed by the mother, slept with his
mother, and was extensively handled by people other than his mother.

Dr. Guilders' hunch proved out. In societies with high separation scores,
meaning that mother and infant were separated comparatively often
during the first 2 weeks, the average height of adult males was 65.8 inches;
in the other societies it was 63.7 inches. (Males were used because the
figures for them were more often available, but much the same difference
was found for females.) As in the first suivey, this difference was
statistically significant at the .001 level; the likelihood that it had
occurred by chance was 1 in 1 ,000.

The correlation between mother-infant separation and adult height held
good whether or not a society subjected the young child to a physically
painful type of stress. But where both types of experience occurred—
separation from the mother, which is presumed to be stressful, and either
piercing or molding— the difference in adult stature was especially marked.
In societies whose infants were not stressed in either manner the average
male height was 62.9 inches; it was 65.9 inches where the infants were
stressed in both ways.

Recent work elsewhere with rats, the Harvard group notes, supports the
notion that periodic separation from the mother during early infancy is
stressful. In this research the crucial factor apparently was not the
separation itself but the accompanying drop in body temperature (about 4
degrees C.) when the infant rats were exposed to room temperature.
Exposing the rats to low temperature in the presence of the mother
produced the same effects as removing them from the mother. Separating
them from the mother under conditions-in an incubator-designed to
maintain the body temperature produced no effect.

Physical stress and periodic sepaiation from the mother in early infancy
lead not only to taller men and women, John Whiting found, but also to
an earlier start of menstruation. In a sample of 50 societies, the average
age at mcnarche in those societies whose child-rearing practices include
neither form of stress is 14 years; where one of these forms occur, the
average is 13 years, 6 months; where both forms occur, it is 12 years, 9
months. Incidentally, no evidence at all was found to bear out "the folk
belief that girls in the tropics mature early.*' So far as age at menstruation
is concerned, neither diet nor closeness to the equator could be found to
make any difference.

Dr. Guilders now wondered if there might not be a relationship
between recent increases in height in a number of countries-including the
United States and Japan-and the growing tendency over the last few
decades to have babies born in hospitals, where they are separated from

175

their mother more frequently than they would be at home. She has been
trying to answer this finding among the Israelis who emigrated from
Yemen. Befoie the move, a baby was usually bom at home and spent the
first several weeks with the mother, on hei bed, during whicli time the
mother was fieed of all household duties. After their arrival in Israel,
however, Yemenite women were gradually prevailed upon to give birth in
maternity wards, wheie a baby is geneially brought to his mother only at
4-hour intervals and only during the daytime. Dr. Guilders has started a
long-teun study of 300 Yemenites, born in Isiael between 1950 and 1957,
about half of them in maternity centers and the rest at home, where they
were cared for dining the first weeks in the traditional way. So far the
investigator has analyzed the weight data for the fiist 4 years. At the age
of 1 month the homeborn children were slightly heaviei than the hospital-
boin, but the difference was not statistically significant. At the age of 1
year, the average hospital-born child was 363 grams heavier than the child
born at home, and by the age of 4 this difference had increased to 1,097
grams or about 2'/2 pounds. Dr. Gundeis believes she has ruled out all
possible reasons for this highly significant weight diffeience-including
possible differences in diet or in mothering-except early stimulation
resulting from periodic mother-baby separation during infancy.

The two groups aie to be followed through the years and compared on
a number of measures. The hope is to leatn if early separation, like that
customa'iy in most maternity wards, has psychological as well as physical
consequences, as it has been shown to have in lats.

John Whiting intends to do a similai study in the United States. He
points out that some hospitals now permit mothers to keep their new
babies with them most of the time, and he thinks it will be possible to
compare these children with those who spend most of their first few days
in the usual cential nursery. He expects to find both physical and psy-
chological differences.

Research Grant Mil 1096
Date of Interviews' Nov 1 , 1966

References

Burton, R. V., and Wlritting, J. W. M. The absent father iind cross-sex identity. Men-ill-

Palmer Quart. Behav. Develpm., 1961 , 7, 2, reprinted as A-277, "The Bobbs-Mcrrill

Reprint in tlie Social Sciences," Bobbs-Mernll, Indianapolis,
Gunders, Sluilarmth M., and WhHting, J. W. M. The effects of periodic separation from

the mother during infancy upon growth and development, Congr. Anthrop. &

Ethnol, Sci., 1964.
Guilders, Shulamith M., and Whitmg, J. W. M. Separation from mother during infancy

and physical growth— a cross-cultural study. Unpublished manuscript, 1966.
Landauer, T. K., and Whiting, J. W. M. Infantile stimulation and adult nature of

human males Amer. Antrop., 1964, 66, 5.
Minturn, Leigh, and Lambert, W. W. Mothers of six cultures: Antecedents of child

rearing. New York: John Wiley & Sons, 1964.
Munroe, R. LM Munroe, Ruth H-, and Whiting, J. W. M. Structure and sentiment;

Evidence from recent studies of the couvade. Amer. Anthrop. Assoc. meeting,

1965.

176

Whiting, Beatrice B (Ed.), Six cultures' Studies oj child reanng. New York: John

Wiley & Sons, 1963.
Whiting, Beatiice B. Task assignment and character development. Unpublished

manuscript, 1962.
Whiting, Beatrice B Sex identity conflict and physical violence: A comparative study.

A met. Anthiop., 1965,67,6
Whiting, J. W. M Socialization piocess and personality In Francis L K Hsu (Ed ),

Psychological anthropology approaches to culture and personality Homewood,

111.- The Dorsey Press, 1951.
Whiting, J W. M Sorcery, sin, and the superego A cross-cultural study of some

mechanisms of social control. In Nebraska Symposium on Motivation, Umv of

Nebi. Press, 1959.
Whiting, J. W M, Resource, mediation and learning by identification, In Ira Iscoe and

H. W. Steven (Eds.) Peisonahty development in childien. Austin, Tex. Univ of

Tex. Press, 1960

Whiting, J W. M. Effects of climate upon certain cultural practices. In W. Good-
enough (Ed.), Explorations in cultural anthropology. New Yoik' McGiaw Hill,

1964.
Whiting, J. W. M. Menarcheal age and infant stress in humans In F A. Beach (Ed ), Sex

and Behavior New York. John Wiley & Sons, 1965
Whiting, J. W. M., Chasdi, Eleanoi II., Antonovsky, Helen F., and Ayies, Barbara C.

The learning of values. In E. Z. Vogt and Ethul M. Albert (Eds.), People of

Rimrock, a study of values in five cultures. Cambndge: Ilaivarcl Univ Press, 1966
Whiting, J. W. M., and Whiting, Beatrice B. Contributions on anthropology to the

methods of studying child rearing. In P. II. Mussen (Ed ), Handbook of reseat ch

methods in childhood development. New Yoik: John Wiley & Sons, 1960.

77

Investigator

BettyeM.Caldwell, Ph.D.
Syracuse University
Syracuse, N.Y.

Prepared by-
Herbert Yahraes

A Syracuse research team has studied the learning characteristics c
children from 1 month of age to 3 years and hopes to relate thes
characteristics to certain features of the home envnonment and also to th
children's mental development. To prevent the retaidation often seen i
children from the lowest socioeconomic level, two types of intervcntio
are being tested: an excellent day-care program beginning when a child i
as young as 6 months and a parent-education program that includes
moving picture, produced under the grant, showing motheis what a bab
can do and how his family can help him develop interests and skills. Earl
results indicate that about half the childien under 6 months respond t
conditioning procedures, that the day-care project quickly leads to soni
increase in IQ's and that IQ increases also occui among babies living i
homes rated high on a stimulation inventory.

When chitdien are very young, one group will perform much the sum
as another on developmental tests, regardless of social or racial origii
Beginning somewhere between 18 months and 2 years, however, the cuiv
representing the peiformance of children from the lowest socioeconoini
level Begins to drop and fiom then on these children as a group scor
significantly lower than other children on measures of ability and achieve
ment.

Commenting upon these findings by a number of investigators, a Nci
York State research team points out that evidently something happens, c
fails to happen, during a critical period early in life to stunt the into
lectual development of disadvantaged children. Consequently, they cnU
school with a handicap many of them can never overcome.
_ Why the difference? What happens or doesn't happen? How can th
situation be changed?

The answers are being sought in a many-angled research program undei
taken by Dr. Bettye M. Caldwell, professor of child development an
education at Syracuse University, and Dr. Julius B. Richmond, dean of th

*Now at the Center for Early Development and Education, University of Arkonso
tie Rock, Arkansas

College of Medicine and chairman of the Department of Pediatrics,
Upstate Medical Center, State University of New York. If the answers can
be found, these investigators point out, they can be used to develop more
effectively one of the Nation's most vital resources, the intelligence of its
people, and at the same time to promote mental health by combating an
important source of dissatisfaction with one's self and with society.

One bianch of the program seeks information on how the learning
process develops between the first month and the third yeai of life and
how differences in development are influenced by differences in the
pattern of family care. Concurrently the investigators are testing one
proposed means of preventing deficits in learning ability. This is through a
program of "massive intervention," as Dr. Caldwell describes it, in the
form of excellent caie 5 clays a week in the research group's children's
center.

The day-caic pioject, made possible by a grant from the Children's
Bureau, diffcis substantially from most other emichment programs for
deprived childien in these respects:

1. The project sets the minimum age for admittance at 6 months
instead of the usual 3 or 4 years. This is because the investigators are
convinced that sensitivity to enrichment declines with age. "The vciy
early years represent a crucial period for the prevention of learning
deficits," they report. "Instead of devising methods of reversing the
decline, it would be wiser and perhaps more economical to devise ways of
blocking the process of decline before it has begun to alter the organism's
adaptive capacity."

2. The project offers "programmed care" by a staff of teachers and
nurses fai larger, in proportion to the number of children, than usual: one
adult to about every four children. The goal is to help each child become
as aware as possible of the world around him, eager to participate in it,
and confident that what he does will have some impact on it.

Preliminary findings suggest that this particular form of manipulating
the environment does influence the learning process. Approximately 30
children who had attended the center for at least 3 months showed an
average 1Q gain of six points. When these childien entered the program,
they ranged in age from 15 to 32 months and their IQ's averaged about
J03; on the rctest, the IQ's averaged about 109.

There is no strictly comparable control group al this time. However, the
investigators do have test results at 6, 12, 18, and 24 months for 23
children, also from the lowest sociocconomic level, who did not par-
ticipate in the day-care program. At 6 months, these children's IQ
averaged almost 120. At each rctest from then on it dropped; at 24
months it was about 100.

Behind the day-care program lies the conviction that low-income
families often do not provide the stimulation that even a very young child
needs if he is to develop fully his capacities for perceiving and reasoning.
But there may be another approach to the problem-parental education.
To test this possiblity the investigators have made a moving picture, "How
Babies Learn," which will be shown to some of the mothers of the

179

children being studied but not to others The motheis who see it will be
encouraged to discuss the ideas presented and to ask questions, Perhaps
these mothers will then provide a more stimulating atmosphere foi their
children, the mvestigatois reason, and if they do, it may be possible to
measure the results as the study proceeds. Such inteivention will be far
less expensive than the day-care piogram; of couise, the question is
whether or not it will be effective.

The study on infant learning follows a child until he is at least 3 years
old. It keeps tabs on his physical and mental development, puts him
through experimental procedures from the very beginning in an effort to
study his patterns of learning, and closely observes his physical and social
environment. Where this environment rates high in stimulation value, a
preliminary analysis shows, a baby's IQ increases between the ages of 6
and 1 2 months; where the stimulation value is low, the baby's IQ is likely
to drop.

The project now includes 50 families, some white and some Negro,
most of them living in public housing. Soon, under the plan, it will add 50
babies from families living in university-operated housing for mairied
students. The new families, too, will have meager incomes, but their social
and economic backgrounds and outlooks presumably wili differ con-
siderably from those of the families now in the study. The idea is to look
for differences both within and between the two gioups on a num. bet of
matters believed to affect a young child's learning ability.

Each group will be divided three ways so that two levels of manipula-
tion—day-caie and parental education— can be tested, and the children in
these subgroups compared with those who are merely tested and observed.

Studying the Learning Patterns of Infants

Not all children aie motivated to learn by exactly the same proceduics.
Promise one child a nickel for doing something, Dr. Caldwell points oul,
and lie does it; promise another child a nickel, and he doesn't stir. One
child may be crushed by a spanking, another gets up and says, "Ha-ha-ha
-you didn't hurt me." A mother who resolves to have infinite patience
and never to use punishment may easily stick to her resolutions with a
docile child; she may be tempted to abandon them, however, after a
different child for the third time has deposited a bowel movement on the
living room or supermarket floor.

Apparently, says the investigator, each child has a certain learning
pattern which results from the interaction between his inherited charac-
teristics and the way he has been dealt with since birth and which differs
at least slightly from the patterns of other children. To test this idea, the
researchers are studying each child's behavior in an array of learning
situations. "Assuming that a child does have a particular pattern/' Dr.
Caldwell observes, "it may well influence what he does on developmental
measures, such as the Cattell Infant Intelligence Scale and the Griffiths
Test of Mental Development, For example, if a child is very sensitive to

ISO

social reinforcements and happens to be examined by someone who is
friendly and smiling and full of encouragement, the child is likely to give
his veiy best performance. But a child who regards smiles and piaise as
almost an intrusion on his own cffoits to solve a problem may well
respond negatively in the same situation and thus not reveal his true
capabilities,"

In this study the early learning situations are conditioning piocedures,
which are applied foi the fiist time when the child is a month old (and has
been brought to the well-baby clinic at the Children's Centei). It is now
known that some children condition dining early infancy and some do
not; what the Syracuse group wants to find out is the relationship
between eaily conditionability and later learning. It also wants to relate
the findings from the conditioning tests to such matters as the home
environment and the parents' ideas about molding a child's behavior.

In the first proceduic, the researcher holds a yellow disk where the
baby can see it and simultaneously squirts a little air at the baby's face.
The baby blinks. Aftei a few trials, some babies become conditioned: they
have learned to blink whenever they see the yellow disk, whether or not it
is accompanied by a puff of aii. Some of the others will learn to associate
the two stimuli when they are next tested, at 2 months.

At 3 months, when sounds begin to be important, the psychologist in
charge of this part of the research-Stanley Moldovan— changes the con-
ditioning technique and uses a tone to signal that the baby's foot will be
tickled. Again, some babies after only a few trials will flare the toes, or
cuil them, or move Ihe foot— without being ticked— as soon as they hear
the bell. Some of the others become conditioned this way the second time
around, when they arc 5 months old, and some do not.

Pieliminaiy findings— concerned with about 35 of the babies now in the
study, all from families low on the socioeconomic scale-show that:

• About half the babies tested at any given age can be conditioned.

• Most of those conditioned to respond to a visual stimulus, the yellow
disk, at 1 month can be conditioned to respond fit 2 months. And most of
those conditioned to icspond to an auditory stimulus, the tone, at 3
months can be conditioned to respond at 5 months.

« But there seems to be no connection between a baby's readiness to be
conditioned to a visual stimulus at I and 2 months and his readiness to be
conditioned to an auditoiy stimulus at 3 and 5 months.

Among older children, some are known to acquire information more
readily through their eyes; othets, through their cars. The Syracuse
findings may be an indication that the same clivision-into what some
investigators call "visiles" and "audiles"— holds true very early in life.

Usually when an investigator applies conditioning techniques to very
young childicn, he is inteiestcd mainly in learning whether children of a
given age can be conditioned in a given way. The present work, though, is
directed toward learning what relationships may exist between early con-
ditionability and later behavior. For example, will children who can be
conditioned early turn out to be relatively quick learners later on? Will

181

there be any connection between early conditionability and later IQ?
Between eaily condition ability and socioeconomic background?

The answer to that question on the effect of social class will be no, il is
believed, because conditionability very early in life is probably determined
by the biological characteristics of the child and not by what has been
happening to him. Babies from middle-class families tested by Moldovan
for another Upstate Medical Center research pioject showed much the
same conditioning pattern as those in the most recent work. By the time
children are a year or so old, howevei, diffeiences related to socio-
economic class-or, perhaps moie accurately, to differences in the patterns
of parental caie-are expected to appear.

The procedures using visual and auditory signals are classical or
Pavlovian conditioning, in which one stimulus comes to be associated with
another. The Syracuse group is also studying the children's response to
operant conditioning, in which behavior is shaped by the giving or with-
holding of rewaids or punishments. Here, too, little is known either about
the factois influencing conditionability in children or about a possible
lelationship between conditionability and later learning.

The first operant piocedure tries to influence a baby's vocalization rate
at the age of 4 months. As Dr. Lois Henning, the psychologist in charge,
explains, infants between the ages of 3 and 5 months babble a good deal
even when no one is around. This is spontaneous vocalization. After about
5 months, the child tends to limit his babbling to those occasions when
there is someone to notice it-and to reward him for it with smiles and
talk and play-all of which constitute, in the terms of learning theory,
"social reinforcement."

There is some evidence that babies from middle-class families aic
rewarded more often for their cooing and babbling and attempts at talking
than babies lower down on the socioeconomic scale. And this is one of the
reasons, it is theorized, that babies in the first group get a hcadstart
verbally and icach school age considerably better equipped to benefit
from the formal educational process.

The Syracuse group asks whether or not there are marked differences iit
4 months in the capacity of children to raise their vocalization rate in
response to social reinforcement. If differences do exist, ate they mainly
between classes or among individuals regardless of class? Do the results at
4 months predict anything about verbal ability and learning ability later
on? If accurate predictions can be made about these abilities, what
happens when there is some sort of intervention early in the child's life-
specifically, parental education by way of the moving picture the research
group has prepared, or exposure of the child to the enriched environment
of the day-care program?

To establish a baseline, the baby's vocalizations are recorded fora few
minutes as he rests in a quite room. The experimenter is present and in the
baby's sight but makes no response to his babbling. During the con-
ditioning period, the investigator smiles at the baby whenever he babbles
and gives him a warm "yes, yes" and a friendly pat on the tummy, This is
the kind of reinforcement, the research team believes, that operates in the

182

natural environment of children and may have a critical role in the de-
velopment of their vocal behavior and of language.

There are no answers yet. Some of the babies— again, about half-do
increase their rate of vocalization; the others show no response to the
social reinforcement. The investigators think that responsiveness may well
be environmental-that the baby who responds in the laboratory situation
is the one whose mother talks to him a good deal when he talks, so he
learns earlier than the others to use vocal behavior to accomplish social
ends.

Whether or not a child is conditioned, the investigators have a measure
of his spontaneous vocalization. They will try to relate this, also, to his
use of language latei on.

Dr. Caldwell and Dr. Henning want to learn, in addition, if this type of
conditioning would be a useful training device in enrichment piograms for
underprivileged children. So they are using it with a few of the infants in
the day-care program, making a systematic effort to increase their vocal
behavior in the belief that language development may thereby be
facilitated.

From the time a child is half a year old, most of the early-learning
experiments involve an apparatus that can be programmed to give him one
of several kinds of rewards (a bar of music, a trinket, a bit of food, a voice
saying, "uh-huh, uh-huh, very good, that was fine, do it again") in return
for pressing the coirect lever or pattern of levers, The leveis can be dis-
tinguished by their position or by the size or color of their identifying
symbol-for example, dots. The point here is to study differences in the
rates of learning and the effects of a given icwarcl and then to learn
whether or not these differences arc related to the children's backgrounds
and to performance on IQ and other tests.

In related work, Moldovan recently found a difference in the reward-
seeking behavior of boys and girls when the reward was a toy and when
there was a delay between the time of the currect response and the
presentation of the reward. Both boys and girls would start out by ex-
perimenting with the three levers of the apparatus until there was a
payoff. Then the boys would continue to experiment until they had
found the correct lever. The girls, on the other hand, would try to follow
the same pattern that had led them to the reward in the first place. Each
girl had her own pattern— "her own superstitious way"— of tiying to get at
the reward, and eventually, through a series of modifications, she would
make the pattern pay off. The investigator cites this as evidence of differ-
ences in the ways people learn. In this case something associated with the
sex of the children made a difference; in the work going on, additional
factors may show up.

Studying the Stimulation Value of the Home

Four times a year a public health nurse on the staff of the Children's
Center visits each family in the infant learning study and evaluates it on

183

the basis of Dr. Cald well's "Inventory of Home Stimulation." The in-
ventory contains 72 items, all of which grew from certain assumptions by
Dr. Ctildwell-based on her own experience, on research by other investi-
gators, and on expert opinion-about the conditions that foster a child's
development.

The development of a young child, Dr. Caldwell assumes, is fostered
by.

1. A relatively high frequency of adult contact involving a relatively
small number of adults (the mother, the father, and, when the mother is
away, one of not moie than three regular substitutes).

2. The provision of a social learning environment that both stimulates
the child and responds to him. (For example, the mother reads to the
child at least three times a week, responds to him verbally when he
vocalizes, tells him the names of things and people, encourages develop-
mental advances such as waving bye-bye and saying his name, and supplies
toys that challenge him to develop new skills. She gives these toys added
value in his eyes by demonstrating her own interest in them.)

3. An optimal level of need gratification, defined as sufficiently
prompt attention to the child's needs so that the young organism is not
overwhelmed, but not so prompt or complete that budding attempts to
meet his needs himself are aborted or extinguished.

4. A positive emotional climate— an interpersonal situation through
which the child learns to trust others and himself. (For example, during
the visit of the public health nurse, the mother spontaneously praises the
child's qualities or belutvioi, does not shout at him or express annoyance
with him, caresses him at least once, and reports that no more than one
instance of physical punishment occurred during the preceding week.)

5. An environment that contains few unnecessary restrictions on the
child's early exploratory attempts. (The child is kept in playpen 01 jump
chair no more than an hour a day, is taken promptly from his crib when
he awakens from a nap, is not slapped ot spanked for spilling or spitting
food or drink.)

6. The provision of rich and varied cultutal experiences. (The investi-
gator is interested in learning, foi example, whether or not the child cats
at least one meal a day with his parents, is taken into a grocery store at
least once a week, goes on an outing with his family at least every other
week, is taken to church by a member of the family twice a month or
more.)

7. A physical environment containing modulated amounts and varieties
of sensory expeiience. (For instance, the house is not overly noisy and is
neither dark nor monotonously decorated, and the family has at least one
pet, one house-plant, and 10 books.)

8. Access to certain kinds of play materials. For a child under 1 year,
these include a cuddly toy; items, like beads and blocks, that go in and oul
of a icceptacte; a push or pull toy; a fit-together toy, and one or two cloth
or cardboard books. For a child between 1 and 2 years of age they include
a child-size table and chair, a ride toy such as a scooter or kiddy car, large
blocks or boards, bang and hammer toys, access to a record player and to

184

children's records. For a child between 2 and 3 years of age, they include
simple wooden 01 heavy rubber puzzles, medium-size wheel toys; role-
playing toys, such as those used in playing at being a cowboy or a mother;
and at least 20 childien's books.

A home's stimulation scoie is the total number of items checked "yes"
by the nurse as the result of her observations during the visit and her talk
with the mother.

The Home Stimulation Inventory, Dr. Caldwell notes, is an experimen-
tal technique. There is yet no proof that any of the items comprising it can
influence a child's development. However, a recent comparison of stimula-
tion scores and of changes on the Cattell Infant Intelligence Scale between
6 and 12 months showed that positive IQ changes had occuned in
children from homes earning high scores and that negative changes were
the general pattern in children from low-score homes. Seveial years from
now the investigators will know more about the impact of this variable.
They expect to be able to say, by looking back at what each child had in
his home environment at various periods, which items and groups of items
are the most sensitive indicators of developmental progiess in later years.

The Effect of Parental Theories About How To Influence Behavior

In a second approach to assessing the family environment, the research
team is studying each mother's ideas about the most effective ways of
shaping the behavior of her child. These ideas, the investigators point out,
represent a mother's theory of how children learn, and it would be helpful
to know whether or not one theory has greater effect than another on the
learning ability of a young child. It would be helpful to know also
whethci or not there is any difference between social classes in the
pjevalence of a given theory.

The research team has prepared lists of behaviors characteristic of many
children-one list each for 1 -year-olds, 2-year-olds, and 3-year-olds. Among
the 45 behaviors listed for 1-ycar-olds are, for example, climbs out of bed
after being put to bed, pokes finger or object into an electrical outlet,
imitates words 01 sounds, and acts afraid in the doctor's office. The
mother is asked in each case whether she would encourage or discourage
the behavior-provided it made any difference to her at all-and how she
would go about doing so.

Dr. Caldwell describes the courses open to the mother as:

1. Manipualtion of privileges and tangibles. -The mother can give the
child a cookie, for example, take him on a trip, let him stay up late-or
she can refuse to do so,

2. Manipulation of maternal emotional responses. -The mother can kiss
the child, smile at him, thank him, frown or glare at him or become upset,
or make no response at all.

3. Manipulation of child's emotional sm/e.-The mother can enhance
the child's esteem by praising him, undermine it by ridicule or by forcing
an apology, undermine it through a scolding, or undermine it by physical

185

punishment-cither mild, like a swat on the bottom, or severe, like a
spanking.

4. Manipulation of input—The mothei can point to herself, her
husband, or another child as a model of behavior for the child to imitate.
Or she can provide a verbal explanation of why he should behave in a
given manner. Or she can demonstrate what he should do.

5. Manipulation of the environment. -The mother can establish n
schedule to include, for instance, putting the child to bed at the same time
every day. Or she can change things-for example, placing a plant out of
reach if the child picks it leaves-so that an unwanted behavior cannot
occur. She can also get at the cause of an unwanted behavior and correct
it,

6. Mandate, -The mothei can tell a child-through a simple command
or through insistence or threat— what to do or not to do.

The Syracuse group thinks that all these ways can be effective but that
the skillful mother probably uses Nos. 4 and 5 moie than the others. The
investigators expect to find positive manipulations, such as granting
privileges and giving praise, more commonly used by middle-class than by
lower-class mothers. If so, Dr. Caldwell notes, the middle-class mother
would have the support of learning theory, which says that rewarding a
desired response is a far moie powerful way of shaping behavior than
punishing an undesired one.

"Maybe only 50 percent, or even considerably less, of the way a child
develops can be influenced by his environment, no matter what this en-
vironment is or what is done to improve it," Dr. Caldwell remarks. "But
environment is the only thing you can do something about; and to do it as
effectively as possible, you have to understand what factors in it are most
influential." Parental theories about how to shape a child's behavior, she
suspects, will prove to be among the important factors.

Some Maternal Behavior Factors That Influence Development

In a related study,1 the same investigators have been concerned with
learning how a child's early experiences influence his social and
personality development. The two dozen children in this study are all
from low-income families, so any differences reflect differences between
families rather than between social classes. Among the data now available
are the reults of developmental tests given the children at approximately
6-month intervals during the first 2 years of life and ratings covering a
broad range of maternal behavior as observed at intervals running from 3
to 6 months.

The investigators find that those children have reached a higher de-
velopmental level whose mothers:

I. Gave them more warmth and affection (the influence of these
quaUties was not clear until the tests at 18 months),

'MH 0852, "Infancy Experiences and Early Child Development," Julius B.
Richmond, M,D,, principal investigator.

186

2. Specifically expressed a desire that their childien do well in school
(the influence of this "maternal need for achievement" was seen at 12
months and persisted).

3. Showed an ability to plan for the care of their children and to
respond adaptively to the suggestions of child-care authorities.

4. Maintained physical order in the home.

The number of cases is small, the investigators point out, but the find-
ings are what might have been expected. Also, they support the idea that
the value of a given home as a child-rearing environment cannot be deter-
mined alone from knowledge of the family's socioeconomic status.

A Movie To Help the Child by Educating His Parents

The actors in the moving picture, "How Babies Learn," aie children
being studied by the Syracuse gioup, plus their mothers or caretakers and
other family members. They are unrehearsed. The narrator is Dr. Caldwell.
"You would be amazed," she says in discussing the film, "at how many
people regard the first year of life as a vegetative period when all you have
to do is feed the baby, change him, keep him quiet, and let him sleep well.
The experiences during the first year, however, may well be the most
ciucial learning encounters that the child ever has."

As the picture opens, the narrator makes two main points; Here are
some things a baby does in his first year; and here are some ways you can
help make it possible for him to develop his mind and his skills,

A baby can suck, react to light, hold on tightly to something placed in
his hand, move his head when placed face down, and make other
responses to his environment, the movie points out. Perhaps his most
powerful tool in adjusting to the environment and adjusting it to him is
his ability to cry. The message of the cry is not very precise, so the mother
has to try something; then if the baby doesn't stop crying, she has to tty
something else. "If he does stop," the narrator says, "the mother has
probably interpreted the message correctly. For the baby, this makes
possible a very simple but important type of learning. The baby learns
that when lie cries, something happens."

Commenting on this point, Dr. Caldwell says that the first step in a
learning situation is to realize that what one does makes some difference.
One of the few things a baby can do is cry. If he docs it and gets no
response, the result is "a step on the road to apathy." He doesn't have the
adult's capacity to tolerate frustration and delay, and he soon quits tiying.

Before long, the narrator goes on, the baby may learn that a smile brings
more love and attention, and perhaps accomplishes more, than a cry ever
could.

As the baby gets his eyes under reasonable control, he begins to look at
everything about him. But gradually he becomes more particular. The
thing that seems to have the most appeal is someone else's face, especially
if that face looks back.

187

By 6 months of age, the movie demonstrates, babies are interested in
almost anything. A grasped object is almost certain to be put into the
mouth. Holding two objects is difficult, not only because the hands have
difficulty but also because the attention is in short supply; holding three
objects is almost impossible. And when the mothers hides one object
under a pillow, the baby shows no further interest in it.

Between 6 and 9 months, however, important changes take place.
Nine-month-old twins are shown sitting up, balancing themselves easily
while they examine their environment. They handle objects and taste
them. When the tattle is hidden, the little girl goes after it: "She has
learned from experience that things hidden are still there." Handling two
things at once is now easier; when one is dropped, the eyes-but not always
the hands-follow it. The boy twin crawls, pursuing toys 01 people thai
interest him. His twin has not yet begun to crawl. But here is a 10-month
girl who can crawl and who spends much time pulling herself to a standing
position. And here is an 11 -month-old girl who walks while holding her
mother's hand, or the furniture, and can even grab a toy while she is on
her feet.

A number of other instances are given of how behavior changes as the
baby develops. "Perhaps this description of infant learning has sounded as
though it all occurred automatically, as though it all came from within,"
says the narrator. "Nothing, of course, could be further from the facts.
At every step in the process the people who are important to the baby
play a major role. From his parents, his brothers and sisters, his grand-
parents, and interested friends must come the stimulation necessary to
develop fully his ability to learn. Little things can make a big diffeience,
Consider something as apparently insignificant as the position into which
the baby is habitually placed during its waking hours. He stays put
wherever he is put down. If kept on his back, and if no people are in sight,
he may well have a monotonous view. (The film shows a baby who has
nothing to look at except a blank ceiling,) Until he can turn himself over
and look around for something interesting, his seeing and looking can be
given an assist by putting him down on his stomach part of the time,
From this position he can practice some of the movements he needs to
master befoie being able to crawl or walk and can choose what he wants
to look at to some extent. Mother's face is far more interesting than the
ceiling. Or he can be propped in a comfortable position and permitted to
examine his surroundings. In this position he can see and can also use his
hands to practice reaching and holding.

"During the first half year or so," the narrator continues, "we know
that the baby needs people and he needs a variety of experiences to help
him learn- But some time between 6 months and a year he seems to need
something else: he needs special people. Not just anybody will do
anymore. In particular, he seems to need his mother. For some reason,
babies reared without this special attachment do not seem to learn as
well,"

188

The film shows a 10-month-old girl responding uncertainly and
reluctantly to the smile and outstretched arms of a stranger But when the
mother beckons, the baby's face lights up and she crawls joyously to her.

Now another 10-month-old girl, who has not seen hei mothei for 3
months, appears on the screen. She is thin and looks frightened. Though
she goes fiom one stranger to another with little hesitation, she does not
respond enthusiastically to any of them. She passively accepts whatever
happens. When a cookie is handed to her, she reaches for it. But when it
falls to the floor, the least bit out of reach, she does not try to get it.
When a toy is hidden, she shows no interest in finding it. She can stand,
but her balance is poor. "In such cases," says the nariatoi, "if the separa-
tion from the mother does not last too long, or if a substitute can be
arranged who will give the baby a lot of tender, loving caie, the learning
deficit associated with prolonged separation of a baby from its mother can
usually be corrected."

The movie emphasizes the importance of other people in the process of
learning to talk. Long before a baby can learn to talk, the narratoi
remarks, he must be talked to—must hear sounds made in i elation to
objects in his environment and in relation to his own needs. Parents often
say that older children will teach the baby how to talk, but this apparently
is not so. Second and latei-born children are, on the whole, slowei to learn
to talk than the firstborn. "Fiom the standpoint of teaming language,"
the narrator says, "there seems to be no effective substitute for the ex-
perience of being talked to by loving and attentive adults."

Play materials can help a baby to learn shapes and colois and to
improve the coordination between eyes and hands. But the materials need
not be fancy or expensive. Empty food cartons supply a variety of colors,
textures, and shapes. Old magazines have pictures that encourage a baby
to improve his perceptual skills and to use his developing powers of
speech. And nothing could be better than mother's pots and pans: "Such
sounds and tastes and such a sense of being involved with objects that
have meaning for the whole family!"

As the film shows a baby playing happily with pots and pans, the
narrator continues: "Thus in their everyday routines, parents can nourish
the learning careers of their babies. In humble but significant activities,
parents can help their babies learn to learn."

The little things, the small personal touches, the narrator emphasizes,
are the ways that enable a baby not only to grow but also to thrive-
"stopping to talk to or smile at the baby while working; playfully en-
couraging him to try new things; helping him to achieve new and more
mature postures; helping him muster the courage to take those first steps,
and helping him back up, and reassuring him when he tumbles. By such
participation and encouragement the parents invite the baby to move on
to a higher level."

Prints of the movie, which lias been praised by pediatricians and other
child health workers, are now in such demand that the investigators, in the
interest of saving the time of everyone concerned, have turned the picture

189

over to the New York University Film Library, New York City, for salt
rent to interested groups.

An Enrichment Experiment With Very Young Children

All but one or two of the 24 youngsters in the day-care program
also in the infant-learning study. When a child in this study reaches the ;
of 6 months, he is given priority foi admission to the day program shoi
a vacancy occur. Those now in the program range in age from 6 months
almost 3 years and are described by Dr. Caldwell as "extremely higli-i
children for whom some environmental enrichment is essential if ilcvck
mental decline is to be prevented."

As an example, she tells about Alberta, who was a sluggish a
apathetic child of 10 months when taken into the day-care project at (
urging of a public health nurse who had been helping the family for yea
Alberta's mother is retarded y presumably as the result of cerebral unox
or loss of the brain's oxygen supply, sustained as a baby when her lie
was caught between the lailings of her crib. The mother, who is i
married, also has a 6-year-old son, and he, too, is retarded, although Hit
is no record or evidence of biological defects.

For a while, Alberta's mother treated her much like a doll, cuddling ai
rocking her for hours on end. When the girl was about 3 months ol
however, the mother seemed to lose interest and began keeping her in
crib in a darkened room most of the day, supposedly to protect hor fro
the extreme hyperactivity and destructiveness of her brother. At 0 nionl
the public health nurse reported that the child seemed to be diiftu
downward. At 8 months, when the girl was given developmental losls
the Children's Center, the investigator recalls, "It was as though she find
veil over her face and wanted to keep the environment out "

In this situation were all the ingredients for producing culturally dck
mined mental retardation: a disorganized family, a dearth of percept u
and cognitive stimulation, and emotional deprivation abruptly followir
emotional support. And there was a strong suspicion that the brollusi
history had contained the same ingredients.

"Perhaps unfortunately," Dr. Caldwell observes, "there is no literal
test for motherhood. We say that mothering is essential and imply Hi;
any mothering will do. We seem to assume that anything that sustains lil
is adequate during the first few years, and we have been very timid iwoi
trying to change the environment of very young children. Yet not n
parents are qualified to provide even the basic essentials of physical an
psychological care."

In the case of older children, Dr. Caldwell continues, society often ste]
in to help shape lives. In fact, to provide a richer environment for lonrniiif
society forces all children over 6 years of age into an institution, th
public school system. But much of a child's ability to learn, insofar as thi
is influenced by the environment, seems to have been shaped long befor
he was 6. The IQ differences between deprived and privileged childrc

190

show up in the second and third years. So the time to provide an enriched
environment, Dr. Caldwell reasons, begins when the child is about 6
months old.

The day-care program at the Children's Center is thus "an exercise in
circumvention rather than in remediation." It is trying to prevent the
deceleration in the rate of development that seems to occur in many
deprived children very early; it is also trying to maximize each child's
potential.

The children in the program are divided into three groups: those from 6
months to about 18 months; those from about 18 months ("when they
are walking and showing some interest in toilet training") to about 2'/a
years; and those older than 2V&. Under the original plan a child left the
piogram when he was 3 years old, but the investigators have now obtained
financial support-from the Children's Bureau (Child Welfare Research
and Demonstration Grant D—156R— enabling them to keep him another
year or two. With the additional support they hope also to compare the
effectiveness of the present program with that of a day-care center admit-
ting children no younger than 3 years of age. A piogram for infants and
toddlers is much more expensive than one for older children, Dr. Caldwell
points out; she thinks it is also much more effective, but she would like
some hard-and-fast evidence.

With the children in the youngest group, the emphasis is on individual
attention. At least once a day for at least a few minutes, each baby
receives the undivided attention of one of the staff members-usually the
same one. During these periods the baby is encouraged to reach for, go
after, and handle different objects, and he is talked to and stimulated to
respond. Whenever he is awake and not in his reclining chair or being held,
he is placed on his stomach in order to encourage visual and motor ex-
ploration of the environment. The teachers talk slowly and distinctly.
They repeat the names of objects as the child plays with them and
describe an activity as the child engages in it. Learning games, similar to
laboratory tests of visual and auditory discrimination, are played several
times a week. Particular attention is given to tasks-for example learning
that a reward is hidden under the larger of two blocks— that help a child
acquire concepts and the ability to think abstractly.

Children in the older groups receive similar training in sensory dis-
crimination and concept formation, capacities in which children from
culturally deprived backgrounds have been found inferior. The children, in
small groups, are read to at least once each morning and afternoon. They
are given magazines and books and encouraged to look at the pictures and
to repeat the names of what they see. Crayons, paint, other artistic media,
and musical instruments are made available, as are toys that help a child
develop an awareness of color, texture, shape, and sound. There is a
well-equipped outdoor play yard. Self-initiated activities are encouraged,
for they help in the development of self-confidence; at the same time the
children are expected to learn to respond to the house rules.

In this environment, where the adults are friendly and accepting, and
the child is encouraged to be curious, to make explorations, and to

191

develop his abilities, even Alberta has made decided progress. After 2 or 3
weeks she became much more responsive. She babbled once in a while and
she learned to pull heiself up on chaiis, and then to walk. Now, after 5
months, she acts most of the time like a normal child. Something happens
over the weekends, though, Dr. Caldwell reports. On Friday the girl is
outgoing and bouncy; on Monday she is once again apathetic and needs ji
warmup period of a day or two before responding to either the nursery
school workers or the toys. What the end will be the investigators cannot
predict. She responds to the environment at the Center but also to the one
at home. Peihaps after seveial years in an enriched environment, Dr.
Caldwell observes, Alberta and other such children will have developed (lie
intrinsic strength to sustain themselves no matter what the home atmos-
phere. This is one of the project's majoi questions.

Research Grant :IAH1649
Related Grant" MH8542

Date oflnteiview: June 14, 1966
References'

Caldwell, Bettye M The effects of infant care. Review of Child Development Re-
search: Volume 1. New York: Russell Sage Foundation, 1964.

Caldwell, Bettye M. What is the optimal learning environment for the young child'
Presented at annual meeting Amer. Orthopsychiat. Assn., 1965.

Caldwell, Bettye M., and Drachman, R. II, Comparability of three methods of assessing
the developmental level of young infants. Pediatrics, 1964, July.

Caldwell, Bettye M., and Richmond, J. B. Programmed day caie for the very young
child-A preliminary report. /. Marriage and the Family, 1964, November.

CaldweH, Bettye M., and Richmond, J. B, Social class level and the stimulation
potential of the home. Presented at meeting Amer. Psychol. Assn., 1964.

192

Investigator

Howard A. Moss, Ph.D.

Chief, Section on Parent-Infant Behavior

Child Research, NIMH

Prepared by:

Antoinette Gattozzi

To explore the nature of early experience, a research group directed by
Dr. Howard A. Moss is studying parent-infant interaction, especially
mother-infant interaction, during the first 3 months of life. The ic-
searchers are tracing sequences of maternal and infant behaviors in order
to calculate the probability of one action leading to another and to tease
out from the total complex the contributions made by parents and those
made by the infant. By determining the detailed patterning of the inter-
action, the investigators may discover how it relates to the child's con-
genital characteristics and to parents' psychological makeup, their early
marital relationship, and parental expectations.

Abundant evidence supports the concept that the nature of early ex-
perience exerts profound effects on developmental processes. For
example, when Dr. Moss was working at the Pels Research Institute in
Yellow Springs, Ohio, he and a colleague, Dr. Jerome Kagan, reported the
findings of a child development study conducted there. The study verified
that many adult behaviors are rooted in early childhood; moreover, for
certain fundamental dimensions of behavior, earliest influences produce
the most enduring effects. Ratings of maternal treatment of children from
birth to 3 years old were found to be more predictive of later childhood
and adult behaviors than were the same assessments made when children
were older. Maternal protectivencss up to the age of 3 years, for instance,
showed better correlation with a child's passivity behavior during the
years 6 thiough 10 than did protcctiveness assessed when the child was
actually in the age group 6 through 10 years. Some maternal behaviors, in
fact, had a prolonged sleeper effect in that their influence was not dis-
cernible until children had reached adulthood.

Drs. Moss and Kagan postulated that the mother's influence is most
apparent (and perhaps strongest) during the first years of her infant's life
and that her behavior then is more reflective of her own attitudes and
values than it is an accommodation to her child's behavior. As the child
matures and asserts his individuality, however, he tends increasingly to
evoke maternal treatment that is influenced by his unique characteristics.

What are the precise timing and behavioral pathways of these dynamic
intereffects? How does the child's sex affect his behavior and influence

i

193

maternal treatment? What are the attitudes that shape the mother's treat
ment of her infant? Dr, Moss and his associates currently are addressmj
themselves to such questions, which bear implications for child menta
health. Such basic developmental research issues as the evolveinent oi
attachment behavioi, the establishment of the prototypic learning frame
the relations between the mother's role in mediating stimulation, and the
child's coping style are only meagerly understood. Normative studies such
as Dr. Moss is directing are needed to explore these areas so that we may
learn how better to help troubled children.

Many of the families that Dr. Moss is studying also are participating in
the Child Research Branch's overall longitudinal program, which is
designed to yield a multidimensional scheme of early family formation.
"For example," Dr. Moss explains, "in the study of married couples,
emphasis is placed on the physical quality of the relationship and the
degree of affectional contact. In the newborn infant project, special at-
tention is given to congenital differences in skin sensitivity. In the parent-
infant project, we can attempt to determine whether children with low
tactile thresholds seek or are soothed by greater physical contact with the
parents, or if parents with certain needs or experiences have a greater
proclivity to handle or caress their infants."

The principal method of research is direct observations, made in the
home in two clusters during the infant's first and third months. Each
cluster consists of two 3-hour and one 8-hour periods. For the 8-hour
stretch, the observer uses a stopwatch and sheets of paper ruled into
time-sampling forms that list, separately for mother and infant, 30 dif-
ferent behaviors. Some examples of maternal behavior are: Holds infant
close, feeds, stirnulates-arouses, imitates, stresses musculature. A few of
the infant behaviors are: Cries, fusses, sleeps, is awake and passive,
vocalizes, mouths. Every minute, the observer marks the form to show
what mother and infant are doing and in what order. Behaviors were
selected on the basis of their presumed relevance to aspects of maternal
contact or because they reflected the state of the infant. Other researchers
have found that the infant's state, or level of arousal, which is a con-
tinuum ranging from quiet steep to agitated crying, is an important
influence on the nature and quality of his experience.

During the 3-hour periods, the observer works with a portable keyboard
connected to an electrically powered event recorder. Each key represents
one of the behaviors. As the keys are depressed, they activate pens in the
recorder that leave ink tracings of the coded behaviors on paper. The
resulting yards of tracings can be deciphered to produce a detailed de-
scription of mother-infant interaction during the period. Both methods
are arduous for the observer, demanding maximum attention and con-
centration coupled with an informal, unobstrusive manner.

About 1 ,000 hours of observation have been clocked so far, all of them
by Dr. Moss. Data collected during the 3-hour periods from a sample of 29
mother-infant pairs are now being analyzed, in part by programming a
computer to extract sequences of action from the mass of separate

194

tracings. The time-sampled data from these mother-infant pairs have been
compiled and are the basis for the findings discussed below.

Variability. -A. strikingly wide range appeared in the amount of time
devoted to various behaviors. Great variability was evident at 3 weeks and
3 months, but was more pronounced when the sample was younger. Dr.
Moss suggests that, for the infants, later differences in social behavior and
learning style-aspects of development that have their genesis in the first
weeks of life-are both reflected in and influenced by these revealed dif-
ferences in behavior. The infant's behavior pattern reflects his unique
constitution at the same time as it influences his mother's treatment of
him. The squally, fussy child provokes a patern of maternal caretaking
quite different from that which the placid, easily soothed infant elicits.
Variabilities shown by the mother's behaviors similarly work to differ-
entiate interactional patterns and their developmental products.

Stability. -From 3 weeks to 3 months, marked shifts occuired in most
behaviors. At 3 months, infants were less irritable, awake longer, and
spent more time smiling, vocalizing, and looking at their motheis than
they had at the age of 3 weeks. Compared to the earlier period, mothers
devoted less time to feeding and in close physical contact at 3 months, but
increased their total attending (nonholding) behavior through social
contacts such as talking, smiling, imitating, and generally stimulating the
baby. These shifts attest to the enormous rapidity of infant growth and
maturity during the first 3 months of life and to the adjustments mothers
made, in part in response to the changing infant. The mother's behavior
also changes, it seems likely, in accordance with her attitude to the infant,
whom she increasingly regards more as a lovable person in her life and less
an an animated bundle of total, demanding responsibility.

Sex differences.- Mean scores computed separately for boys and girls
revealed several significant differences in both infant and maternal be-
haviors. These sex differentials also shifted over time: At 3 months, the
infant variables showed smaller but persistent sex differences, and matern-
al vaiiables were not as differentiated by sex as they had been at 3 weeks.
The findings augment evidence from other studies of children that, as Dr.
Moss puts it, "males are more subject to inconsolable states." Overall,
males fussed and cried more and slept less than females. Not unexpect-
edly, males as a group were held by and attended to more by mothers;
mothers also spent more time with boys stressing musculature,
and looking at, talking to, and stimulating them. With their mothers, girls
interacted more than boys only on such variables as stimulating, feeding,
and imitating. Females vocalized substantially more than males at 3
weeks, but not at 3 months when the sex difference in this infant behavior
disappeared. Another variable showing an intriguing sex difference was
mouthing, which tallied the time an infant mouthed an object (such as bis
fist or a rattle) other than while feeding. The item represents a level of
integrated behavior. At 3 weeks, the mean was 36.8 for boys and 30.6 for
girls out of a total of 480 minutes; at 3 months, mean time close to
doubled for males (61.2) and nearly quadrupled for females (1 16.2).

195

Sex differences in infant behavior and in maternal treatment of info
pose a difficult problem of interpretation. Temperamental differcn
between sexes date at least from the Garden of Eden, and explanatic
have been sought for almost as long a time. Probably sex characterist
encode quintessential biological principles. Whatever nature's contributi
may one day prove to be, there is no question that societies of men lit
always embellished the differences. The familiar verse below is part c
servation, part expectation.

Snips and jsaails and puppy dog fails

And such are little boys made of.

Sugar and spice and all things nice

And such are little girls made of.

The data from this study accord with the findings of other researchci
that demonstrate the very early occurrence of sex differences. The phi
siological origins and biological functions of these differences are no
compEetely known. There is some evidence that male organisms nr
geneially less viable than female organisms at birth; this would lead to ,
tendency for males to be either more aroused or more lethargic thai
females. Further, it is reasonable to assume hormonal differences, perhap
related to the moi phological differences. The effects of the routliw
surgical procedure of circumcision, usually performed within the firsi
week of the male infant's life, are thought not to extend beyond 12 hours

Parental expectations, cameos of the enveloping society, overlay UK
infant's biological substrate and contribute to sex differences in behavior
In paradigms, parents encourage their daughter to be complaisant and
serene, their son to be staunch and vigorous. The Pels longitudinal study
discussed earlier yielded evidence that the child's sex influenced maternal
treatment. Mothers were moderately consistent in protecting and accelera-
ting boys but not girls, and much more consistent in restricting girls than
boys. Differential effects of maternal treatment on boys and girls were
found also. Mothers who were highly protective of a male infant from
birth to age 3 years were likely to foster in the child a drive for intel-
lectual mastery, the appearance of such a drive in girls and young women,
however, was linked to early maternal hostility.

Parents1 modes of treatment also stem from conceptions of their own
sex and familial roles. Some of the differences in treatment emanating
from these sources were exposed during a procedure conducted by Dr.
Moss's research group. Parents were asked to elicit specific behaviors from
their 7-week-old infant. In trying to get the baby to smile or vocalize,
parents spent more time with females and mothers tried longer than
fathers with both sexes. Fathers participated longer than mothers in get-
ting the baby to grab a bell, and they tended to spend a little more time
with male infants in this test of motor skill.

The Cry as Maternal Stimulus

Correlations were computed between a score of infant irritability (the
total time spent crying and fussing) and a maternal contact score, which
was derived by summing the time mothers devoted to holding and attend-
ing infants exclusive of behaviors associated with feeding. For the females
of the sample, the correlation was positive and significant at 3 weeks and
at 3 months, which is to say that these categories of behavioi occurred
together more often than would be expected by chance. For the sub-
sample of males, the correlation at 3 weeks was positive but not
significant, and at 3 months it was negative. This indicates that maternal
contact with males— who were, as a gioup, substantially more irritable
than females as a group— was somewhat random in occurrence at 3 weeks
and, by the third month, that mothers tended to spend inveisely less time
with males who exhibited greater initability.

In terms of the total sample, however, correlations were positive and
significant for both periods of time. Such an association suggests a causal
relationship. Drawing on these observations and on findings from other
studies, the investigators postulate that the infant cry is a potent stimulus
shaping maternal behavior. The hypothesis embodies an explanation of
the sequence of events between mothers and males as well as the rather
different sequence enacted between mothers and females. The investi-
gators' reasoning may be conceptualized as follows.

The mother learns how to care for her infant in a round-the-clock
course of trial and error. She regards the cry as a signal for her attention;
her responses are as ingenious and varied as she can make them. With
practice and experience she learns how best to respond, and attains some
degree of success in reducing what is to her the noxious stimulus of the
cry and in gratifying her maternal needs. If her infant is a male, she learns
by the third month that she often is unable to quiet him. More and more
frequently, she may avoid the possibility of failure by not responding at
all, or by delaying her response to his cry. If her infant is a gill, however,
the mothei learns a different lesson. Because her daughter is usually
soothed by her attention, the mother feels fairly confident of her ability
to succeed and responds consistently to the cry with caretaking activity.

This hypothesis is based on the correlations obtained between infant
irritability and maternal contact, and on the finding that males were
considerably more irritable than females during both periods of observa-
tion, which may imply that males had not been as uniformly soothed as
females by maternal contact.

The interpretation provides a way of viewing the first phases of
socialization, the gradual but ineluctable process through which an
individual is transformed from a relentlessly egocentric newborn into a
more or less cooperative member of society. In the beginning, the
investigators suggest, the infant arbitrates the pattern of interaction
through the powerful stimulus of the cry. As time passes he associates
succor and comfort with his mother, and this increases her effectiveness in
regulating his behavior. Imperceptibly, the two move toward a juncture

197

wmch the larger role in determining the interaction shifts to the
mother. At this point, socialization may be said to begin.

If this is a valid notion, infants who are usually soothed by the mothers'
responses to their cries should be more susceptible to social learning than
infants whose mothers do not answer their cries consistently or who are
not quieted by maternal care. In the sample under study, the first class of
experience corresponds to the females as a group, and the second to the
mate group. Indeed, this view of one aspect of early socialization is in line
with results of other studies and fits the common observation that girls are
more sociable than boys,

Psychological Factors in Maternal Treatment

Infant irritability is one important influence shaping mother-infant
infant interaction that can be isolated and measured. Parceling out its
effects may expose the presence of other influential factors. The le-
searchers derived a measure of maternal responsivity from the data by
totaling the amount of maternal contact that was not in response to
irritability. This tactic uncovered two things. First> it was possible to
classify a mother as an over-responder or an under-responder, depending
on whether her predicted contact score (based on her infant's irritability)
was above or below her actual contact score. Second, even accounting for
greater male irritability, mothers were found to have significantly more
contact with boys than with girls on such variables as "attends" and
"stimulates-arouses;" on the social item "imitates," girls showed the
higher mean score. This finding of sex differentiation in maternal treat-
ment also was made when the same type of analysis was done while
controlling the amount of time the infant spent in sleep. In this instance,
males showed higher mean scores on the variables "stresses musculature"
and "stimulates-arouses," and females scored higher again on "imitates,"
Thus, sex differences in maternal treatment that stemmed from mothers
were discovered by controlling the effects of sleep and irritability, the
salient aspects of the infant state.

The investigators are seeking antecedents that might link to this finding
in the data they have gathered from parents. In addition, parents of 23
infant subjects had been extensively interviewed long before their baby's
birth for the Branch study of early marriage. Two variables relating to
potential parenthood that were extracted from this latter material have
proved germane to the mother-infant study: "Acceptance of nurturant
role" and "degree baby seen in positive sense" are prematernal variables
that may be predictive of maternal responsivity.

The researchers also have located tentative links between a woman's
maternal behavior and her childhood and early family experiences. The
maternal behavioral composites of affectionate contact and responsivity
appear to be related to a woman's recalled attitude toward the emergence
of her secondary sex characteristics and the degree of identification she
had with her father. Further substantive details like these, delineating

198

continuities in attitudes that structure maternal behavior, may be found
by the current search for relationships between early mairiagc and
mother-infant variables. The investigators are similarly analyzing the data
collected by the icsearchers who studied their infant subjects as newborns;
heie they are looking for relationships between an infant's congenital
behavior and the behavioi displayed in that infant's inteiaction with
parents. Meanwhile, Dr. Moss and his associates have begun another study
with a larger sample of parent-infant pairs. They hope to replicate the
findings of the first study, and to follow sequences of action with greater
refinement.

Dr. Moss is also interested in the mother's role in mediating stimulation
for her infant. Past studies have accumulated evidence indicating that
infants have a need for stimulation, and that its quantity and quality aie
influential along many channels of development. Too much or too little,
too intense, chaotic, or monotonous stimulation, as uniquely experienced
by the individual infant, may have deleterious effects. Clearly, mothers
play a central role in balancing stimulation for their infants as they arouse
or quiet them and icgulate their autonomically generated stimulation. In
the mother-infant pahs studied, motheis offered male infants gi cater
amounts of motor stimulation. At the same time, as one possible inter-
pretation of the data suggests, males were experiencing heightened levels
of internal and self-stimulation. If this pattern is found again in the new
study, its meaning and implications will wanant further exploration.

Dr. Moss believes that apart from survival, coping with stimulation is
the infant's consuming developmental task. In this context, a mother's
part in mediating stimulation is a fundamental task for her. Studies of
parent-infant interaction may demonstrate how different mothers handle
stimulation for their infants, how differences in infants affect the stimula-
tion balance, and what relations exist between the kind of stimulus
dependency established and later developmental configurations of social
behavior and learning style.

Intmmwal: NIMH

Date of Interview: Mar. 23,1966

References:

Kagan, J,, & Moss, H, A. The stability of passive and dependent behavior from child-
hood through adulthood. Child Develpm., I960, 31, 577-59 1 .

Kagan, J., & Moss, H. A. Birth to maturity: A study in psychological development,
New York: John Wiley, 1962.

Moss, H. A. Methodological issues in studying mother-infant interaction. Amer. J.
Orthopsychiat,, 1965, 35 (3), 482-486.

Moss, H. A. Coping behavior, the need for stimulation, and normal development.
Merrill-Palmer Quart. Behav. Developm,, 1 965 , 1 1 , 1 7 1 -1 79 .

Moss, H. A. Sex, age, and state as determinants of mother-infant interaction. Merrill-
Palmer Quart. Behav. Developm., in press.

Moss, H. A., & Kagan, J. Stability of achievement and iccognition seeking behaviors
from early childhood through adulthood. J. abnorm, soc. Psycho!., 1961, 62,
504-513.

Moss, H. A., & Kagan, J. Report on personality consistency and change from the Pels
Longitudinal Study. Vita hum., Basel, 1964,7,127-138.

199

Investigator.

Nahman H, Greenberg, M.D.
University of Illinois College of Medicine
Chicago, III.

Prepared bv

Herbert Yahraes

Through an intensive study of physiological patterns during infancy ami
of the interaction between infants and their mothers, a University of
Illinois psychiatrist is seeking hard and fast information about the origins
of emotional difficulties and psychosomatic illness. The investigator is
Nahman H. Greenberg, M.D., director of the Child Development Clinical
and Research Unit at the University's College of Medicine, Chicago.

Dr. Greenberg points out that a baby's development— and, according to
psychoanalytic theory, its emotional health in later life as well is
profoundly affected by the quality of its mother's love. The investigator
wants to find out how this love operates, so he focuses not only on how a
mother feels or says she feels but also on what she does. Certain types of
maternal activity, he believes, either satisfy or fail to satisfy certain of the
infant's biological needs and may influence either for better or worse the
infant's developing social relationships. Consequently, if mothers' actions
and infants' responses, and vice versa, are cataloged and analyzed,
eventually it should be possible to relate a given pattern of action-mid-
response in infancy to a given type of disturbance in later childhood find
perhaps in adult life. The relationships can be established, of course, only
by such long-term studies ns the investigator has embarked upon.

In one part of his program, Dr. Greenberg is determining how various
kinds of activity on the part of an infant affect its physiological state as
shown by respiration, heart rate, body motility, and body tonicity. This
Information will provide a baseline for measuring the effect of llic
mother's behavior upon these same characteristics,

The babies being studied this way include a so-called normative group,
born to the wives of medical and dental students at the university, and an
institutional group born to unwed mothers-mainly white girls, of nil
socioeconomic levels-and living in a foundling home,

200

The measurements are made in the nursery of Dr. Greenberg's psycho-
physiology laboratory for at least an hour at a time as the baby sleeps,
tosses, plays, smiles, cries, or does whatever else comes naturally. By
pushing a button, an observer indicates— on the magnetic tape which is
recording the measurements— the kind of activity going on at a particular
time. Moving picture cameras, one for closeups and one for distant views,
are available for filming all or part of a session. At one point in each
observation period the experimenter introduces a "sensory event" and
notes the results. This point occurs when the baby fusses because he is
hungry, and the "sensory event" consists of attempts to pacify him in
various ways.

The ways in whicli mothers and babies interact are observed and filmed
in the same laboratory. In addition, the mothers are studied through
psychiatric interviews and psychological tests.

The institutional group of babies was added to the research plan so that
infants who were all being reared under the same conditions and, of
course, in the absence of their natural mothers, could be compared with
the others. Through a data collection system at the foundling home, the
investigator receives a 24-hour-a-day record on tape showing foi each baby
when it moved, cried, and was lifted from the crib. He also receives a
filmed record-made by preset cameras at various times— of the activity of
the aides caring for the babies.

In a second part of the research, the investigator is concerned with
infants—and their mothers— in whom developmental disorders are already
apparent. Several of these babies have appeared in the so-called normative
group; the others-there are now 45 in all-have been referred to him over
a 2-year period from the university's department of pediatrics and the
well-baby clinics of the Infant Welfare Society of Chicago. When Dr.
Greenberg first saw them, they ranged in age from 3 to 23 months. All
had been raised in their own families. They were not mentally retarded
nor did they have neurological defects.

Among the symptoms of abnormal development were disorders
described in general as hypermotility, including head-rolling, body
swaying, and some forms of self-injury, mainly head-banging, nutritional
anemia, pica and other feeding disorders, and failure to thrive.

As Dr. Greenberg points out, it is generally impossible for an investi-
gator who is stuyding disturbed adolescents or adults to dig back and get
accurate data about their infancy. But given an infant showing a develop-
mental disorder, it may well be possible to find out what has gone wrong
so far in his very short life, and then perhaps to correct it or at least to set
up warning signs for other families and physicians.

"The effort is to work back and reconstruct the developmental
history," Dr. Greenberg explains, "and learn what the mothers did to get
something like body-rocking or head-banging, or eating plaster off the
wall, or refusing to take anything but a bottle and developing an anemia as
the result." As a major part of this effort, the investigator and his as-
sociates observe and record the behavior of mother and baby during
extended visits to the laboratory nursery. The researchers are particularly

201

interested in the mother-baby relationship at feeding time and the bab)
behavior dining what would ordinarily be a period of mild stress, win
the mother leaves him alone and then a stranger comes in.

Preliminary Findings

The rationale of this research program and the findings to dale ar
presented in some detail in the succeeding sections. The main finding
may be summarized as follows:

1. At 1 month of age, the institutional babies observed during a pilo
study could be pacified more easily than the others and, unlike the others
equally well by any of the methods used. Pacification brought a grcatei
change in their heart rate. At 2 months, the institutional babies had gainec
only half as much weight as the others. These differences seem traceable
to the amount of stimulation provided in the foundling home.

2. On the average, the institutional babies were handled~for feeding,
bathing, and all other purposes-only about 90 minutes a day. Never-
theless, the institutional environment is less uniform than had been
expected, for some babies were found to get considerably more attention
than others, (The differences are going into the babies' records, and an
effort will be made to study their effect.)

3. The abnormal behavior of the infants with developmental disorders
appears to arise from serious disruptions in the infant-mother relationship,
and these disruptions apparently occur because the mother is emotionally
disturbed. For the most part, the mothers are considered to be either
psychotic or borderline. "Commonly there has been severe trauma in their
own lives," the investigator reports, "and they identify with the baby in
terms of their own traumatic childhood and mothering."

4. Most of the infants with developmental problems show an ab-
normality in the way they pacify themselves. Finger-sucking is un-
common, and many of the infants have developed an aversion to bottles.
To reduce tension, the investigator reports, these infants tend to engage in
hypermotility, such as head-rolling or body-swaying.

5. In their relations with these infants, the mothers appear to be either
detached and neglectful or overstimulating and harsh. When the mothers
feed their babies, they show little or no tenderness. Some of the same
mothers, though, handle their infants for inordinately long periods of
time. Their behavior in this respect, the investigator reports, over-
stimulates the infant and is inappropriate to its needs,

6. Separated from their mothers, these babies do not show the typical
response of sobbing or crying. In the hospital, away from "psychotoxic"
influences, the babies tend to improve quickly.

7. Most of the mothers are resistant to initial attempts with psy-
chotherapy.

8. There is nothing to indicate so far whether or not the developmental
problems and the disturbed infant-mother relationship will affect

202

personality and emotional health later m life. The investigator believes
that they will.

9. The great majority of the infants in the atypical group come from
families at the lower socioeconomic levels. The investigator hopes to
check reports that the pattern of disturbance among infants from families
at the higher levels is different.

The Interactions of Mother and Infant

Dr. Greenberg defines psychosomatic differentiation, his field of study,
as the piocess of growth by which individual psychophysiological patterns
develop. This process is most vulnerable to outside influences at the stage
of least differentiation, and this stage— with respect to the influences of
motheiing-begins at birth. Because the mother in large part creates the
infant's sensory environment, it is she who most strongly influences the
developmental process. But the investigator points out that the infant
itself, through its inborn propensities, may influence both the amount and
type of maternal behavior. An infant who sleeps most of the day can be
expected to influence a mother's behavior differently from an infant who
has been irritable-in the sense of being peculiarly susceptible to
stimuli-since birth.

In studying an infant's heart rate, muscle tension, and other measures
of physiological activity, and how these are affected by the mother, the
researchers distinguish six basic behavioral states or degrees of internal
tension: Sleeping, drowsy, awake-inactive, awake-active, fussy, and crying.
To a great extent, the investigator explains, a mother can regulate these
states through such activities as feeding, holding, talking to, and smiling at
her baby. A particular activity by the mother calls into use and helps
develop a particular sensorimotor mechanism in the infant. In other
words, mother and infant are two psychophysiological systems, which
interact through specific mechanisms of stimulation and pacification. As
these mechanisms come into play, there is a change in the degree of the
infant's inteinal tension, and of the mother's as well.

A mother reacts to the different behavioral states in different ways, and
one mother's reaction to a given state may differ from another's. In one
case, for example, Dr. Greenberg found that the mother tended to avoid
her baby when it reached the crying state, but when the baby was quiet,
would begin talking to and playing with it to the point of oveistimulation.
At least party because of the mother's way of reacting, the investigator
believes, the baby developed a serious nutritional disorder and gave
evidencc-for instance, by showing greater distress than normal when the
mother left the room-of emotional difficulties.

Requirements for Normal Development

As Dr. Greenberg sees it, an infant's responses to its mother's stimula-
tion have two functions. In the first place, they satisfy needs. Sucking, for

203

example, Is necessary to satisfy the need for nutrition; visual SL'i
are necessary if the visual apparatus is to mature and serve thu inl;in
normal needs to lecognize those close to him and then to exploit1 itud
learn. In the second place, the infant's responses serve to pacify ;ii
soothe him. Up to a point, sucking-even when no food is heir
received -serves as a normal pacifier or a way of reducing tension. 'M
same is true of movements of arms, legs, body.

For development to proceed normally, the investigatoi hypothesise
the infant's feeding experiences must be such that the organism remnnh i
equilibrium. On the one hand, the organism must be sufficiently annm.1
by the mother's actions to respond to sensory stimuli; on the other Itinn
the amount of tension generated must stay within bounds. When tli
equilibrium is upset, a variety of behavior distortions may arise
them, insufficient sucking, hyperirritabilily associated with
regurgitation, a fearful avoidance of food, and aversion to dietary clungi-
A mother who does not establish a warm relationship with her inl;m
through feeding, the investigator goes on, may attempt to enj>;ige lihi
through visual interactions such as smiling, or through excessive I»K|>
stimulation in the form of rocking and carrying. In either case, [hi* stimuli
the infant receives aie probably insufficiently diversified for UK" IUA!
possible development for his nervous system.

Unable to cope naturally with the feeding situation, a mother may uko
withdraw from the infant and avoid any but the most necessary conMci.
In this case, lacking appropriate levels of stimulation from on (side, (\\c
infant becomes irritable and excited and may turn to various toclini(|iiis
of stimulating or pacifying himself. In the earliest months, Dr. (JtviMihu'ij?
reports, he may engage in excessive sucking, particularly of his hfinris
Later on he may turn to body-rocking, head-rolling, or other rhythmic
motor activity-all in the interest of ameliorating tension.

In sum, under this hypothesis, when the mother's behavior inipumor
overtaxes the normal sensorimotor routes for satisfying needs, the inluut
uses certain of his developing functions maladaptiveiy— that is, for tension
reduction. As the individual develops, Dr. Greenberg believes^ he will mil-
grow a particular maladaptation but, if stressful conditions continue lake
on another. Thus the infant with a developmental disorder such as luMd-
banging may become, unless his distress is removed, the child or uilult
with a character problem, a psychosomatic illness, a psychosis, or MUM
other impairment.

The investigator emphasizes that he has no evidence that an inftnil who
bangs his head, bites his hand, eats dirt, or shows other developmental
abnormalities is more likely than other infants to be headed toward severe
emotional difficulties later on. Dr. Greenberg merely hypothesizes ehut
tin's is so. He believes that only a longitudinal study-one that follows mi
infant through the years rather than one that tries to look back from l:ilcr
life to infancy-can say for sure.

"Hopefully," he says, "we may begin to develop objective means of
describing psych ophysiological development and look at some factors in
maternal behavior influencing them. We shall be learning more about the

204

characteristics of the 'dark age' of infancy so that our notions about
health and disease in terms of this period of life will be based upon
objective data."

Troubled Mothers

The mothers of the infants with development problems— and particular-
ly the mothers whose babies rolled their heads or manifested disturbed
mottlity in other ways—were found to have difficulty in managing their
aggiessivc impulses. In response to an infant's display of vigor, Dr.
Greenberg xeports, these mothers tended to show hostility and at times to
engage in almost assaultive activity, Fuither, if a baby tried to suck a hand
or a thumb, the mother would draw it away, often lather roughly. At
feeding time the mothers tended to act in an impersonal, get-the-job-done
manner described as "institutional." But sometimes they went to the
opposite extreme and strove to attract the infant's attention to themselves
by facial gestuies and excessive handling.

Could it be that such mothers just don't know how to handle babies?

"Most of these motheis come from the lower sociocconomic levels,"
Dr. Greenberg answers, "so there may be an educational factor involved.
But I don't know that to be a good mother requires formal education; I
think it icquires an adequate personality." Most of the 45 mothers in the
group studied so far have been found to have major emotional
disturbances. They are usually irritable, distant, neglectful, depressed, and
markedly anxious, the investigator finds. They go from one crisis to
another.

Troubled Children

When the mother of a normal infant leaves him alone in the labora-
tory's nursery and then a stranger comes in, the investigator's films show,
the child is apt to cry or sob for a while. Not so an infant with develop-
mental problems. Such a child either continues to do whatever he has
been doing, without any sign of acknowledging the separation; or he
retreats, or sits or lies motionless, or begins body-rocking; or he engages in
extreme crying and cannot be comforted by being held, played with, or
even fed. The crying in the last situation is not weeping, Dr. Greenberg
observes, but straightforward shrieking.

Unlike most children who have been raised at home, however, these
babies with developmental problems apparently reach the age of 8 or 9
months without having established object specificity, meaning the ability
to show a preference for one or more of the persons around them. When
one of these infants has to be separated from his mother in order to be
hospitalized, he generally accepts the ward and cheers tip more quickly
than other babies. In Dr. Greenberg's words, these infants have a
gregariousness that lacks definition and singularity, with the result that

205

everybody becomes an acquaintance but no one a solid friend. Whether o,
not such a state, with its portent of trouble later on, will continue remaitu
to be seen.

The most marked symptom in several of the children is pica, a craving
for unnatural foods. This has become a public health problem of some
size, according to Dr. Greenbeig, because the afflicted babies often eat
paint-covered plaster and then develop lead poisoning, which affects the
central nervous system and may result in convulsions and death. (In an
attempt to find the lead-eateis early enough, teams of doctors and nurses
are going into homes in some parts of Chicago and collecting urine
samples for analysis.)

Pica children apparently feel deprived and hungry, but just what
happens in the mothei -infant relationship to give rise to their condition,
says Dr. Greenberg, is not yet clear. Interestingly, the films taken in the
laboratory nursery do indicate that the mothers of pica babies— in contrast
to most of the other mothers—act normally in the feeding situation.

The investigators have encountered even worse problems. One of the
first children referred to Dr. Greenberg, for example, had been admitted
to the Univeisity of Illinois Hospital at the age of 7 months with a com-
mon trouble, failure to thrive. "She was quiet and somewhat detached and
withdrawn," Dr. Greenbeig reports, "and frequently nodded her head in
the typical movements of spasmus nutans (head-rolling). She turned away
from the observers, although she was attentive to auditory and visual
stimuli. She showed no reactions of pleasure, and we were unable to make
her smile. She sucked eagerly when she was held and fed, but was
interested only in the bottle and paid no attention to the feeder's face.1"
Moving pictures made in the laboratory nursery indicated that Betty
was getting a fantastic type of attention. On one occasion, for example,
the mother reached into the crib, pulled Betty's fingers from her mouth
and began kneading the baby's face and head almost as if they were dough
01- clay. Betty began fussing. The mother explained that she was ac-
customed to rubbing the baby's face and body for hoius at a time "to
build her up."

Betty soon gained weight in the hospital and improved in other ways,
She became curious and responsive, smiled, reached for toys and people.
Her head-nodding stopped within a week-and-a-half,

After Betty was discharged, the mother brought her back several times
so that the team could continue to study the mother-infant relationship
and try to help the mother with her own emotional problems. When she
complained of brutality to Betty by the people they were living with, the
doctors and their staff even helped the family move into a separate apart-
ment, A month later, however, Betty was returned to the hospital in a
coma and died of a fractured skull.

Since then, several other battered babies, among those admitted to the
hospital, have been added to Dr. Greenberg's cases. Brutality to infants isj

Greenbeig, N. H. Origins of head-rolling (spasmus nutans) during early infancy.
Psychosom. Med., 1964, 26, 2.

more common than generally realized, and its impact on mental health in
later years may well be shown— through studies like Dr. Greenberg's-to be
grievous. The investigator calls attention to suggestions that schizo-
phrenics have experienced not only harshness during infancy but also
physical cruelty.

Treatment

If we know how emotional disorders originate, perhaps we can step in
to prevent them or at least to shorten the treatment and make its effects
more enduring. And the logical approach in cases like those understudy,
where physical symptoms point to psychological problems, Dr. Greenberg
believes, is to try to restore the babies' normal developmental channels by
removing the pathogenic factors in the infant-mother relationship.

Many of the mothers in this particular study, though, show no
conscious concern for their babies and therefore resist therapy for them-
selves. About the only thing psychotherapy has accomplished so far, Dr.
Greenberg finds, is to help some of the motheis get through periods of
extreme anxiety. If the study clearly demonstrates that certain maternal
responses are harmful, lie hopes eventually to let mothers see in moving
pictures how they are reacting to their babies' needs. Then perhaps the
point can be made that certain aspects of their behavior are not ideal.

When the infants arc removed from what the investigator terms their
psychotoxic environment, many improve quickly. The most dramatic
results occur in babies only a few months old who have feeding disorders.
Within a week after they have been hospitalized, these babies often stop
regurgitating, begin eating normally, and gain in weight. Further, they
show less withdrawal and apathy and more curiosity and social responsive-
ness.

Among the unanswered questions: How long will the improvement last?
Why do some of the babies, particularly those with motility disorders,
show no substantial change in the new environment?

Dr. Greenberg points to another aspect of the situation: some mothers
subconsciously do not want their children to become completely well—
they need sick babies.

Institutional vs. Family Babies

In preliminary work several years ago, Dr. Greenberg found decided
differences between institutional babies from the foundling home, and
family babies being reared by their own parents. One difference lay in the
babies' responses to pacification attempts when they were in distress. The
institutional babies as a group were easier to pacify than the others, and at
one month of age could be pacified equally well by any of three tech-
niques-feeding them, giving them a nipple pacifier, or holding them in a
sitting position in their cribs. The family babies, on the other hand, could

207

be smoothed much more easily by either of the first two methods tlun
the third. As the babies were pacified, their heatt rates dropped, hut th1
heait rates dropped considerably lower among the institutional babies.

These findings indicate, the investigator suggests, that family babk
may organize their response patteins earlier-that is, show a prefcrciitt
among pacification attempts earlier. In addition the findings may intfieali
that the babies raised in the nuisery (I) are more sensitive than the other;
to pacification effoits, and (2) have less effective feedback mechanisms
for controlling the heart rate. The increased sensitivity could be explained
by the conditions of sensoiy lestriction under which infants arc being
laised. In this study the attention given a nursery baby— for feeding,
batliing, and ail other purposes-was always less than 2 hours, and some-
times as little as an hour and a quarter, in a 24-hour day. The aides rarely
sang or spoke to the infants, and the fcedings-for which nipples with
large holes were used— were often very rapid.

Another difference was in weight. Even though the institutional
were on a more than adequate diet and were getting excellent
care, by the end of the first 2 months they had gained only half IKS much
as the other babies. Dr. Greenberg believes that this difference, too, is
attributable to the restricted sensory environment of the nursery infants
At tins point there is no evidence to suggest that the observed dif-
ferences are permanent or that they have implications for personality
development. In the matter of weight, in fact, there is evidence-given (he
investigator by a number of adoptive mothers-that institutional babies
quickly catch up after they have been adopted. The findings do indicate
to Dr. Greenberg that the nature and the extent of the opportunities for
communication between the infant and its surroundings can affect
physiological function and behavioral activity in infancy.

Tliis earlier work is being repeated and extended with the groups of
infants now under study.

For the Future

In order to study the frequency of developmental problems, the ic-
search group is sending pediatric checklists to each mothei who dclivcis nl
the University of Illinois Hospital. The lists enable her to describe quickly
virtually all aspects of her baby's behavior. Lists go out at 3, 8, and IS
months, and many of the mothers check and return them. If the baby's
behavior appears to be abnormal, the mother is asked to bring him in for u
checkup, and she may then be invited to participate in the study being
conducted by Dr. Greenberg.

In still another effort to get precise information about the conditions
leading to emotional disturbance, Dr. Greenberg's team is beginning to
survey 2,000 babies a year born to families using the well-baby clinics of
the Infant Welfare Society. Dr. Greenberg explains that the society
sponsors one of the few chad guidance centers in Chicago that treat
emotionally disturbed children of preschool age. Eventually a few of the
babies being surveyed will develop emotional disorders and be referred to

208

this center for treatment. Then the research group will be able to look for
a relationship between the circumstances of infancy and the types of
emotional disturbance that develop— or, as the investigator sums it up,
who gets what.

Most of the babies now under study because they have developmental
disorders come from families at the lower socioeconomic levels. Would
babies fiom other types of families show the same troubles? Dr.
Greenberg doesn't know, but talks with other psychiatrists and with
pediatricians have given him the impiession that babies from middle and
upper middle-class families have a much lower incidence of feeding
pioblems and of hypermotility. Developmental disorders in these babies
seem to be related somehow to sleep disturbances. Sometime in the future
the investigatoi hopes to check this point by studying infants from
Chicago's North Shore, where most of the residents aie professional
people.

Adopted Children

Studies in child guidance clinics and mental hospitals indicate that
adopted children have a higher incidence of a variety of psychiatric prob-
lems, but these studies generally have not taken into account either the
age of adoption or the time that may have been spent in an institution
before adoption. Dr. Greenberg believes both factors to be critical. The
babies in his institutional group, to be followed through the years, will all
have been adopted at the same age of 2 months. Eventually he hopes to
compare their emotional well-being not only with that of the normative
group, but also with that of some of the children from the foundling
home who were adopted just a few days after birth. In the case of the
2-month gioup, he also hopes to throw light on a largely unexplored area,
the nature of the parent-child relationships in adoptive families.

Rescaich Grant: MH 5527

Careei Program Development Award: K3-MH 13,984

Date of Intei view: June 5, 1964

References:

Greenberg, N. H. Studies in psychosomatic differentiation during infancy. Arch, gen,

Psychiflt.t 1962,7,389.
Greenberg, N. H. Origins of head-rolling (spasmus nutans) during early infancy.

Psychos-am. Med,, 1964,26,2.
Greenberg, N. H. Developmental effects of stimulation during early infancy. Ann N. Y.

Acad, Sci,, in press, 1965.
Greenberg, N. H., Cekan, P., & Loesch, J, G. Some cardiac rate and behavioral

characteristics of sucking in the neonate. Presented at annual meetings Amcr.

Psychosomat.Soc,, Atlantic City, 1963.
Greenberg, N. H., & Loesch, J. G. Celiac disease. Draft, 1963.

209

Greenberg, N, H., &. Loesch, J. G. A comparative study of some behavior;

physiological activities m nursery and family-ieared infants during the f

months of infancy. Presented at annual meeting Ainer, Orthopsychiat.

Chicago, 1960.
Greenberg, N. H., Loesch, J. G., & Lipgar, R. Preliminary observations on symi

foiination during infancy. Dratt, 1 964.

Investigator;

Margaret S.Mahler, M.D.

Masters Children's Center

New York,N.Y.

Prepared by:

Herbert Yahraes

Several times a week, a number of young mothers in the Greenwich
Village section of New York City spend a morning or an afternoon in an
old brownstone house on Horatio Street. Their children play or sleep. The
mothers care for them as usual, talk to one another, or read. All the while
they are being closely observed by a research team, and the mothers know
it. The young women who serve as playroom teachers are actually trained
observers, and one of them slips out every half hour to dictate a report on
the activities and attitudes of one of the mothers and her child. Other
observers, out of sight, make notes from booths. Even the room where the
diapers are changed has an observation window.

Never before, it is believed, have normal mothers and noimal children
been studied so intensively over the period in which the investigators arc
interested-frorn the age of about 5 months to the end of the third year.
The findings are expected to increase our undeistanding of why some
children become schizophrenic or show other signs of emotional dis-
turbance and to improve our ability to prevent such developments.

The brownstone house is the home of the Masters Children's Center, an
organization for the study and treatment of disturbed children and the
study of normal children. Dr. Margaret S. Mahler, the internationally
known children's psychoanalyst who directs the center's research program
(and is clinical professor of psychiatry at Albert Einstein College of
Medicine), believes that the roots of much childhood mental illness are to
be found in the relations between mother and child during the period,
beginning when the child is about 5 months old, in which their initial
oneness begins to slowly come apart. Dr. Mahler calls this theseparation-
individuation phase of child development, meaning the phase in which the
child gradually separates from his extremely close, symbiotic union with
his mother and comes to recognize himself as an individual. In the
investigator's words this is the period in which the child hatches from the
symbiotic membrane and becomes as individuated toddler. If something
goes wrong with the hatching process, Dr. Mahler theorizes, emotional
problems develop; if something goes very wrong, the problems may be
severe and in extreme cases may lead to the type of schizophrenia she has
described as symbiotic child psychosis.

211

H was through hei work with psychotic children that Dr. Ma
became aware of the impoitance of this period of development. In in
cases of symbiotic child psychosis, the symptoms could be explains
seemed to her, only by supposing that the child for some reason had
successfully come through the hatching process and was trying frantic;
to regain union with his mother.

But just what had gone wrong? And when? And might it have b(
prevented?

She found she could offer only theoretical answers, because this peri
of child development had never been studied intensively enough
provide a detailed description of what happens normally. To help supp
the needed infoimation, some of the staff of the Masters Children's Cent
made a pilot study of mothers and children and reported that in tl
normal course of events the separation-individuation phase apparent!
comprises four major parts. In each of these the child is developing bol
physically and psychologically; in each he is called upon to accompli?
certain tasks.

The current study was undertaken in 1963 to verify these findings am
to set forth in detail the mother-child relationship and the child's boliavio
patterns that ate characteristic of each subphase. Heading the staff assist
ing Dr. Mahler and Dr. John McDevitt, a psychoanalyst, and Dr. Kitty
LaPerriere, a psychologist. The information is collected by observing
mother and child at the Center, interviewing the mother every week,
making a long, informal visit to the home every 2 months, and occasional-
ly observing father and child. The babies are tested periodically, and
moving pictures of mother-child interaction are made both at the Ceuta
and in the home.

The families in the study, all of whom live within walking distance of
the Center, are described as middle class and Protestant; most of the
fathers and many of the mothers are college graduates. The mo the is m
happy to participate in the research, Dr. Mahler reports, because the
Center provides excellently equipped playrooms for their youngsters and
emotional support to themselves, and because the mothers like to fed tlial
they and their children are serving the cause of mental health. The Center
is open to them 3 days a week; most of them come in for two or three
mornings or afternoons. By the time the project ends, in 1 968, more than
40 children and their mothers, half of them in the course of the less-
detailed pilot study, will have been observed during the separation-
individuation process.

One of the objectives of the research is to point to the danger signals in
each phase of that process~the indications that child and mother are not
successfully coping with the problems natural to a given stage of develop-
ment and that they are, consequently, storing up problems for later stages.
It is hoped that the children now under study can be compared with one
another at various periods, into adulthood. Some of the children appear to '
be proceeding through the separation-individuation phase more successful-
ly than others, and a variety of at least temporary problems are being
encountered. A long-term follow-up, the investigators explain, would be

the best possible way of establishing a link between certain events or
patterns in this phase and the quality of mental health in later life.

As discussed later, the investigators think they have already spotted a
few of the danger signals. The chief general conclusion emerging from the
work so far is that the child's psychological development depends in large
measure upon the "emotional availability" of the mother. In all phases of
the separation-individuation process, he needs to be sure of her interest
and love.

The Roots of Childhood Schizophrenia

Discussing her earlier work with psychotic children and their mothers,
Dr. Mahler recalls an experiment by another investigator, in which chick
embiyos and newly hatched chicks were pricked with a pin. The chicks
that had hatched showed only infinitesimal damage but those still in the
shell developed majoi anatomical defects.

"The lesson is applicable to the child's psychological development," Dr.
Mahler observes. "Theie is a natural timetable foi the maluiation and
development of the controlling, steering, integrating part of the peison-
ality, the ego. The earlier in life that this timetable is interrupted, the
more detrimental to the total personality." From her previous research,
the investigator believes that in childhood schizophrenia the timetable has
been intenupted very early, either before or not long aftei the beginning
of the separation-individuation process.

Because of disturbed or insensitive motheis, some of the children who
later developed schizophrenia had suffered numerous and severe frustra-
tions and emotional trauma during the first few months of life. At one
extreme there was the depressed mother who could show no sign of
affection for her daughter. At the other extreme there was the smothering
and overwhelmingly affectionate mother who constantly overstimulated
her son and showed no comprehension of his need to experience life at his
own pace.

But Dr. Mahler found just as many cases of childhood psychosis in
which the mother had been of at least average competence and devotion.
In some of these cases the child had shown such an extreme vul-
nerability—had behaved so abnormally almost fiom birth— that not even
the most favorable environmental situation, the investigator is convinced,
could have prevented psychosis. In such cases, she says the vulnerability is
"seemingly intrinsic," or congenital.

Whether tl;e disposition to childhood schizophrenia is innate or
acquired, Dr. Mahler believes that the core of the problem is the same; an
inability on the part of the child to use the mother as a "beacon of
orientation." She points out that an infant's mental appaiatus is too
undeveloped to organize and act upon the stimuli he receives from within
and without; hence his survival depends upon his mother. His psycho-
logical development also depends upon her. Even after he has passed the
completely helpless stage, he looks to her foi guidance and strength in a
world that can be terrifyingly— as well as enticingly— strange.

213

in me cype ot childhood schizophienia described asattfism, Dr. M;
theorizes that the baby's personality development has failed to n
beyond (or else has regressed to) the ''normal autistic phase" of dcve
merit. During this phase, lasting from birth until sometime in the sec
month, the infant seems completely unaware of any distinction beUv
himself and the world around him. The most conspicuous symptom of
autistic child, says Dr. Mahler, is his apparently complete failure
perceive his mother as a living being who represents the outside world i
serves as a beacon and guide. Mothers say: "I never could reach

baby " "He never greeted me when I entered, he never cried or ci

noticed when I left the room. . . ." "She never made any peisonal tipp
for help at any time." Such a child seeks refuge in an "autistic shell," t
investigator believes, because he appaiently has experienced reiiJity-a
even his" mother-as an intolerable source of initation; he has defend
himself by warding off reality and withdrawing into an iitler
constricted, deammated world.

In symbiotic child psychosis, according to Dr. Mahler's theory, fl
trouble with the baby*s personality development arises at a later stage an
results from an imbalance between the rates at which the child is mntitrin
physically and emotionally.

As the investigator explains it, a maturational spurt puts the nonnt
toddler, in his second year, in the position of relatively advanced physica
independence. But in some cases his emotional development seems lolai
far behind- Physically, in such a case, he is able to move away from hi'
mother; emotionally he is not. So he is bewildered and panicky. Miuoi
frustrations common to the early part of the separation-indivMuation
phase bring extreme reactions. He may give up walking for montlis
because of a fall. Typically, the break with reality is triggered by some
event-such as enrollment in nursery school, the birth of a sibling, hos-
pitalization-that makes him fear loss of or separation from his mother
and throws him into a panic he cannot handle. In his hallucinations he
appears to be trying to restore the delusion of an earlier period that lie
and his mother are cue. Since constant panic is unbearable, the child
retreats into "secondary autism" and cuts himself off from the world.
Though his behavior is then like that seen in autism, says Dr. Mahler,
successful treatment depends upon realizing that his personality hns
developed further, to a stage in the process by which a child separates
from his mother. The goal of therapy, she believes, is to enable the mother
and the child to reestablish the symbiotic tie, and then to help them move
forward through the separation-individuation phase.
During the earlier research, Dr. Mahler and her associates were able to
lieve this goal in about half of the dozen child-and-mother cases treated
tir of the others withdrew from treatment against professional advice).
erapist saw mother and child together -several times a week and for
as I W~ at a time. The therapist's first task, the investigator
e her presence felt -to allow the child to experience it
e without having to acknowledge her existence as a
er presence became a soothing phenomenon, and the

child felt more comfortable with it than without it. Then the child came
to use the therapist as an extension of himself— a tool to reach something,
a soft platform to learn against. Eventually he allowed her to meet his
needs more actively— to feed him, to play with him.

"The theory is that the therapist represents a mothering principle, not a
distinct human object," Dr Mahler says. "The situation seems to be
comparable to that stage in development when the baby dimly recognizes
that ministrations iclieving distress come from outside himself. It is likely
that the therapist's comforting piesence helps to amelioiate some of the
child's intense aggression and destructiveness and, therefore, reduces the
level of anxiety." Gradually, and in the face of considerable resentment at
having to relate to her preschooler as to an infant, the mother was led into
the kind of relationship with the child that had been established by the
therapist.

The present reseaich, with normal children and their mothers, was
undertaken both to check Dr. Mahler's ideas about the critical importance
of the separation-individuation phase and to help establish— as an aid to
workers in the field of child mental health— the normal course of events
during that period.

Stages of Development

From the work to date, Dr. Mahler offers the following picture of a
child's development following the "normal autistic phase" of the first few
weeks of life.

Beginning about the second month and continuing into the fifth, the
infant is in the symbiotic phase. The boundaries between himself and his
mother still tend to merge but he seems to be dimly aware-particularly
when he is being fed, changed, or actively cared for in other ways-that his
needs are being met by something outside of himself,

By the fifth or sixth month, the infant seems to recognize that the
object through which his gratifications are provided and his discomfort
relieved is his mother. When her face is near, lie tries to touch and
investigate it; he watches her play peek-a-boo and then plays it himself.
This is the beginning of the separation-individuation phase, the subject of
the current research.

The first 4 to 6 months of this stage are found to comprise a fairly
distinct subphase which Dr. Mahler labels differentiation. It is marked not
only by explorations of the mother's face and mouth and hair but also by
a turning to the outside world for pleasure and stimulation. The infant
looks beyond neaiby objects. He takes a more active interest in toys. He
finds pleasure in using his whole body. He begins to creep, climb, and
stand up. He makes progress in coordinating the use of eyes, hands, and
mouth. During this period the children at the Center were seen to be
active for a longer time and to be more vivacious when their mothers were
close. They showed a distinct preference for playing at their mothers' feet.

215

Beginning as early as the 10th month and as late as the 12th comes (
subphase desciibed as practicing during which the child takes delight
trying out his new skills, particularly his ability to get around by hiinsc
Eager to explore his environment, he ciawls or walks farther and nut!
from his mother's feet and often becomes so absorbed in his own activiti
that for long periods of time he seems to be oblivious to his mollio
presence.

The infant's dominant mood during this practicing period, the inves
gator repoits, is elation. As another authority on children has expressed
the toddler is now beginning his "love affair with the woild." There
such wonderment to be explored that he often disiegards bumps, fall
and other frustrations. Sometimes at the Center he slips out of the infii
room, where his mother is sitting, and makes his way into the adjacei
toddler room which is equipped with toys and apparatus for the olcf
children. But in the midst of his investigations he stops as if sudden
aware that something is wrong and goes back rapidly to his mother, onl
to venture forth again a little later.

A child during this phase seems to need peiiodic physical contact wit
his mother. "We see babies crawling to the mother, righting themselves o
her leg or touching hei in othei ways," says Dr. Mahler, "or just stnnclfn
and leaning against her for 'emotional lefueling.' Even a wilting, fatigue
infant perks up in a very short time upon such contact."

The Period of "Rapprochement"

At 18 months, or a little eailier 01 later, the child gradually passes ind
a stage when he is more subdued and even a little troubled. He has bcei
acting during the practicing period as though he were omnipotent and tin
world were his for the taking. Now, Dr. Mahler reports, it dawns on bin
that he has to cope with the world as an individual who is very small
relatively helpless, separate, and lonely. Watching him in the playroom
the investigators notice that he is no longer either relatively unaware of life
mother or unmindful of frustrations. And they interpret some of hi1
expiessions as indicating surprise at finding himself a separate being, Foi
example, says Dr. Mahler, when he hurts himelf he looks perplexed
because his mother is not-instantly and automatically-with him. As UK
toddler realizes his physical ability to move away from his mother, he
generally seems to have a great need for her to share every new skill und
experience. The observers' notes and the moving pictures show that
children during this third stage of the separation-individuation process are
continually concerned with their mothers1 whereabouts.

Those watching a youngster week after week report than in this phase
of development the toddler's pleasure in functioning on his own is
proportionate to his success in eliciting his mother's interest and participa-
tion in his activities. When the mother leaves the room the child-
sometimes only by a fleeting expression of sadness or anxiety, sometimes
by a tantrum-displays unhappiness. Sometimes he goes and stands by or

216

climbs onto her chair, taking it over for a while as a substitute for hei. He
does not easily accpet physical contact with people who offer themselves
as substitutes. Because of this behavior, Dr. Mahler calls this third stage
the period of rapprochement For most of the youngsters studied, it is a
time when indwiduation proceeds rapidly, but separation is resisted
strongly. It is a period that is conducive to misunderstandings between
mother and child. Some mothers cannot accept the child's demandingness,
cannot undei stand why a toddlei who is obviously more capable and
independent than he was a few months ago during the practicing period,
must now insist upon her sharing every aspect of his life.

As the investigate! explains it, the child needs his mother's active
emotional support at this time in older to prevent seiious injury to his
self-esteem. He has had inklings that he is not really the omnipotent being
he had fancied himself. If he can count on backing from his mother, he
comes giadually to accept this fact and to pour his energy into normal
psychological development. "If the mother is quietly available," Dr.
Mahler says, "if she shares the toddling adventurer's exploits, playfully
reciprocates and thus helps his attempts at individuation, the relationship
between mother and toddler progresses to the point where verbal com-
munication takes over. Emotional paiticipation that the toddler can count
upon seems to facilitate the rich unfolding of his thought pioccsses.

"The less emotionally available the mother has become," The investi-
gator repoits, "the more insistently and even desperately the toddler
attempts to woo her." He won't accept comfoiting from anyone else; he
can't seem to lose himself in play. If emotional supplies are not forth-
coming, he seeks substitutes in eating and sucking, He also turns to such
aggressive behavior as throwing things and hitting people. All this may
drain so much of the energy available for development, Di. Mahler
theorizes, as to hamper psychological growth.

One sign of an unusual degree of conflict in the youngster, the investi-
gator believes, is a more than average amount of "shadowing"-that is,
keeping the mother in his sight even in the midst of play, and at other
times following her around. Another sign is an exaggerated use of the
game of daiting away in order to provoke the mother into pursuing him
and scooping him up. Both types of behavior have been observed in some
of the children under study. Other children have shown a third danger
signal-severe and protracted "separation anxiety" when the mother leaves
the room.

The Final Subphase of Separation Individuation

The fourth subphase begins about the end of the second year and
continues well into and sometimes all the way through the third year. It is
a complex stage which the researchers describe as "the period in which
object constancy is attained." They mean by this that during the final
phase of the separation-individuation process the child develops the ability
to retain mental representations of himself and his mother as distinctly

217

separate individuals. When she is absent, the investigators explain, he win
picture her as being away from him and also as returning to him; lici
continual physical presence becomes less impeiative.

As an indication that this stage has begun, the child makes his way to
the toddler room For prolonged peiiods and accepts the nursery teacher us
a partial substitute for his mother. For example, he takes food irom lit'f
and he gladly joins with her in play. But under many circumstances
including aggiession by or against another child, weariness, a bump, soiled
diapers-he calls for his mother and seeks her out. And if he is in that kind
of emotional state he will insistently demand her when someone else
enters the room, perhaps the opening dooi evokes the anticipation ot the
mother's appearance, the research team suggests, or perhaps the stress ol
having to cope with an additional, less familiar person makes the need for
mother more acute.

The child shows an inci easing inteiest both in his playmates itntl in
adults other than his mother. He is mildly or moderately ncg«livis!ic,
showing resistance quite often to the demands of adults-a cluiracterMk1,
remarks Dr. Mahler, that seems essential for the development of a sense- of
identity. He begins to develop an awareness of time, commonly itbsenl in
the schizophrenic child. Along with it he shows an increased nbilily tn
endure separation and delays. Concepts like "later" and "tomorrow11
generally first associated with the activities of his mother-come lo !H!
used as well as understood.

Crises arise in this period, Dr. Mahler reports, when the mother cannot
accept the child's negativistic behavior and his more or less frequent
display of "primary process" type of thinking, manifested in primitive,
illogical talk and actions, and seen at its most extieme in schizophrenics.
This kind of thinking is common among youngsters during this period,
says Dr. Mahler, but even some apparently normal mothers are unable lo
deal with it calmly and help the child progress to the rational "secondary
process." One mother, for example, found fault with her youngster for
eating well at Grandma's but poorly at home. Angeicd, the child toUl her,
"I eat you up. Then I spit you out. Then I put you in the garbage puil,"
The mother was hurt, angry, and disturbed. She and the yaimgsU-r
bickered for an hour. Finally the girl said, "Then I put you back loj'elher
again." It would have been more conducive to emotional health, Dr.
Mahler observes, had the mother said at the very beginning something
like: "Now you know you are just being angry at Mommies, Miiyto: I
shouldn't have said anything about how well you eat at Grandma's, Any-
way, you know very well that nobody eats people and puts them in
garbage pails."

Bridging the communicative gap between the child's world tmd Ihe
adult's, says the investigator, requires the deciphering of the child's
primary-process language and actions, playing along with them, ami
gradually offering him more logical expressions and more rculislic
solutions to his problems. Fortunately, most mothers do this. To find and
help those who don't, Dr. Mahler and her associates believe, would be one
ot the ways to help reduce the incidence of emotional disorders.

218

Variations in Behavior

The types of behavior characteristic of one subphase are not limited to
it. In other periods, though, says the research team, either they are seen
less frequently or they are less important to the particular developmental
tasks with which the child is dealing. For example, a joyful exploration of
the woild is characteristic of the second, or practicing, subphase but of
course occurs during the rapprochement period as well. Its occurrence
during that period, though, is less significant than the behavior by which
the child is working through the job of finding himself an individual,
separate from his mother, yet renewing on a more advanced emotional
level his bond to her (rapprochement).

Again, during the final subphase, even a youngster who has been eagerly
coming to the toddler room for a long time may suddenly start clinging to
his mother when she leaves for the infant room, where mothers are en-
couraged to stay. Or he may let her go but keep running back and forth,
as he did for a time months earlier.

Nor does every child follow the pattern most characteristic of a
particular stage. For some children in the rapprochement period, the need
to assert separateness over-shadows the need to establish closeness. While
one child seeks out his mother with every new toy or activity in order to
engage her attention and participation, the researchers report, another
child may be primarily engaged in making sure that this very thing does
not happen. He will hold a toy in his hand, look at his mother, and veer
off to seek out a different adult, or to remain by himself. This is most
likely to be so if his mother has been overprotective and intrusive; in this
case it is she, rather than the child, who becomes the "shadower."

The Mother's Influence

The research is documenting what most mothers know-that different
children do not proceed through a given development stage either at the
same rate or in the same way. And it is uncovering evidence that the
motheis themselves influence the pattern.

Early during the first stage of the separalion-individimtion process, for
example, Bernie showed great interest in locomotion and would per-
sistently try to crawl and to pull himself up. Stuart, on the other hand,
usually just lay on the floor and looked; though obviously interested in
the things and the people around him, he made little effoit to get to them.
Constitutional factors may have been at work, say the investigators, but so
were the attitudes of the mothers. Bernie's mother often appeared listless
and apathetic during the earliest months but cheered up—and encouraged
the boy-when he began moving around. Stuart's mother, though, liked to
keep him close. Instead of encouraging him to shift for himself, she met
all his needs so promptly that he never had to exert himself to get what he
wanted. On the developmental tests when they first came to the center,

219

Bernie and Stuart rated about the same. A few years later, however, Stuart
did not do nearly so well as Bernie.

Two sisters, Ann and Susan, are offered as particularly good illustra-
tions of the strength of the maternal influence during the separation-
individuation phase.

When Ann was 9 months old, she was often observed sitting at her
mother's feet, looking up at her and patiently begging for attention. She
got very little of it and in consequence, the investigators say, had little
psychic energy for investing In the activities normal to the next, or
practicing, subphase. She would make only brief forays from her mother;
her "love affair with the world" was subdued and of unusually short
duration. Throughout the final sub phase, Ann was an unhappy little girl
who could not easily endure physical separation from her mother, did not
get along well with other adults and children, yet showed little joy when
the mother returned after brief, everyday separations. In one camera-
recorded scene, she has a tantrum when her mother starts to leave the
room; the child insists on going along but then gives up and just stands
there, suffering. Finally she regresses by retiring to the play area for the
youngest babies. She turns her back on the other people and is clearly
hurt and angry,

Like a few of the other children in the study, Ann is described as
already vulnerable— already in trouble. The investigators believe that these
vulnerable children, unless further environmental experience amply
compensates, may well develop emotional problems rooted in the unmet
needs and unaccomplished tasks of the separation-Individ nation phase.
After the present study has been completed, the researchers hope to
predict what types of problems are likely to develop in a given child under
given circumstances and then to keep in touch with each family foi some
years.

With Susan, Ann's younger sister, the mother had mellowed even
though she was the same somewhat aloof, self-centered person. At the
center, every so often she would put the baby down and bury herself in
the newspaper. But Susan was a more outward -going and determined baby
than Ann, and when she wanted her mother's attention, she knew how to
go about getting it. In one scene, she tugs at her mother's dress, beseeches
her with her eyes, and finally starts to pull herself up to her mother's
knee. The viewer can almost hear the mother say, "Oh, the heck with it,"
as she puts the paper aside and lovingly picks up the baby. In a later
subphase, Susan looks distressed when her mother leaves the room but,
unlike her sister, soon turns happily to playing with the other adults find
children. She is joyful when her mother comes back. A child who has a
good relationship with her mother, Dr. Mahler emphasizes, shows relative-
ly little separation anxiety.

The difference between Ann's experiences and Susan's, the investigators
suggest, may have had a genetic basis, since the younger girl was able to
command the mother's attention and draw upon and be secure in her love
while the other was not. But they believe that Ann's progress through this

220

critical period would have been less stormy with a mother moie attuned
to the child's emotional needs.

With Genie, Danny, and Matthew— three children of Mrs. A., all of
whom have been studied at the center— the similarities have been more
striking than the differences. All three have been conspicuously motor
minded. They undoubtedly had an inborn motor proclivity, says Dr.
Mahler, but they also had a powerful secondaiy impetus: the mother's
great interest in having them become independent as soon as possible. The
incessant crawling, climbing, seesaw balancing, and such activities were
not only pleasant in themselves but also brought the mother's approval.
Particularly in Genie's case, the investigator believes that the hyperactivity
probably served another purpose as well: it seemed to make up for an
unmet need arising from the mother's tendency to ward off physical
closeness and cuddling.

"The research has clearly established," Dr. Mahler reports in summing
up the work so far, "that the emotional availability of the mother is
necessary for the optimal unfolding of the child's innate potentialities."
This is so, she explains, because the child, throughout the process of
separating from the mother, is emotionally so dependent upon her. If he is
left markedly unceitain about her emotional availability, signs of potential
trouble aieseen.

Dr. Mahler adds that the research also points to the sturdiness and
adaptive capacity of the normal toddler. "Even against considerable
odds," she remarks, "he is usually able to extract from his mother the
necessary emotional supplies."

When the cunent work has been completed, the investigators expect to
write a detailed description of mother-child interactions during each part
of the separation-individuation phase and to list and explain the danger
signs—those behaviors and attitudes believed to warn of impending
emotional trouble for the child unless corrected.

Research Grants: MH 8238, Mil 3353
Date of Intei view: Mar. 17, 1966

References:

Mahler, Margaret S. Thoughts about development and individuation. Psychoanal. Stud,

Child, 1963, 18.
Mahler, Margaret S. On early infantile psychosis: the symbiotic and autistic

syndromes./ Amer. Acad, Child Psychiat., 1965,4,4.
Mahler, Margaret S. On the significance of the normal separation-individuation phase,

Drives, Affects, Behav,, 1965, 2.
Mahler, Margaret S. On the development of basic moods, and the depressive effect. In

Solnit, Schur & Loewenstein (Eds.), Heinz Hartmann Festschrift. New York; Inter-
national Universities Press, 1966.
Mahler, Margaret S., Purer, M., & Settlage, C. FF. Severe emotional disturbances in

childhood; psychosis. In American handbook of psychiatry. New York: Basic

Books, 1959.
Mahler, Margaret S.f & Laperriere, Kitty. Mother-child interaction during separation

individuation. Psychoanal Quart., 1965,34.

221

Investigator:

Richard Q. Bell, Ph.D.,
Chief, Infant Development Section
Child Research Branch, NIMH

Prepared by:

Antoinette Gattozzi

Most psychologists would agree that social behavior develops out oflh:
interplay between the individual and his environment. From birth to <MtK
childhood the most significant elements of the individual's environmcnl
are probably his parents. The relationship between child and ptircnls.ontt
thought to fit a simple stimulus-response paradigm, generally is rc^rited
today as a complex interaction whose nature is affected by both child ar.J
parents. The contributions of each coalesce with gieat subllely iintl be-
come more and moie difficult to distinguish as the relationship mmi>
through time, In a research piogram designed to circumvent thisdifficu!
ty, Dr. Richard Q. Bell and his associates are looking for newborn hchavicr
characteristics that are precursors of social development at Ihe prcschorf
age. The investigators are studying infants before they begin to inters!
with parents, then observing the subjects' social behavior when they rva^
the age of 2/2 years.

Dr. Bell became interested in the direction of effects at work h
parent-infant interaction tlirough research he did some time iigo with Or
Earl S. Schaeffer of the NIMH Laboratory of Psychology. Tim luo
scientists devised a questionnaire to compare attitudes of molhm of
mentally normal and abnormal children. Attitude patterns of mo liters of
schizophrenic children could not be distinguished from those of niolhwi
whose children were mentally retarded because of brain damage orotrm
birth defects. Dr. Bell inferred that the same factor-reaction to the diiU'i
condition-might be the significant one. This line of reasoning led hiniK
questions about the child's effect on the parent and to the search f«
congenital characteristics.

Increasingly in recent years, clinicians have postulated tluit faclon
contributing to temperament differences among individuals m;iy te
present at birth, and that these early differences may be implicated in to
later development of mental illnesses. No clear evidence has been four-J
however, that indicates what these factors are or how they are dGlckriQu*
The investigators' efforts to identify congenital contributions to SOCL!
development, although not directly concerned with etiological problem*.
may show how to begin to answer such questions.

Two aspects of behavioral development were of special interest to the
investigators— the child's tendency to stay near adults and his ability to
cope with sudden, unexpected changes in the environment. Attachment to
adults, in a sense, is the binding force of the parent-child interaction as
well as its most important product. The mother may have no task of
gi eater consequence during the first few months of her infant's life than
to make her presence and actions important to him. Both attachment to
adults and coping ability are highly salient features of human behavior,
the development of which can be observed and measured in very young
children. Moreover, these behaviors show a wide range of variability in any
random sample of childien. "They are fundamental dimensions of normal
development," Dr. Bell notes, "and they find a prominent place in
descriptions of both normal and aberrant personalities."

The research proceeded in two parallel projects. In one, the scientists
observed neonates to record characteristics of ^individual infants before
cultural forces had begun to shape them. These children were studied again
when they reached preschool age. In the other project, the investigators
observed the play of preschool children to detect patterns of social attach-
ment and coping ability. Questionnaires were used in both projects. Each
mother of a newborn subject filled in a questionnahe about her infant's
behavior when he was 1 month old, then completed the same question-
naire when he was about 2'/2 yeais old. Mothers of the preschool children
not studied as newborns completed the same questionnaire; that is, each
recalled her child's behavior as an infant. Dr. Bell's design of the research
included both prospective observations and retrospective questionnaiies
because he hoped to demonstrate how the two approaches might be com-
bined in a longitudinal study to accelerate data analysis. This is another of
Dr. Bell's research interests; it is a central concern in difficult, time-
consuming longitudinal studies.

The major outcome of the program to date is the tentative
identification of a congenital behavior syndrome, a pattern of newborn
behavior that is linked to a pattern of social behavior at age 2V4 years, The
investigators found that a newborn male who shows high formula
consumption relative to birth weight, whose respiration is rapid, and
whose mother describes him as restless and hard to soothe at age 1 month,
will be, at the age of 2Vfc years, relatively nonsocial. He will be rather
disinterested in other children and relatively independent of adults.

Research results also include, in brief, the discovery of biological
consequences associated with birth order and sex, the identification of
biological effects of family size and structure on preschool social behavior,
and the establishment of a tentative link between patterns of preschool
social behavior and styles of cognition at age 6 years.

The Study of Newborns

Newborn infants are notably difficult subjects to study. In the welter of
behavior they exhibit, many items are affected by temporary conditions

223

of this period. Above all, change in the newborn is swift. To
meaningful and replicable issults, factors known to affect neoi
behavior tiansiently either must be accounted for or their eff
eliminated from the data. Elements that change rapidly have tc
avoided. Dr. Bell comments: "As an uncivilized human being the newt
is intriguing theoretically, while being at the same time one of the r
difficult subjects as far as abstracting stable measurements is concert!'
Many past studies of newboins reported conflicting or inconck
results because they failed to isolate influential variables. Dr.
compiled the following list of factors to be controlled in studies of inf
and young children. It leveals how demanding such research is, espec
studies carried out during the first hundred horns of life.

1 . State of arousal— There is a continuum of aiousal that ranges I
quiet sleep at one end to the excitation of robust crying at the other
The infant's position along the quiescent range of the continuui
particularly difficult to judge, yet his arousal state clearly affects
behavior.

2. Complications of pregnancy and delivery.- These exert n
effects, depending on their severity. In this connection it is worth nc
that complications are significantly fewer the higher the socioecom
level, and that males are more subject to complications than females.

3. Parity. -Evidence exists both documenting certain tran
congenital differences between fiist-and later-born children and hinti
the possibility of lasting differences. For example, first-borns have
shown to have higher levels of L-lactic acid and 1 7-hydroxy corticoste
during the first few days of life.

4. Analgesics and anesthetics.-ThQsz effects may originate in
prenatal or postnatal periods. Barbiturate medication give predeliver)
instance, has a pronounced dampening effect on the newborn's fei
adequacy and attention to visual stimuli up to the fourth day of li
mother who is breast feeding and on sedation may pass residual sed
to her infant.

5. Age and sex-Theie are key parameters in themselves, as w<
factors that interact with others.

Over a period of 2 years, the investigators studied a total of 75 in
in the nursery of Suburban Hospital in Bethesda, Md. All infants
white, full-term, and free of complications of pregnancy and birth, s
as their records indicated. They came from families of various s
economic backgrounds, and none was a first child. Observations
made on the third and fourth days of life-as far as possible fron
confounding conditions concomitant with birth and as close as possil
the time infants were to be taken home. Among the measures tl
searchers obtained were rate and regularity of respiration, mouth i
ments, eye movements behind closed lids during sleep, rate and pattc
non-nutritive sucking (on a sterile nipple stuffed with cotton and of
ju&t before regular feeding), rapidity and vigor of response to in term
of feeding, amount of time and height to which an infant raised his
while prone, tactile sensitivity, and various body dimensions. Of t

224

neonate characteristics measured, the investigators were interested only in
those that showed wide range across infants and stability within individual
infants, or those linked to known factors such as sex or parity. Stability
was judged by the consistency of the measure from test to retest.

Interaction between sex and type of feeding. -Measures were made of
the prone head reaction (PHR), which is presumed to be a correlate of
muscle strength. The infant was placed supine for 55 seconds, then turned
to the prone position. A soft rubber device was held near the infant's head
to measure how high the chin was lifted during the next 60 seconds. On
the basis of past studies, the researchers expected males to show higher
PHR scores than females, indicating gi eater muscle strength in males.
They also expected that PHR would correlate negatively with skin
sensitivity, which was being tested separately.

Results were no more than suggestive concerning the hypothesis of
greater muscle strength in males, and the data relating muscle strength to
skin sensitivity also were inconclusive. Although the association was in the
expected direction— the greater the muscle strength, the lower the sensi-
tivity of the skin— correlations did not reach significance.

Perhaps the most valuable result of the PHR measurements was the
discovery of a transient interaction between sex and type of feeding. On
the first round of measurements, breast-fed males and bottle-fed females
scored the highest PHR values; 13 hours later, at the time of the second
observation cycle, this phenomenon receded and males scored generally
higher than females regardless of the way they were fed.

The investigators found the same interaction between sex and type of
feeding in their study of tactile sensitivity. This was not unexpected in
view of the relation, even though it is of low order, between muscle
strength and skin sensitivity. Unlike the experience with the PHR
measures, however, the effect of the interaction did not disappear on
retest. Breast-fed males and bottle-fed females still showed less skin
sensitivity. These results confirmed the finding that type of feeding can
interact with sex to produce an effect during the neonate period; the
interaction affects, at least, PHR and skin sensitivity, Studies of newborns
that unknowingly sampled different proportions of breast-fed males and
bottle-fed females could encounter contradictory data.

Yet another surprising effect related to type of feeding was noted,
After infants were handled for body measurements, breast-fed babies
became and remained more active than bottle-fed infants. Dr. Bell offered
two possible explanations. An electrolytic difference between the two
kinds of milk may differentially affect arousal centers; or perhaps a
breast-fed infant is more responsive to handling because he is usually more
hungry than one fed by bottle. This may occur because his mother's milk
is low in volume or late in starting.

Relation of skin sensitivity and arousal characteristics.— The original
impetus behind the study of skin sensitivity was to search for sex dif-
ferences, There is evidence that adult females have greater skin sensitivity
than adult males. Such reports contributed to a hypothesis developed
from observations of primitive human societies. The hypothesis is that

225

males are equipped congenially with gi eater muscle strength and higlu
tactile thresholds than females in order to fulfill the adult male role c
gross motility in space and low response to pain. The principal adul
female role, on the other hand, is the place-bound care of offspring, <i rol
enhanced, according to the hypothesis, by the female's lesser muscl
strength and higher skin sensitivity

Investigations of newborn skin sensitivity have not been able toga the
definitive evidence demonstrating a sex difference. It is still not known
therefore, to what extent the greatei skin sensitivity of adult females is j
congenital attribute and in what degree it is culturally induced. Tc
discover how nature and nurture interact, scientists must isolate anc
measure each influence separately.

The investigatois tried three different kinds of skin stimuli: Removal oi
a coveiing blanket, a fine jet of air, and application of the nylon points ol
an instrument known as an aesthesiometer. The first two of the three test;
confirmed the hypothesis; their results pointed to greater skin sensitivity
in females than in males. Females were more sensitive to stimulation ol
the abdomen, for example, than were males. The interaction of sex and
type of feeding was found in the trials using the aesthesiometer.

The procedure with the aesthesiometer followed this form. As the
infant slept, one of a set of increasingly thick nylon filaments was pressed
against the heel of the left foot until the filament bent, at which point (he
pressure being exerted was exactly known. When the pressure of a
filament elicited a response— anything from the flexing of two toes lo
mass body movement-the value was recorded as threshold. The procedure
was followed three to five times duiing a sleep cycle (the infants slept
right through the tests), and the final score was the mean of the individual
trials.

As noted above, the tactile sensitivity data showed sex diffeiences in
some tests and on some subjects but not others, and the question of
congenital sex diffeiences in skin sensitivity could not be settled. How-
ever, two consistent relations appeared that opened an avenue for follow-
up studies. Skin sensitivity scores were found to relate to other charac-
teristics of the infants, Newborns with low skin sensitivity showed a low
level of arousal as indicated by lespiration characteristics while asleep and
by slow response to interruption of sucking. Thus the data contributed to
one of the research program's preliminary goals, the description of new-
born behavior patterns that might be related to later behavior patterns,

Effects associated with sex and parity.— \l became obvious early in the
course of the newborn study that precise location of an infant's position
on the aiousal continuum was essential to clarify other variables. Arousal
levels were monitored by subjective measures when the infant was awake
and by respiration counts when asleep. The investigators were not entirely
satisfied with the subjective measures and their use made replication of
the study by others hard to do. To find out whether an objective measure
of arousal might be used instead, Dr. George M. Weller undertook a
separate study of infants' galvanic skin response. (Dr. Weller was a Ph. P.

226

candidate working at the Child Research Blanch at the time. He is now on
the faculty of Temple University in Philadelphia.)

Forty newborn infants from the nursery of the U.S. Naval Hospital
Medical Center in Bethesda were the subjects. They were screened to
exclude any with prenatal or perinatal complications. Only bottle-fed
infants were chosen, and both first- and later-born were included. Skin
conductance was measured by an electrode disc taped to the bottom of
each foot. Activity was observed by Dr. Weller and recoided on a manual
counter. Some items in the activity inventory were agitated crying, head
movements, arm and hand movements, movements of the eyes behind
closed lids, and no movement. These items defined six states of arousal
langing from deep sleep to marked agitation. Visual counts of respirations
were made and then galvanic response was measured.

Dr. Weller found conductance and activity levels positively correlated
more strongly in the waking than sleeping states; respiration, too, after
special statistical treatment of the data, showed a positive correlation with
conductance. The lesults indicated that galvanic skin icsponse might be
used as an objective measure of infant arousal levels. Dr. Weller also
discovered that females and later-born infants showed consistently higher
skin conductance and displayed highei levels of activity than their
counterparts. He cautioned that before any implications may be drawn,
the findings must first be confirmed (the finding on females has since been
located in data fiom another study), and then shown to extend beyond
the first days of life. Nevertheless, the study found the first unequivocal
evidence of congenital differences between males and females and
between first- and later-born infants. Because differences in conductance
are associated with differences in maturation during this very early period,
it can be said that females and later-borns arc congenially moie mature
than their counterparts up to the fourth clay of life.

The results may be pertinent to the question of how nature and nurture
interact. Girls are treated differently from boys by theii parents; second-
and later-born children not only experience the company of siblings
absent from a first-born's early months of life, they also get different
treatment from parents now versed in child rearing, These cultural differ-
ences frequently are cited to explain the persistent and often very obvious
differences in social behavior between males and females and between
first- and later-born children. Dr. Welter's findings indicate that biological
factors may play a role in these differences.

Stable characteristics of newborns.-ln the end, the voluminous data
accumulated in the newborn study were analyzed to extract items that
were stable from test to retest. Respiration rate during sleep and tactile
sensitivity as measured by the aesthesiometer were the most stable charac-
teristics. One other measure readily available from hospital records proved
very stable; this was the amount of formula consumed by bottle-fed
infants. Sucking patterns, total crying in response to sucking interruption,
and PHR were next in order in stability. Borderline stability was shown by
measures of respiration variability, pattern of crying, and reaction time in

response to sucking interruption. It seems doubtful that many measures o
arousal level and spontaneous skeletal muscle movements will be useful ii
studies of differences among newboins because of the instability o
individual scores on these items.

The Study of Preschool Children

The preliminary study of preschool social behavior began with sepaiati
samples of 2-year-old boys and gnls for whom no newborn data wen
available, and continued adding subjects as the children completed the I
month of study and the investigators shaipened their methods. Childrci
selected for these preliminaiy studies were second- or later-borns. Eacl
mother completed a questionnaire that asked her to lecall her child1;
behavior as an infant. From this questionnaire the researchers drew uj
"clusters" of infant behavior characteristics. These recalled chaiacteristic!
were then lelated to clusters of preschool behavior displayed by tin
children. The analysis gave the investigators their first leads to the new-
born and preschool relations they might find when they followed up the
newborn subjects.

Later, when the children who were studied as newborns were near the
age of 214 yeais, they were brought to the nursery school daily in groups
of six or seven of the same sex. Just as in the preliminary studies, they
were studied at play by two teachers who participated in their activities
and by two observers seated behind a one-way vision window. None knew
what newborn factois had been singled out for possible relations to [he
behavior they watched, The observers counted and timed aspects of play
and the teachers rated the children on such items as amount of physical
contact initiated with teachers, ability to be soothed by teachers when
upset, and relations with peers. Boys and girls displayed different patterns
of play, an observation made by many other students of child behavior.
Girls were more aware of others than boys and were more shy with strange
adults. Boys were more interested in toys than people. Girls seemed more
mature than boys in speech development and motor coordination.

Maternal recall as a mediator variable, -Dr, Bell devised a complex series
of analyses employing maternal recall as a mediator variable to link the
prospective and retrospective data and test the feasibility of this method
of accelerated longitudinal analysis. The series began, as noted above, wilh
the first samples of the preschool study by testing for relationships
between sets of measures made by observing a child's behavior in nursery
school and sets derived from the mother's retrospective report of his
infancy. The comparison yielded the first potential relations between
infancy and preschool behavior,

The next analysis used questionnaires completed when their infants
were 1 month of age by mothers of subjects studied as newborns. Sets of
items located on these were tested for relationships with sets from the
(same) questionnaire retrospectively completed by mothers of preschool
subjects not studied as newborns. In this way, the questionnaire itself was

228

validated and measures were located in the 1 -month questionnaires that
could be related to measures made in newborn observations

A third analysis located relations between sets of newborn measures
and sets of items from questionnaires completed at age 1 month. The final
step was to test foi relations between newborn and pieschool chaiac-
teristics in a common sample. One set of newborn characteristics does
appear to be associated with a set of pieschool social characteristics. This
is the congenital behavior syndrome in males discussed earlier in this
leport. It traces continuity between newborn behavior and social lespon-
siveness; no clear relations have been found yet between newboin charac-
teristics and later coping ability. The findings that led to the description
of the congenital behavior syndrome cuirently are being prepared for
publication, along with a report of relations found between all measures.
The latter analyses do not rely on mediator variables such as maternal
report or an index of family size and structuie. This index, in addition to
its potentiality as a mediator variable, was found to be related to aspects
of dependency behavior.

Effects of family size and stmcture.-hi\ arresting featuie of behavior
had appeared in the preliminary sample of preschoolers. Boys from large
families with short intervals between children were consistently higher
than other boys on the rating of teacher contact. Mrs. Mary F. Waldrop,
director of the nursery school and one of the teacheis, explored this
obseivation. She noted studies of young animals made elsewhere that
leported an offspring's need for maternal physical contact. Other re-
searchers showed that a child depiived of protective adult support will
increase his dependency behavior when he does have contact with adults.
These reports suggested to Mrs. Waldrop that a child has a need for
physical contact with a mother (or a mother substitute such as a female
teacher) and that if the need is not met, he increases his efforts to achieve
physical contact. A mother of many children closely spaced cannot always
be available to any one child; indeed, other investigators have found that
mothers of many children tend to become unconcerned about the
younger ones. With these clues, Mrs. Wakhop began an analysis to test the
hypothesis that preschool boys from large families would show more
contact-seeking behavior with a female teacher than boys of the same age
from smaller families. Further, by using some information about maternal
availability in the home, she hoped to get a lead on whether a huge family
was associated with low-maternal availability.

The first step was to quantify the size and structure of a child's family
in an index of family size and density (FSD) composed of four weighted
factors. These were number of children, timespan between the child and
his next younger sibling, timespan between the child and his next older
sibling, and average timespan between all births. The higher the FSD
index, the larger and more dense was the family into which the child was
bom.

An FSD index was computed for each of 44 boys and compared with
the rating given him on the behavior "child initiated contact with female
teacher." The postulated relation emerged. Moreover, a prediction based

229

on an alternative hypothesis was not suppoitcd. The behavior could no
be explained by a child's greater experience of social contacts because o
the number of siblings he had: There was no correlation between FSD an<
the rating "friendliness with peers."

A significant negative correlation was found between FSD and a ratiiij
made in the child's home of maternal initiation of contact. This findinj
raised the possibility that low-maternal availability in large, dense families
may be a factoi in the heightened contact-seeking behavior of childici
born into these families.

Are there other factors that might cause such a child to seek mow
contact with adults than do other children? Phrasing the question in the
investigators' teims, aie such children congenitally different from children
born into small, low-density families?

To answer this question, Mrs. Waldrop turned to the analysis of a
different set of data that had just become available. This was a sample of
boys and giils for whom both newborn and nursery school measures were
available. From the newborn data the investigator constructed a lethargy
scoie for each child. It was obtained by adding the infant's ranks on (1)
suck rate, the number of sucks counted during 8 minutes of non-nutritive
sucking; (2) total crying, the number of cries during 60 seconds following
removal of nipple; and (3) reaction time to nipple removal. The higher the
score, according to the theory, the more lethargic and ineffectual as a
feeder was the infant.

A significant positive correlation appeared between newborn lethargy
and preschool contact seeking. The higher the lethargy score, the gicatcr
the amount of contact seeking. Another correlation tended to substantiate
these findings. The higher the FSD index, the less was the child's ability to
defend himself against the aggression of peers. These results we it*
indicative of a distinct newborn behavior pattern in children born into
Jarge, dense families. The pattern is characterized as lethargy and is
predictive of heightened contact-seeking behavior at the age of 2Vz years.
By looking at the lesults of the two FSD studies together, it is possible
to say that children born into large, dense families will tend at the
preschool age to be dependent in this one sense. One component of the
tendency is biological, a congenital pattern of lethargy, and the other may
be social, low-maternal availability.

In one way these findings simply take a place in the large body of
evidence detailing the risk of deleterious effects on children born into
families with many siblings closely spaced. Other lesearchers have shown,
for example, that lates of prematurity, maldevelopment, and mental
deficiency increase with an increase in parity, especially after the third
and fourth child. Infant mortality was found to be highest when the birth
interval was less than 2 yeais; poor physical status and low IQ scores have
been associated with an increase in family size, One epidemiological study
discovered a strong correlation between paranoia and the presence of a
sibling less than 2 years older.

In another way, however, the findings add a critical new idea to the
study of family effects on child development. Most investigators have

230

assumed that the controlling factor was social class because large family
size has been associated repeatedly with low socioeconomic status. But
Mis. Waldiop found no relation in her sample between FSD and a widely
used index of the father's occupational-educational level. The results
indicate, then, that families on the higher socioeconomic levels, if they are
large and dense, are not exempt from the consequence of later children
being more lethargic and less effectual feeders than their siblings. It might
be well to note explicitly that these findings apply to large numbers of
subjects, but not necessarily to an individual child in any one family.

The immediate value of this part of the program lies in the success of its
unique methodology. The investigators demonstrated how to relate
piospective and restrospective data. Their experience leads them to believe
that mediator variables in addition to maternal leport and FSD— for
instance, social class-might be used to link newborn and preschool
behavior in longitudinal and cross-sectional samples.

Dr. Bell and his colleagues plan to complete the data analyses of the
newborn-preschool measures already in hand, and to make behavioral
measures of infants born to couples studied in early marriage by other
investigators of the Child Research Branch. They are also raising new
questions about early social development. Does social responsiveness vaiy
in infants as young as 3 or 4 months? What are the effects on mothers of
specific newborn characteristics? In particular, how docs a restless, hard-
to-quiet infant affect the patterning and quality of his mother's care and
peculiarly shape the initial opportunities of learning? Are there any
characteristics of sleep during the first months of life associated with the
congenital behavior syndrome? Having located some beginning and end
points, the researchers now would like to trace processes involved in the
transition.

The observations and ratings made in the nursery school also weie
employed to test another hypothesis, one relating social behavior to later
cognitive development. Although this research question was not part of
the original program goal, the necessary data were there so the investiga-
tors took the opportunity to use them in a new departure.

Social behavior and cognitive sfy/e.-Eailier studies of cognitive style
had found that a maternal attitude fostering independence and interaction
with the environment was associated with nonverbal skills; maternal
fostering of dependence tended to develop a child's verbal abilities. A
child*s style of dependency has been related to another aspect of
cognition, the way disparate objects are described and grouped. Two of
investigators. Dr. Frank A. Pedersen and Paul H. Wender (the latter is
working currently in the NIMH Laboratory of Clinical Psychology)
wanted to see if the social behavior measures of preschool children were
predictive of the children's cognitive functioning at age 6.

The investigators gave tests of cognition to a group of 6-year-old boys,
graduates of the first nursery school sample. One test was the Wechsler
Intelligence Scale for Children, which measures verbal and nonverbal
abilities. Another was a new sorting test designed to elicit a child's style of
categorizing. (It was devised by Dr. Irving Sigel, of the Merrill-Palmer

Institute.) The third was the Cluldien's Embedded Figures Test, wlw
measures the ability to extract a relevant item from an irrelevant sii
rounding field. It tests peiceptual aspects of cognition and is thought 1
index an individual's dependence on context in acts of perception an
judgment. In a seiies of studies that has extended over several years, I)
Herman A. Witkm, of the State Umveisity of New York, has shown th:
those who are good at picking out the figure embedded in the surroundin
field (field independent) tend to be socially autonomous and to take n
active, analytical view of their woild. Field-dependent subjects, coi
versely, are likely to be socially dependent and to see their world i
diffuse, global terms.

Each boy's peiformancc on these tasks was compared with his ratinp
on four different kinds of behavior reflecting dependence or autonomy
measured when he attended the nursery school; Contact seeking, al
tention seeking, orality (thumb sucking, mouthing), and sustained
directed activity. These were the findings:

• More contact-oriented, attention-seeking boys in the prcschoo
period showed lower nonverbal IQ's at age 6 years than did boys no
high on those preschool ratings. Those who had high ratings or
sustained, directed activity in the nursery school were high on ncm
verbal IQ's at the later age.

• Boys lated high on contact seeking and orality in the earlier study
were, at 6 years of age, high on the use of the categorization
response that reflects the global approach.

• Those high on sustained, directed activity in the nursery school wore
more likely than the others to use a superordinate concept in their
categorizations (recognizing, for example, that a saw and a pair of
pliers were both tools), a style the investigators suggest is a high
ojder of abstraction for 6-yeai-olds.

Results of the Children's Embedded Figures Test correlated significant-
ly only with the early measures of orality, but all associations were in (he
predicted direction.

The investigators concluded that preschool dependency behavior is
indeed predictive of certain kinds of cognitive functioning at age 6 years.
Theirs is the first leport of these relationships at so early an age.

On the basis of the data available to them, Drs. Pederscn and Wonder
weie not prepared to say just what social or biological factois might be
involved in the relation between social behavior and cognitive style. They
were able, however, to rule out the influence of socioeconomic class. In
their sample, family socioeconomic levels were not decisive relative to the
differences in cognition displayed by the boys, Perhaps maternal or
paternal attitudes are primarily mediating factois, as earlier studies
indicate; or perhaps, as the investigators suggest, congenital factois make
important contributions.

They made this suggestion after carrying the study one step further.
Dr. Wender took a close look at the family reports of four boys who were
extreme on social and cognitive measures, hoping to turn up some
nUusible leads on the parental characteristics differentiating the boys'

polar positions. Not only did definite patterns fail to emerge, on the
contiary, the boys came from remaikably diverse backgrounds. The
investigators know that their review of the families is not reason to
discount the influence of parental characteristics on the social-cognitive
developmental axis. For one thing, their family information was based
largely on contacts with mothers and did not include data for an analysis
of paternal effects. For another, current learning theory suggests the im-
portance of specific patterns and timing of rewards; this area of the chil-
dren's background was unknown to the investigatois. Still, the failure of
the data to yield clues implicating parental characteristics calls into
question hypotheses about the effect on cognition of maternal vaiiables
such as affection and neglect. An alternate hypothesis worth considering,
the investigators say, is that "congenital differences in children generate
the relationships between social behavior and cognitive functioning."

Intramural: NIMH

Date of Interviews Mar. 24, 1966

Refeiences

Accelciated Longitudinal Studies
Bell, R Q An experimental test of the accelerated longitudinal approach. Child

Developm., 1954,25,281-286.
Bell, R Q. Retrospective and prospective views of early personality development.

Men ill-Palmer Quart Behav. Developm., 1960,6,131-144.

Effects of Children on Parents

Bell, R. Q. The effect on the family of a limitation in coping ability in the child: a
research approach and a finding. Merrill-Palme} Quart, Behav. Developm., 1 964, 1 0,
129-142.

Bell, R. Q. The problem of direction of effects in studies of parents and children.
Paper presented at the Conference on Research Methodology in Parent-Child Inter-
action, held under auspices of the Department of Pediatrics, Upstate Medical
Center, State University of New York, Syracuse, N.Y., October 1964.

Newborns

Bell, R. Q. Some factors to be controlled in studies of the behavior of newborns.
Biologta Neonatorum, 1963, 5, 200-214.

Bell, R. Q. Level of arousal in breast-fed and bottle-fed human newborns. Psychosom,
Med., 1966,28, 177-180.

Bell, R. Q., & Costello, N. S. Three tests for sex differences in tactile sensitivity in the
newborn. Biologia Neonatomm, 1964,7, 335-347.

Bell, R. Q., & Darling, J. F. The prone head reaction in the human neonate; relation
with sex and tactile sensitivity. Child Developm. , 1965,36,943-949.

Weller, G. M., & Bell, R. Q. Basal skin conductance and neonatal state. Child De-
velopm., 1965,36,647-657.

Relation of Newborn and Preschool Behavior

Waldrop, M. F , & Bell, R. Q. Relation of preschool dependency behavior to family
size and density. Child Developm, , 1964,35, 1187-1195.

Waldrop, M. F., & Bell, R. Q, Effects of family size and density on newborn charac-
teristics. A mer J. Orthopsyclriat,, 1966,36,544-550,

233

Social Behavior and Cognitive Functioning

Pedersen, F. A., & Wender, P H. Early social correlates of cognitive functioning i
young children, Paper presented at the Society for Research in Child Developmeii
Minneapolis, March 1965.

234

Investigator

Fred L. Strodtbeck, Ph. D.

University of Chicago

Chicago, III.

Prepared by

Herbert Yahraes

In the give and take of family discussions an investigator supported by
the National Institute of Mental Health is seeking information on factors
that influence the development of a young person's intelligence and
personality.

Dr. Fred L. Strodtbeck and his research group at the University of
Chicago use what they call the technique of revealed differences to study
how meinbeis of a family interact. The father, the mother, and one or
more teenage children are given copies of a set of questions and asked to
respond individually. Some examples:

Should the parents of a 13-year-old girl have any say about who is
invited to a party at the girl's house, or should the decision be
entirely up to the girl?

A high school junior likes math and would like to take a geometry
course but thinks she won't do well. Should she take the course, or
should she take a course that she would be more certain of doing
well in?

Having a nice personality and being well liked are more essential
for success than any particular type of skill. Do you agree?
The families are then asked to talk over those problems— always
numerous—on which differences of opinion have been revealed and to try
to agree on an answer. At this point the experimenter leaves the room, but
the discussions are recorded and later analyzed. Some of the cases on
which there was disagreement are discussed by three or more members of
the family, some by only two.

Members of the family, the grantee explains, have sized up one another
over the years and now have certain expectations of how each will behave.
Because of these expectations, a family discussion is likely to bring out
typical behavior. Consequently, the discussion, or interaction, is analogous
to a sample of blood: Analysis provides clues in the one case as to how the
body is functioning, in the other, the family.

The grantee believes that the study of family interactions will get at
some of the basic reasons for the ways, healthy or unhealthy, that
children adjust to life and go on to influence the adjustment of others, in
particular their own children. Findings will be checked, it is planned,
when the children in the families studied have become adults,

235

The judges who listen to the tape recordings of the discussions give each
participant a power score. On a given question, for example, if only the
father gave a certain response but then, during the discussion, won the
others to his way of thinking, he gets four points and the others none, If,
instead, the mother and child won the father over to their side, they each
get two points and the fathei none.

The judges also score the participants on such matters as the conviction
with which a position was held; the contribution made by each pctson
toward (a) reaching the decision, and (b) preventing and relieving strain
during the discussion; self-assurance; and the warmth and understanding
shown to each of the other persons.

Among other points considered during the analysis:

• The family's health, or the quality of family relationships, as infencd
from how well the family went about reaching consensus on a given
problem.

• The family's success in dispelling any tension so that members were
free to move on to the next question with a clean slate.

• The masculinity of the father and the femininity of the mother.

High I Q vs. Average I Q Teenagers

In a study neanng completion, the University of Chicago group has
used the revealed-difference technique to study families having an adoles-
cent child between 13 and 15 with very high intellectual ability. Five of
the children were girls> five were boys. They had IQ's higher than 1 60.

The controls weie classmates of average ability. Their IQ's ranged from
110 to 120. Their fathers had been matched by religion and socio-
economic status with the fathers of the high IQ group.

One of Dr, Strodtbeck's giaduate students had theorized, in suggesting
the study, that the children with extremely good intellects would tend to
be more isolated and withdrawn than the others and would thus make for
poorer family relations. But the opposite has proved true: Family health
was considerably better and the relations much warmer in the families
with a brilliant child. This was not owing solely to the direct contribution
of the brilliant child, for when the parents were discussing a problem
alone, their warmth and helpfulness were even greater than when the child
was present. In the other families, when the father and mother were alone,
their relations giew rougher.

The high IQ family, Dr. Strodtbeck reports, has much greater problem-
solving abilities. When the parents and the teenager talk with one another,
they show greater precision in identifying their differences, and greater
clarity in reconciling them. Presumably this is explained in part by the
high intelligence of the youngster and perhaps of his parents as well. At
the same time, Dr. Strodtbeck believes that growth of intellectual
competence is fostered by the kind of warm family relations he finds
displayed by the high IQ families in this study.

236

"We think that if a person's high intellectual capacity is to be realized,"
the grantee says, "he will have to be relatively free of internal concerns, so
that he can continue to take in information about the world and to
stimulate other people to talk in ways that enable him to learn from them.
If a child's family life fulfills his needs, his energy is released for learning
more about the outside world. But if he is tied up in neurotic conflicts in
his family, he will not have the zest for taking in information from
outside."

Recent studies have shown, the grantee notes, that the IQ's of children
hospitalized as schizophrenics had started sliding down before the trouble
became apparent. "When you're working on an extremely tough
problem," he explains, "you just don't have as much energy for encoding
information from the world around you."

Using the revealed-difference technique, another investigator has
compared parents of children hospitalized for schizophrenia with parents
of children hospitalized for tuberculosis. He found, Dr. Strodtbeck
reports, that the husbands and wives of the schizophrenic children failed
to reconcile their views-after originally giving different answers-five or
six times as often as the parents of tubercular children. Dr. Strodtbeck
believes that studies underway or planned by his research team will help
uncover some of the elements in family relations responsible for such
differences.

In addition to the teenagers and their parents in the 1Q study, Dr.
Strodtbeck's long-term research program embraces half a dozen groups,
among them:

1 . A number of young men who, with their parents, were first studied
by Dr. Strodtbeck in 1952, in New Haven, when they were from 13 to 15
years old. The fiist study showed that the higher a father's power score,
the lower the achievement values expressed by his son. The followup
study finds that this effect apparently has a lasting influence. Young men
whose fathers scored high in power tend to have chosen, or drifted into,
occupations lower than their fathers'. On the other hand, the sons of
fathers who were low in power tend to have outpaced them. Though other
findings await analysis of the data, the grantee believes that one of them
will verify and extend a finding of the IQ study. The New Haven study
will show, he believes, that warmth in family interactions relates not only
to functioning at higher IQ levels in early adolescence but also to higher
accomplishment as an adult.

2. A group of "pathological families"— families with children in their
teens who had needed treatment for a behavior problem from 3 to 1 0
years before the study.

3. A number of two-child families in which the parents are relatively
young, about 40 years old when their first child is a high school junior,
and a number in which the parents are 10 years older. This study is
underway. Among the findings to date: The older parents permit their
children to exert more influence than the younger, as measured by the
power scores achieved during the family discussions.

237

The Acquirement of Masculinity and Femininity

The investigators are particularly interested in the processes by whi
child takes on masculine or feminine attitudes, since the way a pe:
identifies his sex role can be an important factor in his emotional he
throughout life.

Dr. Strodtbeck explains that masculinity is taken to imply I
headedness under pressure, the ability to strike out effectively win
crisis is at hand, an unwillingness to be involved in trivial concerns,
ability to act considerately and helpfully without the fear of being •
sidered effeminate, and curiosity about what's going on.

On the other hand, femininity is taken to include a greater inleiu
persons than in things, concern for the quality of the relations betv
persons, the management of a paired rclationship-as that betu
husband and wife-so that the members are complementary nHhurl
competitive, patience in matters involving the welfare of others,
concern and support for members of one's group in time of crisis.

Among the families in the high versus low IQ study, the grantee rept
brilliant children displayed to a relatively high degree characteristics of
role commonly ascribed to the opposite sex. That is, a brilliant boy
judged to be more like his mother than an average boy, but at (he s
time not any the less like his father. A girl with a high IQ was judged l<
more like her father than an average girl, but at the same time nol any
less like her mother.

"We believe," says Dr. Strodtbeck, "that the greater warmth of the 1
IQ families makes it easier for a child to identify with the parent of
opposite sex and to do so without becoming anxious. Thus a child
behave more flexibly."

As evidence for such identification, the study has determined
correlation between the power scores of each of the parents and the lo
rating the child has given to himself. The parent whoso power is corrcl;
most positively with the child's own sense of potency, the study ussui
will be the parent with whom the child most closely identifies.

The findings;

• Boys with average IQ-siighl positive correlation with father's pin
negative with mother's.

• Boys with high IQ-positivo correlation with mother's pin
negative with father's.

• Gills with avciagc IQ-slighl positive correlation with moth
power, negative with father's.

• Girls with high IQ-positive correlation with father's power, ncgu
'lh mother's.

1J Oirodlbeck emphasizes that these findings are based on a stud]
~ses. But they make sense, he says, if one of the component
*encc is "freedom from phobic avoidance of demon Is eha
ipposite sex culture."

is investigator sees it, the development of high inlclligi!
ind at the same time contributes to, family warmth

understanding. In such an atmosphere the gifted child is freer both to
develop his capabilities and to adopt a broader set of worthwhile attitudes
and qualities even though some of these are generally considered charac-
teristic of the opposite sex.

Research Grant: MH 5572

239

Investigator.

Daniel Offer, M.D.

Michaei Reese Hospital
Chicago, III,

Prepared by

Herbert Yahraes

Psychiatrists try to help sick people become well, but what is welt
What do we mean by normal mental Health?

Psychiatry has too long neglected this question, says Dr. Roy R
Grinker, Sr., an eminent psychiatrist and director of the Institute fo
Psychosomatic and Psychiatric Research and Training, Michael Recsi
Hospital and Medical Center, Chicago. Hence, as one part of a broac
lesearch program, membeis of the Institute's staff have been trying to fim
some of the answers. Their work is supported by the National Institute ol
Mental Health,

This report presents highlights of three studies: First, a continuing
investigation of normal high school students; the work of a team headed
by Dr. Daniel Offer, the institute's assistant director; second, a study,
nearing completion, by Dr. David Marcus and several associates, including
Dr, Offer, of what seems to be an important difference between families
that do not have a disturbed adolescent child and families that do; third, a
completed study by Dr, Griuker of a group of normal colleges students.

f. Psychiatry Views the Normal Adolescent

After some years of daily contact with adolescents who were disturbed
or delinquent, or both, Dr. Offer decided to take a good look at the other
kind, the normal ones-their backgrounds, personalities* viewpoints,
worries, and behavior.

In adolescents, he points out, it has been unusually difficult to dis-
tinguish health from illness, normal turmoil from pathological process.
Even in the case of psychotherapists, ideas about the characteristics
making for normality in teenagers have been generally based on memories
of a person's own adolescence and observations of his own children. With
the understanding to be gained by a systematic study, Dr. Offer reasoned,
psychotherapists would have a more valid baseline for judging disturbed
teenagers, and families, schools and society in general would be better
able to handle problems presented by teenagers and to recognize, prevent,
™A ""*"* ""*u abnormal behavior.

Dr. Offer is now halfway through an intensive 3-year study of 84
normal adolescents— boys from middle-class families in the Chicago
metropolitan area selected fiom two public high schools during their
freshman year.

Among the findings to date;

1 . The normal adolescent, like the disturbed one, has feelings of shame,
guilt, depression, or anxiety. But he is less afraid to look at himself and to
admit his feelings. Psychiatrists would be happy, Dr. Offer remarks, if
patients even at the end of therapy were as aware of their problems as
these normal boys are of theirs.

2. Many of these atomic-age boys are worried about the same things
that boys have worried about for generations— including sex, religion, and
money. Only a few are worried about the state of the world. A teenager's
three most difficult problems, these boys say, are to do as well educa-
tionally and vocationally as his family expects, to contiol his impulses,
and to get along with other people.

3. The boys have a conservative sexual code and, through the
sophomore year at least, they behave conservatively. Many of them day-
dream about a specific girl but do not readily admit it. Five percent go
steady; 35 percent group-date only; 35 percent do not date at all. Typi-
cally the boys are concerned with how to behave when out with a girl.
Most are interested in sports— as a means of displaying masculinity and of
sublimating, Dr. Offer believes, both aggressive and sexual impulses.

4. About 20 percent smoke. A few diink.

5. Many of these normal boys— generally when they were only 12 or
13— have performed minor delinquent acts and have associated with delin-
quents. Twenty-five peicent of them, in fact, have been involved with the
police over such incidents as stealing from a drugstore, fighting, throwing
bottles on highways, or overtmning garbage cans. After a boy has been in
trouble once or twice, though, he seems to have learned his lesson, He
docs not make delinquency a pattern. Nevertheless he sympathizes with
juvenile delinquents and ascribes their troubles entirely to their parents
and to society.

6. These normal teenagers are not inclined to rebel against either then
parents or their parents' generation. They see clearly what values their
parents hold, and they tend to hold the same ones themselves. When
adolescents and parents disagree, it is on such matters as the use of the car
and the time to come home at night. The boys find their fathers reliable,
their mothers understanding. They feel closer to their mothers and can
more easily discuss emotional problems with them.

7. Members of the group express definite ideas about what they are
going to do when they have finished school. But they tend to change these
ideas as time goes by.

For the most part these findings stem from four 40-minute psychiatric
interviews with each boy, spread over a period of a year and a half. Four
more such interviews are planned, as are less intensive interviews with the
parents. The complete description of these normal adolescents will also
take into account teacher ratings, school performance records, and the

241

results of projective testing as other means of getting information a bo
hopes, fears, and anxieties.

How does one pick a normal adolescent? Dr. Offer and his associat
did it by going to two high schools m the fall of 1 962, one in a suburb ju
north of Chicago and the other in a suburb just south, and giving sever
hundred fieshmen boys two tests. One was the Self-image Questionnaii
for Adolescent Boys, developed by Dr. Offer and an associate, Dr. Mclvi
Sabshin, now head of the Department of Psychiatry of the University c
Illinois College of Medicine; the other, the Bell Adjustment Inventor;
Each comprised a number of scales, or subtests, and each scale \va
intended to measure one aspect of how the boy regarded himself or hi
world. Taken all together, the giantee believes, they provided a goo.
picture of a boy's ability-in terms of his emotional adjustment-to mee
his problems,

Of the boys tested, Dr. Offer selected the 114-about a third of (In
total— whose scores on all the subtests had been closest to the average
Thus he eliminated the boys who scored highest (and were, presumably
extremely well adjusted) and those who scored lowest. He also eliminatec
those who scored much higher than average in some areas and much lowei
than average in others. So in this study the normal adolescent is what Dr.
Offer calls the modal, or average adolescent.

After making his selections, the investigator checked with the school
authorities and found that 3 of the boys were serious behavior problems
(in a randomly chosen group of the same size, 15 would have been). These
were dropped from the study. A few boys declined to participate and a
number moved.

Dr. Offer emphasizes that most of the boys being studied come from
families at one or another level of the middle class. About 10 percent
come from stable, working-class families, as would be expected ftom the
proportion of such families in the population represented, and about 7
percent are Negroes, as again would be expected. Presumably the study's
findings would hold true for middle-class communities in any metro-
politan area.

None of these well-adjusted boys has lost a parent because of divorce,
only three because of separation, only two because of death; in sum, only
6 percent come from disrupted homes. This compares with 9 percent of
the general population of teenagers (aged 18) in the two communities
studied and with 45 percent of the delinquent adolescents studied by
another investigatoi.

More than half of the selected group are either the oldest or the only
child in the family— a striking statistic, says Dr. Offer, but one that may
only reflect the tendency of families to move to the suburbs when the
first child is ready for school.

It developed also that the group had its share of honor students-12
per cent —during the freshman year, but no failures.

When Dr. Offer was organizing his project, some of his associates
doubted that it could succeed. Normal teenagers, they argued, would
never show up for appointments with a psychiatrist -they'd feel disgraced.

242

As a matter of fact, more than half of the group did rniss theii appoint-
ments during the early months and had to be scheduled again. This
proportion has now dropped to one-fifth, which the investigatoi consideis
good, because the interviews are held after school hours and frequently
involve two round trips for the parent who does the driving.

Teenagers are extremely egotistical, Dr. Offer notes. They participate in
the project because someone is interested in learning what they -and not
adults-think about teenageis and the world. Teenagers are also extremely
altruistic. They participate because they have been told that in the long
run the project will help other teenagers.

What does he think of these normal youngsteis? "Oh my!" the psy-
chiatrist says, "They're tremendous!"

IJ. Teen-Agers in Trouble: A Communications Breakdown?

Parents requesting psychiatric help for an adolescent child, psychiatrists
at Michael Reese observed over a 3-year period, complain most frequently
about'

1. Delinquent behavior, ranging from assault and major theft to
isolated incidences of vandalism in school.

2. Difficulty in making and keeping friends.

3. Inability to adjust to school situations.

4. Inability on the part of the parents to "manage" the adolescent.
The adolescents, on the other hand, most frequently complain that their

parents do not understand them; hence their problems. Their second most
frequent complaint is that they do not understand their parents.

The investigators asked themselves: In a family where the teenagers are
apparently normal, have the parents and youngsters been understanding
each other better than in a family where a teenager is disturbed?

Twenty middle-class families were studied, each intact and each
including at least two adolescents ranging in age from 14 to 19 and in
good physical health. Half the families were classified as normal, or non-
disturbed: no member had, or had had, any obvious emotional trouble.
The other half were classifed as disturbed: one of the adolescents had
been hospitalized for psychiatric reasons, the diagnosis in five cases was
schizophrenia and in five cases character disorders that had led to such
actions as car stealing, assaulting parents, and, in the case of one girl,
becoming pregnant.

Then the Q-sort technique was used to obtain from each adolescent
both a description of himself and an account of how he thought his
mother would describe him. Similarly, each mother described her son
both as she would have liked him to be and as the thought he would
describe himself.

Analysis of the data leads to these principal findings:

1 . In normal families, mothers and children were in good communica-
tion. An adolescent understood his mother's expectations for him and was
able in turn to convey his own viewpoints to his mother.

243

2. In disturbed families, the patient and the mother were in poor ecu
munication. The lines between the other disturbed adolescent and tl
mother were open but not to the same extent as in undisturbed fa milk

In general, it has been the experience at Michael Reese that the moth
of a distuibed child explains that the child is rebellious toward h
ideas-an explanation that the Michael Reese psychiatrists are inclined 1
disbelieve. In describing behavior as rebellious, they suggest, perhaps
mother is excessively suspicious or distrustful. Or perhaps she has bcc
making inappropriate demands on the adolescent. Or perhaps, as in til
study, the lines of communications have broken down : the adolescent h;
not known where the paient stood and the parent has not known wliei
the adolescent stood.

Lack of communication, the investigators repoit, may apply pailiculai
ly to adolescents who are delinquent but not psychotic. In the prcsen
study, in any event, such teenagers said they had been tiying through Ihoi
behavior to foice open the communications at home, particularly vvitl
their mothers.

III. Normality in Young College Men

Several years ago, as part of a research pioject on how the botlj
lesponds when emotions are repressed, Dr. Grinker had occasion to stud)
65 normal young men. They were freshmen at George Williams College in
Chicago where the basic goal is "to provide professional education foi
Christian leadership, primarily for Young Men's Christian Associations,"
but where "men and women of all faiths and races who seek to prepaid
themselves in a Christian atmosphere. . ." are welcomed.

About half the group had been selected as normal on the basis of
various personality tests; the others were judged normal on the basis of u
psychiatric interview and answeis to an extensive questionnaire.

After interviewing the selected students, Dr. Grinker's findings were
startling. "Here was a type of young man I had not met before in my role
of psychiatrist and rarely in my personal life," he repoited. "On the
surface they were free from psychotic, neurotic, or disabling personality
tiaits. It seemed that I had encountered some mentally 'healthy' men who
presented a unique opportunity for study." As the investigation continued
he came to feel that the entire student body enjoyed unusual health.

Typically the healthy young men in the study came from small- or
medium-sized Midwestern cities. Their fathers were semiskilled or while-
collar workeis eauiing a little more than $5,000. The parents had been
loving but strict. Family quarrels, if any, had been generally over money.
The boys had gone to work early. They had loved sports. They had had
rigorous training in religion.

Factors emphasized by Dr. Grinker as having contributed to the boy's
mental health include;

1, Sound physical health from birth onward.

2. Average rather than high intelligence.

244

3. Warm relationship with both parents.

4. Parental agreement about bringing up children-including the setting
of definite and understood limitations on behavior.

5 Reasonable and consistent punishment.
6. Sound early religious tiaining.
7 Part-time jobs when young.

8. Strong identification with the father.

9. A viewpoint (picked up at home, at church, in boys' clubs, at the
Y.M.C.A.) that sees the world as calling for action, not introspection: a
person does something about problems.

10. Ideals centered on doing the job well, doing good, being liked,
achieving contentment and sociability, and succeeding at what one
chooses to do nithei than striving foi either social or economic prestige.

The investigator divided his healthy young men into three groups—the
very well adjusted, the fairly well adjusted, and the marginally adjusted—
and then studied differences in their background and behavior. In general,
the parents of the very well adjusted group had more often been in agree-
ment about then children's unbiinging, had more openly expressed their
affection for each other, and had shown a less rigid concern over such
problems as dating, smoking, and chinking. The motheis had been waimcr,
closei, and more relaxed, and more of the fathers had given all the love
the students recalled having wanted. The very well adjusted individuals
had done better in school and had been more active socially during adoles-
cence. They were more specific than the others about what they wanted
to do-go into Y.M.C./1 work, most of them -after college. They were less
frequently anxious, embarrassed, or depressed. When they weie angered
they tended to speak out- the marginally adjusted ones, to keep quiet.

Even members of the very well adjusted group have had and do have
problems. Dr. Grinker reports, "Like anyone living, they have had
conflicts, established defenses, and have had to sacrifice potential assets in
tlie process of adjustment." He notes in particular a nairowed range of
interests and a tendency toward some anxiety about failing. But in general
they and the others who were studied work and play well, cope
realistically with experiences that rouse them emotionally, and have had
warm, human relationships with parents, teachers, friends, and girls. They
also feel good, and have hopes for the future— among them, "doing the
best I can."

When he describes these young men to social and professional groups,
Dr. Grinker is often told, "Those boys are sick; they have no ambition,"
He disagrees because he thinks "doing the best I can" is an ambition.

Intense commitment to change, the investigator says, may in itself be
one of the elements in neurosis-building. Neither the men in this study nor
their parents have shown much interest in moving fast; they go ahead at a
pace that does not overstrain. They are not creative, not explorative; to
many persons they might appear dull. But Dr. Grinker believes that they
and people like them give our society "a solid steady core of stability."

The investigator observes that what he considers mental health or
normality in these young men is of one type and that research among

245

many kinds of populations is necessary to delineate other types and find
what they have in common.

Research Granf MH4870

References-

Grmker, Roy R., Sr. (with the collaboration of Roy R. Grinker, Ji., and Jolm

Timberlake). "Mentally healthy" young males (homochtes). Archives of General

Psychiatry, 6(6), 1962.
Offer, Daniel and Sabshin, Melvm. The psychiatiist and the normal adolesctiiit.

Archives of General Psychiatry, 9 (5), 1963.
Offer, Daniel, Sabshin, Melvin, and Marcus, David. Clinical evaluation of uorm-il

adolescents, Presented to American Psychiatric Association, May 1964. A men fan

Journal of Psychiatry, 121(9), 1965,

246

Investigator:

Albert Bandura, Ph. D.

Stanford University
Palo Alto, Calif.

Prepared by,

Clarissa Wittenberg

One of the fundamental means by which human behavior is acquired
and modified involves vicarious learning. Both children and adults learn
modes and standards to a great extent by observing the behavior of others.
A great deal of attention has been focused in this country upon the
television industry as this fact has become more and moie apparent. In
addition to television, motion pictures, books, and other reading material
are visual media effective in producing vicarious, as opposed to diiectly
experienced, learning. The variety represented by television alone demon-
strates the difficulty of making simple judgments about the effect on the
viewer. The variations in viewers and the surrounding ciicumslanccs in
which they see the television again points out the complexity. The most
insistent public focus on TV has been with regard to violence and its part
in the violence observed in our society. An investigation of violence per se
seems less likely to yield information on the underlying process of
influence than is the study of the visual media with regard to its impact on
learning in general. Dr. Albert Bandura has spent ovci ten years studying
the effects of such media upon children and adults in a variety of sit-
uations. His work yields the definite finding that such media are powerful
methods of teaching and important sources of influence, but that many
lessons may be taught and they may not be the obvious ones. It becomes
obvious that the environment in which the media is viewed is very impor-
tant.

This research has demonstrated the stimulation of aggressive behavior
by the viewing of aggressive acts on a screen. Variables have then been
explored with attention given to the characteristics of the model and the
attributes of the observer, and the consequences accompanying the
demonstrated patterns of behavior, The difference between acquisition
and spontaneous performance of aggressive acts has been examined.
Further research has been done with the use of the visual media as
therapeutic tools to relieve longstanding and serious phobias and to
improve the social adjustment of withdrawn children. The common thread
throughout the research of Doctor Bandura and his associates is the
concept of observational learning and the effectivity of the modeling
process.

247

Doctor Banduia states that research bearing on modeling processes
demonstrates that, unlike the relatively slow process of trial -and -err or
learning, patterns of behavioi are lapidly acquired observationally in large
segments or in their enthety. The extent of this form of learning can be
seen in children's play when they reproduce parental behavior, including
the appropriate mannerisms, voice inflections, and attitudes. This process
in a more general way is leferred to as "identification."

Doctor Bandura became involved in this subject in 1958 when lie
conducted with Richard Walters, research on the family conditions which
gave rise to extreme aggression in children. The focus was the adolescent
from the "good home" who became antisocial and delinquent. Although a
great deal of research had been done about the effects of poor and adveise
family and social conditions, not much had been done to explain the
reasons why affluent young men were becoming delinquent and antisocial.
Families were selected who looked well integrated and socially we]l
adjusted, but whose children were being followed by the probation
department in the San Francisco area. A matched control group was also
interviewed. Two cential factors emerged. Many parents of the delinquent
boys weie models for antisocial attitudes and aggressive behavior despite
their smooth social exterior. A second pattern was that the parents,
especially the fatheis, often would not permit aggression towards
themselves but would encourage and rewaid their son in fights outside tlie
family 01 defend the boy's right to "raise-." In sharp contrast, the non-
delinquent boys were encouraged more to defend themselves with their
ideas or in the nonphysical spheres. Aggression of a physical type was
consistently discouraged through nonpunitive means in these families.
These boys, who were not on probation, had been taught through
example and precept a different way of solving their interpersonal
problems than the aggressive boys.

Another incident occurred that dramatized the influence of demon-
stration or modeling. In 1961 the San Francisco Chronicle reported that a
boy had been seriously knifed during a reenactment of switchblade
fight the boys had seen the previous evening on a televised rerun of the
James Dean movie, Rebel Without a Cause. This was a vivid illustration of
the imitation of film stimulation and stirred considerable speculation. The
form of the aggression had been so clearly shaped by the film that it gnvc
rise to the idea that aggression viewed through pictorial media may be
influential in shaping the form of aggression when the person is in a
provocative situation. The importance of the visual media in stimulation
or instigating aggression also became a focus on this research.

Transmission of Aggression

One set of experiments was designed in 1961 to determine the extent
to which aggression could be transmitted to children through exposure to
aggressive adult models. In this early experiment children observed an
adult who exhibited relatively unusual forms of physical and verbal

248

aggression towards a large inflated plastic "Bobo" doll. A second group
watched a very subdued and inhibited model. The control group saw no
model at all. Half the children in each experimental condition saw models
of the same sex and half observed the opposite sex. Later the children
were mildly frustrated by having toys restiicted for their use and then
their behavior was recorded in a new situation where they could behave
either aggressively or nonaggressively. The results showed that exposure
to aggressive models heightened the children's aggressive responses to sub-
sequent frustration in new settings in which the model was absent.

In 1963 this investigation was extended by Banduia, Ross and Ross.
The effects of real life models and filmed models were compaied. The
children in this project, as well as in many of these studies, were diawn
from the Stanford University Nursery School. In this study they ranged
from 35 to 69 months of age, with a mean age of 52 months. There were
48 boys and 48 girls who took part. Two adults, a male and a female,
served as models. A female experimenter conducted the study with all of
the children. A "Bobo" doll was again used as the subject of aggression.

One group of children observed real models in the room with them,
behaving aggressively toward the doll. A second gioup saw a film of the
same models performing the same acts. A third group saw a cartoon of an
aggressive figure. The control group was not exposed to any of these
stimulations. Again half of the children saw models of theii same sex.
These children had previously been rated in terms of their normal ag-
gressive behavior by their nuiscry school teachers and they were matched
to the control group on this basis.

The children who saw the real life aggression weie asked into a room
and invited to join a game. The child was shown a table with a variety of
activities. The model also worked at a small table doing tinker toys. Then
the model turned to the "Bobo" doll and kicked it about the room, sat on
it and punched it in the head, pummeled it, and hit it with a mallet among
other things. The sequence of acts was repeated three times and was
accompanied by verbally aggressive comments such as "Sock him in the
nose * * V "hit him down '* * *," "Pow." These acts were not those
usually performed spontaneously by children with a "Bobo" doll.
Although the doll is designed to be hit, the usual play involves poking it or
trying to knock it over rather than "beating it up." The main inteicst of
the research was not whether the children hit the doll but whether they
adopted the unusual modes of aggression demonstrated by the adults.

The movie sequence was identical except it was presented on film.

The cartoon sequence was presented in a TV console and the ex-
perimenter introduced it as a color TV cartoon program. A film was then
presented of a female model costumed as a cat performing against a
brightly colored and fantastic setting. A title and a picture of a stage
introduced the production. The cat figure then performed the same acts
with the "Bobo" doll. Music was played accompanying the film.

Fallowing the exposure, the children were tested for the amount of
imitative and nonimitative aggression in a different experimental setting
without the presence of models.

249

In ordei to clearly differentiate the exposure and test situatio
subjects were tested for the amount of imitative learning in a diffen
experimental room which was set off fiom the main nursery scln
biiilding.

The childien, both control and experimental groups, were mile
frustrated before they were biought to the test room by having t
children begin to play with attractive toys and then being told that th
were the experimenter's best toys and that she was saving them. T
childien weie then taken to the testing room and the experimenter stay
with them, but did papeiwork off to one side.

The testing room contained a variety of toys, some of which could
used for imitative or nonimitative aggressive acts, and others which te
to elicit predominantly nonaggressive forms of behavior. The aggressive
oiiented toys included a "Bobo" doll, a mallet and pegboaid, dart giti
etc. The othets included a tea set, ciayons and paper, a ball, two dolls, e
Play mateiial was also arranged in such a way as to eliminate any var
tions in behavior due to mere placement.

The subject spent 20 minutes in the testing room during which time
was observed through a one-way mirror and his behavior was rated. Tl
20-minute session was divided into 5-second intervals, and so a subject w
scored 240 times. The judges reached high levels of reliability in thi
scoring. The following response measures were obtained; imitati
aggression; partially imitative responses; mallet aggression; sitting <
"Bobo" doll; nonimitative aggression; aggressive gunplay.

Results

Exposure to aggressive models increased the probability that subjec
will respond aggressively when instigated on later occasions. Furth
analysis shows that subjects who viewed the real life models do not diff
from those who viewed the filmed or TV models in total aggressivene;
but all thiee experimental groups expressed significantly more no
imitative aggressive behavior than the control subjects.

The exposure to aggressive models is a highly effective method f
shaping subjects' aggressive responses. Experimental subjects displayed
high level of imitative physical and verbally aggressive acts whereas contr
subjects rarely behaved in these novel aggressive ways. Thus exposure
aggressive models not only reduced children's inhibitions over aggressi1
behavior that they had previously learned, but also taught them new wa;
of aggressing,

A prediction had been made that imitation is positively related to tl
reality cues of the model and this was only partially supported. Whi
subjects who observed the real-life aggressive models exhibit*
significantly more imitative aggression than subjects who viewed tl
cartoon model, the live and film, and the film and cartoon models i
creased nonimitative aggression in the children to the same degree. Da
indicated that of the three experimental conditions, exposure to fnimai

on film portraying aggression was the most influential in eliciting and
shaping aggressive behavior.

The Effect of the Sex of the Model

The boys exhibited more total aggression than girls, more imitative
aggression, more aggressive gunplay and more nonimitative aggressive
behavior. The girls, for instance, were more likely to sit on a "Bobo" doll
and lefrain from punching it.

Subjects who were exposed to male models as compared to female
models expressed significantly more aggressive gunplay. The most marked
cHffeiences in aggressive gunplay were, however, found between the girls
who had been exposed to the female model and males who had observed
the male model. The girls who saw the female model tended to reproduce
more partially imitative acts than the boys who saw the male model and
were more likely to repioduce the large: actions.

The sex of the child and the sex of the model have an effect upon the
degree of influence of the models, and this influence is determined in part
by the sex appropriateness of the model's behavior.

Another section of this experiment dealt with the possibility of
cathartic action upon the viewing of aggressive film material. The subjects
were first frustrated and then provided with an opportunity to view an
aggressive film following which their overt or fantasied aggression was
measured. Many parents and educators encourage hyperaggressive childien
to participate in aggressive recreational activities, to view highly aggressive
TV programs, and to be aggressive in therapeutic settings in order to
"discharge" their aggression. Bandura's work and the work of other in-
vestigators demonstrate that the provision of aggressive models and the
inadvertent reinforcement of aggression which occurs in these situations
act to encourage aggiessive tendencies rather than dissipating them. On
the other hand, providing aggressive children with examples of alternative,
constructive ways of coping with frustration can be very successful in
helping them modify their destructive behavior patterns. Already
frustrated children show more aggressive behavior after viewing live or
filmed aggression than do frustrated children who are not shown a film,
The filmed aggression does not fill theii need, nor does it diminish their
aggressive tendencies.

The view that social learning of aggression through exposure to ag-
gressive film content is adopted by only deviant children, also finds little
support in Doctor Bandura's research. The children who participated in
this experiment were all considered normal; yet 88 percent of the subjects
in the "Real-Life" and in the "Human Film" condition, and 79 percent of
those in the "Cartoon" condition, exhibited varying degrees of imitative
aggressive behavior. In assessing the possible influence of televised stimula-
tion on behavior, it is important to distinguish between learning and overt
performance. Although children may learn whole patterns of behavior by

251

watching TV, they do not ordinarily perform indiscriminately the t»
havior of televised characters, even those who they regard as highly attnn
tive. The responses of the parents appear to be very important in discou
aging overt imitation. The investigators stress that the behavior is learnct
however, even if parental disapproval inhibits it being performed and
may be elicited on future occasions. Indeed, recent research demonstn
tions show that children will not exhibit disapproved aggression in th
presence of the prohibitive adult, but that they are inclined to perfori
such behavior when the disapproving adult is absent.

Children who had been previously rated as more aggressive than th
others by their teachers did not differ in their aggressive reactions in th
experimental setting.

The investigatois have formulated a theory of social learning of aj
gression (Bandura and Walters, 1959) that would suggest that most of th
responses utilized to hurt or injure others, such as kicking or hitting, wcr
learned as exploratory asocial acts. For instance, the infant who learns I
control his legs and kick is exercising and exploring his own movement!
but not being aggressive. When frustrated, however, he may call on thi
response as one that can express his intense feelings, and then the kickin
becomes involved in social interaction. On the basis of this theory, i
would be predicted that the aggressive responses acquhed imitntivcly
while not necessarily for aggressive goals in the experimental scttinj
would be utilized to serve such purposes in other social settings. It wonli
also be predicted that children in the experimental settings would use thi
behavior aggressively more frequently than children in the control groups

These previously mentioned experiments were primarily designed ti
measure the extent to which children learn by observing the aggressive
action of adults. A second major question is whether exposure to ag
gressive models influences the harshness with which people treat otheis
To study interpersonal expiession and aggressive behavior requires stndic
in which people are provided with opportunities to behave punitive])
toward another person after viewing aggressive or nonaggrcssive models, /
study conducted in Doctor Bandura *s laboratory by Donald Hnrtmat
reveals that aggressive models not only foster learning of aggrossivi
behavior, but can also increase interpersonal aggression.

The catharsis hypothesis has generally assumed that viewing aggrcssior
reduces aggressive tendencies in observers if they experience anger at tlu
time of exposure, but that it may increase aggression in nonangcrctl
viewers. To test this idea, Hartman conducted an experiment thai
proceeded in the following manner. One group of delinquent adolescent;
underwent an anger-arousing experience, while a second group had an
essentially neutral experience. The boys then observed one of three
movies. In the control film two boys engage in an active but cooperative
basketball game, whereas in the other two films the boys get into an
argument that develops into a fist fight. The instrumental-aggression film
focuses on the behavior of the attacker, including his angry facia! ex-
pressions, flying fists, foot thrusts, and hostile remarks. The pain-cues film

252

focuses almost exclusively on the reactions of the victim as he is
pummeled and kicked by his opponent.

Major obstacles arise in the study of interpeisonal aggression because a
socially significant measure would involve injurious behavior which cannot
be used for humane and ethical reasons. This major obstacle has been
overcome by several researches by creating a situation in which one
person can administer shocks of differing intensities and durations of his
own choosing to another person. However, the electrodes are not con-
nected to the victim so that he in fact docs not suffer any pain. After
viewing the films, the boys in Hartman's study were provided with oppor-
tunities to shock a victim. The intensity and the duration of the shocks
administered were recorded.

Boys who had observed either the aggressive acts or the pain-cues films
selected significantly highei shock levels, both under angered and
nonangered conditions than boys who watched the control film.
Moreover, angered viewers behaved moie punitively than nonangered
viewers following exposure to the aggressive films, a finding that is
directly counter to the prediction of the usual catharsis hypothesis. Boys
behaved most aggressively when they were angered and witnessed another
person beaten severely.

Vicarious Reinforcement and Learning

In 1963 a study was reported by Banduia, Ross and Ross which
explored the issue of vicarious leinforcement; that is, the changes in the
behavior of observers resulting from witnessing the consequences ex-
perienced by others. In this study > nursery school children witnessed a
variety of situations. The prime issue was whether or not they viewed an
aggressive model being rewarded for his acts. The major finding was that
children who witnessed the aggressive model rewarded showed more
imitative aggression, and preferred to emulate the successful aggressor
than children who observed the aggressive model who was punished. This
last group both failed to reproduce his behavior and rejected him as a
model. Control over aggression was vicariously transmitted to the boys by
the administration of negative responses to the model and to the girls by
the presentation of socially incompatible examples of behavior.

Interviews with the children at the completion of the experiment dis-
closed that although children in the aggression-rewarded conditions voiced
disapproval while they watched the acts that they nevertheless emulated
his behavior on the basis of its success. The key issue was that they
admired the power the model gained over reward resources through his
icprehensible behavior. The investigators noted that the children stated
that physical aggression and forceful confiscation of the property of
others is wrong and they criticized the model for doing it. Therefore,
when these same children later copied this type of behavior, they can be
expected to experience considerable conflict and discomfort. They did
not resolve this conflict by praising violence, but tended to do it by

253

criticizing the victim. They viewed the victim as weak or provocative,
ungenerous or unshaimg, and thereby in a sense "bringing it on himself."
In situations where the aggressive model was punished, even when the
victim was quite provocative, the victim was not criticized and the ag-
gressor was considered bad, Successful "pay off" of aggression rather than
its intrinsic desirability served to stimulate imitation.

The implications of this finding in terms of the attitudinal and
behavioral effects of television would indicate that successful hostile ag-
gression would outweigh even the previously established values of right or
wrong for the viewer This study involves only a single aggressive incident
that was rewarded or punished. In most televised programs the "bad guys"
gain contiol over important lesources and win considerable social and
mateiial rewards through aggressive acts, and punishment, if any, is
delayed, as Dr. Bandura says, "until the last commercial." Many episodes
which are antisocial and "pay off" are viewed before the punishment
occurs.

Bandura and his associates find that feai of a punitive or aggressive
model is not a necessary factor in identification and adoption of aggressive
behavior. The success of the aggressive act rather than the fear of the
aggressive agent is seen as more influential. This has relevance to the
concept suggested by Freud of "identification with the aggressor," which
postulates that a perceived threat by a punitive agent is the primary
motivating force in the assumption of aggressive traits.

Social Power, Status and Identification

Although it is often assumed that social behavior is learned and
modified through direct reward and punishment of responses, informal
observation and studies suggest that the "power" of the individual
involved may also be influential. A child who perceives his mother as a
prime source of rewards in the family may identify with her rather than
with the father who he may see as occupying a subordinate position, and
even compete with him for rewards.

A study was devised to set up conditions with nursery school children
and female and male models to reproduce possible family constellations of
power and reward structures. One of the adults assumed the role of
controller of a fabulous collection of toys and offered to go shopping for
such highly desirable items as two-wheel bicycles for the children. The

Models who were seen as having the power to reward elicited twice as
much imitative behavior as models who were perceived by the children as
possessing no contiol over the rewarding resources. Power inversions on
the part of the male and female models produced cross-sex differences,
particularly in girls. It was found a differential readiness existed between
boys and gills in the willingness to imitate behavior by an opposite sex
model. Boys showed a decided preference for the masculine role, whereas
ambivalence and a masculine role preference were widespread among the
girls. The investigators suggest that these findings probably reflect both
the differential cultural tolerance for cross-sex behavior displayed by
males and females, and the privileged status and relatively gieater positive
reinforcement of masculine role behavior in our society.

The research team further suggests that although failure to develop
sex-appropriate behavior has received consideiable attention and is
often assumed to be established and maintained by concepts of depend-
ency, psychosexual threat and anxiety, external social variables may also
be important. For instance, the distribution of the rewarding power
within a family may be very important. Although the small child has great
contact with this mother he also has ample time to observe his father's
behavior. Also children do not adopt "wholesale" the tiaits of one model.
A child exhibits a relatively novel mix of behavior in his own repertoire.

The makeup of the family constellation is also important. This research
shows that in a three-person group, for instance, if one person is denied
access to rewards, the others may experience negative evaluations of the
rewarding model and thereby decrease his impact as a modeling stimulus.
The introduction of each new person and his treatment at the hands of
the model may produce new shifts in the relationships.

Reinforcement by Self-Approval

People tend to set for themselves certain standards of behavior and
respond to their own actions in self-rewarding and self-punishing ways in
accordance with their self-imposed demands. This is a major difference in
human and animal learning studies. Even children can decide whether
their own performance is creditable or not. Speculation about how
children develop these internal standards was transformed into an ex-
periment. The children were given the opportunity to observe models
performing, and then permitted to evaluate their attainments according to
high or low standards and reward themselves accordingly . It was predicted
that children tend to adopt the standards of self-reward exhibited by the
models they observed, but that children in the control group who saw no
model would have no consistent pattern of self-reinforcement. It was also
predicted that the subjects would adopt the self-reinforcement patterns of
the same sex model to a greater degree than that of a model of the
opposite sex. It was also predicted that children would match the self-
reinforcement patterns of adult models more closely than those of peers.

This study reported by Bandura and Kupers had a group of boys and
girls from a summer recreation program as subjects. It was designed to be

255

an investigation of the transmission of standards of self-reward. The
children ranged from 7 to 9 years in age. Adult and child models per-
formed a bowling game in which they adopted either high achievement
standards or a low standard for self-ieward. On games in which the models
attained their standard, they praised themselves and treated themselves to
candy; but when their attainments fell short of their adopted standards,
they appeared self-critical. Later, the children who had observed, played
the same game alone and the scores for which they rewarded themselves
were recorded. The control children saw no models at all.

Children who saw no models or who saw models with low standaids
tended to reward themselves generously following a mediocre attainment.
Children, who saw models set high standards for self-rewards, rewarded
themselves sparingly and only when they attained a superior performance.
This suggests that the behavior of the models is influential in the develop-
ment of self-control as well as in the transmission of standards for self-
rewards. The children tended to match the patterns foi rewards set by the
adult models more closely than those set by peer models. The results
showed that patterns of self-reinforcement can be acquired imitativcly
through exposuie to models, without the subjects themselves being admin-
istered any direct differential reinforcement by external agents.

Another study was devised to further examine the formation of
personal standards. In the experiment previously described, the children
modeled their own standards after those of the model. It was thought that
this was partly because the performance scores had little absolute value
and therefore the evaluation of the model served as a primary basis for
judging what might constitute an inadequate or superior performance.
This study was a!so designed to see if observeis would select models who
were similar to themselves in ability and reject those who were markedly
divergent. It was predicted that subjects would adopt the self-
reinforcement standards of the model whose ability or competence was
similar to their own, They would disregard the examples of those whose
accomplishments were too different from their own and adopt a more
reasonable standard for themselves. The investigators hypothesized Hint
even low or meiely adequate performances by adults would be highly
regarded, and if a child matched or exceeded the performance of an adult
that it would raise his own self-esteem. It was hypothesized that children
attaining the achievement level of even an inadequate adult would tend to
reward themselves highly.

In this study, groups of 80 boys and 80 girls ranging in age from 8 to 1 1
years were given a series of tasks and then they were either told that they
were successful or not. Then they observed a model displaying competent,
superior or inferior performance. The superior model adopted an
exceedingly high standard for self-reward, and the inferior model set a low
one. The children assigned to a control group saw no models at all.

Dc!sults showed that children who observed inferior models tended to
t lower standards for themselves and rewarded themselves more
ously than children who were exposed to more competent models
higher standards. Children tended to scale down the achievements of

the adults to a lower standard more commensurate with their own
abilities.

This experiment also examined whether children's willingness to adhere
to high standards is affected by their prior success or failure experiences.
Children who have had failure experiences tended to reward themselves
less than their successful counterparts, a finding that was most noticeable
among children exposed to the inferior model. Control subjects who had
experienced failure displayed a higher rate of self-reinforcement at lower
levels of performance than did chikhen who experienced past success. The
investigators suggest that under some circumstances self-giatification may
primarily serve a therapeutic rather than a self-congratulatory function.
The same principle is seen when a person "treats" himself to a play or a
special dinner to help himself over a difficult experience.

Another finding was that boys and gills diffeied significantly in the
frequency of verbal self-piaise, but not in the incidence and magnitude of
self-administered material rewards, such as candy. Boys weie more
generous in commending themselves for equivalent achievements.

Contiguity and Other Factors

It is often assumed that the occurrence of limitative or observational
learning is based on the observer experiencing reinforcing consequences.
This does not account for the learning of imitative behavior when the
observei does not perform the model's responses during the process of
acquisition, and where neither rewards nor punishments are given to either
model or observer. It is suggested that in these cases a contiguity theory
can best account for observational learning. Contiguity means that events
or objects in a seiies or close to each other in time or space become
associated. When an observer then witnesses a model exhibit a sequence of
responses the observer acquires through the principle of contiguous as-
sociation of sensory events certain perceptual and symbolic responses that
cue other responses even aftci time has elapsed.

Bandura states that the acquisition of matching responses may take
place through contiguity, whereas the reinforcements administered to a
model exert their major influence on the performance of imitatively
learned responses. Several of Bandura's studies have shown that even
children who do not reproduce the aggressive behavior of models were
able to desciibe the behavior in great and accurate detail. When these
non performers were rewarded they would readily reproduce the modeled
behavior. However, these children usually failed to reproduce the entire
behavior pattern and this indicates that factors other than exposure to
models or contiguity influence response acquisition. It appears that
observers attend to models that arc most relevant to them. Both prior
experience and the distinctive qualities of the modeling example are im-
portant in determining the attention paid to it by the observer.

Social behavior is generally highly complex and composed of a large
number of different behavior units combined in a particular manner.

257

Ban dura points out that such responses are produced by combinations o
previously learned components which themselves may be intricate units
The rate of acquisition of new responses will then be partly determined h;
the extent to which the necessary components are contained in Hn
repertoire of the observer. For instance, small children may be more iibli
to reproduce motor behavior than verbal behavior.

Learning-Pain, Fear, and Other Emotional States

Studies of vicarious emotional learning show that people develo}
emotional reactions to certain places, people, or events through observing
others undergoing emotional experiences. However, the findings reveal
wide individual differences in the degree to which people aic affected bj
the emotional aiousal of others. Bandura and Rosenthal reasoned Iliitl
observers who are easily susceptible to emotional reactions and who ;irc
emotionally aroused at the time of exposure to the affective expressions
of others will show the strongest emotional learning. To tost UK
hypotheses, adults observed a model performing a task when a buv./ei
sounds and then the model feigns an expression of pain as though he Imd
been shocked. Throughout this period, the observers' physiological
responses weie measured to determine the degree to which the formerly
neutral buzzer had taken on negative emotional value for the observmas
a result of the other person's adverse experiences.

Prior to the emotional conditioning phase of the study, observers ex-
perienced different degrees of emotional arousal produced both phys-
iologically and psychologically through the administration of epinoplirine,
a sympathetic stimulant. Before the study began all subjects comnli'K'tl
the Taylor Manifest Anxiety Scale to provide a measure of their genenil
susceptibility to emotional arousal.

The results show that the observer's level of emotional arousui is n
significant determinant of vicarious emotional learning. Observers who
experienced either very low or very high arousal displayed the weakest
vicarious learning of emotional responses, whereas those who were under
moderate arousal were affected most. Interviews conducted with the
adults after the experiment was completed disclosed that those in the hijih
arousal groups neutralized the emotion-arousing situation by diverting I heir
attention from the model's distressing cues and by conjuring up posllive
or relaxing thoughts. Further study of these cognitive activities may throw
light on how people insulate themselves against the distressing experiences
of others. In this experiment some subjects felt extremely cmpathic wilh
the model and others derived considerable satisfaction from witnessing
pain being inflicted upon the model.

Therapeutic Applications of Modeling Procedures

Research in this area has shown the potential of modeling influences for
changing people's attitudes, behavior, and even their personal standards of

258

self-evaluation. It is also a potent means of treating powerfully charged
patterns such as phobias and fears of long standing. One experiment,
reported in 1967 by Bandura, Grusec, and Menlove, treated children who
were extremely afraid of dogs. During the couise of treatment, the phobic
children who observed a bold peer-model handle dogs comfortably and
appropriately, lost their fears.

After being referred by their parents, the children weie given stand-
ardized perfoimance tests, on the basis of which 24 boys and 24 girls,
ranging in ages from 3 to 5 years, were selected.

The initial selection test included a graded sequence of tasks which
involved increasingly intimate interactions with a dog. Initially, the
investigators brought the children into a room where a cocker spaniel was
confined in a modified playpen. Later, tasks were required which tanged
from walking up to and looking at the dog, to finally climbing into the
pen with the dog, petting her, scratching her, and then remaining alone
with the dog in the room.

The children were then assigned to one of the following treatments; a
modeling-positive-context where a fearless peer model exhibited progres-
sively bold interaction with a dog in the midst of an enjoyable partylike
atmosphere; a modeling-neutral-context where the subjects observed the
same type of brave behavior modeled, but in a neutral atmosphere; an
exposure-no-rnodel condition (here the children saw the dog, but with the
model absent); a positive-context group which participated in the party,
but were never exposed to either the dog or the model.

The day after the treatment series was completed the children were
tested with the experimental animal; then about IVa hours later, with an
unfamiliar white mongrel. The dogs had been tested prior to the
experiment, and it was established that they were about the same in terms
of activity level and attractiveness. Half the children were tested with the
familiar animal first, and then with the unfamiliar white mongrel; for the
remainder, the sequence was reversed.

A month later a followup evaluation was done, and the children were
again tested to determine the stability of the treatment effects, as
manifested by the children's willingness to interact fearlessly with the
dogs. The two groups of children who observed the peer model interact
fearlessly with the dog, achieved and retained substantial reduction in
their fears of dogs. In an effort to minimize the cognitive aspects, all the
children were Informed that the test animals were harmless. After the
experiment was over, the children were told that while most dogs were
friendly, before petting an unfamiliar dog they should ask the owner. This
was done to reduce indiscriminate acts by the children toward strange
dogs,

The effect of the modeling was obvious; the atmosphere, whether
partylike or not, had a minor effect, if any.

The investigators speculate that several factors are involved in the
extinction or disappearance of the avoidant behavior. One is simply that
as the child acquires more information about dogs and about contact with
dogs, he becomes less fearful. The nonoccurrence of anticipated adverse

259

consequences to the model, plus the pleasure the model has from contocl
with the clog, may help extinguish the fear leaction as well.

Treatment of Dog Phobia Through Filmed Modeling

Another experiment was done (Bandura and Menlove, 1968) with
children who were seriously fearful of dogs. One group was shown a movie
which demonstrated how a single model would display progressively less
fearful interactions with a dog, as in the pieceding experiment. Another
group observed a movie which showed boys and girls of varying ages
interacting positively with a variety of dogs of different sizes and dis-
positions. Children in the control group were shown movies with no
canine characteis.

Results showed that children who observed approach behavior which
resulted in no adverse effects to the model, displayed enduring and
generalized reductions in their own concerns about dogs. Controls showed
no change. Comparison of the final step achieved (i.e., staying with a dog
alone) by children who had seen the single model and those who had seen
the movie with many models showed that the latter approach was
superior. However, although modeling was equally effective regardless of
the severity of the children's phobic behavior, those who manifested a
wide variety of fears benefited somewhat less from the multiple modeling
technique than cliildren who had fewer fears.

The control children were shown the multiple model film after the
main experiment was completed. They were markedly more able to
handie dogs after this.

The investigators found here that the symbolic portrayal is less
powerful than live demonstrations. A single model seen live is more
effective than a single model shown in a movie. However, a movie can be
made more powerful by using multiple models and a wider variety of
objects than it is usually practical to provide live.

Snake Phobia Project

This project, conducted by Bandura, Blanchard, and Ritter, was carried
out with adolescent and adult subjects who were terrified of snakes. In the
area of California where Stanford is located, snakes are prevalent enough
to seriously limit the life choices of any adult who is severely frightened
by snakes. For instance, it would mean he couldn't be in any job where he
would be required to inspect houses, read meters, show real estate, do
plumbing, or any activity where he might be out of doors or in basements,
He would be limited as to the location of his home and be largely unable
to participate in many popular local sports, such as hiking or camping.

Fear of snakes is considered by psychologists to be a relatively stable
fear and for that reason is often used in laboratory experiments.

The subjects ranged in age from 14 to 60 years of age. Some of the
phobias had existed for 1 5 to 20 years. In the initial phase of the experi-

260

ment, the participants were administered a behavioral test that measured
the strength of their avoidance of snakes. In addition, they completed a
comprehensive fear inventory. This inventory was then available for deter-
mination later to determine if reduction in anxiety about snakes brought
about other changes.

The cases were individually matched on the basis of their avoidance
behavioi and assigned to one of four conditions. One group participated in
a symbolic modeling treatment where they would run for themselves a
film depicting young children, adolescents, and adults engaging in progres-
sively more threatening situations and interactions with a large (about
4-feet long) king snake. The subjects were taught to be relaxed during the
film. They were told to stop the film when scenes made them anxious,
reverse it to the beginning, and watch it over. They were asked to attempt
to achieve deep relaxation at the same time. They were to view the
threatening scene repeatedly until it was neutralized for them.

The second group, receiving live modeling with guided participation,
watched a model handle a snake in increasing proximity until it was
wrapped around him. The subjects were then aided in perfoiming with the
snake. The model held the snake and had the subject touch it, stroke it,
and then gradually hold it until anxieties about contact were gone. Then
the subject and the model performed the tasks together until the clients
were able to hold the snake in their laps, to let it crawl around, and finally
to retrieve it.

The third group received a form of desensitization treatment. Deep
relaxation was paired with imagined scenes of interactions with snakes. As
in other conditions, the treatment was continued until the clients'
anxieties had disappeared or until the maximum time of 6 hours allotted
had passed. This time limit was imposed upon all groups.

Subjects in the control condition participated in the behavioral and
attitudinal assessments without receiving any intervening treatment.

In the assessment phase, all initial tests were readministered. In order to
test the generality of extinction effects, half the clients in each of the
conditions were tested with the now familiar brown-striped king snake
and then with an unfamiliar crimson-splotched corn snake that appeared
strikingly different. The rest of the groups saw the snakes in reverse order.

The subjects weie asked to look at, touch, and hold a snake with bare
and gloved hands; to remove the snake from its cage, let it loose in the
room, and then replace it in the cage; to hold it within 5 inches of their
faces, and finally to tolerate the snake in their laps while they held their
hands passively at their sides. Before and during these tests clients rated
the intensity of their anxiety on scales.

Control subjects remained unchanged in their ability to handle the
snake. The subjects who had symbolic modeling and desensitization had
substantial reductions in phobic behavior, and live modeling combined
with guided participation proved to be an unusually powerful treatment
that eliminated snake phobias in virtually all subjects (92 percent). The
modeling procedures not only extinguished avoidance responses of long
standing, but they also neutralized the anxiety-arousing properties of the

261

phobic objects. Both of the modeling tieatments achieved marked
decrements in anticipatoiy and peiformance anxiety. Although subjects
who had received desensitization treatment also experienced less
emotional arousal when approaching a snake, the magnitude of theii fear
reduction was significantly less than that shown by their counterparts in
the modeling conditions.

It was found that attitude changes toward snakes occurred. The more
potent the treatment and the moie changed the subjects ability to handle
the snake, the greater the positive change in attitude.

In addition, other fears were affected by the removal of the snake
phobias. Fear of other issues was relieved in proportion to the potency of
the treatments employed. For instance, live modeling with subject
paiticipation effected widespread fear reductions, not only related to
animal anxieties, but also in relation to a variety of threats involving both
interpersonal and nonsocial events. The investigators note that this seems
to involve two different processes. The first involves generalization of
extinction effects from heated stimuli to related anxiety sources. In other
woids, being relieved of one serious fear makes a person generally less
fearful and more able to cope realistically with other concerns. The
second entails positive reinforcement of a sense of capability. Having
successfully overcome a phobia that had plagued them for most of their
lives, subjects reported new confidence that they could conquer other
problems and successfully deal with other anxiety-arousing situations.

A 1-month followup assessment revealed that the beneficial changes
produced in behavior, attitudes, and emotional responsiveness were ef-
fectively maintained. The clients also displayed evidence that the behavior
improvements had been canied over from the therapeutic to the real-life
situations. They were able to hike, garden, and even help frightened
friends 01 childien overcome fear of snakes.

It is the conviction of these investigators that any type of phobic
disorder can be successfully treated by this method with considerable
success. Subsequent experiments show that information alone docs not
contribute to therapeutic change. It was piimarily through a combination
of demonstration, information, guided performance, and the control over
observational experiences that this success was achieved.

Other snake phobia treatment experiments have been done with
children as subjects, and these have been equally successful,

A slightly different type of therapeutic program was developed by
Robert O'Connor working with Doctor Bandura to improve the social
behavior of withdrawn children in a nursery school setting. A group of
withdrawn children was shown films of children playing together and
having a very good time. Another group, as controls, was shown a movie
about Maiineland, instead. The group of withdrawn children, who had
seen the movie designed to help them overcome their social inhibitions,
showed demonstrable improvements in social interactions; those who hud
seen the other film, showed no change in their behavior.

262

Conclusions

There is little doubt that filmed or televised images have tremendous
power to shape attitudes and behavior. That this deserves investigation can
hardly be questioned when we realize the almost universal contemporary
exposure to TV. The complexity of the problem and the successful study
of TV and movies are best approached by isolating factor after factor, and
then painstakingly evaluating the results. These studies are even more
striking when it is clear that they, for the most part, deal with the impact
of single incidents with relatively little reinforcement, whereas the average
commercial-viewing fare is repetitive and often highly glamorized. The
multiple violent techniques demonstrated by a wide variety of relatively
unpunished people on TV can be expected to be highly effective in
teaching, and even in eliciting violent and aggressive behavior in the
viewers. On the other hand, there is the same potential for influencing
viewers toward positive action and more constiuctive methods of problem
solving. Unfortunately some of the problem-solving forums that are
televised, such as the U.N. in critical debates, are repetitious, monotonous,
lacking in the pace and focused force of the usual piogramming, even
though there is no denying their importance. Television can contribute to
the dissemination of information and be highly influential in developing
awareness. However, this research points out that, at times, new informa-
tion without some guide as to action can arouse increased anxiety. The
stream of information about slums and racial tensions without
constructive proposals may illustrate this phenomenon.

Since people can imitate more successfully acts that are within their
own range, they may imitate the more direct acts rather than the more
abstract ones. This is likely to be especially true of children who, for
instance, are more likely to have extensive physical vocabularies than
verbal ones.

Doctor Banduia's research has dealt with the simpler units of behavior,
and he indicates that he feels that more research should be done in terms
of subtle factors. Physical violence is not the only kind of destructive act
and, perhaps, even more benign than some types of interactions between
people. For example, the portrayal of racial prejudice, the dramatizing
and romanticizing of poor marital interactions, lying, and cheating may
be more important than the number of fist fights and murders that are
seen. In addition, Dr. Bandura states:

"All the laboratory studies that I have reported deal with the
immediate impact of a single exposure to aggression on the viewer's
attitude and conduct. While the questions about immediate effect
have been clarified to some extent, we need much more research on
the cumulative impact of television, and the way in which the
medium combines with other beneficial or adverse influences in the
shaping of people's thoughts and actions."

He also points out that results of recent studies of therapeutic ap-
plications of modeling show that such influences can produce generalized
and enduring effects.

263

Another factor is the number of people exposed to essentially the sail
stimulation. The same visual images are seen by people who ordinari
would not come into contact with the same influences. The cultm
spread is much larger than ever before in history. Certainly, an u
precedented audience all over the world witnessed the Apollo 1 1 walkc
the moon.

This body of research points up the fallacies in several popular idea
One is that violence only affects those who are already violent or deviai
and involved in aggression. This has not been borne out. All viewers ten
to be affected. Normal children also learn and are encouraged to peifon
aggressive acts by viewing them under certain circumstances. Another idc
is that if parents instill in their children adequate standards of what i
right or wrong, the violence they see will "wash over them." It was clcarl;
demonstrated that even where children can label behavior as bad o
wrong, if it was successful, they may imitate it and the conflicts would b>
resolved more often by a revaluation downward of the worth or the roll
of the victim. Whether or not the observed aggressive acts are successfu
becomes more important than the moral value of these aggressive acts.

Perhaps the most prominent idea which has been questioned is thatoi
catharsis. There is no evidence that viewing violence, at least in mosl
forms, dissipates aggressive drives and makes a person more healthy. In
fact, it has been demonstrated that a frustrated viewer watching violence
would become less inhibited and more likely to act on violent impulses.

The difference of sex roles and the impact of male and female models
have been briefly discussed and seem important. Clearly, our society
works toward helping girls inhibit aggression and to enhance masculine
roles at the same time. The boys are given more latitude towards ag-
gressive expression and less toward the emulation of any feminine traits.
Still the effects cannot be oversimplified, as can be seen in the new
phenomenon of unisex clothes and the girls who picket, swear, and attack
the police; and the boys who embrace nonviolence even to the point of
choosing prison over the army. What part in this was played by TV is not
fully understood. This is the generation, however, that is often referred to
as the "TV generation" and one of the first groups to have been exposed
to its influence during the entire span of their lives. Doctor Bnndura
points out:

"It is evident that observers do not function as passive videotape
recorders which indiscriminately register and store all modeling
stimuli encountered in everyday life."

The tremendous prices commanded by advertising time on TV would
alone testify to the power that both the industry and the public attribute
to it.

Learning often takes place in a neutral setting and even with strict
prohibitions, and then acts are later effectively performed. The police
recruit learns to shoot on a range, the army enlistee at a camp. Later they
shoot people, Undoubtedly parents can have a considerable effect on their
children's activities, either by monitoring what is seen or by encouraging

264

or discouraging imitation. This research shows, too, that the learned
behavior may still be retained. However, it will take more to break down a
parental prohibition if it is firmly expressed on the part of the parent than
if no intervention is attempted.

Many issues, such as moral and personal achievement standards, often
considered the province of the parent, school, and church are now being
directly and powerfully influenced by other sources, such as TV.
Certainly, censorship seems a limited answer. It is doubtful, for instance,
that merely banishing cigarette commercials from TV would have been as
effective as the antismokmg campaign has been. It may be that the key
lies in the presentation of a broader variety of ideas and more objective
information.

Research Grant. MH 5162
Date of Interview: October 1968

References'

Bandura, A. Behavioral modifications through modeling procedures. In: Krasner, L.,

and Ullmann, L. P., eds. Research in Behavior Modification. New York: Holt,

Rinehart and Winston, 1965, pp. 310-340.
. Influence of models' reinforcement contingencies on the acquisition of

imitative responses. Journal of Personality and Social Psychology, 1 (6). 589-595,

1965.

..Vicarious processes; a case of no-trial learning. In: Berkowitz., L., ed. Advances

in Experimental Social Psychology. Vol. II. New York; Academic Press, 1965. pp.
1-55

Role of modeling processes in personality development. The Young Child:

Reviews of Research. Published by the National Association for the Education of
Young Children, Washington, D.C., 1967.

A social learning interpretation of psychological dysfunctions. In: London, P.,

and Rosenhan, D., eds. Foundations of Abnormal Psychology. New York: Holt,
Rinehart and Winston, 1968, pp. 293-344.

.Modelling approaches to the modification of phobic disorders. Ciba Founda-

tion Symposium. The Role of Learning in Psychotherapy. London; Churchill,
1968, pp. 201-2J7.

.Principles of Behavior Modification. New York: Holt, Rinehart and Winston,

1969.

.Social-learning theory of identificatory processes. In: Goslin, D.A., ed. Hand-

book of Socialization Theory and Research. Chicago; Rand McNally, 1969. pp.

213-262.
Bandura, A.; Blanchard, E.B.; and Ritter, B. The relative efficacy of desensitization

and modeling approaches for inducing behavioral, affective, and attitudinal changes.

Journal of Personality and Social Psychology, 1969, in press.
Bandura, A.; Grusec, J.; and Menlove, F. Vicarious extinction of avoidance behavior.

Journal of Personality and Social Psychology, 5(1): 16-23, 1967.
Some social determinants of self-monitoring reinforcement systems. Journal of

Personality and Social Psychology, 5(4): 449455, 1967,
Bandura, A., and Harris, M.B. Modification of syntactic style. Journal of Experimental

Child Psychology, 4; 341-352, 1966.
Bandura, A., and Kupers, C. Transmission of patterns of self-reinforcement through

modeling. Journal of Abnormal and Social Psychology, 69(1): 1-9, 1964.
Bandura, A., and Menlove, F. Factors determining vicarious extinction of avoidance

behavior through symbolic modeling. Journal of Personality and Social Psychology,

8(2): 99-108, 1968.

265

Bandiira, A , and McDonald, F. Influence of social reinforcement and the behavior o

models in shaping children's moral judgments. Journal of Abnormal and Socia

Psychology, 67(3)' 274-281, 1963.
Bandura, A , and Mischel, W. Modification of self-imposed delay of reward througl

exposure to live and symbolic models. Journal of Personality and Social Psy

chology, 2(5}' 648-705, 1965.
Bandura, A., and Perloff, B. Relative efficacy of self -monitored and external!;

imposed reinforcement systems. Journal of Personality and Social Psychology

7(2)" 111-116, 1967.
Bandura> A., and Rosentlial, T.L. Vicarious classical conditioning as function o

arousal level. /ouMa/ of Personality and Social Psychology, 3(1): 54-62, 1966.
Bandura, A ; Ross, D.; and Ross, S. Imitation of film mediated aggressive models

Journal of Abnormal and Social Psy chology, 66' 3-11, 1963
Bandura, A.; Ross, D., and Ross, S.A. Vicarious reinforcement and imitative learning

Journal of Abnormal and Social Psychology, 67: 601-607, 1963.
Bandura, A., and VVhalen, CK. Influence of antecedent reinforcement and divergent

modeling cues on patterns of self-reward. Journal of Personality and Social

Psychology, 3(4): 372-382, 1966.
Hart ma nn, D.P. Influence of symbolically modeled instrumental aggression and pain

cues on aggressive behavior. Journal of Personality and Social Psychology, 11:

280-288, 1969.
O'Connor, R.D. Modification of social withdrawal through symbolic modeling. Journal

of Applied fiehavior Analysis, 1: 15-22, 1969

266

Investigator

Theodore Lidz, M.D.

Yale University School of Medicine

New Haven, Conn.

Prepared by:

Herbert Yahraes

An intensive study of families with a schizophrenic offspring has led
Yale investigators to characterize schizophrenia as a deficiency disease.
Schizophrenia, they conclude, results from a deficiency in the nurturing
supplied by the parents, plus a deficiency in the transmission from parents
to child of the basic techniques, particularly those dealing with language,
that he needs for adapting himself to the world when he leaves the family.

On the basis of their findings, Drs. Theodore Lidz and Stephen Fleck
suggest that schizophrenia may be passed along almost as truly as certain
physical characteristics but without any involvement of the genes. Dr.
Lidz is professor of psychiatry at the Yale Univeisity School of Medicine
and a career investigator of the National Institute of Mental Health; Dr.
Fleck is professor of psychiatry and public health and psychiatrist-in-chief
of the Yale Psychiatric Institute, where the patients were hospitalized.

Many authorities suspect that an inherited biochemical abnormality
plays at least some role in schizophrenia. The Yale investigatois do not
deny that this may be so, but they find no reason to think that it must be
so. They believe that schizophrenia can be fully explained on the basis of
what happens to a person within his family during the first two decades of
his life. In general this depends upon the adequacy of his mother and
father to fill their roles as parents, and this in turn is influenced strongly
by their experiences as children with their own parents.

In every family studied, at least one of the parents was judged to be
seriously disturbed. Many of the other parents were judged to be, if not
seriously disturbed, rather difficult and peculiar. Generally the problems
of the parents were found to have antedated the marriage.

Dr. Lidz and his associates, principally Dr. Fleck and Alice Cornelison,
a research social worker, began their investigations in 1953. They wanted
to learn whether or not there was something specific within the family
circle that might be responsible for the appearance of schizophrenia in an
offspring.

Their interest in the family sprang from the concept of schizophrenia as
a condition in which a person fails to achieve a workable integration of his
personality by late adolescence or early adult life. Unable to direct
liimself, he then retreats into asocial ways of living. Characteristically he
does so by breaking with the way the people of his culture think and

267

communicate. His failure, the investigators reasoned, might result from
failures in the way he had been prepared to take up life as a reasonably
independent adult. Since the family is the fundamental training place and
the parents the most important influence upon the developing child, an
intensive study of family backgrounds appeared essential. Indications
from recent studies that schizophrenic patients had grown up in seriously
disturbed families reinforced the investigators' line of reasoning.

The research team selected only families in which it could study, \\\
addition to the patient, at least the mother and one brother ot sister;
almost always the father was available, too. These and other relatives,'
together with teachers and friends of the patient, and friends and as-
sociates of the parents, then and earlier, were interviewed at length in tin
effort to get an intimate and detailed family picture. No other series of
families, the investigators believe, has ever been so thoroughly studied.

Single, specific causative factors have not been found. Schizophrenia
developed in the families studied, the investigates conclude, because the
parents had failed to carry out the tasks essential to the adequate biinginp
up of children. The study groups these tasks into three functions: (1)
parental nurturance, meaning normal love and care; (2) the proper
structuring of the personality, which is achieved through a family struc-
ture in which the parents support each other, carry out the roles ap-
propriate to their sex, and respect the boundaries between generations; (3)
the transmission of the techniques essential for adaptation to the culture
in which the developing human being finds himself. In the families
studied, the investigators report, there was a failure to carry oul
adequately not just one but all three of these functions.

"There is nothing so mysterious about schizophrenia," Dr. Lid/
remarks. "It is only mysterious when you go to the back wards of ;i
hospital and see the people who have been there for many years why arc
just standing rigidly or jabbering to themselves. But if you watch young
people coming into the hospital and take your time to get their slory, il
makes sense. If a person really understood human development, in certain
families he would have to start looking for something like schizophrenia
even though he had never heard of that condition."

The findings of the present study, the researchers believe, provide Ideas
both for preventing and treating schizophrenia. There is some indication
that they may also lead to a means of predicting in which families
schizophrenia is likely to develop.

The sample investigated is small, 17 families, so the researchers point
out that validating studies are essential. It is also unrepresentative, since
the families could afford care-costing thousands of dollars a year in ;in
outstanding private mental hospital. All the families except two wm
rated either upper class or upper middle class, although schizophrenia Es
more prevalent at the lowest socioeconomic levels. The researchers explain
that they wanted to avoid the complexities created by economic distress
and related conditions as factors contributing to whatever family defici-
encies might be found. As it happened, the series differed from the cases

268

usually studied in that most of the families were intact, but this circum-
stance, too, the investigators think, helped them clarify the conditions
essential for the development of this most common of mental diseases.

The researchers believe that the findings based on the sample studied
will stand up. "We've had at least 200 schizophrenic patients at the Yale
Psychiatric Institute since the study began," Dr. Fleck repoits, "and
everything we know about those 200 fits in with what we've learned in
our intensive study of the 17."

A recent French study of primarily lower class families is reported to
supply at least some evidence that the Yale team's findings are applicable
to families at all levels.

When the patients were taken into the study, they were in their teens or
twenties. Some had only recently broken down; others had been sick a
long time and had been treated in other institutions. They and their
families were studied over periods ranging from 4 months to 10 years. In
some cases the number of factfinding interviews— in addition to the
therapeutic interviews with the patient and, sometimes, other members of
the family -ran into the hundreds.

Members of the research team have been publishing their findings in
professional journals as the work has progressed. The present report is
written as the main study neais its end. It summarizes the principal
findings, notes some of the research problems encountered, and discusses
work that has grown out of the investigation.

Troubled Marriages; Two General Types

In more than half the families with a schizophienic offspring, the
research team reports, the parents' problems had led to marital schism.
This is defined as a state of seveie chronic disequilibrium and discord,
which aggravated the personality troubles of each parent and constantly
threatened the marriage. In the other cases the parents' problems had led
to marital skew, a state in which, though the marriage was not constantly
threatened, the more normal parent allowed the psychopathology of the
other one to dominate the home and thus distort the child's development.

As an example of schismatic marriage, the investigators discuss a couple
they call Mr. and Mrs. Grau,1 the wife Catholic, the husband bitterly
anti-Catholic. Soon after marriage, Mr. Grau informed his wife that the
children's religion would never cause trouble because there weren't going
to be any children. When, nevertheless, Nancy was born a few years later,
he refused -to his wife's distress-to let her be baptized.

Looking back during her late teens, after schizophrenia had set in, this
girl could remember no period when her parents had not been fighting
openly and threatening to separate. Her recollections corresponded with
those of her younger sister, Ellen. At issue besides religion had been Mrs,
Grau's child-rearing practices (she was overly protective of Nancy as a

1 All names are fictitious.

269

young child and then, as the girl grew up, highly intrusive), the propei
amount of formal education (at least 4 years of college, asserted Mr. Gran,
deriding his wife because she had only two), and a number of othei
matters. The investigators describe the husband as chronically irritable and
paranoidly suspicious. Although his wife had been unusually insecure
about motherhood, she might have functioned more adequately, they
suggest, had she received at least a reasonable amount of emotional
suppoit from her husband.

Mr. and Mrs. Newcomb are presented as one example of the other
general type of marriage-skewed. Even though Mrs. Newcomb was
exceedingly difficult, she and her husband got along well because he was
constantly grateful for her attention and deferrent to her judgment.

When the first child, Jack, was born, the father bathed him because the
mother feared she might accidentally drown him. Jack and a younger
sister were virtually isolated from other children until they went to
school. Then Mrs. Newcomb pestered the teachers with demands that the
children's special abilities be recognized. When Jack and his sister
quarreled with their mother in later years, Mr. Newcomb 's only advice was
to do as he did and never oppose her. Jack became psychotic during his
freshman year at college.

Whether skewed or schismatic, all the marriages studied had the same
general effect: the production of an environment so deleterious to normal
development that a person need not look beyond it, the investigatois
believe, to explain why at least one child in each family became schizo-
phrenic.

Faults in Family Structure

The research group has drawn out the factors considered common to
each maniage and of significance in producing the harmful environment.
Some of these factors are viewed as deficiencies in the organization or
structure of the family; the others, as disturbances in the way the parents,
largely unconsciously, conveyed the essential techniques for adaptation.

In matters of family structure the following deficiencies were found:
I. The Parents Failed To Form a Coalition

husbands and wives have differing roles and functions, the
-joint out, these should interrelate to form a unit in regard to

When a marriage is working properly, each parent supports the other's
role, thus providing some of the assurance and strength of other must
have to perform it. Further, as the result of this mutual support, the
investigators explain, the child's natural tendency to divide the parents
and, in fantasy, shove one out in order to have the other for himself-the
Oedipus complex of psychoanalytic theory -is naturally frustrated, so the
child's development proceeds in the normal direction. He grows up to
value maniage and to see it as a union in which each person works for the
other's satisfaction as well as his own.

This structural necessity, parental coalition, may often become
weakened in normal families, the investigators assert, but in the families
investigated it had either collapsed or become ciitically distorted very
early, if indeed it had ever been achieved.

In the schismatic marriages, husbands and wives criticized and devalued
each other, thus placing an almost unbearable burden upon the children;
for in the natural course of development, Dr. Lidz explains, a child
identifies himself with, and models himself upon, the parent of the same
sex. The other parent serves as a model, too— of the kind of person with
whom the child will seek to unite when he leaves the family. (The
transformation of the person he selects into the kind of person he has
been unwittingly seeking, Dr. Lidz notes, is more readily achieved "when
his perception is blurred by sexual impulsion.")

But it is terribly difficult for a boy to accept his father as a model-in
spite of the very strong natural tendency to do so-if the father is
constantly being devalued by the mother, whose love the boy seeks,
Similar difficulty and confusion arise when a girl hears her mother
criticized and disparaged. The trouble is compounded because the parent
who normally is a primary or basic love object, or desired source of
affection, is constantly tearing down the parent with whom the child
identifies.2

Even in those families having little overt conflict between husband and
wife, the study finds a marked failure of the parents to support each
other. Mrs. Newcomb, for example, did not belittle her husband to his
face, but she made clear her expectation that Jack would have a brilliant
career in art rather than, like his father, a moderately successful one in
business. Beyond this, she ran the family. Though the father made the
money, he had abdicated the other obligations of his parental role.

//. The Parents Failed to Maintain the Essential Boundaries Between

Generations

Generally the failure to maintain the boundary between generations
was marked by the effort of one parent -and sometimes both-to satisfy
through the child an emotional need not being met by the other parent. In
the Nussbaum family, following a bitter and protracted quarrel, the wife

2 Licte, T. The Family and Human Adaptation. New York: International Universities
Press, 1963.

271

held herself aloof from her husband, who then became excessively at-
tentive to their daughter. He would cuddle her until she went to sleep, and
would even sleep with her on nights when she woke up and was afraid.
This near-incestuous lelationship, as the investigators describe it, was
broken off by the girl late in adolescence in sudden terror that she was
pregnant. A little later she became openly schizophrenic.

Seveial of the mothers said they lived only for their children. They
lived through them as well, the study finds, unable to differentiate clearly
between their own needs and anxieties and those of their offspring. Such
mothers closely supervised the children's activities, fought or sought to
fight all their battles, were constantly intruding. Some mothers behaved as
though their children were little more than extensions of themselves,
living-as at birth-in a symbiotic relationship. One mother, for example,
when she needed a laxative, gave her twin sons a laxative also; when the
doctor prescribed a sedative for her, she gave it to the boys as well.

Sometimes one parent competed with the child for the other parent's
love and attention. In the Lamb family, for example, the husband,
resenting from the start his wife's efforts to care for their son, acted less
like the boy's father than a jealous older brother. Later he boasted of his
athletic record and appeared to want his son to be an athlete, too. But he
criticized the boy's efforts and lost his temper over them. Such rivalry, Dr.
Udz comments, leads a boy to stop trying to acquire masculine assets lest
they arouse the father's hostility. At the same time Mr. Lamb belittled his
son's interests and achievements in other fields. If anyone was going to be
encouraged and admired, it had to be the father.

When the generation boundaries are confused, the researchers note, the
child's place in the family is disturbed, and energies that should be going
into furthering his own development are drained off to provide emotional
support to a parent, or to struggle with a rivalry imposed by a parent, or
merely to survive. His emotional development is warped ; he has difficulty
gaining his own identity.

III. The Parents Failed to Maintain the Sexual Roles
Appropriate to Them

The sex of a child, the investigators assert, is the most decisive factor
entering into the formation of the child's -and adult's-personality charac-
teristics. Confusions and dissatisfactions concerning sexual identity can
lead not only to perversions but also to neuroses and character defects.3
In the patients studied such confusions were also found to be part of the
complex of problems leading to schizophrenia.

A person takes on the attributes appropriate to his sex, Dr. Lidz points
out, not simply by being born a member of that sex but by being con-
fronted with the appropriate expectations from infancy onward nncl by
identifying himself with the appropriate parent.

3Lidz, T. The family and Human Adaptation, Cited earlier.
272

If the parent of the same sex as the child plays an inappropriate role,
the child's development is likely to be warped. In some of the families
studied, the role reversals were obvious. One husband mothered the
children and took care of the house while his wife ran a business. Another
husband went to his law office every day, but it was his wife who actually
earned his office rent and the money for all the other bills. Naturally, say
the investigatois, the children in these families grew up with distorted
views of masculinity and femininity.

In most cases the failure to maintain the appropriate sex-linked roles is
reported to have been less obvious but just as real. The psychiatrists
explain that a girl needs to grow up in the company of a warm, expressive,
helpful woman if she is to have the childhood experiences that will enable
her to fit easily into a woman's role herself, but the mothers of the
schizophrenic girls in this study were found to be distant and cool, toward
their daughters at least. A boy, if he is to fit readily into a man's role
himself, must grow up in the company of a man strong enough to
represent his family to the outside world, to live without being over-
whelmed, and to let his family feel his love. But the fathers of the
schizophrenic boys in this study tended to be either weak and ineffective
as husbands and parents (though usually successful as moneymakers) or
else aloof.

The failure of a parent to maintain an appropriate role can usually be
traced to that parent's personality problems, the research team reports,
but these problems are often aggravated by the other parent. As an
example, Mr. Forel married one of three sisters who were openly
contemptuous of men. (His wife boasted to her sons that as a teenager she
had dated boys mainly for the pleasure of standing them up.) Mrs. Forel
laughed at her husband's efforts to make decisions for the family; refused
for years— until threatened with divorce and a reduced income—to move to
a city where his career would be advanced but where she would be a long
way from her sisters; teased him sexually but denied him satisfaction. Of
the two sons, the older grew up trying to please his mother and aunts, and
women in general, by his effeminate interests; the other grew up clinging
to his father and fearing all women. It was the younger child who became
schizophrenic.

In this case, the investigators suggest, Mr. Forel may have been suf-
ficiently weak and masochistic to bring trouble upon himself and his
children no matter whom he married, but the trouble would have been
less severe had he married a more nearly normal woman.

Defective Transmission of Cultural Techniques

Ordinarily the family provides the child not only with the models to
follow but also with tlie fundamental skills and techniques necessary to
live as an independent human being. The most important of these are the
techniques for communicating with others, mainly through the use of
language. The investigators point out that a person must acquire the

273

language of his culture in order (a) to think, (b) to acquire most of the
other techniques, and (c) to associate constructively with other people.

In the process of thinking, the researchers explain, an individual uses
words to build a symbolic veision of the world, which he then manipulates
in imaginative trial and error in order to arrive at the most desirable course
of action. The extent to which his representation of the woild conforms
with reality depends importantly upon the meanings of the verbal symbols
with which he builds it. Hence his thoughts and actions are deeply
influenced by the language he learns.

The schizophrenic patients in the Yale study had all learned English,
but they had learned it in environments that provided, in the investigators'
words, training in inationality. Some of the paients were delusional. A
man who had made a fortune in business spent most of his spare time
isolated in his bedroom reading stock market reports and Eastern
theology. He, his wife, and the governess all believed him to be the ic-
incarnation of an Asiatic god, and they bi ought up the children in the
same belief. A mother, writing in her diary after her son was born,
confided hei hope that she had given birth to the Messiah and for years
recoided the family's unhappy life in idealized terms. Another mother
believed her telephone was tapped. Mr. Gran preached a world conspiracy

of Catholics.

In most cases, however, the parents were simply distorting reality, un-
consciously, to meet their own emotional needs. The Lerners pretended to
the world at large that Mr. Lerncr was a busy and respected lawyer,
although in fact virtually all of his clients had fallen away after his
partner's death some years before. A fathei wrote each week from a
distant city that he would soon come home. He never appeared, but his
wife kept assuring the children for years that daddy would be there "next
week". Another mother kept promising lovely trips, but did not keep her
promises.

A young patient remarked during the couise of a family therapy session
that her recent visit home had been marred by her father's nagging. "Your
father never nags," the mother snapped, though ho had been nagging the
daughter just a few minutes before. Later on the patient turned lo her
mother and said, "I find that I'm often uneasy with you." "If you arc,"
the mother replied, "you're the only person who is." Yet this mother
consistently upset almost everyone she talked to at the hospital.

Under such conditions, the research team notes, much of the com-
munication within a family must be unreal, with the result that children

even more unmindful of messages the child did not put into words. The
impervious parent is not rejecting the child, the investigators believe;
ratlier, he is rejecting anything that threatens his own equilibrium. "We
feel that many of the parents have a very narrow base of stability, that
they are sort of hanging on-limiting their environment so that they can
cope with it," the study reports. By so doing, the parents also limit what
the child is able to see, perceive, or do, and they create in him despair
about the validity of communication.

The conclusion that the patients had been trained in irrationality is
based not only on the researchers' own expeiience with the parents and
on the recollections of other members of the family but also on the results
of two projective tests-the Rorschach, in which a person tells what he
sees in a series of ink blots, and the Thematic Apperception test, in which
he tells what he sees in a series of pictures.

Dr. Margaret Singer, a psychologist who made a blind analysis of the
parents1 responses— made it, that is, without knowing anything about the
cases— almost invariably and in consideiable detail confirmed the picture
that had been built up by direct observation and history taking. As an
example, in interpreting Mrs. Newcomb's responses in the Rorschach test,
Dr, Singer wrote; "She takes a negativistic viewpoint and kills off
meanings by saying that she feels nothing. She keeps conveying that
meaning is hardly worth seeking because one cannot find anything
likeable or clear, and furthermore, she will not try. She will not talk
directly about anything ... At the same time that she blurs meaning-
fulness, she creates an aura of being a nice, sweet person ..."

After reading Mrs. Newcomb's responses to the Thematic Apperception
Test, the psychologist reported: "She is agonizingly contradictory, and
when the tester inquires about an inconsistency, Mrs. Newcomb simply
slaps down the examiner and further blurs meaning by stating that the
picture does not make any sense . . .Nothing about reality seems to please
her or seems right, logical, or consistent. Sexuality is among her worst
topics. People are both male and female at the same time."

The investigators report a tendency to minimize the extent to which
children are exposed by their parents to serious distortions of meanings.
"Parents who are borderline schizophrenic or somewhat paranoid," the
study says, "are not counted in the statistics of mental illness. A vague
and rambling mother who is more or less schizophrenic may obscure
meanings to an extent that even a psychiatrist has difficulty in com-
munication with her, but her children have been exposed to her blurrings
and inconsistencies of meanings since they were born. A father who is
only considered somewhat rigid and overbearing by his business
colleagues, at home dominates the behavior and thinking of the family
with his paranoid rigidity and distrust. Such circumstances are apt to be
more malignant when the deviances are not sufficiently pronounced to be
categorized as 'crazy,' or when the distortions of the disturbed parent are
accepted by the other parent, or when the parent holds a place of esteem
in the community and therefore must be right."

275

The investigators report that some of the parents also failed to transmit
basic nonlinguistic skills. One patient, an artistically gifted young woman,
for example, had never learned how to buy and adjust a brassiere, or even
now to put on her stockings properly. She had never done any cooking,
and she had never shopped in a grocery store.

Skills, customs, and social amenities are picked up not only from
parents but also from peers. The young people who developed schizo-
phrenia in the families studied, however, tended to have associated less
than usual with other children. Thus, the investigators point out, thcie
had been less chance for eccentricities learned at home to be corrected.
Since these eccentricities presumably had helped keep the child apart
from his peers, a vicious circle had operated. Most of the patients became
overtly psychotic, the study emphasizes, only after leaving theii restricted
world, bumping into customs and ideas that seemed strange and becoming
aware of a confusion -sometimes terrifying-about their roles as men or
women. Universities with a good psychiatric staff, Dr. Lidz remarks,
expect to find every year, a few months after the start of school, a gioiip
of students requiring hospitalization. Detected and treated early, moslof
these can return to college within a year.

Words, Meanings, and Schizophrenia

In schizophrenia, the investigators believe, the patient alters his repre-
sentation of reality-a representation built with words— in order to escape
from a world that has grown untenable. Faced with conflicts to which he
sees no solution, he changes his perception of himself and others ami
abandons the meanings and logic of his culture. He thus finds room for
living and a kind of self-esteem.

This distortion of the thought processes without loss of intelligence
potential, according to the investigators, is what distinguishes schi/o-
phrenia from other types of mental illness. Generally the distortion occur*
in only certain areas of thought, "Provide really good care for patients,"
says the research team, "and they don't continue looking very schizo-
phrenic."

(One of the patients in this study learned analytic geometry while lie
was hospitalized; another composed intricate music; a third analyzed the
stock market and selected a portfolio of stocks that would have paid off
handsomely had the psychiatrist followed his patient's advice.)

The abandonment of meanings, with the consequent distortion of
thought, the research group explains, isolates the patient and tends to
make his condition self -perpetuating. He no longer tries to match 1m
concepts with those of other people and thus to learn whether or not his
concepts are the ones required for living normally.

Why don't other persons facing conflicts that seem insoluble take Hie
schizophrenic's way out? Because they cannot, the investigators suggest.
The schizophrenic can take it because he has never attained a firm ami
useful system of meanings. More than other people, he has encountered

276

serious difficulties in understanding and coping with the situations in
which he finds himself. He has never been sure, at least in certain areas of
living, just where he stood. He has grown up in an environment where
words were used to mask or even deny reality.

Mothers and Mental Health: A Revised View

In general, Dr. Lidz points out, psychoanalytic theory has held that if a
child is to be schizophrenic, something must have gone wrong with the
mother-child relationship at the very beginning. "We don't go along with
that," he says. "In these families, some of the children did not become
schizophrenic even though the early mother-child relationship was bad.
Others became schizophrenic even though they did not have particularly
devastating experiences in early childhood."

The investigators emphasize that the functions of a mother extend over
many years and that her ability to carry them out depends upon a variety
of emotions, attitudes, skills, and ways of communicating, which have
been shaped by her relationships with her own parents and brothers and
sisters and with her husband. The first year or so of a child's life is only
one of the critical periods. Another occurs when he starts to attend
school. His security then depends upon having a firm base at home from
which he can move outward, and this base includes a mother who can
encourage him to surmount the inevitable difficulties rather than convey
her anxieties or her own distrust of the larger world.

Another critical period occurs at puberty. For a child to develop into
an emotionally healthy adult, Dr. Lidz holds mothers-and fathers as well-
must have the proper attitudes in regard to a child's changing needs over
the first two decades of life.

Question: If a mother has trouble getting along with a child who later
develops schizophrenia, may not the original difficulty have been caused
by some inherent deficiency in the child? The investigators remark that
this suggestion is often advanced and that it does seem to hold true with
certain autistic children. But no evidence to support it was found in the
cases studied. Rather, there seems to have been "a disharmony in the
mother-child relationship." The disharmony arose because of the mother's
own difficulties, those of her husband, and the problems of the marriage.

Psychiatry has been overemphasizing the role of mothers in the
development of schizophrenia, the study suggests, because "the patho-
logical characteristics of some of these mothers make a lasting impression
upon the psychiatrists whom they harass," with the result that the
psychiatrists tend to generalize from these mothers to all mothers of
schizophrenic patients.

Everything considered, women probably do have a more important role
than men in the production of schizophrenic children, the investigators
suggest. For one thing, women have a greater influence upon children. For
another, statistics indicate that schizophrenia develops somewhat later in
women than in men. Since women can remain sheltered longer, this may

277

mean that more mentally disorganized women will marry. But in general
the investigators believe that it takes two persons to produce a child and
two to produce the distortions in family structure and communications
that can make a child schizophrenic.

The Parents' Backgrounds

The clinical workers in this study had an unusual expeiience: they got
to know the parents of schizophrenics and in many cases to help them,
and they wound up feeling as much sympathy for them as for the
patients.

Among the parents were a number of individuals who weie pursuing
successful careers in the fields of business, industry, education, and the
arts. Fifteen of the seventeen families came from the upper socioeconomic
classes. As long as relationships with other people were fairly formal, Dr.
Lidz explains, even the most peculiar and disturbed of these individuals
managed to get along. It was in close relationships, notably those within
the family, that the peculiarities became strikingly evident.

These parents had not tried to hurt their children, the research team
reports; they had done everything they could to help-everything within
their abilities and the limits placed by their own emotional difficulties.
"Too often," the team observes, "the psychiatrist forgets his psychiatric
understanding when dealing with parents and expects them to have been
able to be different from what they were, or to change through reading a
book or just because he tells them to behave differently. They, too, areas
much bound to their unconscious conflicts as the patients and could not
have been other than what they were." 4

The problems of the parents had been exacerbated but not caused by
the marriage. Mrs. Newcomb had grown up feeling unwanted because she
was a girl. Her father preached the value of education but refused to let
his daughter go to college. He also refused to let her boy friends come into
the house. Mrs. Newcomb's mother had made clear her feeling that her
own beauty and money had been wasted on an unsatisfactory marriage.

The mother of another boy who developed schizophrenia was schizo-
phrenic herself. She was the daughter of an anxious, overly protective, and
confused woman. And this woman, the patient's grandmother, had been
permitted to grow up believing herself to be her mother's younger sister.
Here was evidence, the researchers point out, of serious pathology across
four generations.

In the case of Mrs. Forel, who had been dominated first by her mother
and then by two sisters and who, in turn, dominated her husband, the
investigators found a pattern of female domination going back to her
great-grandmother.

' ^ v"' *•!"" R" Fleckl S" and Terry> Dorothy- The
of the schizophrenic patient; I. The Father. Psychiatry, 1957, 20, 4,

278

Another mother recalled that her fathei had been a slave to routine.
Every night he would read to her from 7 to 7:45, precisely. He committed
suicide when she was 6.

The husbands, too, aie reported to have had difficult backgrounds. One
man's father had been an alcoholic. Another husband was closely tied,
even after years of marriage, to an almost psychotic mothei . The father of
Mr. Grau, the man who became irrational on the subject of Catholicism,
was described as having been stubborn, dominating, and perhaps paranoid.
Mr. Newcomb, grateful for any crumb of affection his wife could offer,
was an orphan.

"We almost feel," says Dr. Lidz, "that schizophrenia is the end result of
cultural deviation increasing gradually over the generations. Then you get
two paients who are unable to straighten out each other's distortions— and
you end up with a child so deviant that he virtually leaves society."

So what can be done?

"That is essentially what the mental health movement is about," the
investigator answers, ". . . to stop this vicious ciicle of unstable parents
having unstable children." Among the approaches, he lists premarital
counseling "to try to keep some of these people from getting married "-
an effort that is not usually successful— and marital counseling, to help
husbands and wives better understand each other and themselves. Another
approach is to educate paients about bringing up children. "The parents
will still have their emotional problems," he observes, "but they will do
better with their children simply because of knowing what to do." He
believes there should be more marriage counseling centers and more
family planning and well-baby clinics, and wider use made of them.

The Brothers and Sisters

The investigating team had set itself two major questions:

First, what factors in the family environment are responsible for, or
contributory to, the development of schizophrenia?

Second, if certain factors in the family environment lead to schizo-
phrenia in a particular child, why are the brothers and sisters not affected?

The answers put forward to the first question have been summarized in
preceding sections. The Yale group's belief in their validity is reinforced
by the findings, noted below, reported as answers to the second.

1. Most of the brothers and sisters of the children who became schizo-
phrenic developed serious problems themselves.

The patients in the study had 24 brothers and sisters. Three of these,
investigators found, were also schizophrenic. Seven others were considered
to be borderline schizophrenics and eight others to be emotionally
disturbed. Only 6 of the 24 were judged to be either adequately or well
adjusted, and all but 1 of these, the investigators report, suffered from
constricted personalities, marked by limits on their emotional maturity,
their perceptiveness, and the use of their intellectual resources.

279

2. The children who became schizophrenic had been brought up
circumstances that differed from those affecting their brothers and sisters

In half the cases the circumstances were markedly different, Dw
patient, for example, had been conceived during a brief reconciliation
between his parents 1 1 years after his mother -Mrs. Forel— had ended the
marriage relationship. He was mothered by his father and an older biotliei
and sister. After Mrs. Forel was injured in an accident, hci husband
became more subservient to her than ever and she became more opcnlj
contemptuous of him. The older children left home when the boy was 6,
and the father died when he was 1 1 . Mrs. Forel then went to live with om
of her beloved sisters, who forced her to boaid the child elsewhere. Tilt
Forels offered the most dramatic example of changing circumstances, and
even children close together in age and of the same sex were found to haw
been subjected to different influences.

Nancy Grau was caught in the middle of the conflict between lur
parents, mainly over religion. Ellen, born less than 2 years later, escaped
the woist of the battle— paitly because Mr, Grau assumed that his wife
would not dare let Ellen become a Catholic and partly because Hltcji
skillfully avoided situations that she knew, from watching Nnncy, would
bring down trouble.

Even a pair of identical twins faced different developmental pressures
As the investigators explain it, the mother came to prefer Peter, who had
been born first, and to identify him with her envied twin sister, who also
had been born first. Through this boy the mother fancied she would liv
out the dominant, aggressive role to which she had always aspired, Ih?
other twin, Philip, came to represent the passive and feminine aspects sli?
despised in herself.

Peter grew up to be grandiose, antisocial, in constant need of admit*
tion, and unconsciously terrified of all women because he viewed Hicnih
overpowering and engulfing, like his mother. He struggled wilhudikinini
common in adolescents who become schizophrenic, the researches
In that he was supposed to be a part of his mother-to achieve
order to complete her life— and at the same time was supposed lolwi
man. He chose to go to a distant college in order to gel away from h
mother's seductive behavior* When his mother then turned lowiinl Ph-i
however, Peter felt betrayed, "and murderous impulses mingled with if:
incestuous, creating panic.'1 He tried to fill his need lor love in ;i hom>
sexual relationship but lost his partner to another boy.

Philip, on the other hand, early relinquished his mother to Peter, Ihi
investigators explain, and thus protected himself from her
Peter's hostility. This twin, too, tried homosexuality, but in Im ft
he was a girl— and therefore not in competition with his brother and hi
wandered about the town in girls' clothing.

Peter became psychotic at college and had to be hospHaltocd. Phil;
broke down, too~in part, the investigators explain, because ho nceJd
Peter to bear the brunt of the mother's attcntion-but was
treated while living at home*

280

3. As a group, children of the same sex as the child who became
schizophrenic were clearly more disturbed than children of the opposite
sex,

The nine male patients in the study had eight biothers and six sisters.
The only healthy sibling was a sister. The 16 female patients had 7 sisters
and 3 brotheis. Two of the brotheis but only one of the sisters weie
emotionally stable.

These findings are explained, the investigators believe, by the findings
reported earlier -the failure of the parents to maintain their appropriate
sex-linked roles. The fathers of the boys formed poor models for their
sons; the mothers of the girls, poor models for the daughters. One or
more, and commonly all three, of the following reasons were involved:
The parent's own serious psychopathology; the parent's unnatural
attitude—intrusive, rivalrous, or aloof —toward the child ; a parent's loss of
worth resulting from the other parent's depreciating attitudes and
behavior.

Under the circumstances, the Yale group explains, a child who wanted
the approval and affection of the parent of the opposite sex, tried to
differentiate himself from, rather than identify himself with, the parent of
the same sex. Thus the child lacked a model to follow in order to gain
maturity as a man or a woman.

At the same time, the parent of the opposite sex, to whom the child
was naturally diawn, produced further confusion by conveying a distorted
picture of that sex. Mothers piesented sons with a model of women as
persons clangeious to males; fatheis presented daughters with a model of
men as unworthy of or even dangerous to females.

It follows that ccitain types of family structure and interaction will be
more dangerous to a child of one sex than to a child of the other. A
mother who cannot establish clear boundaries between herself and the
child will interfere more seriously with a boy's development than a girl's.
This is because a boy, to achieve a firm masculine identity, must break
away more completely than a girl from the initial mother-child symbiosis,
Likewise a weak father is worse for a boy than a girl because a boy needs
to identify himself with a masculine figure.

On the other hand, a cold and aloof mother does more harm to a girl
than to a boy because a girl needs to absorb maternal feelings in order to
develop maternal characteristics herself. Likewise a father who dominates
and belittles the mother interferes more seriously with a girl's develop-
ment than a boy's. In this case the girl, seeking her father's affection, must
differentiate herself from the woman he finds unsatisfactory instead of
pursuing the normal course and endeavoring to emulate the woman he
loves.

Published descriptions of the parents of schizophrenic children often
seem contradictory, the investigators report, but the contradictions may
be reconcilable if the patient's sex is taken into account. "Firm sexual
identity— that is, to behave as it is appropriate for a member of one's own
sex to behave-is one of the strongest foundations of a stable personality,"

281

the study says. "A child who grows up insecure in his essential identity is
likely to be in tiouble."

In sum, if family conditions are such that one child becomes schizo-
phrenic, this study finds, other children are more likely to be affected
deleteriously if they are of the same sex as the patient. Not all Hie
affected children become1 schizophrenic, because family pressures differ
from child to child, and also because the child who becomes schizophrenic
often lessens the impact of the parental disturbance upon the other
children-by seiving as a target, for instance, or as an example of how not
to get along.

The Patients Improve

Of the nine male patients in the study, only one is still hospitali/ed,
elsewhere. The others are reported to be fending for themselves, most of
them quite well. However, one young man who went back to college after
several years of hospitaliz-ation and another who is working are noi
considered very successful patients.

The young women haven't done so well. The investigators don't know
why but call attention to a finding by Dr. David Rosenthal of NIMH, thai
schizophrenia tends to become more chronic in women than men. The
research group notes also that five of the eight female patients in the
study were rather chionic cases when fast hospitalized-' 'the parents
being less alert to their abnormalities or less apt to secure optimal euro."
Two or three of the young women still have to be hospitalized from thno
to time.

In the group's judgment, only a few of the patients would be good fouls
as parents themselves; it is hoped that most of them won't have children.

Implications for Therapy

The investigators believe that they themselves and the doctors as-
sociated with them are now better able to treat schizophrenia because Ik
study has made them more aware of the results of family interrelations.
The patients' problems and communications have become much mow
comprehensible. As Dr. Flecks puts it: "We now understand quickly
things that before this study might have taken us months to figure oul."

As the result of this work and of research elsewhere reaching the sanw
general conclusion that schizophrenics have abnormal family backgrounds,
the research team expects to see an increasing emphasis on efforts lo
change a patient's environment.

The tendency at the Yale Psychiatric Institute has been to keep schizo-
phrenics long enough so that when they leave the hospital they can work
and live on their own, away from the families in which they were brought
up. But the Institute also tries to change the environment by changing Ifw
parents through psychotherapy-individual, family, parents alone, and
group. Dr. Lidz believes group therapy, in which parents of schizophrenic

282

children meet together, to be particularly effective. It lessens the parents'
feeling of guilt, gets them over the difficult period of adjusting to their
child's hospitalization, enables them to see what has gone wrong in other
families and therefore what may have gone wrong in their own and what
they can try to do about it. Where patients can be hospitalized only a
month or so, he points out, group therapy offers virtually the only hope
of effecting an environmental change.

On the basis of his experience during the study, Dr. Lidz doubts that
any great change can be worked in the things basically wrong with the
family. The main hope lies in changing attitudes— those of the parents
toward the patient and those of the patient toward the parents. "In the
case of the patient," he says, "our main effort is to free him of the
obligation to feel and think as he did in the past, in order to satisfy the
needs of a disturbed parent or to model himself on that parent." If the
patient had been able to see his family truly and to accept its peculiarities
without becoming enmeshed in them, Dr. Lidz believes, he would not
have become sick.

The psychiatrists who conducted the study say they would now prefer
to keep young patients hospitalized a longer rather than a shorter time—
"because we have come to realize that the young schizophrenic has a great
deal to learn, mainly about how to get along with people. In a way, he
now has to go to school— a school for resocialization— and he cannot finish
in a couple of months."

In numerous other mental institutions the effort is to get the patient
out of the hospital as soon as possible. "You hear it said again and again
that if we use drugs, we can get the patients back in the community in a
short time and they can get along," the investigators note. "That's true, to
an extent. At our general hospital, Yale-New Haven, we are interested in
that approach. We feel some things can be modified— perhaps the parents*
attitude toward the child and toward each other -so that the patient can
go home and get along with outpatient care."

"But here at the Institute we aren't very excited about doing that.
We're not interested in keeping patients out of the hospital, nor in sending
them back home fast, into what we feel is a pathogenic environment.
We're interested in finding out how we can get them to lead reasonably
satisfactory lives. And that's a very different goal."

"You see,*' Dr. Lidz adds, "there are different ways of treating
different patients, according to the conditions surrounding them and ac-
cording to what can be provided for them."

The research group sees another implication in its findings: that psy-
chiatrists should take a different attitude toward the parents and not
consider them either malicious or downright rejecting.

Birth and Progress of a Research Project

As a medical student (College of Physicians and Surgeons, Columbia
University) and an intern (New Haven Hospital and National Hospital,

283

London), Theodore Lidz recalls having been particularly interested in
birth disoiders affecting the central nervous system. These were organic
disorders; damage or rnaldevelopment could be demonstrated 01 surmised.
He became impressed with the idea that schizophrenia was "a different
kind of biith disorder."

The idea was strengthened during his psychiatry residency at the Johns
Hopkins Hospital, 1938-41, under the late Adolph Meyer, one of America's
great psychiatrists. Dr. Lidz helped care for Dr. Meyer's private patients,
"I got to know the relatives of the patients quite well," he recalls, "and
became impressed how really disturbed many of them were. The more 1
listened to them, the more certain it appeared that an individual in those
families couldn't possibly have grown up to be a stable person. Some of
the siblings felt the same way and told me why they thought they were
not schizophrenic but the patient was."

Dr. Lidz wrote a paper on five of the patients and their families and was
thinking of sending it off for publication when he noticed that the initials
of their last names went L, M, N, 0, and P. "I thought that if the initials
could occur this way by coincidence," he relates, "it could also be coin-
cidence that the families of my schizophrenic patients were disturbed."

So he and his wife-~Dr. Ruth Lidz, whom he describes as a congenital
psychiatrist, inasmuch as her father was piofessor of psychiatry at Heidel-
berg—set out to learn whether or not family environments were worth
investigating for their role in the development of schizophrenia. They
compared the histories of the latest 50 schizophrenic patients admitted to
the Phipps Clinic at the Johns Hopkins with those of the latest 50 manic-
depressive patients. Even though the histories were far from complete, the
husband-and-wife team found marked differences in the family back-
grounds of the two types of patients.

Publication of the findings was delayed 8 years. The first interruption
was the war, during which Lidz served as an Army psychiatrist, rising to
be a lieutenant colonel and chief of the neuropsychiatric service of the
18th General Hospital. Later, as the Lidzcs compared notes with psy-
chiatrist friends who, like themselves, were treating schizophrenia, the
observation that schizophrenics come from disturbed families "seemed so
obvious that we saw no reason for reporting it." Then they realized that
while the finding might be obvious, nothing like it had been published. So
in 1949 out came the results of the study the Liclzes had done in 1941.5

Dr. Lidz went to New Haven in 1951, determined to undertake the
present study if he could find financial support, which he did 2 years later
from NIMH. (The other Dr. Lidz went to New Haven, too, mainly as wife
and mother and assisting in the study only part time, also maintaining an
office at home as a practicing psychiatrist.)

Dr. Fleck, who had worked with Dr. Lidz at the Johns Hopkins Hos-
pital, left his teaching position and part-time practice in Seattle and joined
Dr. Lidz at Yale to pursue this study and carry out their ideas on how a

5Lidx, Ruth W., and Lidz, T. The family environment of schizophrenic patients.
Amer. J, Psychiat., 1949, 332-345.

284

mental hospital should be conducted. Dr. Fleck assumed responsibility for
the hospital. Miss Cornelison, a social worker, took the major respon-
sibility for assisting the parents and gaining information about the
families. Typically, a research project begins with the formulation of a
specific hypothesis and continues with attempts to demonstrate its
validity —thro ugh experimentation, observation, and other types of study.
But Dr. Lidz explains that in this case it seemed wise to start with only a
very general hypothesis and to explore the field. "We might have set up
the hypothesis that schizophrenia is caused by a certain type of mother,
for a number of observations seemed to point in that direction," he says.
"But then we should have been studying mothers, and we had reason to
suspect that fathers also were important. In the beginning we just had the
idea— look, children grow up in families; then thinking has to be deter-
mined to a certain extent by what goes on in the family; they need certain
kinds of care in the family. We also knew that most, if not all, schizo-
phrenics grow up in disturbed homes. We sought to learn if something in
the radius of the family might be responsible."

The investigators had given themselves 5 or 6 years. (As in most such
cases, the research job for the senioi membeis of the team was only part
time. The first obligation was to run the 44-bed psychiatric hospital. They
also taught. The proportion of their time they could give to research was
never as much as 50 percent and frequently was much lower.) But the
exploration pointed to so many potential sources of trouble that Drs. Lidz
and Fleck wondered if they would ever identify the essential ones.

Everywhere they looked they discovered something amiss. Usually both
parents were disturbed, but generally one of them more than the other.
Some of the patients had lacked proper nurturance during infancy but all
of them had lacked it later on. Some of the families had been in an almost
constant tempestuous state; others had been relatively though only super-
ficially calm. Communications within families were distorted, though in
different ways.

Another problem, in addition to the confusing wealth of material,
popped up. As patients selected for the study began improving, they
began showing hostility to their parents, and the paicnts reacted by trying
to take them out of the hospital. The team lost several subjects just as it
felt it was beginning to know the family. It solved the problem by telling
the parents ahead of time what to expect and by providing them with the
emotional support they needed.

The investigators had originally dared hope they would find a specific
cause of schizophrenia. Instead, to fit their obseivations, they found it
necessary to develop ideas about the structure and functions of the
family. "It didn't take us 12 years to collect our material," Dr. Lidz
remarks. "Much of the time went in trying to find a way of thinking
about it— in trying to develop concepts. We also had to learn a great deal
about how families function and the functions of families."

When the parents fail to establish the essential structure and fail to
carry out the essential functions, the investigators concluded, schizo-
phrenia can develop in the offspring most affected by the failures. The

285

chcumstances leading to these failures, but not the general nature of the
failmes themselves, differ fiom family to family.

The exploiation, then, has led to the hypothesis that schizophrenia
develops as the result of "deficiencies in family organization and function-
ing as noted in failuies of the family's nurturant capacities, structure, and
capacities to transmit the essential adaptive techniques."

"The patients in the study had not gained the essentials for existence as
independent adults," the investigators sum up. "They had not gained
adequate trust in otheis, confidence in themselves, a stable personality
organization and structure, or the linguistic tools they needed in order to
think clearly for self-guidance, to understand others, and to relate to
others in cooperative endeavors. With such deficits, their integration and
adaptation were bound to be highly tenuous."

The research group believes that the case histories of the families
studied-and also the histories of other schizophrenics treated by the same
investigators-support the hypothesis. But the group recognizes that
important questions remain to be answered.

A Study of Sociopaths

If the hypothesis is correct, the group observes, abnormalities in the
families of children who are emotionally disturbed but not schizophrenic
should differ from those of the families that have been discussed. As a
partial check of its findings, therefore, the team instituted a study of
individuals described by some psychiatrists as sociopaths and by others as
persons with a character disorder. These persons have calloused con-
sciences or, as Dr. Lidz prefers to put it, consciences that operate ineffec-
tually. The behavior of such persons bothers society but not, apparently,
themselves.

The investigators looked for sociopaths from families at the upper
socio-economic levels— as were most of the families in the schizophrenic
study— and found them. By and large the boys had come to the attention
of some public agency for auto theft, stealing from stores, or assault; tlie
ghls, for sexual delinquency.

The big problem in this phase of the study has been the dropout rate,
The parents have cooperated so begrudgingly, in general, that the re-
searchers have not gotten to know that families with anything like the
thoroughness desired. However, from a preliminary study of the material
now available, the psychiatrists report themselves reasonably sure that the
two sets of families do indeed differ, Superficially, at least, the parents of
a soclopathic child seem to be marked by these characteristics;

1. An unawareness of what the child has been up to. As an extreme
example, one boy sneaked a girl into the house and kept her in his room a
week, with his parents completely unaware,

2, A willingness to accept sociopathic behavior until the situation
reaches the point where the police, or some other agent of society, steps

286

in. For example, one mother knew that her boy kept stolen money in the
house, but she did nothing because, as she explained, she was afraid of
getting him into trouble.

3. A concern with appearances rather than more basic values.

4. Often a disregard of social and ethical values.

One underlying fault in the families of sociopaths, the investigators
suspect, may turn out to be that the parents have abdicated authoiity to
the children. This too, would be a breaching of the boundaries between
generations, but quite different from the one found to occur in the
families of schizophrenics.

A Research Problem: Selecting Families for a Long-Term Study

Dis Lidz and Fleck would like to test and elaborate their hypothesis
about the causes of schizophrenia by closely observing families in which
schizophrenia is considered likely to develop. Such a predictive or
longitudinal study, they point out, would turn up more nearly accurate
information about family interactions during the early lives of the
children than even the most painstaking restrospective study, such as the
one just ending.

But how do you select the families? If you make a blind start, observing
any new families that agree to cooperate, Dr. Lidz points out, you may
not get the job done in a lifetime. Figuring the incidence of schizopmenia
at something more than 1 percent and figuring 3 children to a family, he
notes, you would have to take in about 30 families in order to get 1 likely
to have a schizophrenic offspring. "Maybe the moie disturbed families
would avoid coming into the project," the investigator goes on, "so you
could very well study 50 families and not come up with a schizophrenic.
If I'm to study families for 15 01 20 years, I need greater assurance that
Pm dealing with a high-risk population."

Hcie Dr. Lidz puts his finger on one of the major problems of clinical
research: the comparative slowness with which human beings develop and
therefore the length of time that must be given to such studies as the
relation between aberrant developmental factors and schizophrenia, or
other mental disorders.

Like other investigators facing this problem, Dr. Lidz is looking for a
shortcut-in his case, some screening procedure for selecting high-risk
families so that the ratio of families studied to families in which schizo-
phrenia develops is not greater than that about 5 to 1 . He is hopeful that
current work with a test of thought disorders in families that have
produced a schizophrenic child will provide the answer.

A Possible Solution: Testing for Disordered Styles of Thinking

The finding that schizophrenic patients had been exposed all their lives
to disordered and irrational modes of thinking and communicating within

287

the family gave the research group a number of questions; (1) Is dis-
ordered thinking indeed a characteristic of all families producing a schizo-
phrenic offspring? (2) Will a simple test for disordered thinking distinguish
these families from normal families? (3) Will it distinguish them from
families in which a child develops a psychiatric problem other than
schizophrenia? (4) Can families In which schizophienia is likely to develop
be spotted beforehand through such a test7

Recent woik by Dr. Margaret Singer, the psychologist who was
mentioned earlier, and Dr. Lyinan Wynne at the National Institute of
Mental Health provides evidence, in addition to that of the Yale group,
that the answer to the first question is yes. Using projective test material
gathered at NIMH and at Yale, those investigators found that disordered
thinking in schizophrenic offspring was related to peculiarities in the
styles of thinking and communicating of the parents. It proved possible on
the basis of the projective tests, all by themselves, to differentiate parents
of patients from parents of normal persons. Because the type of peculiarity
differed from family to family, it was even possible to say which patient
was the offspirng of which parents.

The Yale group meanwhile has been working with a simpler instrument
known as the Object Sorting Test. In the first part of this test a person is
given something-such as a toy, a dish, a pack of matches-and directed to
place it on a table along with whatever other objects on the table belong
with it. Then he is asked why all the objects in a group belong together.
This process is repeated a number of times. In the second part of the test
the person is shown in succession a number of different groups, all tlie
objects in a given group having something in common— they are all red, for
example, or all toys, or ail smoking materials. Each time the person is
asked why the objects belong together. Everything he says during the test
is recorded.

This version of the test is the one developed a dozen years ago by an
Australian psychologist, S.H. Lovibond, and used with schizophrenics,
The scoring system was based on the idea that in schizophrenic thinking a
person is unable to suppress, out of all the material that comes to mind,
the unessential irrelevant, and illogical. In the Object Sorting Test, such a
person might group a toy cigar and a ball together because both were
made of rubber, but he might then add the packet of matches to the
group because matches can be used to light cigars. The highest scoring
patients-that is, those who most often made inappropriate groupings-
were found to be those whose schizophrenia had been rated most severe.
Later, another Australian investigator used the test with parents of
schizophrenic patients. He reported that 60 percent of these parents,
including at least one parent of each patient, scored high in contrast to
only 9 percent of a group of parents who did not have a schizophrenic
child.

This was exciting. When Dr. Lidz and his associates repeated the
experiment, however, they got much less striking results. Where was the
trouble? "We thought we could run it down in 4 months/' Dr. Lidz
recalls, "but it took 4 years. In the beginning we couldn't say that the

288

people m Australia were wrong because, as we went over our techniques
and results, we found where we ourselves might have been wrong. For
example, there was a weakness in the contiol groiip that might possibly
have accounted for the difference. So we kept working, and writing back
and forth, and eventually concluded that the investigator down there
hadn't run a proper control group and hadn't had anybody check the
reliability of his scoring.

"Nevertheless, there was something in it— not as good as he had put it
but good enough to be highly interesting to us. When we finally finished,
we found that the mothers of schizophrenic patients scored significantly
differently on this test from the mothers of controls, but that there was
no significant difference between the two groups of fathers.

"Our psychologist kept saying, however, that there really was some-
thing wrong with the way the fathers went at this blamed thing and that it
wasn't being picked up by this particular scoiing method. So then we
started working up different scoring methods."

When Dr. Singer demonstrated that the parents of schizophrenics could
be identified on the basis of projective tests, Dr. Lidz sent her the records
of the Object Sorting Test as administered by the Yale group. She was in
California. The tests had been administered in New Haven by other
psychologists, and Dr. Singer was aware only of the husband-wife pairings.
Nevertheless she identified 80 percent of the couples correctly. On the
Object Sorting Test as on projective tests, she reported, the responses of
the parents of schizophrenics were marked by the fragmentation and
blurring of attention and meaning.

Using Dr. Singer's comments, Dr. Cynthia Wild at Yale has developed a
new means of scoring the Object Sorting Test for use by the Yale group
and other researchers in studying schizophrenic families. The responses are
scored in several categories, of which the first is An Inability to Maintain a
Consistent Attitude to a Task. The types of responses scored in this
category are desciibed as:

1. Fragmentation of attention.— This includes the introduction of an
extraneous topic (such as a personal experience) or of behavior that
interferes with the testing (such as getting up and walking around the
room). It also includes "shifts of context of reference," meaning that the
subject has a piecemeal approach to the objects within one group and
seems to shift fluidly from one frame of reference to another. For
instance, when shown a circle of red paper, a red eraser, a red rubber ball,
and a red book of matches, he says that the eraser doesn't belong except
that it could be used to eradicate anything written on the paper, that the
matches could be used to light the paper, and that the ball doesn't belong
either except that, like the eraser, it is rubber.

Meticulousness to a peculiar or bizarre degree is also scored. As one
example, a subject looks at a group of objects, all of them made of rubber,
and points that the toy cigar has a paper band, the eraser contains
abrasives, and the sink stopper has a metal handle.

2. Inability to maintain the role of a subject being tested. — Here the
subject wants to take over the tester's role. He reaches out and adds

289

another object to the group the examiner has place in front of him, or he
badgers, criticizes, or lectures the examiner for not following the subject's
idea of how the test should be given.

3. Negativism.— "The subject does not accept the basic assumption of
the test that there is some reason why certain of the objects belong
together. Asked why a group of round objects go together, for example,
he says, "Who says they do!"

The second main category is Blurring of Meaning. Here a subject gives
the right answer, but adds several otheis and doesn't indicate which lie
considers best. Or he gives the right one with an air of extreme un-
certainty or qualification, He does not let any response stand. For
example, he puts the red ball with the red paper circle and says, "Both
reddish and round, and otherwise I don't see any leal connection with (lie
other things."

The final main category is Peculiars It includes: (\) peculiar verbaliza*
(ions, such as stilted language and made-up words; and (2) imprecise
referents, meaning vague statements that could be applied to almost any
group of objects-for instance, a subject groups the pieces of silverware
because "you use them together."

After satisfying itself that the scoring system was reliable, meaning Unit
clinical psychologists, working independently, tended to arrive at much
the same score in each case, the Yale group began applying it widely, Qiw
of the fiist findings pointed to a relationship between scores on the Object
Sorting Test and both age and education. There was a tendency for older
parents to make higher, or worse, scores. There was a considerably mom
marked tendency for parents with the least education to make the woisl
scores.

"In these studies of thinking," Dr. Lidz comments, "people with loss
than a high school education lespond much more poorly than we
expected. We believe this can be explained by the ways of thinking they
learned. The poorly educated are accustomed to viewing things very
concretely, they have difficulty thinking conceptually.'1

This relationship between test scores and education, the investigator
suspects, ties in with the widely reported observation that schizophrenia is
most prevelant among people at the lower socioeconomic levels. These
people are also the ones with the least education and therefore, Dr. LidK
believes, the ones most likely to have trouble viewing their difficulties in
an organized way and teaching their children to think clearly. (Another
factor in the greater incidence of schizophrenia at the lower levels, Dr.
Lidz points out, probably is the greater incidence of disorganized
families.)

The new scoring system has now been applied to the transcripts ol
approximately 200 parents, including some who had been tested by
investigators at the National Institute of Mental Health. For a clear-cul
comparison between parents of schizophrenic children and parents ol
apparently normal children, however, the subjects were matched on
education and age, fathers and mothers separately, a process that cut the
sample to about 100 individuals.

290

Major findings follow:

1 . Of the parents of patients, 75 percent had high scores (above the
group's median); of the control group of parents, 31 percent.

2. In 58 percent of the couples with schizophrenic offspring, both
parents had high scores. The corresponding figure in the case of the
control couples was 12 percent.

3. The two groups were discriminated better when compared on the
basis of the total scores for each couple. In the patient-parent group, 79
percent of the couples had high scores (as compared with 75 percent of
the individuals). In the control group, 82 percent of the couples had low
scores (as compared with 69 percent of the individuals).

In view of number 3 and of related findings-notably that in control
couples where one parent scored high, the other almost invariably scored
low— the investigators suggest that a healthier parent may offset the effect
of the sicker one, so that the family style of communication is not
seriously disturbed.

The question now is: Will the Object Sorting Test differentiate the
parents of schizophrenic children from the paients of children having
other serious emotional disturbances? Work on this has been started. If
the answer is affirmative, mental health researchers will have something
they have long needed— a relatively simple technique to provide a first step
in screening out families with a high risk of having a schizophrenic child.
Those families can then be observed through the years to learn all the
circumstances entering into the development of schizophrenia in one of
the children.

A Basic Problem in Clinical Research

Dr. Lidz believes that too little research is being done on schizophrenia,
apart from studies on biochemical or pharmacological matteis. As he
states it, the problem is circular. Good clinical research in schizophrenia
demands a good facility— that is, a hospital treating psychotic patients-
having a stable staff of experienced clinicians. Only under such conditions,
he believes, can enough patients be studied intensively and long enough to
yield meaningful results. But most experienced clinicians are not prepared
to give the bulk of their time to inpatient service and research, and most
inpatient service would not begin to pay them adequately. Further, not
enough people are getting the kind of training~-"thorough training, going
beyond a couple of years of inpatient experience" -that will enable them
to provide really adequate care to psychotic patients.

"What I think we need," Dr. Lidz says, "aie more places where people
can work cooperatively on a given problem, as here at the Yale Psychiatric
Institute, In such a place one gets to know patient after patient in many
different situations—not only in individual therapy but in group meetings,
for example, in staff meetings, in rounds. The average psychiatrist does
not have access to this wealth of information; the average resident is
exposed to it for a year or two and then goes off, into private practice."

291

And the man in private practice who wants to do a clinical study of some
condition, the investigator points out, has to work with very few people.
Not more than half-a-dozen institutions in this country, says Dr. Lid;.,
are able to undertake the intensive and long-term clinical research needed
if we are to make sizable advances against schizophrenia.

Research Grant, MH 728

Date of Interview Mar. 22, 1965

References.

Fleck, S. Family dynamics and origin of schizophrenia. Psychosom, Med., 1960,22,5,
Fleck, S., Cornelison, Alice, Norton, Nea, & Lidz, T Interaction between hospital Mnff

and families. Psychiatry, 1957,20,4.
Fleck, S , Lidz, T., & Cornelison, Alice. Comparison of parent-child relationships of

male and female schizophrenic patients, Arch. Gen. Psychiat., 1963, 8.
Fleck, S., Lidz, T., Cornelison, Alice, Schafer, Sarah, & Terry, Dorothy. The inlra-

familial environment of the schizophrenic patient. In J.H. Masserman (lid,),

Individual and familial dynamics. New York: Grime & Stratton, 1959.
Lidz, T. Schizophrenia and the family. Psychiatry, 1958, 21,1.
Lidz, T. The relevance of family studies to psychoanalytic theory./, new. incut. Ms.,

1962,135,2.
Lidz, T. The family and human adaptation. New York: International Universities

Press, 1963.
Lidz, T., Cornelison, Alice, Fleck, S., & Terry, Dorothy. The intrafamilial environment

of the schizophrenic patient: I. The father. Psychiatry, 1957, 20, 4.
Lidz, T., Cornelison, Alice, Fleck, S., & Terry, Dorothy. The intrafamilial environment

of the schizophrenic patient: II. Marital schizm and marital skew. A trim. J.

Psychiat., 1957, 114,3.
Lidz, T., Cornelison, Alice, Terry, Dorothy, & Fleck, S Intrafamilial environ men 1 of

the schizophrenic patient. VI. The transmission of irrationality A.M. A. Arch,

Neural Psychiat., 1958, 79.
Lidz, T., & Fleck, S. Schizophrenia, human integration, and the role of the fiiimly. In

D. Jackson, (Ed. ), Etiology of schizophrenia. New York: Basic Books, 1959.
Lidz, T., & Fleck, S. Family studies and a thcoiy of schizophrenia. Manuscript, l%5,
Lidz, T., & Fleck, S. The mothers of schizophrenic patients. Manuscript, 1965.
Lidz, T,, Fleck, S., Alanen, Y,, £ Cornelison, Alice. Schizophrenic patients and llieir

siblings. Psychiatry, 1963,26, 1.
Lidz, T., Fleck, S., & Cornelison, Alice R, Schizophrenia and the family. New York:

International Universities Press, 1 966. (Monograph series on schizophrenia, No. 1.)
Lidz, T., Fleck, S., Cornelison, Alice, & Terry, Dorothy. The intrafamilial environment

of the schizophrenic patient; IV. Parental personalities and family interaction.

Amer.J. Orthopsychiat., 1958,28,4.

Lidz, T., Schafer, Sarah, Fleck, S., Cornelison, Alice, & Terry, Dorothy. Rgo dif-
ferentiation and schizophrenic symptom formation in identical twins. J, A met.

Psycftoanal, Association, 1962, 10, 1.
Lidz, T., Wild, Cynthia, Schafer, Sarah, Rosman, Bernice, & Fleck, S, Thought

disorders in the parents of schizophrenic patients: a study utilizing the object

sorting test. /. Psychiat. Res., 1962, 1.
Rosman, Bernice, Wild, Cynthia, Ricci, Judith, Fleck, S., & Lidz, T. Thought disorders

in the parents of schizophrenic patients: a further study utilizing the object sorting

test./. Psychiat. res, 1964, 2(3), 211-221.
Wild, Cynthia, Singer, Margaret, Rosman, Bernice, Ricci, Judith, & Lidz, T. Using the

object sorting test on parents of schizophrenic patients. Arch. Gen. Psychiat., 1965,

13(5), 471-476.

292

Investigator

William Goldfarb, M.D., Ph. D.

Henry Ittleson Center for Child Research

New York, N.Y.

Prepared by

Herbert Yahraes

Research getting at some of the basic problems of childhood schizo-
phrenia-including causes and effective methods of treatment-is being
conducted in a mansion high above the Henry Hudson Parkway in the
Riverdaie section of New York City. The mansion is the Henry Ittleson
Center for Child Research, named for the late banker. Twenty-four schizo-
phrenic children, from 6 to 1 1 years old, live and go to school there.
Twenty others come on weekdays for treatment and for classes, which aie
taught by specially trained teachers of the public school system. The
director is Dr. William Goldfarb, a pioneer in coordinating clinical research
and the treatment of disturbed youngsters. He is also associate clinical
professor of psychiatry at Columbia University.

The Center is making a careful study of schizophrenic children over
time, Dr. Goldfarb explains. It wants to sec what changes occur not only
while the children are at the Center but also after they leave; it hopes to
follow them into adulthood. This is believed to be the first study to keep
track of schizophrenic children during the process of growth lather than
to look back at them after years of treatment.

The first findings of this research program, which began a dozen years
ago, are summarized here; some are presented in more detail further on.
They are based on observations of children from intact families only,
because the investigators are especially interested in how the charac-
teristics of the parents, and the ways in which parents relate to each other
and the children can affect the development of mental illness. The
children studied come from all socioeconomic levels.

Major Findings and Conclusions

1. The schizophrenic children under study fall in to two main groups,
labeled by the investigators as organic and nonorganic. Children classified
as .organic are considered to have brain damage, but this is revealed only
by a close examination. Their neurological signs are usually "soft,"
meaning equivocal. (Psychotic youngsters with obvious neurological
defects are excluded and referred to neurological services.) Children

classified as nonorganic have no sign of even slight physiological impair-
ment. Of the 1 29 children studied so far, about two-thirds have been
placed in the organic group.

2. The families of the organic or brain-damaged children are of more
nearly normal adequacy than the families of the other schizophrenics. An
early study found that almost half of the nonorganic schizophrenic
children had mothers who were schizophrenic, too. A new study is it)
process.

3. The symptoms of the organic child do not spring directly fiom liis
physiological impairment but from family-child inteiactions set in motion
when he fails to respond quite normally to his environment. The parents*
response to this failure is to stimulate him too much or too little, general-
ly the former, or to confuse Mm. His response stamps him more than ever
as different. The cycle continues. The product is a child so confused about
himself and the world that he withdraws more or less completely uml
exhibits other characteristics of the childhood schizophrenic.

4. The symptoms of the nonorganic, or apparently undamaged, child
may have been influenced by some undiscovered defect. But they appear
to be completely explainable as maladaptive reactions to an abnormal
environment. This environment, which has been shaped by "parental
perplexity" or "parental paralysis," is marked by indecisiveness, in-
sensitivity to the child's needs, bewilderment in the face of unusual
behavior, and inability to direct the child. The end, again, is a confused
and panic-stricken child whose bizarre behavior, like that of the organic
youngster, is a means of coping with his distress.

5. Under residential treatment, both groups of schizophrenic children
improve clinically, acquire learning skills, and rate higher on a standard
test of intelligence. About three-fourths return to the community. Bui
there is a decided difference in their response to the Center's schooling. A
study of 37 children in residence during a recent 3-year period found that
all those in the nonorganic group eventually gave at least a normal per-
formance in reading and arithmetic. In the organic group, most of the
children improved but reached a peak that was lower than normal and
then leveled off. In sum, as emotional disabilities were alleviated, the
children became free to use their potentiality for learning, which was at
least normal in the nonorganic group but below normal in the organic.

6. For some schizophrenic children, the Center's experimental day
treatment program seems to be just as effective as its residential program.
These are the children in the organic group. The nonorganic children in
the day program do not do nearly so well as those who live at the Center.
Presumably this is because the former return every afternoon to a
pathogenic family environment.

Dr. Goldfarb points out that the division of schizophrenic children into
two broad groups is not yet generally accepted, though evidence of
organic impairment in some child patients has been reported by half n
dozen other investigators as well.

One big impediment to geneial agreement on the causes and nature of
childhood schizophrenia, Dr. Goldfarb believes, is that different investi-
gators use different means of evaluating their patients and different labels
for describing the results. To clarify the Center's research techniques and
findings, Dr. Goldfarb and his associates are making a moving picture
showing some of the children under study and in each case the reasons for
classifying the child neurologically as clearly organic, equivocally organic,
or clearly nonorganic.

Beyond offering information on the causes of childhood schizophrenia,
the Centei's findings have important implications for treatment. They
strongly indicate, for example, that the majority of schizophrenic children
are educable. (Most of the children in the learning study had arrived at the
Center without educational accomplishments; almost all had been
considered unmanageable by the schools in the community— and most of
them uneducable as well.) The study also suggests that day programs for
schizophrenic children can be very successful, particularly if limited to
children like those found by the Ittleson Center to show signs of neuro-
logical impairment. Day programs are far less expensive than residential
programs. The aveiagc annual cost for a child in residence at the Center is
about $8,500 (the bulk of which is met by city, State, and private contri-
butions, these last funneled through the Jewish Board of Guardians, a
charitable organization under whose auspices the Center was chartered).
In contrast, the average annual cost for a child in the day program is
approximately $4,500.

The Ittleson Center has undertaken its long-term study not merely to
learn what happens to childhood schizophrenics, but also to test the
effects of certain treatment procedures. Among the questions it hopes to
answer are: (1) Will treatment directed at correcting a specific problem of
a schizophrenic child-far example, lack of self-awareness-make for
overall improvement? and (2) Will work to improve communications
within the family -which are judged to be abnormal, particularly in the
nonorganic cases— result in improvement in the child?

Dr. Goldfarb hopes also to study a group of children who have neuro-
logical defects but are emotionally normal. The reason they have not
become schizophrenic, he believes, lies in the way they have been handled
at home. Detailed information about this would presumably help
substantiate the Center's findings about the causes of childhood schizo-
phrenia and increase our ability to take preventive measures.

Three Principal Defects of Schizophrenic Children

The children at the Henry Ittleson Center differ widely in capabilities,
behavior, and, the investigators believe, in the causes of their disorder. But
they are alike in displaying extreme emotional reactions that combine
fear, anger, and disorganized motor response. These reactions-which the
investigators consider to be similar to the fear reactions of very early
childhood -seem to spring from the children's overwhelming feelings of

295

strangeness about themselves and their world. And these feelings are
believed to be rooted in the three major defects that the investigators,
originally on the basis of clinical observation alone, found common to all
these children.

The first defect lies in the processing of perceptual information. The
children have rheir senses but do not make normal use of them, hi
particular, they pay little attention to what they see and hear. Some of
them when they come to the Center appear to be deaf.

The second big trouble is a deficiency in the child's awaiencss of self -•
the absence of a sure sense of a unified and intact body. Most of usgiow
up and think we have always had that sense, Dr. Golclfarb lemnrks;
actually we achieved it. This defect is manifested in a variety of ways. One
child was found talking to her hands as if they were beings quite in-
dependent of herself. Another asked if her hands would fall off. Children
cutting paper have to be watched, for several have cut into their fingeis,
unaware that their fingers are part of themselves-and apparently unaware,
too, that the pain of a cut finger results from an action by themselves.
One child asked to have her head opened because there were men running
around in it; a dental examination found a badly decayed tooth. Another
child, walking with a counselor, asked: "Is this a long walk? Am I tiicd?
Do my feet hurt? Will my feet hurt?"

Some of the children are not sure whether they are boys or girls, or of
their physiological needs. They become agitated when hungry or when
under pressure to relieve bladder or bowels. But they have to be told to
eat or to go the toilet; they also have to be told to stop eating or to leave
the toilet, for they have no sense of when their needs have been met. Such
children, unable to find gratification even in satisfying hunger, arc un-
common, but it is characteristic of childhood schizophrenics, Dr. Goldfarb
notes, to be confused about pleasure as well as pain and to find trouble
directing themselves into pleasure-giving activities.

The principal defects noted so far-in self-awareness and in the Ability
to take in and organize what the senses perceive-lead to a confusion jilso
about space and time. To allay the resulting anxiety, schizophrenic chil-
dren use a variety of mechanisms, including withdrawal, a seeking for
sameness, and a compulsive over-concern with time and place, marked by
endless questions about where something is or when some daily event is
going to happen.

The third major characteristic of the schizophrenic children studied is a
difficulty in communicating with people. Like the other characteristics
noted, it occurs universally but in many degrees and forms, ranging from
mutism to talking in such a way that the child cannot be fully understood.

Two Groups of Schizophrenics

Compared With Normal Children

Long observation led Dr, Goldfarb and the <?ther staff psychiatrists lo
conclude that many of the schizophrenic children -surprisingly many,

296

they thought-had inherent defects. Their view was supported by a neurol-
ogist who was not on the Center's staff and who did not know how the
children had been tentatively classified. He found that a majority of the
first 26 children intensively studied gave evidence of neurologic impair-
ment—54 percent in their examination, 58 percent in their history, 65
percent in either examination or history. The neurologist and the psychia-
trist saw eye to eye 8 times out of 10.

The organic and nonorganic groups were then compared with a group
of normal children and with each other on the basis of a number of
neurological and psychological tests. The findings provide important infor-
mation about the underlying reasons for the principal defects common to
childhood schizophrenics. They also show, the researchers believe, that
certain behavioral variations usually consideied characteristic of childhood
schizophrenics in general are, in fact, most characteristic of the group
designated by these investigators as organic.

All three groups of children were found to have the same average sen-
sory acuity— in sight, hearing, and touch. But marked differences appeared
in the children's responses to perceptual tests. Some of these tests dealt
with the ability to differentiate body cues. Esthesiometer tests, for ex-
ample, measured the child's ability to discriminate two tactile stimuli
presented simultaneously; the finger-location test determined his ability to
localize the fingers of his hand (when these, out of his sight, were touched
one by one). These abilities, Dr. Goldfarb points out, help make it possi-
ble for a person to have a clear image of his body. Other tests measured
the ability to discriminate a picture from its background and the ability to
form a picture from fragmented parts.

In all these tests the schizophrenic children were markedly inferior to
the normals, and in all except the figure-discrimination test the organic
group children were markedly inferior to the nonorganic. Evidently, says
Dr. Goldfarb, the schizophrenic youngster takes in messages from the
environment but doesn't handle them properly -doesn't make normal pat-
terns of them or respond to them normally. So he has no clear conscious-
ness either of himself or of his environment.

The results of the delayed auditory feedback test are taken to support
that conclusion. In this test a child's voice was returned to him after a
delay of 0.1 6 second. When a normal child heard his voice after the delay,
his speech became distinctly different but he recognized his voice and
showed no signs of confusion. When the schizophrenic child— of either
group— heard his voice, his speech did not change but he did not recognize
it as his own. Further, he acted as though confused, restless, and under
tension. The investigator concludes that the schizophrenic youngster,
since he did not change his speech, was speaking without hearing or, more
probably, without listening. This finding supports clinical observations
that the schizophrenic child tends to exclude hearing as a means of learn-
ing about and dealing with the world around him. The investigator also
concludes that the schizophrenic's inability to recognize his own voice
shows a limitation in his awareness of himself. The results of this limita-
tion are to be seen in his confusion and restlessness, often amounting to

297

panic, under the conditions of the experiment, and his feelings of
strangeness, fear, and rage in everyday life.

Persons with perceptual impairments of the type shown by the schizo-
phrenic children at the Ittleson Center, Dr. Goldfarb observes, might be
expected to have impairments also in the higher cognitive processes
involved in conceptualization. The tests bear out that expectation. They
show that the schizophrenic child (a) finds it difficult to differentiate
right and left; (b) is poorly oriented for time, place, and person; (c) has
trouble grouping objects by foim and color; and (d) represents the human
body far moie primitively than normal children. In all the tests leading to
these results, the organic group was significantly worse than the non-
organic.

To measure overall intellectual functioning, the investigators used the
Wechsler Intelligence Scale for Children. The schizophrenic children
proved to be inferior to the normal children both in total IQand in each
of its components-verbal and performance IQ. More than half of the
schizophrenic children but none of the others had IQ's in the retarded
range-under 75. More than 90 percent of the normal children but only 23
percent of the schizophrenics snowed at least average capacity— IQ's above
90. Marked differences appeared between the two groups of Ittleson
Center children. The average IQ of the nonorganic schizophrenics was 92;
of the organic, 62 (as compared with 109 for the normal children).

Tests of motor coordination and locomotor balance showed similar
variations. On the Lincoln-Oseretsky test, which provides 36 motor tasks,
the normal children's average score was 70.1 and the schizophrenic chil-
dren's 27.9, but the nonorganic group averaged 48; the organic only 1 7.2.
On the railwalking test, the average scores were 60.2 for the normal chil-
dien and 23.7 for the schizophrenics, but the nonorganic group scored
43,8; the organic, only 1 3.

Another differentiating measure was the whirling test, in which the
child stands erect, eyes closed and arms extended parallel to each other,
and the examiner applies varying degrees of pressure to turn the child's
head, without discomfort. In general contrast to the nonorganic schizo-
phrenic, the organic schizophrenic tended to whirl-that is, to turn his
whole body, as long as pressure was applied, in the direction his head was
turned. Such a response, not shown by children in the normal group, is
considered to indicate immature development of the central nervous sys-
tem.

The investigators were especially interested in the results of the
auditory startle test. When normal children, wearing earphones, were
subjected to a sudden loud tone, they showed that they were aware of it
but not distressed by it. The schizophrenic children gave a wide range of
reactions. At one extreme, eight children showed no evidence of hearing
the tone at all. At the other extreme, six children showed extreme dis-
comfort. With one exception, all these extreme cases were in the organic
group.

Dr. Goldfarb reports that children at the Center who give no evidence
of hearing speech or other sounds on one occasion may react with

298

frenzied distress on others, and that such fluctuation between pseudodeaf-
ness and distress may be seen in the same child on the same day. The
auditory startle test, he believes, indicates that the most marked forms of
hypei sensitivity and sensory avoidance are linked with organicity rather
than nonorgamcity. But some measure of sensory avoidance does, he
emphasizes, frequently occur among nonorganic schizophrenics.

Measurement of speech and voice characteristics, through techniques
developed at the Center, showed distinct differences from one schizo-
phrenic child to another, but not between the two groups. All the schizo-
phrenics had deviances in such matters as volume, pitch, fluency, stress,
intonation, and inflection-characteristics that determine a person's ability
to get across meanings beyond those in the words themselves. Ordinarily
the culture's standards for these characteristics are learned very early in
life. Since the schizophrenics had learned them only partly if at all, these
children had a way of speaking that isolated them still further from
normal activities. Typically the voice was so inexpressive that even when a
child's words could be understood, the person talking to him felt shut out
and unable to get through.

To sum up, children in both the organic and the nonorganic groups
were found to be measurably different from normal children in a number
of tests of perception, cognition, and psychomotor functions. In certain
cases-notably the delayed auditory feedback test and the measures of
communicative ability— the children in the nonoiganic group were found
to deviate as greatly from the normal as those in the organic group. In
general, though, the investigators believe, the results support the conclu-
sion that childhood schizophrenics can be divided into two general
clusters— indeed, that such a division is necessary to an understanding of
childhood schizophrenia,

"What is currently called schizophrenia," Dr. Goldfarb declares, "is a
hodgepodge population that needs subclassification. In mental deficiency
it has become increasingly possible to separate out subpopulations on the
basis of cause, such as phenylketonuria or severe maternal deprivation.
This is likely to be the history of schizophrenia." But in working out a
subclassification, he emphasizes, the relationship between the child and
his family has to be thoroughly investigated even in those cases where an
organic defect is apparent.

Reproductive Complications

The comparison of schizophrenic and normal children has been carried
a step farther through a study of their prenatal and perinatal histories-the
record of events before and during birth and shortly afterward. In this
phase of the research, the investigators studied three groups: (a) 29 chil-
dren in residential treatment at the Center; (b) 39 brothers and sisters of
these children; (c) 34 public school children comparable to the schizo-
phrenic children in ages and backgrounds.

299

The comparison was made on the basis of a "reproductive complica-
tions lecord" listing 83 conditions that had been used by other investi-
gators in studying factors associated with cerebral palsy, mental retarda-
tion, and childhood behavior disorders. Many of these conditions arc
suspected rather than known to be related to fetal damage. A few of (he
items are listed below. Those in the first three groups were scored phism\
considered presumptive evidence of reproductive complications. Items in
the fourth group, The Newborn, were scored either plus or double plus,
generally depending upon the severity of the condition; a double pins
rating was considered highly significant evidence of reproductive compli-
cation.

Historical (+)

Mother's age over 45

Abdominal or pelvic X-ray of either parent

Accidental conception despite using spermicidal contraceptive jelly

Pregnancy O)

Total gestation period less than 8 months

Smoked 20 or more cigarettes daily

Excessive nausea

Severe emotional shock

Any infection

Endocrine disorder

Unusual medication
Delivery (+)

False labor

First stage of labor more than 24 hours

Face, brow, breech, or transverse presentation

Midforceps, high forceps

Precipitous delivery

Placenta previa

Baby held back
The Newborn (+ or ++)

Cyanosis + or ++

Weak cry +

Absent cry ++

Injuries + or -H-

Jaundice + or ++

Vomiting + or -H-

Convulsions -H-

Abnormal reflexes -H-

When information from all sources-mothers, attending physicians, and
hospital records-was used, the schizophrenic children were found to have
had more than half again as many reproductive complications on the
average as their brothers and sisters and the public school children. The
difference in the number of complications, both those rated more signifi-
cant and those rated less significant, was particularly marked in the case of
the boys.

One of the problems in a study like this, the investigators point out, is
that data from the mothers of severely disordered children "are inevitably
selected and colored by such repressive or distracting influences as guilt,
illusory wish fulfillment, or defensive denial." So the investigators place

300

more weight on the analysis of hospital data which had been recorded at
the time of observation and without any knowledge of how the infant
would turn out. This analysis shows that the average number of repro-
ductive complications is significantly greater for the schizophrenic boys
than for their brothers or for the boys in the public school group.
Between the schizophrenic girls and the girls in the other groups, however,
there is no significant difference.

The sample was small, and the Ittleson Center's children may not be
representative of all childhood schizophrenics. Nevertheless, the investi-
gators say, the finding apparently ties in with earlier findings by other
researchers that (a) more boys than girls have childhood schizophrenia and
(6) more boys than girls are affected by disorders such as cerebral palsy,
where brain damage is an undisputed factor, and by mental retardation,
where brain damage frequently has a significant role. The investigators
note also that the ratio of boys to girls at the Center is about three times
as high in the organic group as in the nonorganic.

All in all, this research team is inclined to believe that:

• Cerebral dysfunction as a result of reproductive trauma is more
typical of the boy than the girl schizophrenic.

• The higher proportion of boys among schizophrenic children reflects
the importance of brain damage as a contributing factor to the
adaptive disturbances in these children.

Dr. Goldfarb emphasizes that brain damage, from whatever cause, need
not lead to a behavior disorder— and that even when it does, the disorder
may well not be childhood schizophrenia. Further, childhood schizo-
phrenia can occur even when the brain seems organically normal. Evident-
ly the experience of the child in his family relationships is also important.

The Families of Schizophrenic Children

Before a child is admitted to the Ittleson Center and then again when
he is discharged, a trained observer pays a long visit to the child's family
and notes the interactions between mother and father, both as wife and
husband and as parents; between parents and children; among the
children; and among all the family members. The observer rates the family
on 46 aspects of its behavior, including the spontaneity of interactions,
the methods used by the parents to exercise control, and the ways in
which the children's needs are met. In each case a high score means that
the members are acting in a way that reflects and enhances the mental
health of the group. The sum of the ratings is called the family adequacy
score.

The research group has used this technique to compare the families of
22 children at the Center with those of 22 normal children. The families
of the schizophrenic children in the nonorganic group were found to be
the least adequate, by far. On the average they scored 158 out of a
possible 300. The families of the other schizophrenic children scored 206,

301

25 such visits has been concerned principally with the communicnUoual
behavior of the mother.

Some of the women made little effort to talk with their children
Others talked a good deal, but in a way that was obviously confusing. One
mother, for example, kept saying to her 6-year-old boy, "How nre you?
How are you doing?" at a time when he was crying and saying that he
wanted to go home. At another point, she asked, "What did you have for
lunch, Jimmy -ham?" He answered, "Fish." Then she said, "Jimmy, what
did you have for lunch, huh?" and he said "Fish." Then she said again,
"Tell me, what did you have for lunch, sweetheart?" He stood in front of
her and looked at her and said, "Fish." By laying down a barrage of
topics, this mother, like some others, discouraged the child from com-
municating his own feelings.

When the mother of Mary -a child with a great deal of anxiety about
her body-greeted her, Mary said, "Finger," and smiled and motioned to
her finger. The mother said, "What's that?" Mary replied, "Finger," and
smiled broadly. The mother said, "What is it?" Mary answered, "1 love
finger." The mother made no verbal response. This interchange, the1 in-
vestigators point out, illustrates both the child's deviant means of com-
municating her need for affection and the mothei's failure to do anything
to counteract this deviancy or to meet the need. Earlier, Mary's mother
had told the Center that she and her husband both had been afniid to
touch the child as a baby. "Both of us got up in the night, and one gave
her the bottle and one held the baby," the mother said. "I put a hand-
kerchief in his hand to hold the baby's head because I was afraid thnl if he
would touch her with his bare hands, her head would dilapidate."

A common error reducing communicational clarity was found to be
discordance between the mother's expression and the child's level of com-
prehension. For example, a 10-year-old boy greeted his mother with a
French word, combien. (The investigators learned after the visit that Ihis
apparently meaningless word was an error on his part. He had actually had
a French lesson that day and had meant to say "Comment allez-voux?"in
an effort to win his mother's praise.) His mother did not express her
puzzlement about combien, nor did she insist on an explanation so thill he
would sound less incoherent. And she missed entirely his natural dcsiic for
praise. She said, "Are you speaking French already?" and then, "You
couldn't have learned it all in a couple of days. There must still be sonic
left to do." This remark distressed him because he thought she was saying
he had not studied his lessons, whereas she was actually using an adult
variety of sarcastic humor. He said, "I did really my studies." Eight
minutes later, he persisted, "Mommy, how come you said 1 couldn't learn
it all yet?"

Such errors occur also in interchanges between mothers and normal
children, the investigator notes, but tend to be much milder. This mother,
Jimmy's mother, and a number of others displayed what the investigators
describe as tmpervtousness: they failed to recognize, or at least to
acknowledge, a child's feelings and meanings.

304

Each mother was given a clarity score, based on an analysis of the
transcript of the first few minutes of the visit. In general, the mothers of
the 16 children classified as organic scored higher than the others. This
findings supports the hypothesis, Dr. Gold far b notes, that the roots of
childhood schizophrenia, in the cases where there is no neurological
impairment, lie in parental failure. The normal mother spends much time
explaining novel experiences to a child so that he can giasp them and
relate them to previously achieved understandings. She stimulates and
guides his psychological development, encouraging some of his responses
and discouraging those that are not acceptable. By voice and gesture she
stimulates an interest in acoustic and visual experience. She encourages
the child to be attentive to her communicational clues, and she is attentive
to his. The communicational failures observed during the mother-child
visits at the Center suggest to the investigators that many of the mothers,
particularly those whose children are neurologically sound, did not
successfully cairy out such aspects of the maternal role.

Dr. Goldfarb and his associates recognize that the verbal interactions in
this study may have been influenced not only by the characteristics of the
mothers but also by the setting in which the interactions took place— a
visit to a child in an institution. To obtain comparative data, the re-
searchers have now arranged to study mothers who visit their children in
an orthopedic hospital.

Treating the Child in a Residential Center

As the result of its clinical observations, its tests, and its study of the
families, the Center has developed a treatment program it describes as
corrective socialization.

In one way or another, Dr. Goldfarb notes, the child's home environ-
ment failed to supply what the child needed in order to perceive and
understand reality— that is, to get a true picture of events outside of
himself and inside of himself— and to act accordingly. The failure occurred
in many cases because the children presented a special challenge to which
the parents did not react appropriately. In the other cases the failure
resulted from a high degree of parental confusion and paralysis. Whatever
the primary cause of the failure, the consequence was "a calamitous and
unvarying state of strangeness,"

Dr. Goldfarb believes that this feeling of strangeness explains the states
of panic often noted among schizophrenic children. "The persistent and
overburdening sense of strangeness makes it difficult for the child to cope
with what is unfamiliar," he says, "and restricts the likelihood of the kind
of satisfying complacency which accompanies the perception of what is
familiar." He and his associates call the panic states "primordial" because
they are considered to represent elemental reactions to strangeness (in
contrast to the panic that can arise from unconscious conflict).

The Center's therapeutic approach has these general goals: To help the
schizophrenic child orient himself to the real world and to improve his

305

ability to recognize himself as an individual capable of self-direction -t
realize, for example, that this is his hand and that he can throw a ball wit
it if he wants to. Toward the achievement of these goals, the staff of cigh
psychiatrists, eight teachers, five social workeis, four psychologists, tw
nurses, and nine child care workers 01 counselors-including member
assigned to the day program-tries in many ways to telieve the child'
strangeness and anxiety and to enlarge his range of gratification am
pleasure.

The children are given a carefully structured world, having boundari?
in space and time that they can come to recognize and hold onto. TJn
Center's large grounds have been marked off into small areas havhij
clearly defined functions-a garden area, a playground area, a bicycle area
(Before the grounds were divided, the children wandered over them
aimlessly and anxiously.) Inside, too, the space is carefully delineated in
terms of the function it serves, and the children are encouraged to move
freely within the bounds appropriate to the current activity. At bedtime,
for example, a child may move about his group's bedroom, bathroom, and
playroom. At schooltime, he may move about the classioom. Even the
most jegressed child, Dr. Goldfarb reports, quickly learns the spatial
boundaiies for any given activity.

The children learn, too, that there are set times for specific activities
each day, each week, and each month. Acceptable variants of behavior
with respect to time are taught. For example, pajamas are worn to bed at
night, clothes must be worn during the day, overclothes are worn outside
in winter.

In all talk with the children regarding space and time, the staff tries to
be completely clear. The child is told again and again what he is going (o
do and where and when. He is offered choices, but these are simple-die
choice, for instance, between playing in the bedroom or in the playroom.
Activities, and communications about activities, are linked as often as
possible to rewards; the child is told, for example, "In the morning we
dress and then go down for breakfast." The idea is that the reward en-
courages the development of a sense of anticipation and of the ability to
generalize about time—characteristics that the typical schizophrenic cliild
does not have.

"The climate is child -centered in that it responds sensitively to Hie
child's needs," Dr. Goldfarb reports. "However, it is adult-directed in thai
it is characterized by active, assertive, adult direction, and it is the adult
who determines and delineates the environment for the child. The atmos-
phere thus contrasts sharply with a permissive, nondirective environment.
The child is reminded at all times of adult expectations. The whole ideal*
to correct the child's abnormal ideas and behavior whenever they are
expressed, at the moment they are expressed -all day long."

As an example, a staff member found Mary, mentioned earlier as being
abnormally anxious about her body, in panic because she had bumped liei
arm and had seen a slight red bruise appear. "I'll die," Mary cried
repeatedly. "I'm going to bleed to death." Reassuring words did no good,
so the adult gravely took her arm, bent the elbow gently several limes,

306

examined each finger carefully, and finally informed her he was sine he
could make the arm better in a few minutes. He asked Mary to watch the
red mark get lighter as he patted the arm. She watched eagerly and her
tears subsided, but she continued to ask if she would die. The staff
member informed her authoritatively that there was absolutely no pos-
sibility of her dying from the bruise. As proof, he showed her that the
redness was already disappearing. The girl complained that the arm still
hurt. In that case, he told her, there was one more thing he could do. He
ceremoniously proceeded to put cold water on the injured area. Patting
the arm dry, he asked, "Doesn't that feel better now?" Mary looked at her
arm and then up at him and said, "Yes," and walked quietly to her
classroom.

The recurrent puzzlement, often accompanied by panic, shown by the
schizophrenic child reflects an inability to see a relation between the
experience to which he is being exposed and some earlier experience. It
also reflects, Dr, Goldfarb continues, a desire to see such a relation— to
find a pattern and familiarity in his experiences. So an expression of
puzzlement in the adult's cue to offer explanations and to help the child
evaluate a particular event. "Very much like a parent of an infant," the
investigator says, "the staff member uses warmth, affection, comforting,
and physical contact to reinforce the learnings involved and to improve
the range of pleasure in and tolerance for informational input."

Along with corrective socialization, the Center uses drugs in some cases
and individual psychotherapy in virtually all. And it works with the
parents in an effort to clear up their own puzzlements and anxieties, and
thus enable them to develop a healthful home environment.

Three-fourths of the children, as reported earlier, do return home.
Detailed information about their progress awaits an analysis, which is
under way, of observations over the last 10 years. However, the Center
already knows that most of the children need special help, such as a
special school 01 a special class to maintain them in the community. These
findings, then, are indications at once of progress, of problems, and of
opportunities.

Day Treatment Versus Residential: Comparison of Results

The day treatment center was opened in 1959 in coopeiation with the
New York State Department of Mental Hygiene. It is the equivalent of a
day hospital. It cares for its children from 9 to 4:30, 5 days a week and 9
months a year. During that time it provides the same kind of detailed,
on-the-spot intervention as provided in the residential treatment program.
For approximately 20 children, the day center has 3 child psychiatrists, a
psychiatric supervisor, and 3 teachers. Caseworkers serve the families as
they do those of the other children. Extensive psychological services are
provided .

The two types of treatment have been compared by Dr. Goldfarb and
his associates on the basis of changes in two groups of children over a

307

3-year period-one group living at the center, the other going there days.
Each group had 13 children matched as to sex, age at admission to treat-
ment, neurological diagnosis, and intellectual functioning at admission.

Three of the children in each group were so retarded in intellectual
functioning that no full IQ score could be obtained. Two of these in each
group were classified as organic, while the classification of the third
remained in doubt. On the basis of psychiatric appraisals and the results of
tests, these six nonscorable children showed no improvement over the
3-year program, regardless of the treatment program.

Of the othei 20 children, 6 in each treatment group were classified as
organic and 4 as nonorganic. The following findings pertain to these 20:
• Most of the children classified as organic showed improvement, both
as rated by psychiatrists and as indicated by tests of reading achieve-
ment and IQ. The extent of the improvement was apparently not
influenced by the type of tieatment program. In reading, the organic
children in the day center went from an average standard score of %
to one of 129; those in the residential program, from 74 to 132. The
day children showed an average IQ of 62 upon admission, reached an
average of 74 at the end of the second year, and then dropped to an
average of 65 at the end of the third year. Similarly, the oiganlc
children in residence went from an average IQ of 66 to one of 80 at
the end of the second year, and then dropped to 72. (These drops,
after the initial rises, are unexplained.)

« None of the nonorganic children in the day program showed any
improvement as rated by psychiatrists. Three of the four in the re-
sidential program did show such improvement. The children it) each
treatment group improved in leading, but those in the residential
group far more— from 89 to 182, as compared with from 98 to 137
for the day children. As for the IQ scores, the children in each group
started with an average in the upper 80's and gained a few points in
the first 2 years. In the third year the average IQ of the residential
children shot up to 103, while that of the day children remained 92.
In sum, the bulk of the schizophrenic children who had brain damage,
but were sufficiently intact to be scorable on an IQ test, improved to
about the same degree whether in the day or the residential program. The
nonorganic children, on the other hand, did substantially better in the
residential program.

These differences can be explained, the investigators suggest, by dif-
ferences in the families. The family of the nonorganic child has been a
more fundamentally disturbing influence, from which it benefits him to
get away completely. But the organic child seems not to need complete
separation ; he can progress if he is given an educational and environmental
experience with which he can cope, and if his family is given professional
assistance in contending with his deviancy and meeting his needs.

The most efficient treatment of schizophrenic children, these findings
suggest, calls for two types of facilities: One offering a day program for
children of the organic type, and one offering long-term impatient treat-
ment for those children whose deviant manifestations are explained lo a

308

large extent by psychosocial factors. In making these suggestions, the
investigators point out that the samples in this study, which is being
repeated, were small and that all the children came from intact families.
Other confirmatory studies are urged.

Research Grant. MH 5753
Date of Interview. Dec. 28, 1965

References'

Goldfarb, W. Childhood Schizophrenia. Cambridge: Harvard University Press, 1961.

Published for the Commonwealth Fund.
Goldfarb, W. The mutual impact of mother and child in childhood schizophrenia.

Amer.J. Orthopsychiat,, 1961,31,4.

Goldfarb, W. Families of schizophrenic children. Ment. Retard., 1962, 39.
Goldfarb, W. Self-awareness in schizophrenic children. Arch. Gen, Psychiat., 1963, 8.
Goldfarb, W. Childhood schizophrenia.//!/. Psychiat. Clinics, 1964, 1, 4.
Goldfarb, W. Corrective socialization: A rationale for the treatment of schizophrenic

children. Canacl, Psychiat, Association J , 1965, 10, 6.
Goldfarb, W., & Goldfarb, N. Evaluation of behavioral changes of schizophrenic

children in residential treatment. Amei. J Psycho ther,, 1965, 19,2.
Goldfarb, W., Goldfarb, N., & Pollack, Ruth C. Treatment of childhood schizophrenia.

A 3-year comparison of day and residential treatment. Arch. Gen. Psychiat., 1966,

14.
Goldfarb, W., Levy, D.M., & Meyeis, D.I. The verbal encounter between the schizo-

phienic child and his mother. Presented at the 20th Anniversary Meeting of the

Psychoanalytic Clinic for Training and Research at Columbia University, 1965. In

press, 1966.
Goldfarb, W., & Mintz, I. Schizophrenic child's reactions to time and space. Arch, Gen.

Psychiat., 1961,5.
Goldfarb, W., & Pollack, Ruth C. The childhood schizophrenic's response to schooling

in a residential treatment center. Evaluation of psychiatric treatment. New York.

Grime and Stratton, 1964.
Meyers, D.I., & Goldfarb, W. Studies of perplexity in mothers of schizophrenic

children. Amei. J. Orthopsychiat., 1961,31,3.

Meyers, D.I. & Goldfarb, W. Psychiatric appraisals of parents and siblings of schizo-
phrenic children. Amei: J. Psychiat., 1962, 1 18, 10.
Taft, L.T., & Goldfarb, W. Prenatal and perinatal factors in childhood schizophrenia.

Developm. Med. Child Neural, 1964, 6.

309

Investigators.

William Pollin, M,D.
James R. Stabenau, M.D.
Loren IVIosher, M.D.
JoeTupin, M.D.
Axel Hotter, M.D.
Martin Allen, M.D.
Barbara Scupi, M.S.W.
NIMH

Prepared by:

Maya Pines

When one identical twin become schizophrenic, but the other does not,
what in their environment caused the difference? And when both twins
become schizophrenic, what in their biochemical make-up predisposed
them to it?

By asking provactive questions like these, Dr. William Pollin and liis
associates at the Section on Twin and Sibling Studies, Adult Psychiatry
Branch of NIMH, hope to uncover major pieces of the jigsaw puzzle that is
schizophrenia - a disease which accounts for roughly half of all hos-
pitalized mental patients in the world. They are not primarily interested in
twins, but in schizophrenia, and for the past six years they have been
conducting a series of strictly controlled studies that combine the
techniques of psychiatry, genetics, and biochemistry. These studies focus
on: (1) the specific family patterns that contribute to triggering schizo-
phrenia in one twin, but not in another, and (2) the underlying bio-
chemical abnormalities that are shared by both twins. Thus, they point
the way to two fruitful lines of attack on this widespread disease.

On the biochemical side, they are currently evaluating the possible
significance of their most recent finding on 1 1 pairs of identical twins; the
high rate of catecholamines excreted in the urine of both the schizo-
phrenic and the healthy twin. The catecholamines have two important
functions: Some of them, particularly norepinephrine (noradrenaline)
carry messages to different parts of the brain; and all of them play an

essential role in the body's efforts to cope with stress. For this reason,
previous reports of high catecholamines in psychiatric conditions had been
attributed to the effects of the patients' anxiety. However, stress usually
produces other signs as well - for instance, an elevation of adrenal steroids.
Yet the healthy twins showed no parallel rise in their adienal steroids,
indicating that the high rate of catecholamine excietion was not solely a
response to stress. And, indeed, other analyses in this Lab determined that
the amount of catecholamines in both twins was under genetic control.
This strongly suggested that overactivity of the catecholamine system did
not merely reflect psychosis, but might help to produce it.

The Sibling Studies: Schizophrenia vs. Delinquency

As Dr. Pollin points out, it is very rare in behavioral research to have
really well-controlled comparative studies. This represents a serious
problem, particularly since there are so few well-anchored, reliable
measures of personality or of psychopathology. Psychiatrists can clinically
get the impression that a certain kind of life course is very relevant to a
psychosis, but trying to nail down the mechanisms involved is extremely
difficult.

He began, therefore, by studying siblings, on the theory that children of
the same parents would provide a better basis for comparison. He wanted
to compare the circumstances that led a child in each of three sets of
families to become either schizophrenic, delinquent, or exceptionally well
adjusted, while another child in the same family was just normal. "In
some ways, delinquency and schizophrenia are polar oppositcs," he
explains. "Schizophrenia is a disease in which inner representations of the
external world become completely distorted, resulting in delusions and
hallucinations. Delinquency is a pathology in which one's relations to the
outer world are disordered. It is the difference between acting in and
acting out." However, nonspecific reactions of guilt and shame are present
in the families of both.

The research scheme devised by Dr. Pollin together with Drs, James R.
Stabenau and Joe Tupin fitted in with the trend towards focusing not
solely on the schizophrenic patient, but on the patient and his family as a
unit. It involved three groups of five selected families. The ten youngsters
in each group were carefully matched for age (between 14 and 18), sex,
sibling order, and social class. All were Protestant. All were being raised by
their biological parents. Both parents and both siblings in each family
participated in all phases of the research. The five schizophrenics ("S"
ndexes) had been diagnosed as schizophrenic by at least two psychiatrists
ind had suffered from hallucinations, marked paranoid delusions, or
narked regression for 6 months to 9 years. The five delinquents (*'D"
ndexes) were referred by legal agencies; they had been charged with car
heft, promiscuity, or running away. The exceptionally well-adjusted
/oungsters ("N" indexes) had been nominated by the staffs of a local
unior college and high school for their outstanding academic work and/or
personality adjustment and peer relationships. With the exception of the S

311

indexes, all were seen during outpatient visits, and all participants,
including the 15 normal contiol siblings, were interviewed and tested ill
length.

Fiom the outset, very different patterns emerged among the three sets
of families - patterns which appeared to antedate the symptoms of
psychological disturbance in S and D families. Among the D families, for
instance, all relationships were unstable. There were no fixed roles or
responsibilities for anyone, and it was uncertain who would be the father,
who the mother, who would dominate or be the leader at any time. By
contrast, the S parents seemed locked in a rigid relationship to each other
and to their children. It was a kind of stalemate. On the other hand, the N
parents seemed flexible, empathic, and able to complement one another.
On tests of communication and clear thinking, such as the Object Sorting
Test and the Revealed Differences Test, both the S and D families did
poorly compared to the N families.

The family histories of the three groups revealed even more striking
differences. In nearly every S or D family (but not in the N families), a
major crisis, spontaneously characterized by the family as "the worst
time," had occurred when the index child was between 6 months and 3
yeais of age, usually reaching its peak at about the time he was 1 8 months
old. "One of the important consequences of this differing incidence of
family ciisis appeared to be its potential for serving as the ougin of a
negative identity in a child through identification with a depressed, guilty,
and anxious mother," wrote Drs. Stabenau and Pollin. "In addition, the
disruptiveness of the family life at that time further served to reduce Hie
sense of internal security for the index child."

During these earliest years, too, the parents in the S group often viewed
the child who later became schizophrenic as either physically or psy-
chologically "damaged." (Few of the non-schizophrenic children were
considered defective in this fashion.) And nearly all the children who later
became either S or D had younger siblings born within their first two
years, while few of the controls did.

But why did schizophrenic symptoms develop in one group, and delin-
quent symptoms in another, when both suffered from early family crises?
The researchers point to differing family life styles and, in the case of
delinquents, to an additional crisis: the sudden collapse of the role of the
father, just before the child began to act out. Depending on the timing of
this change, one child within the family would become delinquent while
*i *i.^r wouid not_ jn one cas6) for example, a father who had kept his

from a parent, close sibling or grandparent. Often the mother scrutinized
the index child's activities with phobic concern about where he was and
with whom he played, fearing harm at play or his being killed in the
streets. This resulted in depriving him of the freedom to explore the world
around him - a freedom granted within broad limits to N childien, and
given almost recklessly to D children. Typically, the index child's life style
was, and had been, extremely constricted.

In addition to differences such as these in the patterns of family
relationships, existing evidence of a genetic factor in schizophrenia
suggests that certain organic factots must have been present in the S
families, though not in the others.

An Intensive Study of Identical Twins Discordant for Schizophrenia

As long as the subjects were genetically different, these findings had
limited genera Usability because inherited tiaits might account for most of
the differences between them. In 1962, Drs. Pollin and Stabenau deter-
mined to repeat their study - this time, with identical twins. One-egg
twins, they believed, would represent "the optimal controlled sample," in
which not only genetic factors but also social, ethnic, chronological and
psychological variables were matched to a degree not attainable in other
ways.

Twins have been studied with great interest ever since antiquity. The
ancients who believed in astrology were intrigued by the fact that twins
born at almost the same time could have contrasting life histories. To
explain this divergence, they decided that major changes in the configura-
tion of the heavenly bodies must have occurred precisely in the interval
between the births of Twin 1 and Twin 2. This meant that if they could
locate identical twins with differing life histories and find out the moment
of their birth, they could pinpoint the limits of major astronomical
periods. Therefore, they actively sought out and studied such twins. But it
was only in the 19th century, with Francis Gallon, that the classical twin
method came into use as a method of measuring the relative contributions
of heredity and environment to a specific illness or other condition.

Since monozygotic (MZ) or identical twins are genetically alike, but
dizygotic (DZ) or fraternal twins are genetically no more similar than any
other siblings, the degree of concordance between twins of both types is
meaningful. If 40 out of 50 pairs of MZ twins are concordant for an
illness, but only 10 out of 50 pairs of DZ twins are so concordant, the
difference between these two figures is a measure of the genetic contri-
bution to the illness. However, this assumes that the environment of MZ
twins could be considered to have been constant - a major flaw in the
method, according to Dr. Pollin.

"We set out differently," he explains. "We did not try to evaluate the
relative importance of the genetic factor, but to ignore it, in the sense that
it was controlled for both. We were thus in an optimal position to study
the non-genetic factors involved." It took very extensive recruiting

313

through hospitals, university psychiatric departments, and other sources
to find enough pans of identical twins in which one was healthy but the
other was adjudged clearly schizophrenic by five psychiatrists, and in
which both parents were willing to come to Bethesda to be studied for
two or three weeks at the NIH Clinical Center, together with both twins.
But finally 16 such families were brought to the Center and evaluated,
together with 9 control families.

Having at least these key members of the family present is "almost
essential if one hopes to obtain a meaningful historical leconstruction
concerning the differential experiences of the two twins," declare the
researchers. "Much of the more important data concerns the mother's
piegnancy and the birth and first years of life of the twins, facts known
only to the parents. In the absence of prolonged psychoanalytic relation-
ship with the parents, such material appears to be most accessible via a
family-focused evaluation situation in which four family members are
constantly supplementing, stimulating, and correcting the material that
each recollects and presents." Besides, they could be given complete
psychiatric, psychological, and biological work-ups, including tests of
blood, urine, and chromosomes and some 25 different psychological tests.
The twins' zygesity was determined by investigating 28 blood-group
factors, as well as fingerprints and 10 anatomical features. Over 30 differ-
ent collaborating investigators participated in the study.

The schizophrenic twin was admitted to an inpatient ward. If actively
psychotic at the time of the study, in most cases he received lengthy
treatment at the Clinical Center at no cost to the family. The other
members of his family stayed in a physically identical ward on another
floor. The nonschizophrenic twin was often interested in being evaluated
because he wanted to find out his chances of remaining healthy.

Ever since Dr. Franz Kallmann described his genetic theory of schizo-
phrenia in the 1940's and stated that the concordance rate for schizo-
phrenia was approximately 80 percent - that if one identical twin became
schizophrenic, his co-twin was likely to develop the same symptoms in 80
percent of the cases - these figures have stuck in the public mind,
producing great fear among affected families. However in 1963, after a
careful study of 1 6 pairs of MZ twins in which one was schizophrenic, tlie
Finnish investigator Pekka Tienari reported that none of the co-twins was
affected. Previous studies had traced twins through mental hospitals;
Tienari stalled out with parish records of twin births and followed them
up, This may have accounted for some of the difference in results, since
the overburdened families of two schizophrenic patients would be more
likely to have them committed to a mental hospital than the families of
one. A larger study by Einar Kringlen in Norway followed up all the twins
born in a given area during a given decade and found the concordance rate
for MZ twins to be 38 percent, substantially lower than any reported
before 1960.

Like the Scandinavian work, Dr. Pollin's current analysis suggests that
instead of being concordant for schizophrenia in 80 percent of the cases,
MZ twins are actually discordant for the illness in approximately 75

314

percent of cases. Even at the time of his earliest twin studies, he was able
to reassure the healthy twins about their prospects, and while all members
of the family found the study stressful, in most cases the healthy twins
left the Center with their fears lessened.

These figures focused interest, once again, on the psychodynamic
factois involved in schizophrenia. Tienari had noted that the twins who
were psychologically more submissive were the ones to develop schizo-
phrenia in all but one of his 16 cases. Kallmann and others had made
similar observations. The NIMH study of twins essentially confirms this
insight, and also explains how and why the environment of the stricken
twin differed from that of the healthier twin, almost from the time of
conception.

In a detailed report on the first five families studied in 1964, Drs.
Pollin, Stabenau, and Tupin described "a consistent pattern of historical
events and related familial attitudes" which distinguished the index from
his healthy co-twin:

1 ) The twin who later became schizophrenic weighed less at birth.

2) He was perceived by his parents, particularly by his mother, as
vulnerable, and his survival was thought to be imperiled.

3) He was the focus of more worry, involvement, and attention than his
co-twin.

4) He developed somewhat more slowly.

5) He tended to perform less successfully, and to be perceived as the
less competent and weaker of the twins.

6) He tended to be the more docile and more compliant of the two, was
less independent, and had difficulty in achieving any degree of autonomy
and separateness.

7 ) These relative differences tended to be persistent and unchanging.
This pattern resulted in part from constitutional diffciences, and in part

from "a rigidly 'imprinted' role expectation, initiated at birth, determined
by the constitutional differences, and subsequently reinforced by minor
differences in development," the authors report. They see it as largely the
result of a self-fulfilling prophecy.

In all five instances, the mother had a strong conscious fear of death
concerning the index twin and/or herself, immediately after his birth. In
two of these cases there was a brief period of legitimate concern about the
index twin's survival, as he remained in the hospital for several days after
his co-twin was taken home. In another, however, the mother would not
accept the pediatrician's assurance that both twins were fine and that the
smaller one, because of her size, would necessarily feed more slowly and
take less formula - she had a phobic anxiety that the smaller twin would
die if she did not feed her as frequently and as much as the other.

These worries led the parents to concentrate their efforts on the smaller
twin. In each of the five families, the mother felt that the smaller twin
"needed her more." When she fed both twins, she fed the smaller one
first. If only one could be breast-fed, it would be the smaller one. In two
instances, the mothers reporting pinching and slapping the smaller twin
and using cold water on him to keep him awake so he would eat more.

315

Once established, this pattern persisted as the twins grew up. "The
smaller twin would receive extra praise for things taken for granted in the
larger one," the researchers report. "He would receive additional help with
schoolwork, and would be less expected to dress himself or take respon-
sibility for personal needs or household chores." The bigger twins were
always a bit more successful in school and social life. They played the
leader role and made all the decisions. They were also less docile and
"good." Only the healthy twins had episodes of active rebellion or acting
out. One, for example, engaged in violent controversies with his father,
in which the index twin never participated. The index twins never showed
non-compliance; if they expressed it at all, it was passively, in such a way
that the parents could rationalize it as just another symptom of their
disability.

Since the twins appeared identical in so many ways, "the needs of all
concerned to find distinguishing identities for them led to a sharpening
and highlighting of such initial differences,11 note Drs. Pollin, Stabenau,
and Tupin. They reflect that if the healthy twins had not been present for
nearly constant comparison, the parents might not have perceived the
index twins as so weak or vulnerable, but might have accepted them as
normal. Their anxiety and intense involvement with the smaller twin often
led the parents to ambivalent feelings towards him. They tended to
project the negative side of their self-image onto this smaller twin moie
often than on the larger one.

This general pattern tended to hold true as the study progressed. In a
report on 1 1 families with MZ twins discordant for schizophrenia in 1 965,
the authors declared that each of the 1 1 index twins was the smaller one
at birth. The difference in birth weights ranged from 1/2 ounce to I
pound 1 2 ounces. This lower weight "appears to reflect some lower level
of anatomic development and/or differentiation, and consequently, a
lesser physiological competence and stability present at birth," they
reported. The index twins also had a marked preponderance of such
problems as cyanosis, infantile colic, feeding difficulties, burns, multiple
fractures, and severe illnesses which knocked them flat for months at a
time. Their parents recalled that these smaller twins had "worried more,"
cried more, been more "fussy," and seemed more sensitive. Almost
without exception, and from earliest childhood, the index twins were
described as more dependent, more submissive, more fearful, more
compliant, and more constricted than their heavier co-twins.

To help refresh their memories, the parents were asked to bring in
whatever pictures or home movies they had available. ("It seems all
families have large numbers of pictures in shoe boxes," says Dr. Pollin.
"They may never have put the pictures into albums, but they have
them.") Nevertheless, there remained the problem of retrospective
distortion. To alleviate it, a social worker lived with the families in their
homes a couple of days, then went to see as many relatives, doctors,
teachers, and friends as possible, accumulating up to 25 interviews per
family to get a more objective view of the past. Often she found that
many of those who knew the family recalled the bigger twin as an easier

316

child to deal with. Almost from birth, the smaller twin actually had quite
a different life course - as he must have had a different, less favorable
intrauterine experience.

It is quite normal for one twin to be smaller than the other at birth.
Thus, many parents fall into the pattern described above because of
realistic concern for a vulnerable child. However, they change this pattern
after the child gains in strength and maturity. It is only in certain cases,
when paients are too troubled or rigid to modify their picture of the
weaker twin, that the pattern becomes potentially dangerous. Apparently
something about the weaker twin's situation resonates with a particular
problem in these parents' lives, relighting an intense, unresolved conflict.
No matter how much the pediatrician may reassuie them, they remain
consumed with anxiety about the child. On the other hand, they find
dealing with the stionger twin an easy task - a distinct relief, In this way
the twins experience different models of parental behavior, and have
increasingly divergent experiences.

In a later report (1967), Drs. Pollin and Stabenau discuss two additional
sets of identical twins in which, surprisingly, the twin who became schizo-
phrenic did not weigh less at birth. They find it very significant that,
despite a favorable start, these children suffered from specific stresses
which reversed their relative position and, in effect, made them weaker
than the lighter twin. In one of these cases, the index twin had turned
blue from cyanosis because of poisoning from a defective heater next to
her bed and had been taken to a hospital, near death, while her co-twin,
who slept in another part of the room, suffered only minor effects. In the
other, the twin who was heavier at birth nearly died from a severe case of
Rocky Mountain spotted fever at age 3 and 1/2, after which there was
great concern regarding his health and survival.

Thus, whether it began before biith - through diffeiences in fetal
positioning and consequent crowding, differences in fetal circulation, or
other factois that produced relative physiological incompetence - or
through accident in early childhood, life presented very different
experiences to these "identical" twins. Between the ages of 2 and 6, the
stronger twin was usually the more verbal and the more independent.
From 6 to adolescence, the weaker twin often became increasingly
dependent on his healthier co-twin - who simultaneously began to turn to
others for friendship. In late adolescence and early adulthood, the healthy
twin moved further towards individuality and heterosexual relationships.
This accentuated a sense of loneliness and despair in the weaker twin.

"Often, disorganization, withdrawal, and schizophrenic symptoma-
tology develop in the lighter birth weight twin at just this time, i.e., when
the heavier, more differentiated twin is making a sudden spurt in the
development of an individual identity and the establishment of hetero-
sexual and genital level of personality organization," write Stabenau and
Pollin. To the index twin, the world had always appeared more
threatening - and its stresses mounted all around him as he grew up.

317

The Veterans' Study: 15,000 Pairs of Twins Discordant for Schizophrenia

In 1967, the Medical Follow-Up Agency of the National Academy of
Sciences-National Research Council made available to researchers its
national registry of all the pairs of white male twins born between 1917
and 1927 who served in the Armed Forces during World War II or t!ie
Korean War. It was a list drawn from the 54,000 pairs of twins born
during that decade. It excluded 23,000 pairs of twins who did not serve i\\
the Armed Forces at all, as well as 15,000 pairs in which only one twin
served. This left 15,930 pairs of twins healthy enough for both to liave
passed the physical and mental tests leading to induction into the service.

The records of these men yielded a wealth of information that could be
fed into computer tape: Induction physical examinations; inpatient and
outpatient hospital and clinic diagnoses during their period of services; VA
hospital diagnoses after they left the service; and diagnoses based on
responses to a questionnaire. They were between 38 and 48 years old in
1965, after a follow-up period averaging 1 8 to 20 years.

Among these 15,930 pairs of twins, Drs. Pollin, Martin G. Allen, Axel
Hoffer, and their associates found 338 pairs in which one or both twins
had been diagnosed as schizophrenic at some time after entry to active
duty. In some cases their zygosity was unknown, but 226 pairs could be
identified as identical (MZ) or fraternal (DZ) twins. Among the 80 MZ
pairs, 11, or 1 3,8 percent, were concordant for schizophrenia. Among the
146 DZ pairs, only 6, or 4.1 percent, were concordant for the disease. The
concordance rate for schizophrenia was thus 3.3 times greater among MZ
pairs than among DZ pairs. By contrast, the concordance rate for neurosis
was almost the same in both kinds of twins - 10.7 percent for MZ pairs as
compared to 7.1 percent for DZ pairs.

A re-analysis of 1 8 earlier major twin studies showed a similar ratio in
all but one instance, suggesting "the presence of a genetic factor in the
pathogenesis of schizophrenia, and its relative absence in psych on euro sis,"
report the researchers. "However, since approximately 85 percent of
affected MZ pairs in the NRC sample are discordant for schizophrenia, the
role of the suggested genetic factor appears to be a limited one."

This large sample of twins also allowed Drs. Allen and Pollin > to their
own surprise, to cast doubt on one of the most widely accepted notions of
the psychodynamic determinants of schizophrenia in analytic theory: the
"diffuse ego boundary" hypothesis. This holds that schizophrenia is due
to a "confusion of identity" resulting from weak ego boundaries, and thai
such problems are greater in twins, especially MZ twins. Therefore, MZ
twins could be expected to have a higher incidence of schizophrenia than
T>7 twins, and all twins would be expected to have a higher incidence ofit

hypothesis, they do not actually refute it, either, since possibly other
factors unique to twin personality development might offset whatever ego
boundary defects exist.

The Transmission of Schizophrenia

"It is very easy, in work with family dynamics, to say, Aha, this is the
difference in the behavior of the mother and father that leads to schizo-
phrenia - these are the schizophrenia-producing characteristics of parents.
However, that is not the way we think of it right now," warns Dr. Pollin.
Parental factors are only one among a variety of ctiologic factors that may
play a role in lowering resistance to the disease or triggering it, he
explains.

For years there has been an ideological struggle between psychiatrists
who believe that schizophrenia results from some genetic, biochemical
impairment, and those who believe it comes from faulty child-rearing. Dr.
Pollin sees a great need to integrate both points of view.

So far there is no definitive evidence that parental behavior is the
predominant factor, he points out. Whether or not it is a necessary factor
cannot be determined until enough prospective studies have been
completed. Much of the disturbance seen in the parents of schizophrenics
may be secondary to the disease, rather than a cause of it. All that can be
said with assurance so far is that certain family patterns seem to ac-
company schizophrenia. However, these patterns cannot be blamed for
the disease. It may be that there was some defect in the child from the
beginning - a defect we do not yet know how to recognize or define.

Nor are the patterns of family interaction that Dr. Pollin described the
only possible ones in the development of schizophrenia. They may repre-
sent only one of several different patterns leading to the same effect. By
choosing to work only with families in which both parents and both twins
were willing and able to come to the NIH Clinical Center for an extended
period of time, he narrowly limited his sample. As he makes clear, the
incidence of schizophrenia tends to be high among broken families and
those with a great deal of strife and schizms - the very kind he had
eliminated from his study. "We dealt with families where over-
involvement with a child and over-protectiveness were more likely to be a
factor," he notes, "If we had not required the families to participate, we
might have found other patterns, centered on more overt types of
rejection."

Illnesses can be passed from one person to another in many different
ways. They can be transmitted by a microbe, as in bacterial infection;
genetically, as in phenylketonuria; socially, in the sense that poor, socially
backward slum families live in conditions that are likely to produce a high
incidence of TB; or by various combinations of these factors.

To understand the transmission of schizophrenia, it helps to look at
other illnesses, Dr. Pollin believes. He particularly likes the model de-
veloped by two English researchers to explain the incidence of a very
different illness; congenital dislocation of the hip.

319

Though much simpler and more concrete, congenital dislocation of tli
hip clearly has genetic as well as environmental factors, as does schizc
phrenia. It runs in families, with identical twins concordant for it in 41
percent of cases. Yet the majoiity of twin pairs are discordant for tli-
disease. Drs. Cedric 0. Carter and John A. Wilkinson of the Mcdiea
Genetics Unit, British Research Council, were able to tease apait tin
vanous factors involved.

They found, first of all, that certain aspects of the anatomy of the liij
joint weie controlled by heredity. If the hip joint is visualized as a kind oi
modified ball and socket, the shape of the socket - its depth, and thcsiz*
of its roof - is the key to a good fit. Obviously, the shallower the sockcl
and the shorter its roof, the easier for the head of the thigh bone (the ball
in it) to pop out. Yet this shape was determined by a genetic factor. Thus,
a genetic factor produced an anatomical predisposition for the illness.

Next, investigating the fact the dislocation of the hip was eight or nine
times more frequent among girls than among boys, they found ;i gencwl-
ized laxness in all the joints of female infants, which they traced to a (low
of hormones from the endocrine glands of girl babies just before birtli.
These hormones temporarily loosened the infants' connective tissues.

The condition was also much more common among children born by
breech presentation, which bent their legs in a position that favored the
thigh bone's popping out, The custom of swaddling had similar effects,
mis-directing pressure on the baby's legs; this accounted for the high rate
of dislocation of the hip among certain American Indian tribes.

Among the Chinese in Hong Kong, on the other hand, dislocation of
the hip was rare. The researchers traced this to the custom of carrying
infants in a back sling which, far from loosening the hip joint, tended to
push the ball back into the socket.

When twins were discordant for the illness, it often turned out that one
twin had been carried in a back sling, while the other had not; one Iwd
been swaddled, and the other had not; or else, being of different sexes,
they had had different levels of hormones at birth. With MZ twins, Ilir
most common difference was their manner of birth: because of Ihc
intrauterine mechanics involved, one would be born by breech picscnla*
tion, while the other was not.

Though the differences between schizophrenia and congenital dis-
location of the hip are obvious, Dr. Pollin believes one can draw some
cautious but useful analogies between the two illnesses. In bolh cases,
there are many different pathways which lead to the same abnormal
structure, The development of the human ego, like that of the hip joint,
can be impeded by genetic factors, intrauterine mechanisms, environment
in early infancy, social customs, accidents, or various combinations
thereof,

It should also be noted that "schizophrenia" is a shorthand word for
various forms of mental illness, he points out. When Bugcn Blculer
introduced it in 1 9 1 1 , he used the plural: "The Group of Schizophrenia*."
It is still generally believed that schizophrenia includes several cliffcren!,
though overlapping, disease entities. Eventually, when the total picture

320

becomes clearer, several different patterns of family interaction and bio-
chemical characteristics may be recognized as leading to different forms of
schizophrenia.

The Possibility of Prevention

The fact that the majority of twins with the genotype for schizophrenia
do not become schizophrenic shows that some kind of intervention is
possible: Most life experiences do not lead to the development of schizo-
phrenia. "However, so far we don't even know what specific biochemical
changes take place at the time of the psychotic break," points out Dr.
Pollin. "Nor do we understand why LSD, in microamounts, can cause a
break with reality and a florid psychotic reaction. We are still fumbling in
the dark - as though we were trying to treat heart disease without
understanding the basic mechanism of the heart."

Within ten years, our knowledge of the factors that contribute to a high
risk of schizophrenia should be at about the same level as our present
knowledge of the various factors that contribute to a coronary, Dr. Pollin
speculates. He hopes that it will then be possible to state with some
certainty which combinations of factors represent a risk high enough to
warrant the use of preventive drugs, or which factors in early childhood
need to be modified.

"If you understand the specific steps that lead to a disease, you can
approach its therapy and prevention from a rational point of view,'* he
says. "But our treatment of schizophrenia has been quite the opposite — all
entirely empirical or accidental. People have stumbled on methods of
treatment. The tranquilizers were found during a search for a new type of
anti-histamine. If one depends on such accidents, the odds against finding
the most rational treatment and prevention measures are very bad. We
must define more precisely the specific factors that contribute to a high
risk."

Dr. Pollin cites the work of Dr. Sarnoff Mednick, Professor of Psychol-
ogy New York School for Social Research, New York, New York, with
children of schizophrenic mothers as an example of the kind of studies
which offers the best hope of finding clues to prevention. The children of
schizophrenic mothers are a high-risk population. Dr. Mednick expects
that, out of his first series of 200 children, at least 30 will become schizo-
phrenic within 15 or 20 years. However, instead of having to depend on
their relatives' memories, he will have a complete record of the children's
pre-illness characteristics and of the conditions under which they were
raised. This will allow him to see what differentiates the children who
become sick from those who stay healthy. So far, Dr. Mednick Js results fit
in very well with Dr. Pollin's for he has found a clear correlation between
the development of schizophrenia and the kind of pregnancy and birth
difficulties which might have caused damage to tlie patient's central
nervous system either before or during birth. Given a genetic predis-
position to schizophrenia, such insults to the central nervous system might
well be the factor that triggers the disease.

321

For the past three years, Drs. Pollin, M. Allen, and D. Cohen have *
carrying out a prospective study of their own to better uiuleislimil
development of personality characteristics and family lelationships win
when accompanied by other factors - might play a role in pioilm
schizophrenia. However, unlike Dr. Mednick, they do not anticipate
cases of schizophrenia among their subjects. They are simply studying
origins of certain patterns in the early life of twins. Having eunliu.
obstetricians in the Washington area about women who expected nuilt
births, they asked the parents' cooperation and then tlid [iron,
interviews in the parents' homes. One of the psychiatrists in tin- M?CI
was present at the time of each multiple delivery. If the twins v.
identical, the researchers kept careful jecords of exactly how I lie lu
differed at birth, how the birth process itself varied, the ncumlufi
findings on each, and their behavior in the nursery. They followed up
twins regularly, first every few weeks, then every few months at fun
and from time to time the twins were brought to the NIH Clinical ('en
for extensive tests and films.

"We wanted to define exactly when the differences between iileuli
twins became consistent. We found that one of them became mi
dominant, more skillful with objects, more comfortable with slr.iHfi
less fearful, at a very early age," declares Dr. Pollin. "Our fit si juimp
twins, 10 pairs, is now between 2 and 3 years old, and we aie oitltMi
our second sample of 10. We hope that by the time they are 5 VOM
we will be able to pull together some useful generalizations,"

Biochemical Abnormalities that May Predispose to Schizophrenia

Recent studies by Dr. Pollin have singled out abnormalities in I
catecholamine system as a genetically determined factor whidi in
contribute to the development of schizophrenia - and perhaps loiHli
psychoses. In Leningrad, in the summer of 1970, he ivpnilol I
preliminary findings on 19 pairs of identical twins, of whom II vu
discordant for schizophrenia, 4 were concordant (both selil/oplin-Jii
and 4 were normal controls. All were inpatients at the NIH C'link'ul fun
in Bethesda, where their diets were similar and samples for nearly iill |'j
of twins were obtained on the same day.

Each identical twin excreted nearly the same amount of cnlecliohimr
as his co-twin. The following catecholamines were analysed: l)<H\i?nr
(the precursor from which norepinephrine and epinephrinc lire nwimf;
tured); norepinephrine; epinephrine; normetanephrine; mclanephririiv
VMA. Theintraclass correlation coefficients- which measure
which intrapair similarity in identical twins is greater than the Mi
between persons who are not genetically related - were high and sU
ly significant at values that ranged from <.05 to <.001 , indicating ll>.it
levels of these substances were under a significant degree of ik'ii<-'i
control,

The similarity remained even when one identical twin WJK ^h"'-
phrenic and the other was not. For the 1 1 discordant MZ pairs nCb*)11

322

the intraclass correlation coefficients were +.79 for norepinephrine, +.77
for dopamine, +.62 for normetanephrine, and +.80 for metanephrine.
"The degree of genetic control present is not submerged or obscured by
the presence or absence of schizophrenia," leported Dr. Pollin. "It is
especially high for norepinephrine and dopamine,"

This was particularly interesting by comparison with the MZ twins'
discordant production of adrenal steroids. The catecholamines play two
important roles in the body: (1) Some of them (norepinephreine and
dopamine, those most clearly under genetic control) act as neurotrans-
rnitters in the central nervous system, e.g., norepinephrine in the hy-
pothalamus, the area where emotional activities aie integrated with higher
abstract activities, and (2) All of them are known to rise in response to
stress. However, they usually rise together with other responses by the
adrenal glands. Yet in this case, the healthy MZ twins did not show any
abnormally high level of 17-OH steroids; only their schizophrenic co-twins
did. Thus, the high levels of catecholamines in both twins could not be
attributed entirely to the stress of the immediate situation. In Dr. Pollin 's
hypothesis, they may represent one of the underlying factois that
predispose to schizophrenia.

Researchers have long thought that it would make sense if some of the
steps leading to serious psychopathology were located in the system that
responds to stress. For if one is stressed and becomes anxious, beyond a
certain point this anxiety can itself reduce, rather than increase, one's
ability to cope with stress. As one's ability to cope with stress decreases,
the external threat appears greater, leading to a greater response, and
eventually to a pathological spiral.

A man who sees a lion come charging at him responds with a sudden
rush of epinephrine (also called adrenaline) and other catecholamines.
This brings about many physiological changes. His blood is massively
shunted from other parts of the body into the muscular system, equipping
him to run twice as fast as he otherwise could. His energy is mobilized for
either fight or flight. Useful and adaptive as this response may be to a
caveman who often faces dangers requiring physical prowess, it becomes
self-defeating in our society, where the threats are more complex, subtle,
and chronic. Studies in Dr. Pollin's Lab have shown that high levels of
circulating epinephrine lead to a constriction of the perceptual field -they
make one see less. Again, the ability to focus on a wild animal to the
exclusion of everything else might prove quite useful in a jungle, but in a
complex situation it would be better to increase one's ability to make
subtle abstract distinctions — the kind of ability which is impaired by high
levels of epinephrine. These high levels of epinephrine also diminish the
precision of man's reaction to stress, leading him to over-react in in-
creasingly non-productive ways. In this fashion they can produce the stage
of exhaustion described by Hans Seyle, in which all ability to adapt to
stress is lost.

323

It is therefore quite possible that the hyper-secretion of catechol
impedes the development of the ability to cope with stress, and tin
to schizophrenia. This might be one of the mechanisms involved, t<
with the non-genetic factors discussed above. In cases of discc
among MZ twins, perhaps one of the twins develops techniques t
with the hypothetically higher level of catecholamines, while tht
does not; then, as a result of his larger number of successes, he devc
normal fashion. Meanwhile his less successful co-twin becomes inci
ly dependent, submissive, and constricted, and in the absence of
kinds of intervention from family or friends, the direction of his d<
ment may become irreversible.

Many interesting and important leads tend to implicate the cat*
mines in mental illness, Dr. Pollin points out. For instance, mosl
that are able to cause an artificial psychosis are methylated derivat
amines. Thus, mescaline is a methylated derivative of dop
Methionine, one of the essential amino acids, is believed to be es
specifically because it supplies the methyle group, a very common
which plays an important role in body chemistry. And it was sh<
this Lab nine yeais ago that when large quantities of methioni
-administered to schizophrenic patients, some of these patients will s
severe exacerabation of their psychosis. (This is one of th
biochemical findings about schizophrenia on which there is general
ment.) This leads to the hypothesis that some abnormal, mctti
metabolites of the catecholamines are formed in the body of i
persons because of a genetically determined fault; that one of
metabolites is chemically similar to a psychotomimettc drug; and til
metabolite - perhaps a methylated derivative of epinephrine — can
psychotic symptoms. However, this remains to be proved.

Another lead comes from Parkinsonism. One of the side effects
effective anti-psychotic drugs known to date is that they produce (r
similar to Parkinsonism. It was originally believed that these tremo
rigidity were an allergic response to the drugs. But it has now been s
that if anti-psychotic drugs are given in high enough dosage, sor
percent of all patients will develop Parkinsonism. The biochemical b;
Parkinsonism has recently been established - it is a deficien
dopamine, which among other things serves an essential neurotransi
role in the basal ganglia, the part of the brain associated with ]
activity. So Parkinsonism implies a lower level of dopamine. A lowci
of dopamine implies a lower level of all the catecholamines, since i
body dopamine is the substance from which the other catecholamin
derived. The drug L-Dopa treats Parkinsonism by increasing the le
dopamine. Interestingly, it has had ill effects on some schizopl
patients. In the past two years it has been used on a fair number of
patients, not to treat their illness, but to treat the severe Parkinsonisi
symptoms developed by some 2 or 3 percent of schizophrenics as a
of liigh dosages of thorazine. In a significant number of cases, L-

324

exacerbated the psychosis. This, once again, seems to indicate that in-
creased levels of dopamine and other catecholamines help to produce the
symptoms of schizophrenia.

If indeed an inherited biochemical fault involving high levels of catechol-
amines predisposes certain persons to schizophrenia, it is a fault that
might become manifest only if the system were working at high pressure —
which brings one right back to the intrauterine experiences and family
factors with which Dr. Pollin began.

"There is something unfortunately seductive about biochemical work,"
notes Dr. Pollin, an analyst who does not find the analytic and
biochemical approaches conflicting, but complementary. "At least it can
be quantified ! We have fascinating life histories about our twins, each one
a novella, but presenting it in a hard, piecise way is infinitely more
difficult."

He points out that whether the research is approached from the angle
of biochemistry or that of family patterns, two issues remain unclear:

1 ) Can his findings be replicated with a larger sample? and

2) Which of his findings aie specific to schizophrenia, rather than to all
psychoses?

His next study will try to get at these issues. From the large pool of
twins made available by the NAS-NRC, he hopes to select four matched
groups of twins: one group of twins who are discordant for schizophrenia,
another who are discordant for depression, a third who are discordant for
severe neurosis, and another who are normal. Each group will consist of
about 20 pairs of twins. Through extensive biochemical and psychological
comparisons of these four groups, Dr. Pollin hopes to determine which
factors are specific to schizophrenia, and which arc common to all severe
psychopathology.

Research Grant: Intramural

Date of interviews: December 1970, January 1971

References:

Allen, M., and Pollin, W. Schizophrenia m twins and the diffuse ego boundary hy-
pothesis, American Journal of Psychiatry, 111 (4): 43 7-443, 1970.

Hoffer, A. and Pollin, W. Schizophrenia in the NAS-NRC panel of 15,909 veteran twin
pairs, Archives of General Psychiatry, 23:469-477, 1970.

Mosher, L., Poltin, W., and Stabenau, I. R. Identical twins discordant for schizo-
phrenia: Neurological findings. In press, A rchives ofGeneiaiPsy chiatry.

Mosher, L., Pollin, W., and Stabenau, J. Families with identical twins discordant for
schizophrenia: Some relationships between identification, thinking styles, psycho-
pathology and dominance-submissiveness. British Journal of Psychiatry t 1 18:29-42,
1971.

Pollin, W. The unique contribution of twin studies to the elucidation of non-genetic
factors in personality development and psychopathogenesis. Acta Genetic Medicae
et Gemellogie, 19 (l-2):299-303, 1970.

Pollin, W. and Stabenau, J. Biological, psychological and historical differences in a
series of monozygotic twins discordant for schizophrenia. In Kety, S. S., and
Rosenthal, D. (eds.). The Transmission of Schizophrenia London, Pergamon Press
Ltd., 317-332, 1968.

Pollin, \V. Possible genetic factor related to psychosis. For presentation May 7, 1971,
Annual Meeting of the American Psychiatric Association, Washington, D.C.

325

W. A new approach to the use of twin study data in studies of the pathogen.

of schizophrenia and neurosis. In Kaplan, A. (ed ) Genetic Factors in Scht

phrenw. To be published by Chas. C. Thomas Press
Pollfii, WM Allen, M. G.t Hoffer, A., Stabenau, J. R., and Hrubec, Z. PsychopaJhdc

in 1 5,909 pairs of veteran twins. Evidence for genetic factor in the pathogenous

schizophrenia and its relevative absence m psych on eurosis. American Journal

Psychiatry. 126, #5:597-61 1, 1969.
Stabenau, J. R. and Pollin, W, Early characteristics of monozygotic twins discordant I

schizophrenia. Archives of General Psychiatry., 17:723-734, 1967.
Stabenau, J. R and Pollin, W. Maturity at birth and adult protein bound lodir

Nature, 215.996-997, 1967.
Stabenau, L R. and Pollin, W, Comparative life history differences of families i

schizophrenics, delinquents, and "normals." American Journal of Psyclttatr)

124:11, 1968.
Stabenau, J. R., Pollin, W., and Mosher, L. A study of monozygotic twins disco «l a r

for schizophrenia' Some biologic variables. Archives of General PsycJitatr)

20:145-158,1969.
Stabenau, J. R., Tupin, J,, Wener, M., and Pollin, W. A comparative study offiimilit

of schizophrenics, delinquents, and normals. Psychiatry, 28(l):4S-59, 196S.

326

We shall do so much in the years to come,
But what have we done today?
We shall give our gold in a princely sum.
But what did we give today?

-Nixon Waterman

Investigator:

Joseph D. Teicher, M.D.

University of Southern California

School of Medicine

Los Angeles, Calif.

Prepared by

Gay Luce

Using interviews and psychological tests, a research team has compared
50 adolescents after an attempted suicide, with unsuicidal peers of the
same age, sex, and background. Although economic privation, broken
homes, and disciplinary problems were found in the control group-the
sequence and timing of events occurred at a different phase in the devel-
opment of the child. The profile of the suicidal adolescent includes long-
standing problems with family, a stage of escalation during adolescence,
and a final stage of alienation-a chain reaction that dissolves the adoles-
cent's closest personal bonds. Given detailed biographical knowledge of an
adolescent, this study indicates that it should be possible to pick out the
youth in danger, for adolescent suicide is not irrational but over-
determined by sequences of life events occurring in critical periods.

Background

Adolescent suicide is horrifying, unthinkable, and a little unreal to most
adults, for we tend to be complacent about the troubles of the young. To
the modern adults, Romeo and Juliet may seem only a story. Yet many
adolescents cling to one another in similar love, with the desperation of a
last hope in a lonely world. A modern Juliet is likely to be a frightened
and pregnant little girl; the boy is likely to be rejected, and both may feel
totally alone.

Literary descriptions of childhood suicide seem bizarre, yet they re-
semble modern case histories. In Thomas Hardy's Jude the Obscure, the
restless wanderings and misery of unmarried parents overcome an un-
wanted oldest boy. When he hears that yet another unwanted baby is
coming, he kills himself and the other children. It is not that such events
don't happen, but we are reluctant to believe them.

In 1965, Jacobziner estimated that there were 60,000 attempted sui-
cides among young people under age 20 in the United States each year.

329

Adolescence can be a particularly lonely and difficult period, a time of
biological upheaval and social change. A person is expected to emerge
from the safety and dependency of childhood into responsible maturity.
Even healthy and happy adolescents become moody and oscillate between
passions and depressions in a manner that the older people around them
rarely understand. Most adolescents have fantasies about killing them-
selves in moments of rage and frustration or when they feel totally iso-
lated from their families and friends. This is not surprising. Who has not
imagined, with some glee, the remorse his parents would feel if he killed
himself? Between such imaginings and the act lies the world of pathological
events that Doctor Teicher and his associates have begun to define.

Statistics portray great misery among a large population of adolescents.
Suicide ranks as the fourth most frequent cause of death for young people
15—19 years old. Fortunately, the vast number of attempted suicides in
this age group are thwarted. An estimate of 60,000 suicide attempts a year
may seem exaggerated, but hospital admissions offer a convincingly sad
picture. In 1960, for instance, at New York's Bellevue Hospital attempted
suicide was the reason for admitting 10 percent of the child and adoles-
cent patients. At Kings County Hospital in Brooklyn, 13 out of every 100
children who came to the hospital had attempted or threatened suicide.
Each month, the huge Los Angeles County-U.S.C. Medical Center admits
about seven patients between 14 and 18 who have attempted to kill
themselves; over 80 a year.

The Attempted Suicides

There has been a general tendency to dismiss a suicide attempt in an
adolescent as an impulsive act stemming from a temporary crisis or depres-
sion. Perhaps it is soothing to believe that someone so young with "life
ahead of him" could not have intended to kill himself. He could not have
considered that he might die. On the contrary, Doctor Teicher and his
associates at the Medical Center of the University of Southern California
have found many adolescents who attempted to take their lives more than
once. At first they may have used the drastic move as a threat to draw
attention to their problems. Instead, it generally made matters worse.
After an escalation of long-standing problems and loss of any meaningful
relations, many concluded that death was really the only solution lo
unsolvable, unbearable, and chronic problems.

Beginning with Freud around 1920, many keen minds in the develop-
ment of psychiatry have wrestled with the problem of adolescent suicide,
but inferences drawn from a few cases or psychological studies did not
indicate how to predict a suicide from outside circumstances. In the fall of
1964, the investigator and his associates began to study the life situations
of adolescents who attempted suicide, comparing them with control
adolescents matched for age, race, sex, and family income -control
adolescents who had never attempted suicide, Quite a few interesting
patterns have been drawn from this study of 50 young people who

330

attempted suicide. All were between 14 and 18, None of them was
mentally retarded or obviously pregnant. All had been brought into the
Los Angeles County-U.S.C. Medical Center sometime between September
1964 and May 1965 because of their suicide attempt.

At least one parent, usually the mother, was studied as well. For
comparison there was a control group of 32 youngsters and their parents.
Three-quarters of the attempted suicides were girls. On the average the
suicidal adolescents were around 16 years old. They were white, Mexican,
Negro, Protestant, Catholic, and Jewish.

Procedure: Charts of Life Events

The procedure called for an interview with the adolescent patient
within 24 to 48 hours after the suicide attempt. The parent or parents
were also interviewed. Then, the suicidal youngster's therapy sessions in
the hospital were taped and transcribed for further analysis.

Two biographies were elicited from structured interviews. There was
the parent's version of his child's history, and there was the adolescent's
version of his own life. On the basis of the case histories, a life history
chart was constructed for each suicide attempter and his matched control.
Tliis was done by constructing a chronology (in parallel) on a vertical
continuum that depicted all the experiences of the adolescent from birth
until the suicide attempt. These graphic charts show residential moves,
school changes, the beginnings of various behavioral problems, separation,
divorce, or remarriage of the parents, and deaths in the family. The charts
were put in a sequence that displayed how the events tended to pile up at
a particular point in the adolescent's life. This indicated how the crises
had accumulated during the adolescent's life.

What events distinguished those who attempted suicide from those who
did not? A simple comparison of events in the lives of the control group
and the suicide-attempters might not show that there was a very
pronounced difference. The investigators discerned a distinct process
leading to progressively deeper unhappiness and pessimism. The suicide-
attempters went through a sequence that led to progressive isolation from
the important people in his life. The control adolescents did not. The
process can be summarized in three stages: The suicide-attempters all had
a long-standing history of problems from chidlhood into adolescence.
There was also a period in which problems seemed to escalate, usually at
the very beginning of adolescence. Moreover, the problems mounted in a
manner that seemed to exceed those of peers and friends, Finally, came a
phase characterized by a "chain reaction dissolution of any remaining
meaningful social relationships." This isolation occurred in the days and
weeks preceding the suicide attempt.

331

Sequential Analysis of Life Events

The advantage of looking at things sequentially can be demonstrated by
comparing the two groups. For instance, the life histories of the suicide-
attempters showed that 72 percent of them came from bioken homes, yet
53 percent of the control group also came from broken homes. Former
studies of suicide have emphasized the fact that there were more broken
homes among suicide attempters than "control" adolescents. However
none of these studies examined the broken homes of comparison groups,
If one looked only at the incidence of broken homes and severed parental
relations, there is no great difference between suicidal youths and
comparable nonsuicidal youths. However, by looking at the chronological
biographies of these two groups, the grantees have seen that the relevance
of a broken home depends upon when the instability occurred in Ihc
child's development.

Critical Phase

Although 72 percent of the suicide-attempters and 53 percent of Ihc
control adolescents came from broken homes, the timing of divorce and
remarriage was different. In the suicidal group 58 percent of the pnienls
remarried, but only one-fourth of the control parents remarried. More-
over, these control parents managed to remarry very early in the child's
life and remained married. The parents of the suicidal adolescent cilEter
remarried quite a bit later in his life, or, if they remarried early, they were
subsequently divorced and remarried several times again.

The chronological mapping of biographies shows that the suicidal
adolescents ha'd parents who were divorced, separated, or remarried after
the onset of adolescence. By contrast, the control families experienced
change earlier, if at all. Instability in the home apparently had ;i dif-
ferential effect depending upon the age of the child. Both groups ex-
perienced the instability of a broken home, but the nonsuicidal adoles-
cents had a stable homelife during their last 5 years, while the suicidal
youths had experienced instability then. As the investigators have
written:1

'This is particularly significant, not only because divorce.

person to depression and suicide in later life. Loss of a love object, as the
grantee has remarked, is an important aspect of the process. But loss must
be viewed as a part of the process, and particular attention must be paid
to the time when it occurred. Most of the adolescents began their
maelstrom descent toward suicide after a long period of alienation from
parents. One 14-year-old who had tried to commit suicide twice was asked
why. She replied, "It's my mother."2

Asked what her mother did, she answered, "We just don't get
along. We haven't for 3 years. Before that we were like sisters and
then it seems like since she divorced my stepfather it started a lot of
trouble."

This girl enjoyed being in the hospital and did not want to return home.
It is particularly poignant that she wanted to be committed to a State
mental hospital rather than return home. Many of the young suicide at-
tempters described their alienation from parents as a process in which
either the mother or father would nag them, would cut them off from
their friends, would disapprove of their favorite friends, and thus made it
difficult for them to have relationships outside the home, at the same time
making life very difficult for them within the home. This was their
version.

The Broken Romance

Typically, many of these adolescents had fallen in love and formed very
possessive and exclusive romantic relationships. This actually isolated
them even more. A girl and boy would concentrate so intensely on one
another that they tended to cut off all their friends. Then, if the romance
failed, they would feel hopeless, lost and despairing.

At the time of the interviews none of the adolescents in the control
group was ending a romance, but a number of the "suicidal adolescents"
had just broken a romance. Moreover, five of these girls were either
pregnant or feared that they were pregnant. As the biographies revealed,
pregnancy inevitably led to a great sense of isolation. These girls withdrew
and were rejected by their boyfriends. Usually, they were also rejected by
their parents at this time when they most needed support. The suicidal
adolescents were really in a state of depression compared with their
counterparts, and, indeed, as the grantees point out, this seemed to have
been prompted by their real experiences in life.

The Way They Saw It

Only 38 percent of the suicidal youngsters considered their childhood
to have been happy. But about 94 percent of the control group considered

Unpublished transcript.

333

childhood to have been a happy time for them. In describing the

biographies, the investigators wrote:3

"Judging from the verbatim accounts of the suicide-attempters in
the interviews as well as the suicide notes left by them, and notes
written by other adolescents outside our sample, the decision to
suicide was the result of a rational, decision-making process.
However, the choice of death is not based on a desire to die, They
would, if they could, choose to live, Death, in a sense, is not chosen
at all but results from the progressive failure of adaptive techniques
to cope with the problems of living, where "the problem" is the
maintenance of meaningful social relationships. In short, the
potential suicide felt he had no choice, i.e., death is necessary. It is
from this recognition of necessity that his sense of freedom stems
and immediately preceding the act itself there is often a feeling of
well-being, a cessation of all cares. This is evidenced in the matter-
of-fact presentation found in suicide notes.11

Profile of Problems: Disruption at Home and Discipline

Early in childhood or adolescence the suicidal youngsters usually ex-
perienced the break-up of their home. In some cases this meant the
institutionalization of the child or a family member. Many of them were
placed in foster homes or left with relatives. Many of them changed
schools and residences frequently. Many of these families were very poor.
In some cases, the parents also had been depressed and had attempted
suicide. A sizable percent of the suicidal youngsters had either a parent,
relative, or close friend who had attempted suicide. Seventy-two percent
had one or both of their natural parents away from home, eithei because
of divorce, separation, or death. Most of those living with stepparents felt
they didn't like the stepparent. A great many had a parent who wfls
married several times. In about 62 percent of the cases both parents were
working. Half of these families lived on less than $3,600 per year. The
background is one of poverty, instability, and unhappiness.

The specific period just preceding a suicide is characterized by a vicious
spiral of events. It may begin when a parent feels unable to cope with
some behavior in his or her adolescent. The parent begins to nag and use
severe disciplinary procedures to prevent the youngster from going out
He may resort to physical punishment. Parents of the suicidal adolescents
*u«* <-*-e[T children would get into less trouble if they were watched

view of the adolescents (as revealed on a rating scale), withholding privi-
leges, fussing, nagging, and whipping were considered the worst discipli-
nary techniques. The suicidal adolescents and their nonsuicidal counter-
parts agreed on this rating. At the same time, some of the adolescents felt
they would gladly forego undesirable behavior, and their parents should
have helped them to discourage this behavior. When the parents didn't
intervene, the young people took it as a sign of rejection.

As the parent-child situation got worse, the parents grew frustrated, and
the adolescent felt that his parents couldn't understand and were punish-
ing him inappropriately. The biographies revealed that this impasse led to
the adolescent's rebellion or withdrawal. This stage of deterioration usually
led to a breakdown of communication between parent and child, in which
the youth's withdrawal was a consequence. Essentially, both parent and
adolescent would give up and stop trying to communicate.

Many suicidal adolescents said that they got into the habit of lying and
would simply withdraw into their rooms, or withdraw into themselves in
order to avoid their parents and conflict.

School

A third of the adolescents who had attempted suicide were out of
school at the time. Either they were ill because of pregnancy or because of
an earlier suicide attempt. An astonishing number had already attempted
suicide in the past. A quarter of these suicidal adolescents had been out of
school because they were acting up in class, had shown some emotional
instability, or had been involved in fights. Half of them had been truant
from school during the last 5 years because of lack of interest or active
distaste.

To Whom Do You Turn in Time of Trouble?

When asked to whom they turned when they were in trouble, a quarter
of the suicidal adolescents said there was no one to turn to. None of the
control adolescents felt such isolation. The pathos and the loneliness of
the suicidal adolescent is very dramatically shown in some of the figures.
Of the 46 percent who reported their suicide attempt to other people, less
than half reported it to their parents. Almost two-thirds of them talked to
people other than family members. This is particularly significant since 88
percent of the suicide attempts occurred at home, very often with the
parents in the next room. In every instance, the lack of communication
between family and the child and lack of communication with peers was a
very important factor in the period leading to suicide. On interview, these
suicidal adolescents conveyed the despairing sense that death was the only
solution, there was no other way out. Consider these excerpts from a
letter by a 17-year-old Negro boy to his father. This note was written the
evening before he made his second suicide attempt:

"Dear Father, I am addressing you these few lines to let you know
that I am fine and everybody else is and I hope you are the same.

335

Daddy, I understand that I let you down and I let Mother down
the same way when I did that little old thing [the suicide attenip
that Wednesday night. Daddy, I am sorry if I really upset you, 1>
Daddy after I got back I realized how sad and bad you felt when
came back to California. -I had lost my best girl the week before
did that. I had a fight because some dude tried to take advantage <
her when I sped to the store, so I came back and I heard a lot (
noise like bumping so I run in and there he is trying to rape my gir
my best one too.-Daddy 1 tried as hard as I could to make it chcoi
ful, but it does get sad. Daddy I am up by myself. I've been up &
night trying to write you something to cheer you up, because I couli
see your heart breaking when you first asked Sam's wife if the;
would have room and that Sunday Dad, it was hard but I fought tin
tears that burned my eyes as we drove off and Daddy part of ni]
sickness when I had taken an overdose I did just want to sleep mysel
away because I missed you Dad.

"But when I left I felt like I had killed something inside of yoi
and I knew you hated to see me go, and I hated to go, but Daddy,
well, I kind of missed Mother after I had seen her. I miss yon niul
remember what you said, 'settle down', but Daddy I tried so hard so
I went and bought some sleeping pills and took them so both of you
could feel the same tiling."4

When an adolescent has retreated from family problems into a love
affair, and then the romance breaks up or culminates in pregnancy, then
there is even more isolation than before. A girl is especially alone if her
boyfriend disappears and she has already alienated other friends. Pnrcnls
often become disillusioned and give up at the time their child needs help
the most. In a letter to her former boyfriend, a desperate young girl
showed the lengths to which she would go for a social relationship and a
solution to the problem of pregnancy. She wrote on the night of n suicide
attempt. A short excerpt indicates the tragic sense of rejection and isola-
tion.

"Dear Bill, I want you and I to get an understanding about certain
things because I think you got the wrong impression of me * * * and
believe me it hurt. I knew all the time you were hinting to me I was
too young, didn't know nothing about life, but you were wrong. I
know a whole lot about life. I'm ashamed of the things I know to be
so young. I couldn't tell you this personally, 'cuz I couldn't face
what you might have said and I sure it would have hurt my feelings
badly .^ I'm two months pregnant by you. You don't have to admit it,
1 don t care. You may say anything you like. You don't have to
worry about any trouble. It would be a disgrace forme to let people
know I threw myself on you knowing you didn't care or feel anyway
toward me. Don't worry, no one will ever know my child's father. I
will never mention you to him or her whichever it be "5

CfT' J: D", and Jacobs> Jl Ado^scents who attempt suicide: Preliminary
. American Journal of Psychiatry, 122(1 1):S, May 1966.

Teicner, J. D,, and Jacobs, J. Adolescents who attempt suicide. American Journal
of Psychiatry, 122(11), 1966.

336

Parents and Physicians: Surprised

Despite the history of increasing problems, the families were inevitably
hurt and surprised by the suicide attempt. Parents and physicians who had
seen the adolescents would say "it was so unexpected." Actually, some 46
percent of the suicide-attempters had visited their physicians at some time
before the attempt. Over half had been treated for some physical or
mental disturbance during the prior 5 years. A third had some serious
physical complaint, and a third of them had some family member who
was sick or had been hospitalized. In screening the adolescents to be
included in this study, Doctor Teicher and his associates examined over
100. In the first 30 they found 1 1 with duodenal ulcers.

In spite of the long history of problems, however, the physician and
mothers acted surprised by the suicide attempts. While perhaps expressing
some guilt, the mothers would deny that there was anything in the home
situation that would cause a suicide. The very people who were closest to
the suicide-attempters apparently failed to see the progression of social
isolation: the problems with parents, with poverty, broken romances,
excommunication from school or peers, especially in the instance of preg-
nancy. Since these are problems that most people would be reticent to
discuss with others, adolescents in such predicaments are especially iso-
lated.

After a period of not communicating, their first suicide attempt came
as a surprise to parents, friends, and schoolmates. The physicans who saw
them just after the attempt had been taken off guard perhaps because
suicidal people are not easily distinguished from others with severe prob-
lems. There seem to be no simple and convenient ways of anticipating a
suicidal attempt. No litmus test can determine who is a potential suicide.
Clearly a major reason that suicidal attempts are not warded off is lack of
communication of the real feelings. The true biography of the unhappy
person was not known by anybody around him.

Profiles for Prevention

Adolescence is a time of sufficient duress for parents and youngsters as
new behavioral problems arise. Moreover, many of the suicidal youngsters
in the Los Angeles study also had illness or mental illness in their family
during the preceding 5 years. Doctor Teicher and his associates feel that
various sets of events must be considered in anticipating suicide. Among
them are such factors as economic status, geographic mobility, and the
divorce rate in the home. These alone do not predict suicide. However,
these events seem to occur at particular times in the adolescent's life and
the timing may be critical. Along with an escalation of behavioral prob-
lems, a youth who Is isolated from family and peers may be in danger of
trying suicide.

It should not be surprising to learn that their parents also had unhappy
histories. The mothers often got married only because they were pregnant.

337

Some had illegitimate children. Quite a few suffered depression and were
depressed after giving birth. Tin's was particularly notable among the
mothers of the boys who had attempted suicide. Many had illegitimate
children or had been forced into marriage because of pregnancy. Seventy
percent of them were separated or divorced, a good number of them after
short-lived marriages of convenience. Needless to say, a huge percentage
had suffered from economic deprivation.

Male Suicide

The number of suicides and suicide attempts among girls far outweighs
the number of attempts among boys; and this has been associated with
broken romances, rejection, and unwanted pregnancy. In attempting (o
understand the male suicide attempts, Doctor Teicher and Dr. N.L Mar-
golin did a special study of 13 of the boys in their group. They were
interviewed by one of the authors after their suicide attempt. Identical
questionnaires about parent-child relationships and school, about adjust-
ment to peer groups and career aspirations were given to the boys find
their parents. Both took a battery of psychological tests in addition.

The boys in the control group also came from broken homes. Many had
both parents working and relatives living with the family. However, the
vignettes of the suicidal boys differed in that they showed a repeated
sequence of events which the authors summarize in this order: They had>
first of all> a mother who was angry, depressed, or withdrawn, both before
and after pregnancy. Generally it was an unwanted pregnancy. Then, there
was the loss of some very significant person or persons in the patient's
early life, usually the loss of the father. There was also a reversal of roles
with the mother. At the time of the suicide attempt it had seemec! to the
boy that the mother (or his mother-surrogate) was also going to leave his
life forever. During the boy's period of distress his mother was pre-
occupied with her own depression, up to the time of her son's suicide
attempt.

An 18-year-old Mexican-American boy is typical. His mother never
wanted him. She became very over protective until he was about age 12,
At age 5 his semi-alcoholic father left the home. At this point he andliis
mother began to sliift around from house to house, mostly living with his
grandmother. After the divorce he began to get headaches. His mother
thought he missed his father. He always felt rejected, and he made de-
pressed statements such as: "I wish I hadn't been born." Then at the age
of 15 he was rejected by a girl. This left him emotionally fractured. H«
would get into romances where he was inevitably hurt and depressed. His

"I was a very blind and stupid woman. I didn't realize what I was
doing to Tom, how sensitive and emotional he was. Well, time went
by and Tom started to go to parties and dating, not too often, but he
had started to have friendships on the outside. Soon after I met
someone at work from the same department and we got along real
well. He was divorced also. He has a family of three to support, so we
have quite a lot in common. The man moved in. He liked my son and
went out of his way to cultivate him, but apparently things went
along very well until Tom started to complain that since Sam had
moved in with us he was nothing around the house just in my way,
that I didn't love him any more, but that was not true."6

In a pleading letter to the doctor she asked what he could do to undo
the damage she had done her son at an early age. Here was the tragic
pattern of events— the unhappy circumstances around his birth, the di-
vorce, his father's withdrawal, infantile identification with the mother,
frequent moves, repeated loss of peer relationships, the clinging to an
angry and depressed mother, and, finally, the threatened loss of his
mother to a new man. Case after case revealed this kind of dependency
and frustration in the first years of life. In 1 1 cases the fathers were
physically absent from the home. In eight instances the father had left
home before the child was 6 years old. Almost all of these boys were
prevented from being children. They were thrust into the role of helping
their mothers either because they were the oldest or the only child. In
each case there was also a sense of loss on the part of the child, either
because the mother and father had just recently separated, because the
mother had a serious illness, or because a stepfather had just recently left
home. In one instance, the mother had just recently married, and the boy
had been left by his girl friend.

"On the basis of our data we find that the male adolescent suicide
attempt seems to have its origins in the mother-child relationships of
infancy. Most importantly, these relationships revealed not only
early deprivation, but chronic repeated separation threat or object
loss. This state of affairs leads to continued, intense, archaic identifi-
cation with the mother. The lack of a masculine image in the experi-
ence of these boys together with the ambivalence of the mothers
prevents any working through of the Oedipal phase of develop-
ment."7

A helpless and dependent child needs his mother and cannot "allow" her
to be bad. He then blames himself for anything wrong in the environment,
which allows him to soak up the badness, as it were, making things around'
him all right. The investigators suggest that this situation eventually cre-
ates a self-destructive pattern.

"The early and repeated separation trauma resulting in disturb-
ances in early ego and superego development lay the foundation for
later pathological identification, and leave their marks on character

^Margolin, N. L., and Teicher, J. D. Thirteen adolescent male suicide attempts.
Journal of the American Academy of Child Psychiatry ; 7(2):30I, 1968.

7Op.cit. p,312.

339

formation and personality development. As the child enters adoles-
cence, the conflicts over separation intensify due to a number of
concurrent reasons, all of which essentially have to do with the
biological and psychological need to be autonomous from the
mother. The adolescent male tries to defend himself against feelings
of his helplessness in many ways. He may regress to feelings of
omnipotence and pseudoindependence and seek challenging, danger-
ous situations such as reckless driving, motor-cycling, etc. He may
act out antisocially as a defense to prevent loss of identity. How-
ever, it seems that these defensive attempts cannot be maintained
when actual separation from the mother is threatened. This threat
can occur in the form of the mother's withdrawing because of her
depression, her becoming interested in a new husband, etc, Also
significant is the breakup of the adolescent's romance, i.e., experi-
encing the loss of a mother surrogate. When the mother becomes
depressed and suicidal, the adolescent perceives rightly that his very
existence is a burden upon her, He acts as if he were saying, 'If I
destroy the bad part of myself, then mother will live to care for me.'
"Internally, ego regression with splitting occurs. The split-off
part of the ego, representing the bad self, is rejected and persecuted
by the parts of the ego and superego identified with the rejecting
suicidal mother. This Identification is of great significance in Use
suicidal adolescents. Freud (1923) states that the ego, feeling hated
and unprotected by the superego, will let itself die, a situation that
is similar to the anxiety in infantile separation from the mother."8

In these 13 cases, the boys professed to love their nagging and ambiva-
lent mothers, They did not necessarily feel they were loved, but because
of an infantile dependence, the mother's depression, anger, withdrawal, find
disapproval had a very devasting effect upon them. In many instances, the
mother also had suicidal thoughts, and the boys identified with their
mother's depressed and suicidal state. Interestingly enough, the suicidal
girls described their mothers in uniformly glowing and idealized terms and
denied any flaws, despite the fact that their mothers were often very
hostile.

"The suicide attempt is an overdetermined symptom and whether
it is an attention-getting or an attempt to die it is always serious. It is
an effort to solve a chronic problem, living; a plea for help; an
expression of rage and hostility; and at times a symbolic reunion
with the pre-Oedipal mother or father."9

Therapy

In many ways the therapist in the hospital has proven to be the lifeline
of these youngsters. He maintains his contact with the suicide attempters
from the beginning of consultation until final rehabilitation or referral.

"Op.cit. pp. 312-313.

9Teicher, J. D. The treatment of the suicidal adolescent-the lifeline approach.
Proceedings of the JV World Congress of Psychiatry, p. 749, Madrid, September 1966,
Excerpta Medica International Congress Series No. 150.

340

When they are first brought to the hospital they are shaken, anxious,
depressed, insecure, guilty, and apprehensive because of the anger and
hurt that they've caused. They feel terribly alone, and this is probably
their worst agony. Usually the mother has been angry and sometimes
guilty; her next reaction is usually hostile and she will defend herself with
great denial. The father, or more usually the stepfather, would consider
the suicide attempt a bother and show little concern. Doctor Teicher
recommends that suicidal adolescents should be hospitalized, if only
briefly, and placed in a ward where there are other adolescent patients to
offer warmth, support, and understanding. In many instances the patients
of this study didn't want to leave the hospital, and they would cling to the
staff and other patients. Adolescents will often talk about the precipi-
tating events, such as their parents' refusal to let them go out, or a broken
romance. The rejection by a boyfriend or a girlfriend is a most common
precipitating factor, but this would be taken in stride as an unhappy
experience if there had been some positive experiences earlier in life. The
role of the therapist as seen by the investigators is that of a person who
provides understanding and love. Slowly the therapist can guide a young
person to cope with his conflicts and communicate with his family. Mean-
while he offers support and is always available so that the adolescent
doesn't feel so lonely and isolated.

From this study one may clearly see that youth, itself, is no antidote to
a hostile environment. The old myth that all suicide attempts are impul-
sive and irrational is forever banished, and in this study one can see how
an accumulation of adverse factors at a critical period shapes the biograph-
ical profile of the potential adolescent suicide. This profile might be used
in further studies to predict and prevent suicide attempts.

This brief research has already shown that no simple correlations be-
tween life events can predict suicidal despair in a young person. Yet young
people-in shockingly vast numbers-are miserable enough, and lonely
enough that they are brought to hospitals by the tens of thousands each
year, after attempting to kill themselves, often in a room right next to
their parents.

Further research in this area has implications beyond suicide preven-
tion. The development of biographical profiles may yield techniques
whereby informed doctors, social workers, and school personnel might
spot the precarious young person in time to obtain therapy for him.
However, the import of this research is broader in its implications. It
begins to fold back the curtains upon the circumstances and the timing
that weaken an individual to the stresses of life and alienate him from all
of those who might help him. The chain of misery seems to pass from
one generation to the next, and in each case privation plays its part.
Moreover, the relations of family members show a psychodynamics that
produces instability and separation instead of cohesion and mutual help.
Adolescence can be an especially creative and exciting time of life. In this
particular era, adolescents are having an ever-increasing impact upon
society— they have changed the entire genre of popular music, for
example-but exceedingly great numbers of adolescents are having the

341

opposite experience. Suicide prevention studies among the most unhappy
of these people may give considerable insight into what it takes to deflect
an entire life from misery toward productiveness and participation.

Research Grant: MH 1432

Date of Interview. September 1968

References:

Jacobs, J , and Teicher, J. D. Broken homes and social isolation in attempted suicides
of adolescents. International Journal of Social Psychiatry, I3(2):140-149, 1967.

Jacobziner, H. Attempted suicide in adolescence, Journal of the American Medical
Association, 191(7):1 1-14, 1965.

Margolint N. L.> and Teicher, J D. Thirteen adolescent male suicide attempts. Journal
of Child Psychiatry, 7(2):296-314, April 1968.

Teicher, J. D. The treatment of the suicidal adolescent-the lifeline approach. Proceed-
ings of the IV World Congress of Psychiatry, Madrid, September 1966, Exccrpla
Medica International Congress Series No. 150.

Teicher, J. D., and Jacobs, J, Adolescents who attempt suicide: preliminary findings.
American Journal of Psychiatry, 122(11). May 1966.

The physician and the adolescent suicide attempter. Journal of School Health*

36(9):406-415, November 1966.

342

Investigator:

Elizabeth Elmer, M.S.S.
Assistant Professor of Social Case Work

School of Medicine
University of Pittsburgh
Pittsburgh, Pennsylvania

Co- Investigators:

Grace Gregg, M.D.

Byron Wright, M.A.

John B. Reinhart, M.D.

Contributors:

Thomas McHenry, M.D.

Bertram Girdony, M.D.

Paul Geisel, Ph.D.

Prepared by:

Clarissa Wittenberg

Historically, the terrible toll taken by childhood illness and industrial
accidents overshadowed the risk of children being injured by their parents.
At one time children were believed to be in the grip of the devil because
they had been conceived in sin, and harsh punishment was thought
necessary to save them. Parents "owned" children and almost any
punishment was considered legitimate. As our concepts of child
development have become more sophisticated and our understanding of
learning and discipline has advanced, harsh punishment has become less
and less acceptable. Consequently the parent who beats his child is an
object of censure. Today, we hold parents responsible for the well-being
of children, and, therefore, the malnourished and medically neglected
child becomes a subject of concern. Recognition that parents do abuse
their children has grown, and hospitals and doctors are increasingly aware
of the problem. Studies have been done to help define the problem and its
dimensions, and to record the effects of abuse on children.

Children's Hospital of Pittsburgh is located in a large low-income
district of the inner city and, like many other city hospitals, has an active
emergency ward where many injured children are brought. Some of these
children have multiple bone injuries. Early in the 1960*s the staff began to
systematically study the possibility of abuse in these cases. A research
team headed by Miss Elizabeth Elmer began, in 1962, a study of 50
families with children suffering from bone injuries who had been admitted

343

to the hospital over the previous 13 yeais. This was a follow-up study lo
determine what happened to these children after their initial admission to
the hospital. A second study followed which focused on infant accidents,
and compared infants and families where accidental injury had occuned
with those where neglect or abuse was present.

The problem of diagnosing abuse was attacked in the first study. Oases
were selected for the follow-up study on the basis of their hospital
admission record. Although abuse is a complicated subject involving both
social and medical problems, the criterion of multiple bone injuries was
selected for the purpose of a less controversial diagnosis. The family
history was then examined and the families judged to be either abusive or
nonabusive; those who could not be placed clearly in either group wore
considered unclassified.

TVie first study showed clearly that these children are in scrtous
jeopardy, that many die and many become severely retarded ami/or
crippled and spend their lives in institutions. The first Study of Fifty
Families resulted in an examination of what constitutes abuse. An
examination of the "failure to thrive" child was begun and the rote of
accidental injuries noted. The second study focused on small babies not
yet capable of getting into trouble on their own, thus illuminating the role
of the parents in such accidents.

Both studies resulted in an examination of the theories and accepted
ideas surrounding tliis issue. For instance, the working mother, commonly
felt to contribute to child abuse, did not appear to be important. Neither
were these children typically abused by extramarital partners or
non-related figures; there was no "wicked stepmother" syndrome. Wliile
many parents were found to have serious emotional problems, few wore
mistreating the children for bizarre or extemely sadistic reasons. These
children were rarely abused "coldly," Few of the parents were "bud"
parents and total failures; most stayed with their families and eventually
exhibited some success with their children. Neither did the parents
typically injure all of their children. Abuse has been found to be a
phenomenon related to the child-bearing period of the mother, and often
the mother has been uninformed about contraception.

THE FIFTY FAMILY STUDY

a pediatrician selected 50 former patients for the

'«ln^o/1.

iy x-ray film, indicating the occurrence
rt % in conjunction with -

that might account for the

ui assault or gross neglect, or the absence of a history
showing convincingly that the injuries were accidental or attributable to
an unusually traumatic delivery, A small group of children do suffer
undiagnosed fractures at birth.

344

The Final group was equally divided between male and female subjects,
of whom 36 were white and 14 Negro. This racial distribution
approximated that of the hospital's clientele. A number of the children
had come to the hospital for other complaints, and bone injuries had been
discovered in the course of routine examinations.

The majority of subjects had been young babies at the time of their
admission. Seventeen were under 3 months of age when multiple injuries
were found. Nine were between 3 and 9 months of age, This is in contrast
to the curve for childhood accidents where the incidence rate for
accidents is minimal below the age of 9 months. It then begins a sharp
climb, reaching a peak between 2 and 3 years when it begins to level off.

Fewer than 50 families were actually interviewed - due to deaths,
institutional placements, and refusals. Only families who still had their
children were interviewed. Six families refused to cooperate. Thirty-one of
the children in the original group, plus two siblings found to have bone
injuries, added up to a total of 33 children studied and 31 mothers
interviewed. Seven were foster mothers and one an adoptive mother.
Essentially the families were told that the object of the study was to
examine the hospital's treatment of patients, and an attempt was made to
avoid focusing on the suspected episodes of abuse in order to minimize
suspicion and distortion. It is of interest that the noncooperative parents
were in general better educated than the rest of the group. They may have
been more suspicious of the hospital's motives or more guilty about their
own behavior.

Information was accumulated from hospital records, current
examinations, home visits, and interviews with the mothers. It was
initially anticipated that the fathers would not be available for interviews.
Fathers were not interviewed formally, and it was felt that potentially
information from the fathers would have been of value.

Each of the children was given a current examination which included a
complete pediatric evaluation, psychological testing, a psychiatric
interview, a hearing test, and an x-ray survey of skull, long bones, chest,
pelvis, and spine— with special attention to the sites of old bone injuries.

On the basis of all this information, the children were divided into three
groups. Twenty-two were considered abused, four nonabused, and seven
unclassified. Nonabused children were those whose early bone injuries had
a plausible explanation other than assault by an adult. If agreement could
not be reached as to the cause of the injury, the child was considered
unclassified. For example, one such child had a record of birth injuries
and a hospital admission at 3 months with fresh fractures, but no account
could be obtained of abuse or accident. Unclassified families, then, while
not labeled abusive cannot be considered nonabusive either.

Almost all of the families struggled to live on low incomes. Most had
less than a high school education and correspondingly few job
opportunities. In most cases the families had three or four children. About
a quarter of the families in this study were on welfare; however, none of
the nonabusive families was, The families lived in substandard, but not the
worst, housing. Most lived in private dwellings or apartments, but none in

345

trailers or rooming houses. Many of the families kept their homes in fai
good condition and the mothers tended to be good housekeepers. Physi
squalor was not characteristic of this group.

The study families, particularly the abusive ones, suffered from man
stress. Many couples had been separated and reconciled many tim
without coming to any real resolution of their problems or differenci
The abusive families tended to have more quarreling and drinking than tl
others. Several abusive mothers expressed fear of their husbands, and tl
investigators thought that in general their fear was well justified. Oi
fathei, for instance, had a prison record for murder; another was observt
to blow cigarette ashes in his baby's eyes and then to knock the child
head against a post. It is possible that mothers with poor self-control ten
to be attracted to men with similar problems, or that the mothers war
the fathers to appear in a bad light so as to appear sympathetic b
comparison.

For disciplinary measures most of these families relied on physia
means of control. Whipping and spanking were the most commonly nsci
methods of discipline; scolding, withdrawal of privileges, shaming, am
shaking were also common. Reasoning with a child or avoidance of tin
conflict were methods almost never used, These parents tended to set
even small infants as needing discipline and as consciously and deliberately
misbehaving. It was rare for anyone other than the mother, or the mothei
and the father together, to discipline a child - and very unusual for th(
father to deal with the children by himself.

The nonatmsive families tended to use a few types of punishments
consistently, while some of the abusive families used a broad range of
disciplinary measures that they were searching for some effective way to
manage their children.

Mothers who abused their children felt very negatively toward the child
who had been injured. It is not known if they felt this way about all their
children or only the one who was abused. In one exceptional case, the
mother expressed sympathy for the child who had been abused by her
husband.

The abusive mothers appeared to have more emotional problems of
greater severity than the nonabusive ones". Depression was common with
about half of the abusive mothers troubled by difficulties in eating or
sleeping and having a tendency toward crying spells. The nonabusive
mothers, in general, had fewer and milder symptoms.

Several of the abusive mothers admitted to uncontrollable actions in
the past — including physical aggression against other women; sexual
promiscuity; and secret, compulsive spending. These mothers, who
themselves had serious problems of control, admitted being afraid of their
husbands as well. By their own reports, more of the abusive mothers than
the nonabusive were easily irritated.

The abusive mothers were lonely people, often with no place to escape
from the pressures of home and children. In many cases they had poor
relationships with their own parents. There were no friends or relatives to

346

help. It was noted that the mothers actively discomaged friendships, and
did not join even relatively impersonal groups such as the PTA.

Child abuse is a family affair, however, and regardless of the identity of
the abuser, the rest of the family participates. The other parent is involved
by virtue of lack of interference or tacit approval. In many cases siblings
may have injured the child, but again the responsibility must rest with the
parents. The family dynamics are important in these situations.

The following case history illustrates the type of family problems that
surround child abuse:

A 19-year old mother brought a three -mo nth -old baby, her third, into
the hospital. The baby was wearing a cast, and his weight including that of
the cast was 10 pounds. He had had a birth weight of 5 pounds, 3 ounces.
His x-ray showed that he had an old fracture of the skull, an injury to his
shoulder, fracture of the left arm, multiple injuries to knees, ankles, and
long bones of both legs. In addition, he had a bulging fontanel suggesting
subdural hematoma.

The mother expressed her horror that eveiy time she picked up the
baby he appeared to have something else wrong with him. The child had
been in another hospital at six weeks of age when he had been injured
falling off a bed onto a concrete floor. When the emergency room doctor
saw the baby, he wanted the police called as he thought it obvious that
the child had been beaten. The baby needed two subdural taps at that
time.

The mother's explanation was that she had put the baby in the middle
of a double bed while she went to another room to wash his crib. A
14-month-old sibling was in the room with the baby. She heard a thump
and thought toys had been dropped, and then ran to find the baby on the
floor. She assumed that he had "scooted off" the bed. The father was
critical of the mother for not watching the baby.

The mother had been a favorite child and had attended church
faithfully. Her family had had ambitions for her to get a good education.
At sixteen, however, she became pregnant and was disowned by her
parents. The minister of her church was also very critical of her. The baby
was born after a six-month gestation period and died after three days.
During this crisis the mother was alone as none of her family came to see
her. She married the father, and became pregnant again and had a little
boy. When he was three months old, she became pregnant again and
delivered the baby who was the patient. This added up to three births
within 22 months. Two children had been premature. In addition, her
parents separated and blamed their troubles upon her "disgrace."

The Children

Most of the children were quite young at the tiro
admission to the hospital. This study has shown that mar
suffer grave and irreversible damage, but also that som

this early abuse, and leach a phase where their parents can successfully
care for them and they can attain a reasonably good physical condition.

Eight of the 50 children had died by the time of this study. Most had
been under five months of age at the time of death; two had been slain by
their mothers. Five children were in State institutions for the retaidcd.
Many, who were in basically good health, had scars or deformities, but
considering that they had been at the point of death and had suffered very
serious injuries they were quite well-recovered. One child was suffering
from malnutrition, and several had organic brain defects. A large number
of children were observed to show signs of upper motor neurone disease,
as manifested by hyperactive tendon reflexes as well as abnormal plantar
reflexes. A few children had signs of cranial nerve involvement manifested
by strabismus and nystagmus. These signs appeared to be related to injury
in all of the children born at full leim, except in the case of one who was
jaundiced at birth and had had convulsions prior to the injury.

In the premature children with signs of neurological damage, the effect
of prematurity itself cannot be discounted. Only two of the prematures
were known to have had head trauma and symptomatic convulsions. In
one child, prematurity was the only known condition that could account
for central nervous system damage. The abused children had twice (lie
incidence of neurological signs as was true of the rest of the group.

The investigators found that two of the children had been injured in
substitute homes. In one case the substitute home was arranged through
an informal agreement between the natural and the foster parents, ami in
the other an adoption agency chose the home for an infant who was born
out of wedlock. The latter child was subsequently moved to another
foster home. In all, 1 1 children in the study were moved to snbstiliilc
homes for their own protection, following the abusive incidents.

Birth

Histories were obtained from the mother, and other available sources
such as hospital records. It was found that about a third of the cMhlm
weighed less than 5,5 pounds at birth, indicating prematurity. As Hie
national figure for prematurity is 8 percent, the percentage in this study is
extraordinarily high. It is known that birth weight varies by race and by
socioecnomic status. The national rates are 7 percent prematurity for
whites and 1 2 percent for nonwhites. In this study, however, the higher
percentages of low birth weight occurred among the white families: 8 of
the 24 white and only 2 of the 8 Negro children ha'd birth weights of less
than 5.5 pounds. Although the significance of the large number of
premature babies is not known, one possible explanation is that prem-
ature infants, because of their incomplete development at birth, are
more vulnerable to bone injuries than full-term infants. A pedialric
radiologist, Dr. John Caffey, is of the opinion that there is more
vulnerability in the first few weeks. The bones of a premature baby may

348

be injured even with normal handling, for instance, during diapering.
However, when chronological age plus the number of weeks of
prematurity equals nine months, Dr. Caffey obseives that vulnerability to
bone injury becomes that of any full-term newborn.

The median age of the premature children in this study was 1 1 months
at the time of hospital admission. This would indicate that their injuries
were not due to immature bone development, but to othei factors. Of the
21 abused children whose birth weight was known, seven were premature;
none of the nonabused children were premature. It is known that
premature babies aie more difficult to caie for than full-term ones; they
may be more irritable and cry more due to their immature nervous
systems. The mothers may be more apprehensive about picking them up
because they are so tiny. In addition, the emergency situation that so
often surrounds premature birth may be a serious strain on an already
easily upset mother. Preparations for births are often incomplete when a
prematwe baby arrives, and for a family with only marginal resources the
strain can be severe.

Negroes, who often had extended families or else lived in overcrowded
housing where other women were available, seemed to cope with the
strains of prematurity better, with relatively fewer combinations of
prematurity and abuse.

There is another issue, loo, that must be considered: the more subtle
problem of the mother's condition during pregnancy. A woman who is
unhappy about herself, her marriage, her pregnancy, or her other children
may take inadequate care of herself or be too overwhelmed to obtain
help. In many cases these mothers may not even seek prenatal care.

Other questions arise: For instance, what causes one family to zealously
protect, or even overprotect, a premature infant, and another family to
abuse such an infant. Why, if a couple with abusive tendencies has other
children, is the premature child selected for abuse?

Conditions at Time of Original Admission

At the time of admission to the hospital there was no difference
between the chief complaints of the abused children and the others. The
majority were brought to the hospital because of limitation of motion or
pain in an extremity. The next most common complaint was convulsions,
then "failure to thrive," and gastric symptoms. Convulsions and subdural
liematomas, physical conditions that are often associated with brain
damage, were diagnosed in eight children upon admission. Surgical
procedures connected with subdural hematomas were necessary in seven
cases. Two other children required orthopedic surgery due to bone
injuries. One-third of the group had previous hospital admissions.

Records of growth show that poor growth and abuse are not always
associated. However, many of the children showed an improvement in
appetite and growth while they were in the hospital.
/

349

Condition at Time of Study
Retardation

Forty-five percent of the entire study group had IQs under 80. Twelve
of the 22 abused children and none of the nonabused children fell in this
low IQ group. This is more striking when one realizes that this group does
not include five of the original children who were placed in State institu-
tions for the letardecl. Only children still at home were included in the
study. The investigators stress that they have no way of knowing what was
cause and what was effect in this relationship between abuse and retalia-
tion. Neurological impairment is important in retardation, and many of
these children showed such signs. In addition, many had histories of poor
early growth, a condition thought to be associated with later mental re-
tardation.

Speech problems, which are often associated with both emotional
difficulties and mental retardation, were found in this study to be more
closely related to mental retardation than to emotional problems.

Emotional Characteristics

The abused children had marked difficulty in impulse control ns
compared with others in the study. Many of the children, regardless of
their classification, had poor self-concepts. Even the nonabused childien
had suffered serious injuries, pain, and traumatic experiences at an early
age. Most had scars or physical deformities. It is not difficult to
understand that they might view themselves poorly or feel inferior,
especially if the parents had not been able to help them in a sensitive way,

Eight of the abused children had difficulty in controlling anger, ami
either had outbursts, of rage or serious inhibition of negative feelings,
manifested by very apathetic responses.

General Functioning

The abused children who remained in the same environments had a
substantially greater number of problems than the nonabused children.
Eight of these were retarded. The unclassified children had more general
problems than the abused children who had been moved to foster homes.
Seven children, whose physical development had been poor at the time of
hospital admission and who had been moved from the home, had achieved
an average level by the time of the study. To emphasize the importance of
the home environment, two children who remained in the same poor
homes showed average development on admission but below average
development at the time of the study.

Families at the Time of Abuse

The abusive families by and large lived in far more difficult
circumstances than did the nonabusive families. However, all these
families had often lived under stress, and for some reason abuse was not a
constant process. Rather, it breaks out and then abates. In many cases, the
sex or oidinal position of the abused child had a special significance for
the abusive parent. One child was a second girl, as her mother had been,
and both were family scapegoats. In another family, the two girls were
severely abused by the mother, but never the boys.

The birth of a sibling less than one year before or nine months after the
incident of abuse was found to be important. Nine of the abusive mothers
were pregnant at the time of the abused child's hospital admission, one
abusive mother had miscarried just before the child was admitted, and two
others had borne an infant other than the patient during the year piior to
admission. In only one of the other 1 1 families, unclassified and
nonabusive taken together, was there an interval of less than one year
between the injured child's admission and the birth of a sibling.

The investigators found that the connection between abuse and the
burdens of pregnancy and child-rearing is clear and important. They cite
the theories of Bibring, who identified pregnancy as a biologically
determined maturational crisis that is not always resolved with the birth
of the baby, but usually continues for some time, even in the most
auspicious circumstances. The investigators point out that these families
abuse their children primarily during the child-bearing phase of marriage,
Later they appear to cope in a better fashion. // was found that those
families who had successfully begun to use contraception were able to
recover from their previous strain and to stop venting their feelings of
frustration and rage upon their children,

Several of the mothers who were abusive were quite disturbed, and
some were under psychiatric care. In three cases the fathers were clearly
very disturbed or antisocial.

Substitute Care

A change in environment often saves the life of a child who has been
assaulted. Still, some children were abused while in foster care. It was
found that foster parents who voluntarily took children who were injured,
neurotic, or retarded often had an unhealthy need to have children who
were excessively dependent upon them. Furthermore, while much good
can be accomplished, even the best foster or adoptive parents cannot undo
irrevocable damage already done to the child.

Some foster children, due to their previous abuse, have severe
difficulties even after the original crisis is resolved. These are troubled
children and symptoms can appear long after the original trauma. In
addition, the protective care so helpful in the beginning can cause rebellion
later if the foster parents are unable to modify their methods in

351

accordance with the changing developmental needs of the growing child,
However, most of the children in placement showed marked improvement
in their physical health.

INFANT ACCIDENT STUDY

The majority of the abused children in the Fifty Families Study hud
been brought to the hospital as accident victims even though their injuries
were caused by assault. A few innocent patents of children sLiffcting
authentic accidents had unfortunately been suspected and sometimes
accused of abuse. The masquerading of abuse as an accident itml the
reverse was possible because neither phenomenon was clearly understood.
The investigators decided to study infant accidents, including abmii, to liy
to pinpoint the characteristics distinguishing one from the other.

Subjects were infants under 13 months of age who had been brought lo
Children's Hospital foi x-ray following an impact accident or abusive
incident. Since the younger the baby the more important the role ol'tla1
caretaker, it was felt that this study would yield information about
parental maltreatment and neglect, and not be complicated by (he
considerations of the noimal accidents of the active toddler. One humhetl
and one children were seen, both inpatients and outpatients. Viirimis
issues were explored: for instance, the difference between families of
abused or neglected children and those suffering accidents, will) or
without injury. The differences between retarded children who had been
abused and retarded children who had not were also studied, with spccfol
attention to the mothering received by each group.

Abuse was suspected if the families1 explanation of the injuries weic
not adequate, or if abuse was reported, or if more than one injury WHS
present. Initial assessments were made of family stress. Pregnancy und
very small children were considered stressful. The family was conskleml
to be under strain, too, if either parent had a close relationship wilh
another adult who was unrelated to the child. This issue was consider!
and observed because of a number of mass media reports of alniw
involving step-parents of pa.amours. As in the first study, this did not turn
out to be an important issue.

The family was also considered under stress if the baby had
developmental problems such as a significant deviation in growth,

m0t°r °r S0dal devel°Pment> or such troubles as feeding
or excessive crying,

r? important as Parental reports are heavily
by anxiety, guilt, and concern when a child is brought to a

h 1 atrthe,entire fam^ structure to find the clues dial
t, n T%famiiy: When abuse is ««P<*ted, deficiencies and

S v bf f?nd ^ °ther 3Spects of fc"«y ^. This study
oK^"' intfpersonal relationships, and child care
order to illuminate these differentiating issues. Because the

352

children were so young, the mother was considered the principal caretaker
and her interaction with the baby was carefully examined. In order to
assess the mother-baby relationship, the pair were observed — and also the
mother was interviewed and asked to fill out questionnaires. The
observations were felt to be of particular importance, as the mother's
habitual behavior with the child is likely to be beyond her awareness and
ability to leport. The observations provided data to supplement and
correct the information gained through the other methods.

Many issues involved with mothering weie examined in addition to the
traditional ones of providing food, shelter, and medical attention. The
stimulation given the baby, the verbal lesponsiveness, the quality of play,
and the ability of the mother to assess the changing needs of the child
were all considered. As in the previous study, a high number of retarded
cliildren in tlu's group was noted and some important observations about
their motheis were made.

Attempts to have an equal number of boys and girls, white and non-
white, were unsuccessful as the potential subjects did not fall this way.
Most of the children presented for x-iay were white females. Black female
babies rarely appeared and, in addition, it was inoie difficult to enlist the
cooperation of the nonwhite families in the research. All social classes
were represented, but the majority of the families were in the lower
classes.

Although a few families had refused to participate in the first study, the
refusal rate was even higher in the second. It was suspected that the
increased public awareness of child abuse and the outcry and pressure that
had been building up made some parents less cooperative. Also, even
parents of accidentally injured babies experience great guilt leading to
unwillingness to discuss the event. In the Infant Accident Study, it was
found very difficult to keep the allegedly abusive parents as subjects. They
rarely refused outright to come to an appointment or to allow a home
visit, but they failed to appear or were away from home at the specified
time.

Neglect played an important role, and several families who could not be
considered abusive were still considered deficient in their care of their
children based on observations made during the study. Some parents, for
instance, left their children without competent babysitters when they
were absent for prolonged periods of time, or failed to obtain needed
medical care despite repeated and careful instructions as to the needs of
the child.

The final study group consisted of 100 cases, 78 of which were
followed through all phases of the study and 22 in which families
participated in the initial and final procedures, with only a mailed
questionnaire in the interim. One of the mothers had two children in the
study, making the total number of children included 1 01 .

The methods included initial screeening of x-rays, several home
interviews with the mothers, and observations of the mothers in
examination, feeding, and teaching situations. A questionnaire was mailed,
and several pediatric and developmental evaluations were made. In several

353

situations the mother was put under mild stress. In one instance, where
she was asked to teach her baby to stack a series of blocks, the task was
generally too advanced for the child so as to determine her reaction when
frustrated by the baby. In another, she was asked to fill out a
questionnaire when accompanied by the baby, to learn about her attitude
when she was intent on another task. Four of the research persons saw
each baby and family, and they were seen in as many situations as was
practical.

The babies were evaluated twice in the first two years of life, a time of
extremely rapid growth and development. This allowed for observations
of the effect of the environment in a way not possible at later stages of
life. Effects of poor parental care are obvious very soon during these early
stages.

Of the 101 babies, brought to the hospital, only 10 were entirely
without signs or symptoms. These 1 0 had been brought to the hospital for
examination and reassurance, that despite a potentially injurious event, no
injury had occurred. The other 91 babies displayed a range of conditions
from mild bruising to symptoms related to the central nervous system,
such as momentary unconsciousness with or without vomitting, seizures,
paralysis, and coma. The proportion of abused children without
symptoms was roughly equal to the proportion of nonabused.

In addition to x-ray examinations of the site of the presumed injury, 21
x-ray surveys of the entire skeleton were performed. Ten of these were
part of a diagnostic work-up for failure to thrive, the rest because multiple
injuries were suspected. Eighty-two children had no evidence of fracture,
12 had a single recent fracture, and 7 had multiple fractures. The
proportion of children with multiple fractures was much greater among
the abused than the nonabused children. It was thought that if skeletal
surveys had been universally performed, they might have disclosed other
unsuspected, clinically unimportant fractures, which would have helped to
evaluate the quality of child care. However, it is difficult to justify x-ray
examinations without symptoms of injury.

Twenty-four children were judged to be abused, or to be both abused
and neglected. Ten children were thought to be neglected only. There
were 67 nonabused, non-neglected children. All initial judgments
concerning abuse and nonabuse were reevaluated at the end of the one-
year study.

The research was focused on the effect of abuse on the growing infant,
The main areas of investigation included mental and motor development,
behavioral characteristics, health status, and physical growth. The baby
was seen as being affected by at least two kinds of factors - those that are
relatively unchangeable, such as conditions at birth, and those that are
influenced by the caretaker. It was hypothesized that the abusive group,
in comparisoai to the nonabuse group, would show more stress, less
support, and greater authoritarianism.

As required by State law, when abuse was found, reports were made to
the Child Welfare Services and the parents were informed. Reports were
made on eight children. A few other families were already known to the

354

Child Welfare Services when they came into the study. In several cases, the
mothers had named their husbands as the abuser and had separated from
them. Two mothers overtly rejected the children whom they had
mistreated, and the study personnel helped them arrange for placement
away from home. The protective agency removed eight children from
their own homes.

Because it is often observed that sick children are iiritable and difficult
to care for, it was noted whether or not the child had an acute illness at
the time of admission. Eighteeen of the babies were sick with
gastrointestinal and upper respiratory complaints when brought to the
hospital. A few had anemia, and suffered other problems such as eye
infections. By and large the babies were not suffering from infections, and
the traumatic event was not related to the extra demands and needs of a
sick child.

Twenty-four of the babies were admitted to the hospital, eleven of
these for protection while further investigation of the family condition
was carried out. Thirteen needed hospital medical-surgical care, some for
incidental medical problems and some for injuries resulting from the
accident or abuse.

The "failure to thrive" babies were studied from several standpoints.
For some, metabolic and endocrine studies were clone with inconclusive
results. This condition, defined as occurring when a child has weight and
height below the third percentile for his age and sex, is not well
understood. Rarely are these children seen because of trauma;
characteristically, the mother who brings them in is full of concern
because a child has not reached the expected developmental landmarks.
Often she is anxious because her child is not growing and will assert
strongly that she feeds her baby well.

Home interviews in this study did substantiate that some of these
mothers fed their babies adequately. Medical opinion is growing that this
entity belongs with others where psychological phenomena and physical
development interact pathologically, as in anorexia nervosa or infantile
marasmus. Studies at Johns Hopkins Hospital have indicated that the
problem lay in the hypothalmic area and that it was reversible without
any hormonal or chemical treatment when the child was placed in a
hospital, a relatively nurturing environment. It was postulated that
emotional disturbance in these children may have had an adverse effect
upon the release of the pituitary tropic hormone via the central nervous
system. In the Infant Accident Study, "failure to thrive" babies whose
environment was changed tended to achieve normal growth, but rarely
normal development.

Initial Pediatnc Evaluation

Upon initial evaluation, 54 - or slightly over half the children — were
found to have either no medical problems or only the insignificant ones
expected during the first year of life. Two-thirds of the abused children,

355

benous nealth pioblems. Slightly fewer than half the babies
had a number of actual and/or potential health problems including
prematurity, moderate or high peiinatal stress scores, significant medical
problems, and histories of acute illness, Children in the abused group had
a disproportionately greater number of health problems per child than the
nonabused children.

The abused group was also distinguished initially from the nonabuseil
by their poor physical growth. In part their retarded growth was probably
due to prematurity (9 or 37 percent were premature by birth weight,
gestation* or both), but even giving credit for weeks of prematurity did
not bring them to normal level.

An estimate of how we'll the child was cared for in general was judged
by the manner in which baby appointments and immunizations were
attended to by the caretaker. Nineteen had not been seen regularly, if at
all. Twenty-eight of the mothers had not kept their babies1 immunization
schedule up-to-date, and some babies had not received any immunizations
at all. Most of the faulty child care was concentrated in the abused group.
However, upon questioning it was found that 92 percent of all tlie
mothers were able to recognize symptoms of poor health in their children
and to find suitable medical services for them when they became ill.

Thirteen babies were consideied poorly dressed, dirty, or ill-kempt
when they were brought for their pediatric visit. The number of abused
children in this category was much greater than the number of nonabused.

Accidents Versus Abuse - Initial Findings

Eighty-eight caretakers gave an accident history. Twelve abused children
were among those with credible accident histories, an overlap that
illustrates the complexity of diagnosis in these cases. Thirteen other
children either had totally unexplained injuries or they were x-raycd
because of suspected abuse, but no injuries were found; none of these had
an accident history.

The assumption was made that adequate protection by the caretakers
could completely abolish true accidents. The investigators realized,
however, that this is unlikely and even undesirable, as a child i eared in
such a protective environment might have many other problems.

Three-fifths of the accidents were termed "active" because the baby's
motor activity was an important contributing factor. Active accidents
were subdivided into three categories: "open field" in which the babies
propelled themselves into danger - for example, falling down the stairs
when a gate had been left open; falls from appropriate furniture, such as
couches or dressing tables; and falls from inappropriate furniture, such as
the tops of washing machines - caused for instance, when a baby in an
infant seat wiggles and the seat slides off the slippery top of the washing
machine, It was thought that most of the active accidents might have been
prevented by the use of built-in safety devices, such as belts to confine
babies on dressing tables.

356

Passive accidents were those in which the baby's contribution was
minor and the responsibility of the caretaker greater. Subdivisions of
passive accidents included babies dropped by their caretakers, those
suffering "Act of God" events, such as being hit by a stray baseball, and
those who were admittedly assaulted by another person.

The accidents were described, then the abused gioup was compared to
the nonabused, non-neglected children who had suffered accidents. Points
of comparison, in addition to general health and injuries already
mentioned, were behavioral characteristics, age, and ordinal position. The
families were compared as to social class, stress at the time of the incident,
and health of the mo theis.

The only infants who differed behaviorally were the babies who had
active, open-field accidents. They were piedominantly negative in mood,
not distractible, and moderately or highly active. This combination of
traits can be seen to result in babies who are difficult to protect from
harm. By contrast, the other subjects, including the abused children, were
positive in mood, easily distracted, and moderately active. The babies
represented in the "open field" accidents were also the oldest (median
age, 42 weeks) and, therefore, their motor development was more
advanced.

Most of the babies in the accident group, active or passive, were only
children in their families, while on the average the abused child was the
second child. This suggest that parents of a first child are not as aware of
potential hazards as they might be, and also indicates that in abuse cases -
in addition to evaluation of the stress of having several small children - the
possibility should be considered that one small child might injure another.

Ninety-two percent of the abusive parents were identified with Class V
(low) according to the Hollingshead Two-Factor Index. Forty-eight
percent of the nonabusive, nonneglectful families studied fell in this class.

Regardless of the type of injurious event, the mothers typically had
special stress added to chronic factors of strain. The abusive mothers
mentioned baby irritability generally and the other more often mentioned
disrupted schedules, fatigue, etc. Over 50 percent of the mothers of
abused children had significant health problems, for example, mental
retardation, emotional difficulties, seizures, and heart disease. Such major
health conditions were found in only 20 percent of the nonabusive
mothers.

In attempting to determine the quality of the mother-child relationship,
the investigators studied the caretaking process. This process makes
manifest much about the mother's interaction with the child and gives
some measure of her general ability to function. The ability of the mother
to monitor the environment for her helpless baby changes in relationship
to many things. The mother, of course, operates within her own milieu
and class structure, and this partially defines good mothering for her. The
events that occur, the health and stability of the mother, the abilities she
is potentially able to bring to bear to help her child, and her ability to
perceive accurately his needs all affect her care of the child. Her degree of
affection as well as her convictions about child rearing also enter in.

357

undoubtedly, too, the resources of the mother to provide support,
affection, and aid for her are crucial.

Stress is seen as a major issue in child abuse, the caretaker being under
insupportable stiessin most such situations. In the case of the accidents, it
has been noted that most of the mothers were reacting to stress in varying
degrees — the abusive mothers, however, were under greater stress, had less
support, and fewer personal resources. The interrelationship of stress,
support and the ability to cope can be seen as a continuum.

It must be remembered that the early years of child rearing are heavily
demanding. Little money is available, and many young people aic
unprepared to become parents. Shifts occm in families even when children
are desired and planned for, and greater strain is felt with unwanted
babies. The investigators view most young families as being at a point of
lowest toleiance for stress at a time when they are subjected to the highest
stress during the years when childien are being born. However, it must be
realized that what constitutes stress for one individual can be handled by
another. Unfortunately, some types of chronic stress are brought on by
poverty, which rarely permits growth or learning and usually
undermines a family's ability to function. The investigators were most
interested in everyday stress, as opposed to extraordinary or emergency
stress, because they felt that chronic stress was of key importance in child
abuse.

The demographic data about the abusive families was a documentation
of the degree of chronic stress. Almost half the abused children were
black, and most of the combined abuse and neglect occurred among the
black families. Because of the larger numbers of children with
comparatively few fathers in these homes, the families fall into a gioup
known to be especially vulnerable to many kinds of stress. According to
socioeconomic status, all the families in the combined abuse and neglect
group are classified in the two lowest classes of the Hollingshead scale of
social position.

There were ten families that were considered neglectful but not abusive,
and it was found that in some ways they resembled both the abusive and
nonabusive families.

The nine families demonstrating abuse but not neglect were
predominantly lower class, with two members of the middle and upper
classes; the ten families showing neglect alone included four classes from
Class I, the highest of Hollingshead 's classifications. One of these families
was classified as neglectful because they habitually left the baby in the
company of an active 30-month old sibling without adult supervision,
Another family was called neglectful because their child was encouraged
to perform physical feats distinctly beyond his limited ability, such as
hanging from the pantry shelf by his fingers.

An assessment was made of the medical condition of the child and tlie
mother's reaction to it. It was found that of the ten "neglect only"
mothers, eight showed only slight reactions to medical problems. The
mothers of the abused children reported feeling great stress due to their
babies1 medical conditions. The focus of these mothers on their babies'

358

health was seen by the investigators as being realistic in view of the ex-
tremely poor health of these children.

There was one group of 24 women, primarily made up of black
mothers, who were unusually bland or under-reacting to medical problems
in both themselves and their children. One such mother had noted an
abnormality in her baby's eye for more than a week, but hadn't sought
medical advice about it. The investigators had several ideas, but no
definite answers to explain this attitude. A middle-class mother, they
believe, tends to emphasize her attentiveness to her child whether or not it
is warranted. This is not always the case with the lower socioeconomic
class mother. The investigators also note that many of the "upper
reactors" were poorly educated and suggest that perhaps they did not
really understand the potential hazard of some of the conditions.
However, seven of the "under-reactors" belonged to middle- and upper-
class groups and were well educated. An analysis of the "under reactors"
by social class showed no significant class association. It is also suggested
that the appaient apathy may be a defense against implied criticism and
intrusion by the outsider or a way of coping with what would otherwise
be overwhelming anxiety. It may also indicate a true indifference or a
general state of apathy which includes, but is not limited to, the child.

The type of stimulus perceived as stressful and the reaction to it are
highly individualized matters. To avoid imposing any preconceived
hierarchy of stress, the investigators inquired what had happened to the
mothers or their families since conception for the index child. The events
divided naturally into Hill's four categories: physical difficulties,
separations from persons or possessions, accession events such as a new
person moving into the house, and social disgrace. The mothers' reactions
were dichotomized as 1) mild or non-existent, or 2) strong. The number
of stress events and associated reactions were combined to yield a total
score for each individual. Mothers in each group were ranked and the
groups compaied.

In general, recent events were given highest stress ratings by the
mothers. Acute conditions were reported as more stressful than chronic
ones, and events involving the immediate family as more stressful than
those involving extended family or friends. The proportion of mothers
reporting no stress whatsoever was greater among the neglectful than the
nondeviant, and only 4 percent of the abusive reported no stress.

Accidents, moves, and physical illness were the greatest sources of stress
for the abusive mothers; also, prominent among the accidents were attacks
upon the mothers by others. Several women reported having been beaten
by their husbands. One abusive mother claimed to have been raped on her
way to the hospital with her baby. Although the report seemed
questionable, it was similar to others in its preoccupation with violence,
either factual or fantasized.

A common source of stress for the abusive and other families was a
change of residence. Typically the family moved during the woman's
pregnancy to obtain more room. In both the abusive and nondeviant

359

groups, tlie moves in late months of pregnancy resulted in strain on tJic
mothers and at times brought on premature deliveries.

Illness was often reported and sometimes - especially in the group of
abusive mothers - reported not as chronic conditions, but related to
pregnancy for the index child. The abusive mothers felt very strained by
the pregnancy of the child in question and ranked pregnancy as a higher
stress even than deaths.

According to the physician's rankings of physical disorders associated
with pregnancy, the abusive mothers who reported the most sticss
actually had the least, thus indicating a higher psychological sensitivity.
Six of the abusive mothers who reported difficult pregnancies weic caring
for other young babies when they were pregnant and, also, had fewer
people available to help them.

Potential support factors included a satisfactory male relationship: the
presence of a man in the house — whether spouse, common-law partner, or
father; a continuous association with a male during pregnancy; help fiom
the man in relation to the baby; a stable source of income; a continuous
source of medical care; participation in religious activities and involvement
with neighborhood activities. The abusive families and the others differed
significantly in the amount of support available. The abusive families had
the least support.

Statistically significant factors were continuous association with (lie
father during pregnancy and help from a male — whether husband, father,
or friend - in relation to the baby. The current presence of a male in (he
home on a stable basis did not appear to be a significant positive factor;
nor was marital stress a significant negative factor. During pregnancy (he
help of the father appeared to be mainly psychological, but once the baby
arrived the mother received more support when some male did something
concrete to assist in the care of the baby.

The effects of race on stress and support were assessed. When support
was low, black mothers reported significantly more medical stress Mum
white mothers; they suffered more physical problems than the whites. The
white mothers, however, reported significantly more social stress when
support was low. This comparison would seem to reflect the perception of
the woman, white or black, as she viewed herself in relation to her peer
group.

The mothers were scored for general negative and positive reactions to
their babies, and several trends appeared. The abusive mothers tended
toward extreme reactions, judging the babies to be either all good or all
bad, while the non-abusive parents saw their children more realistically as
both pleasing and annoying. The abusive mothers were relatively silent
with regard to their children's development. Fifteen of the 19 abused
children showed early signs of retardation,

360

Modes of Punishment, Discipline, and Teaching

It was predicted that the abusive mothers would use harsher methods of
punishment and would have less interest in teaching their children than
the other mothers. The results were not so clear cut.

Forty-one percent of all the mothers used some form of physical
punishment with babies less than 6 months old, usually slapping the hands
or the buttocks. At 9 months, physical punishment intensified. By 24
months, 87 percent of all the mothers were using this method of physical
punishment at least part of the time.

The type of behavior punished varied with social class. Mothers in the
two highest classes punished principally for aggressive acts; middle-class
mothers for activity, dangerous or otherwise; and lower-class mothers, for
conduct such as excessive demands, disobedience, or crying. Generally the
abusive motheis, most of whom were lower class, punished for
unacceptable conduct. Across all groups, girls were consistently punished
earlier: by the age of 9 months, 3 1 percent of the girls, but only 5 percent
of the boys were being punished; by 18 months, the figures rose to 70
percent and 50 percent for the girls and boys, respectively.

The investigators had become aware that most mothers are extremely
sensitive to their babies' aggressive acts against them as mothers. When
asked how they would respond if their infants struck them or spat upon
them, the overwhelming majority of the mothers of babies 6 months of
age or older said they would retaliate in kind, **. . .to show him that he is
not to do that kind of thing." Three mothers of babies who were less
than 6 mo nths of age also said the same . Eighty percent of the abusive and
63 percent of the nondeviant mothers said they would hit back against
infant aggression.

Regardless of their social class, most mothers asserted that a baby
should know right from wrong by the age of 12 months, and one-third of
the mothers specified 6 months. This belief implies a common lack of
realistic information about infant development and when babies learn
concepts of right and wrong. These mothers also perceived the babies as
having "tempers" and other directed feelings at a much younger age than
is actually possible.

The mothers involved in this study usually discriminated very little
between discipline and teaching. When asked how they would attempt to
teach the baby some new behavior representing a real learning effort for
him, they most frequently responded in terms of scolding or spanking to
get him to learn after first giving verbal instructions. The investigators feel
that infants are punished physically more often than is realized. When it is
common practice to strike babies, however lightly, with the goal of
teaching them, the laws of probability indicate that some babies are going
to be struck too hard and that some will be injured.

361

neiaiea to Mothering

The mothers were questioned as to their expectations concerning the
child, The majority preferred their babies to be "good"; that is, respectful,
grateful, obedient, and not rebellious. These were the particular goals of
the abusive and neglectful mothers; they were not interested in creativity,
etc.

There were varied opinions among them as to what constituted an
"ideal mother." All the abusive and the neglectful mothers mentioned
keeping the baby clean and giving him material things. A few mentioned
the importance of being a "proper" woman. They described the ideal
woman in negative terms as somebody who does not run around or sit at
bars. The abusive group often described the ideal father in terms of
discipline or financial support. Several women said that the ideal father
should not beat the mother.

Quite a few mothers felt that affection should be restrained: thai (hero
is danger in being too affectionate toward babies. This trait was more
marked in the abusive and the neglectful mothers.

The index of values related to mothering clearly and significantly
distinguished between the abusive and the nonabusive groups, correctly
classifying 77 percent of all the families. Among the abusive mothers,
emphasis was placed on cleanliness and materialistic values. They tended
to perceive themselves and their husbands in stereotyped roles, n
perception suggesting difficulties in forming and maintaining close
relationships. Their fear of showing too much affection toward llicir
babies was another manifestation of the same difficulty. These
characteristics, together with the common need to have an obedient,
compliant baby, established the abusive mothers as more authoritarian
than the nonabusive women.

The Baby

The contribution of the baby to the mother-child relationship is
extremely important. A smiling baby who is responsive may keep even a
detached mother involved. While many types of behavior are important, it
was decided that four behavioral characteristics would be examined:
mood, level of activity, approach or the way the baby related to a new
person or new object, and distractibility. These characteristics were
studied during the initial and the final pediatric examinations.

Among all the children the distribution of positive and negative mood
showed a decided difference according to sex and developmental age,
Regardless of their developmental age, half the boys were positive, half
negative. The girls were strikingly negative when developmentally young
but became positive as they matured. Abused boys were markedly
negative compared to their nonabused peers. Abused girls were more
positive than nonabused girls; however, the abused females as a group
were developmentally older than the nonabused.

362

Eight abused children, four boys and four girls, were separated from
their parents by the time of the second testing. All were predominantly
negative in mood. According to the study data, this seemed associated less
with their removal from home and more with the mood to be expected
from the above findings. The four negative boys were similar in mood to
the majority of abused males, while three of the negative girls were
developmentally young and thus apt to be more negative.

The factors of sex, age, and abuse which affect predominant mood need
considerably more study before the interrelationships will become clear.
Nevertheless, these findings suggest that boys and girls may indeed
respond to abuse in quite different ways.

It was not possible to find associations between mental development
and behavioral patterns. Some of the children were advanced
developmentally, some at age level, and some retarded. Some in each
group were positive in mood and could be distracted.

Distractibility depends upon whether or not a child can be intrigued
away from something he is doing, especially if it is a hazardous activity. In
this case, it is a positive quality, as opposed to hyperactive distractibility
which interferes with concentrated learning.

All of the babies who were positive in mood and distractibility were
positive in approach; that is, interested or curious or pleased at meeting
new people and new things. Among the nonabused children who had
negative or mixed scores on mood and distractibility, a racial difference
occurred on the approach scores. The whites of both sexes were
predominantly positive in approach, while the blacks of both sexes were
predominantly negative. The investigators noted that the examining
doctor was a white woman and wondered if this could be a factor
influencing these results.

Activity levels did not distinguish between abused and nonabused
children. The babies who were both abused and neglected were low in
approach and play behavior and high in negative activity. The "abused
only" or "neglect only" had wider repertoires of behavior. However, of
the "neglected-only" children, a large proportion either remained high or
became high in activity. It was observed that the mothers were largely
ineffectual in controlling their children; they tended to pile on
command-after-command while the children became more anxious, active,
and difficult. The ability of these mothers to limit the activity of their
children, who seemed to be in special need of help in controlling or
directing their activities, seemed very meager.

Observations of Interaction

The mothers were observed with their babies during a feeding period at
home and in a teaching situation in the pediatrician's office. It was learned
that mothers vary considerably in their perceptions of what is dangerous
to their babies. While the mothers were in the doctor's office, the babies
were often attracted to the doctor's kit containing instruments. Most

363

mothers did not permit their children to handle these, yet failed to see the
danger in the sharp corner of a drawer that the children loved to pull out.
The examining table was also a danger, as mothers often turned away
while their babies who were lying on it waited to be dressed.

The actions of each baby and mother were tallied and analyzed
according to content, mode, and context. Mothers of retarded children
concentrated on feeding them and behaved more positively toward the
child as the baby ate. They did not talk spontaneously to the child as
much as the mothers of non-retarded children. The investigators believed
that this reflected the mother's concern that the child eat rather than play
or socialize. There was an overwhelming tendency for mothers of retarded
abused children to show low verbal response to the babies' vocalizations,
but an opposite trend was shown by mothers of retarded nonabitsed
children. The age of the baby was not a factor in the mother's tendency to
verbalize when the baby made sounds. Some babies responded to tlicir
mothers' speech, others did not, and again this was not related to b;iby
age.

It is probable that an involved mother gives her baby many types of
stimulation in addition to the verbal. The verbal response, however, seems
to be a good indicator of the total social environment provided for the
baby. Mothers with a good education were much more verbally responsive
than those with a poor education,

A significant association appeared between mothers who responded
verbally and the higher rates of development among these babies, The
investigators note that such an association has not previously been
reported but they point to several conditions that may affect it. The
children of well-educated, intelligent mothers may have superior genetic
endowment. Also, babies who have had a great deal of verbal experience
. do better on tests, which often require ability to follow verbid
instructions.

With regard to control, the abusive mothers tended to give their
children great latitude until their patience wore thin, when they would
abruptly threaten or strike their children. The neglectful mother seemed
to burden the child with repetitive commands and threats to which lie
paid little attention, apparently sensing that the mother did not know
how to control him, or for some reason was unable to do so. As lie
became more active, the mother became more frenetic.

The abusive and neglectful mothers tended to care for the babies, but
made neither broader responses nor extra reactions to their children. They
tended toward stereotyped responses.

The teaching situation, which was essentially an artificial one, aroused
some anxiety. However, the observers of the feeding, who knew the
mothers, thought that they behaved much the same as they had in the
past. This was substantiated by the significant positive correlations
between the feeding and teaching observations with respect to
maternal-verbal responses among all cases; mothers of nonretarded
children; high social class; and females,

364

Final EvaEuation of the Babies

The most important final difference between the abused and the
nonabused babies appeared in the scores on the Bayley Mental Scales.
There was significantly more retardation among the abused children when
compared with the nonabused. The likelihood of retarded mental
development among the abused children was greatly increased when they
were also judged to be neglected by their parents. The fairly high rate of
mental retardation found in all groups of children in this study may mean
that the hospital outpatient population is biased in this direction.

The final checkups showed little difference in height and weight
increases; between the abused and nonabused children. However, this was
true largely because one-third of the abused childien had been removed
from their homes and placed in benign and maturing homes. All but one
showed remarkable catch-up growth. There was a significant association
between height and mental development ratings, with retarded mental
development occurring more frequently among children below the 10th
percentile in height.

In terms of family characteristics, Ike single factor most strongly
related to the mental development ratings was the amount of income per
person in the household. The percentage of children within the retarded,
normal, and advanced groups coming from families with less than $100 a
month per person was 74 percent, 46 percent, and 7 percent respectively.
Although abused children more often come from families with low
incomes, the relationship between income and mental development was
not altered significantly when controlled for the occurrence of abuse.
When children with retarded and normal development were combined, 79
percent of the abused and 52 percent of the nonabused children came
from low-income families; this represents a statistically significant
difference.

Among those with advanced development, the majority of the children
were white; among those with slow development, the majority were black.

Although stress was found important, no statistical association between
ratings of social stress and ratings of mental development appeared.
However, the number of supportive resources for the mother was related
to mental development. Among the families of retarded abused children,
69 percent were low in support while only 31 percent of the families of
nonabused retarded children had similar low ratings. When stress hits a
family with few sources of support and assistance, then the problems
become more intense.

The presence of the father is also important. The group who were
advanced mentally all had their fathers at home, while only 59 percent of
the retarded had fathers living with them. The father was absent in the
cases of 64 percent of the retarded children who were abused.

Regarding the probability of retardation, three factors in addition to
abuse are important: low monthly income per person, significant physical
problems in the baby, and low verbal responsiveness in the mother. When
any two of these factors plus abuse was present, 1 00 percent of the

365

cmiciren weie retarded. Among children without any of these factors, only
21 percent weie retarded.

The second evaluations showed that, remarkably enough, there are
cliiidren who appear normal despite abuse, and it is also evident that there
is a range of intensity in abuse. Some children are subjected to pervasive
and long-standing abuse, while for others the abusive incident is isolated in
an otherwise favorable environment. The investigators caution that an
overall characterization of the abused child demands both pediatric and
family assessment The physical and mental effects of abuse can be
mimicked by other conditions, and also the physical and mental state inav
not fully expose the abusive atmosphere of the home.

Diagnosis of Abuse

A crucial factor in the diagnosis of abuse is the willingness of tlw
physician to consider abuse as a possible cause of a child's injuries iincl lo
examine him accordingly. Dr. Grace Gregg, a pediatrician and an
investigator in this study, points out that a diagnosis of abuse requires .1
history that fails to explain the injury, the elimination of systemic disease,
and an assessment of the type of care that would allow such a condition
to develop. "Failure to tlirive" children must be looked at with an eye lo
abuse and neglect. It is important where there are multiple injuries thiit
each be accounted for. Multiple bone injuries are considered a key
indication of abusive treatment. It must be remembered, however, Hint
some bone changes do not show up immediately on x-ray and may be
hidden until about 12 to 14 days. Furthermore, x-ray can tell Hie
condition of the bones, but not how they were injured nor the motivation
of the person responsible for the injury. In some cases a parent can
roughly and abruptly grab a child to prevent an injury and accidentally
hurt him. However, while this type of accident can happen once, a scries
of such incidents would be highly suspicious. Also, the idea that siblings
can injure infants is unpopular, but must be considered.

Malnutrition is a key indication of abuse or neglect, but evidence of
malnutrition is difficult to identify when intake becomes adequate, unless
photographs are taken.

Familiarity with the normal injuries of children is indispensable to
adequate diagnosis. Superficial injuries above the elbows, shins, and knees
that do not resemble dermatologic conditions should be examined to sec
if they have been caused by rough handling, human bites, cigarette burns,
etc. All bones and joints should be examined, not merely those pointed
out as injured.

Legal Issues

By June 1967, 52 child abuse reporting laws were in existence in all 50
States, the Virgin Islands, and the District of Columbia. Puerto Rico

366

added a law soon thereafter. In most cases mandatory reporting by
medical personnel, occasionally by schools or social workers, and
investigation by law enforcement agencies was typical. In some states the
professional can be fined or imprisoned if he fails to report a case of child
abuse.

At first glance it looks as though the situation has been acknowledged
and adequately coveied. However, this is far from the case. The Children's
Bureau looks upon child abuse reporting laws as case-finding devices. How
successful they are for this purpose is difficult to assess. Some problems
have, however, been identified. For instance, diagnostic guidelines are not
well-drawn. There is the possibility of an inappropriate accusation,
perhaps a law suit. Medical training is often limited in terms of teaching
the type of social-family assessment that is required in many child abuse
cases. This type of case can be tremendously time-consuming and time is a
rare commodity in most medical practice.

Another problem is that of confidentiality. The child abuse laws place
the child's right to safety above the traditional rights of the patient — in
this case, the parent — to protected communications, A social class
difference may slant the manner in which a case is treated. For example,
in many States hospitals and physicians are required to report a case if an
abused child is brought for care. Private doctors, however, are not exposed
to the same public attention and might - and do - manage private patients
differently. A doctor who reports a patient stands to gain ill-will and lose
the family for treatment. He may feel he can give more help by not
reporting the case and staying involved as an interested and concerned
family doctor. Patients who can afford private care can also "shop"
around. They can go to different physicians and the full extent of the
child's history of trauma might be hidden in this way.

The goal is not, it must be remembered, merely full reporting. The goal
is the protection of the child. The two are related, but not the same.

The investigators feel that professionals who report a family in good
faith should be granted immunity in the event of a law suit. Other changes
should allow for concern for the other children in the family.

The lack of community placements for such endangered children also
make some people reluctant to confront an abusive parent, as it is realized
that all too often the child must return home with an even more enraged
and abusive parent.

There is considerable question as well, as to whether or not the police
should be given responsibility for establishing whether abuse has occurred.
In many cases this type of approach with the goal of proving guilt and
establishing criminal behavior is unfruitful. As this research has shown, the
problem may be subtle; a child may have been left with inadequate
supervision or the parent may have shown poor ability to anticipate the
child's pattern of activity. The caretakers may be extremely immature or
disturbed, and thus cannot be considered directly responsible for injury to
the child. Nonetheless the child may be in great jeopardy. But it is
questionable whether most police have the orientation, time, or training
to investigate these issues.

367

Prevention is largely ignoied, as is appropriate follow-up and assistance,
to families in need of community aid. Expanded protective services ;uc
much to be desired. Assistance with related problems such as
contraception should also be available.

Punishment of the parents or probation, which often means only Ihu
most minimal surveillance, rarely accomplishes much toward the most
important goal - protection of the child. Punishment for doctors or
hospitals may very well be self-defeating and discouiagc reporting. The
most fruitful approach is via education. Physicians need to be sens! li/ccl U)
the issues and hospital procedures need to be changed so as to permit
early identification of the endangered child. Refeirals to social agencies
should be facilitated and child care resources developed. Where a numicr
or brual attack has taken place, then the police are appropriate; where uu
overwhelmed mother has a child who continually injures himself OIK- to
lack of supervision, another resouice such as the help of a trained home-
maker might be more appropriate.

Prevention of Child Abuse

To save a child from the serious effects and irremedial damage caused
by abuse and neglect, it is necessary to recognize the situation when il
occurs. Professionals need to be alert when they notice that yomv!
families are having their children too quickly, with no relief between
pregnancies The danger signs of marital strain, poverty, isolation, and
overwhelmed mothers need to be heeded. Premature births with
indicaoons of family strain should be of great concern; all possible

XLn'V suryeillfnce should be g^en these parents. Patents with
children showing developmental difficulties should be given similar holp.

vJln ™portant. *at education be restructured so that every you up
M? «,£ ™hldf >™ children gr0w and devel°P- Classes should I,S
e L^rii ?*i ' TWy 3t the elem«"*"y and junior high level
nd ?gnot hi hP,rOSPeCtlVe m1°thers Wil1 be at Icast basically informed
S 6Xperience with child c«rc- U cannot bo

^ t0° many women thi»k a Xouna baby is
i CM react ^Wtectually like an adult. The
pUnSlent for infants «*^™ re-thinking in
development. Again, it must De

Phenomenon, ^^^ ,is a/owor-cJ^

taught the importance of tJT I •? " classes' Aud males nuist >c
and'the ^^m^ntrAut^ to *fi stability of the f ami ly

^,made -valbbte if we are ft,

children. Ap^pS^ ^ ^ "«"«* thcir
available. Unwanted children ar^f ?°ds should be made ensily

long-term residual damages ^ danger of ab«se and all iu

368

Education of legal personnel, especially that of judges, is necessary. The
investigators note that even in cases which represented blatant abuse as
manifested by multiple skeletal trauma with central nervous system
damage, when they petitioned for removal of the child from the home the
authorities were more concerned about the rights of the parents than the
welfare of the babies. Some children were returned to the custody of their
parents after their fractures had healed and their general condition
improved without any assessment of the family at all. The courts do not
seem to be truly aware of the risk to the child - that it might not merely
be a question of a single meaningless act, but that abuse may constitute an
active expression of a wish to be rid of the child. The overwhelming odds
against complete recovery from parental abuse seem to escape appropriate
attention. The horror of many situations and the intense feelings they
arouse may cause some people to try and mend the parent-child
relationship in order to wish the whole situation away. It is not always
true that the natural parent is best for the child, nor is it true that any
parent is better than none. Parents, on the other hand, need not be treated
as cuminals because they have abused their children. A total assessment of
the entire situation and its pressures needs to be made.

One issue, that of community support, demands special attention. As
family patterns change and mobility increases, social institutions such as
the church lose their strong hold on family life. People who tend to be
isolated become even more so. It is probably true that many parents who
do not beat their children would also benefit if a neighbor could help
them out when they are overwhelmed, or if a network of friendly visitors
would somehow fill this void. Volunteers could be used to extend the
work of the public health nurses and the hospital clinics to ensure that
both mild supervision and help would be available to young mothers.
Homemaker services can be extremely important. Community programs
and neighborhood associations could also be helpful if the prohibition
about interfering or getting involved could somehow be broken down
constructively. It is possible that an auxiliary to the police department
could be useful.

The interrelationship of poverty, isolation, and too many
Linplanned-for small children is important. Too often, little is done to
reach the very people who are too weak to ask for help. Newspapers and
TV, too, often carry only the sensational story and not the steady
compilation of data that might enable us to make reforms in our welfare
systems, our birth control clinics, hospital regulations, courts, and foster
home programs - data that might enable us to prevent such tragedies and
save these children.

Research Grant: MH 14739
Date of Interview: June 1969

References:

Caffey, J. Multiple fractures in the long bones of infants suffering from chronic
subdural hematoma. American Journal of Roentgenology, August 1946.

369

uc rrancis, Vmcent Child Abuse - Preview of a Nationwide Survey. The Children's

Division, American Humane Association, Denver, Colorado, 1963.
Elmer, Elizabeth. "Identification of Abused Children." Children. U S. Department of

Health, Education, and Welfare, 180-184, September-October 1963.
Elmer, Elizabeth. Hazards in determining child abuse Child Welfare, January 1966.
Elmer, Elizabeth. Child abuse, overview of the problem and avenues of attack. Paper
presented at the 5th Annual Mental Health Institute, St. Louis, Missouri, July 1 966.
Elmer, Elizabeth. Abused children and community resources. JnternationalJournalof

Offender Therapy, 11, 1, 1967.
Elmer, Elizabeth and Gregg, Grace, M.D. Developmental characteristics of abused

children. Pediatrics, 40;4(Part 1): 596-602, October 1967,
Elmer, Elizabeth Children in Jeopardy University of Pittsburgh Press, 1967.
Gregg, Grace, M.D. Physician, child-abuse reporting laws; and injured child,
psychosocial anatomy of childhood trauma, Clinical Pediatrics Philadelphia,
Pennsylvania; J. B. Lippincott Co., December 1968.

Hill, Reuben. Social stresses on the family. Social Casework, 39: 139-150, 1958.
Powell, G. F., M.D., Brasel, J. A., M.D., and Blizzard, R. M., M.D. Emotional
deprivation and growth retardation simulating idiopathic hypopituitarisin: I -
clinical evaluation of the syndrome. The New England Journal of Medicine,
276,23:1271-1278, June 8, 1967.

Powell, G. F., M.D., Brasel, J. A., M.D , Raiti, S., M.D., and Blizzard, R. M., M.D,
Emotional deprivation and growth retardation simulating idiopathic
hypopituitarism: II - Endocrinologic Evaluation of the Syndrome!. The Journal of
the American Medical Association, 88:358-362, April 27, 1964.
U S. Department of Health, Education, and Welfare. The child abuse reporting laws; a
tabular view. Washington, D.C., 1 966. (Reprinted with revisions in 1 968)

370

Investigator;

John M. McKee, Ph. D.

Draper Correctional Institute

Elmore, Ala.

Prepared by:

Herbert Yahraes

Introduction and Summary of Findings

In a sparsely settled agricultural area 25 miles north of Montgomery,
Ala., lies a unit of the State's prison system known as Draper Correctional
Center. There, some 800 men, more than half of them in their twenties or
late teens, are serving sentences ranging up to life for crimes ranging up to
murder.

Under the Alabama parole system, an offender is usually considered for
parole when he has completed one-third of his term. (Inmates joke about
Dne of their fellows who is serving three life sentences that run
consecutively. "When he gets through the first one," the joke goes, "they
ire going to have a - of a time getting him through the next two."
Actually he has served several years and can expect a chance to prove
limself in the outside world in a dozen more.) The parole system helps an
Df fender to get a job; it requires, in fact, that he have a job, or an assured
Mace in school, before he is released. And it tries to keep track of him and
:o counsel him. Nevertheless, it is estimated that from 65 to 70 percent of
lie general run of parolees fall into criminal ways again and find
Iiemselves back in prison.

Among a special group of inmates, however, in a project given impetus
iy an NIMH grant, a preliminary followup indicates that only about 30
percent of those released have had to conic back.

That dramatic cut in the recidivism rate appears to have been achieved,
n great part at least, by applying principles of behavioral science to
;pecific educational and social problems of inmates. The basic tool has
icen programed instruction, under which a person proceeds at his own
lace through a given subject, whether it be writing grammatical sentences
)r building good relations on the job, and wins rewards as he goes along.
The project is conducted by a private nonprofit organization, the
Rehabilitation Research Foundation, chartered "by the State to develop
ind administer programs of research and training in human development.
Offices and most classrooms, all within the gates at Draper, are in a
varehouse-type brick building that once served the prison as a cotton mill.

Dr. John M. McKee, the psychologist who conceived the program and
vho, as executive director of the Foundation, administers it, points out

371

that the participants have been volunteers, so presumably they have been
more highly motivated than the nonparticipants to improve themselves
and to make good in the outside world. Offsetting this at least to sonw
extent, he believes that many and perhaps most of the students originally
wanted only to escape work on the prison farm, or at maintenance jobs,
or in the laundry, which serves the entiie State prison system.

Besides willingness to volunteer, prospective students usually have liad
to demonstrate a sixth-grade level of education, which was considered
essential if they were to benefit from either academic or vocational
training. Beginning in 1968, however, applicants below this level have
been accepted and given 20 weeks of individual and group instruction
directed toward fitting them for an occupational course. Applicants'
criminal records (except in the case of rapists, who were ruled ineligible
by the State Board of Corrections because there are women on the
Foundation's staff) and their records of behavior in the institution \\m
not in themselves affected admissibility. But priority has gone to men
having a reasonable chance of being paroled soon after completion of
training.

Doctor McKee traces the genesis of the Draper project to 1959, when
he was one of a group of psychologists who dreamed of establishing a
center for research in behavior, with emphasis on the application of
learning theory. The psychologists were staff members of the Division of
Mental Hygiene of the Alabama State Department of Health; their
behavior center would be part of the University of Alabama, and one of
its many possible research studies would deal with inmates of Alabama
prisons. When the group was unable to raise the necessary funds for the
whole program, Doctor McKee decided to see what he could dp with (he
correctional project. He felt certain, after several years of directing mentnl
hygiene work in the State's prison system, that if you wanted to pry a
man loose from antisocial activities, you had to show him how to become
and-more important, to get him to believe that he could becomc-a
self-respecting member of society. Programed instruction, Doctor McKcc
believed, offered a highly promising way of meeting these require-
ments.

Draper's warden, John C. Watkins, who had majored in sociology at the
University of Alabama and then completed course work for a masters
degree from Auburn University, strongly encouraged Doctor McKee and
offered to make Draper available as an experimental center if Alabama's
Commissioner of Corrections approved. A. Frank Lee, the commissioner,
not only approved but also scraped together $3,000 to help finance a trial
run of McKee's idea. ("We sold the commissioner," McKee recalls, "by
giving him a little programed sequence that quickly taught him how to
multiply in his head certain 2-digit numbers by other 2-digit numbers,")
That was in 1961. The following year, NIMH made a grant to expand
the work and analyze the results. Institute support, which continued into
1969, went mainly to develop, test, and apply self-instructional programs
in the academic and personal-relations fields. The findings were so
promising that in 1 964 the Office of Education and the Department of

372

Labor granted funds, under the Manpower Development and Training Act
(MDTA), to enroll more inmates and to add vocational subjects. Work
programs under the NIMH and MDTA grants have complemented each
other.

For the most part, enrollees now spend 8 hours a day in the program.
There is considerable variation in how these hours are apportioned. The
student whose aim is a high school equivalency certificate, for example,
will spend most of his time in academic work. Typically, though, two of
the hours are given to basic education courses, including English, remedial
reading, and arithmetic, and the rest to vocational training— in the
classroom for theory and in the shop for practices. Vocational courses
include welding, repair of electrical appliances (including air conditioning
and refrigeration systems), barbering, sign writing, and work in an
automobile service station.

Among the regular staff membeis, in addition to Doctor McKee, are
Donna M. Seay, a vocational educator who serves as the Foundation's
assistant director; four learning managers or supervisors, who in a
conventional educational system would be called teachers; five vocational
instructors; and three counselors, for guidance to men in the program, for
job placement, and for followup.

The staff is augmented by the so-called college corps-junior and senior
college students, majoring in sociology, psychology, or education, who
join the regular staff during summer vacations or for part of a semester
during the school year. Some 50 corpsmen from a dozen colleges and
universities in Alabamp and neighboring States have worked at Draper,
three or four at any given time. They have served as counselors, as aides to
the regular instructors, and, perhaps most important, as "peer
models"— examples of the fact that a young man can do something
worthwhile with his life. "With encouragement from the corpsmen,"
Doctor McKee reports, "many inmates came to believe for the first time
that their lives counted and that they, too, could succeed."

About five "service" corpsmen also are on duty, mainly as assistants to
the instructors. They are relatively well-educated inmates who have shown
special interest in the project and the desire and ability to work with
members of the regular staff and with fellow inmates.

Augmenting the academic and vocational work, the project has set up:

1. A reading laboratory, which evaluates and provides help for each
student who seems deficient in reading skills. Some of the work— notably,
rate-comprehension exercises presented by a variable-speed film
projector— is clone in groups, but most of it is individual. Each student
receives a kit that contains stones and self-graded tests on the content of
the stories. Vocabularies are developed through other self-graded
exercises. Students are encouraged to spend at least an hour a day outside
the laboratory reading any library book of their choice, provided it is not
far below their known level of ability.

2. A seminar program, which provides the group interaction not
present in the programed instruction courses. Group discussion, or
seminars, have dealt so far with English composition, poetry

373

appieciation, human behavioi, great books, creative writing, and
events. Most of the leaders have been volunteers fiom Montgomery
recruited by Doctor McKee because of their interest in the subject matter,
The program has been so successful in awakening and expanding the
prisoner's interests that it will be continued and broadened.

About 100 inmates were enrolled in the Draper project as of mid-1969*
bringing the grand total of participants-including a number in the early
years who were involved for only short periods-to almost 2,000. Better
than 40 percent of the current enrollees are Negroes, a proportion that has
more than doubled since the start of the program. Two thirds of the
institution's population is white.

Doctor McKee reports the following major findings, most of them
based on a followup of 228 graduates of the vocational training program
(out of 290 who had been released) from 1 to 3 years after they hud left
Draper.

• Sixty-eight percent of those released-most of them by parole, the
others by completion of their terms-were still free. Of the other 32
percent, almost half had been returned to Draper for technical violations,
such as failure to keep in touch with parole authorities, or consorting with
known criminals, while the rest had been picked up for new felonies or
misdemeanors. This contrasts with the pretraining record of these men:
fully 70 percent had been recidivists and only 30 percent had been first
offendeis. For the State as a whole the recidivism rate was estimated at
between 50 and 70 percent.

• Of the 288 men in the followup, about 80 percent had initially gone
into jobs directly related to their training and another 10 percent had
gone into such jobs later on. The number of employers who hired
graduates reached 361, since some graduates changed jobs several times.
Two-thirds of those released got jobs through the efforts of the program's
job development and placement officers. Only a few graduates-eight out
of the 187 for whom this information was available— performed
unsatisfactorily in their first jobs,

• The average income of the trainees who have been released— based on
follow-up data from 150-is $3,640 a year. Each man pays an estimated
average of $546 a year in taxes. Moreover, the public has been relieved of
the burden of maintaining the man in prison, estimated at $1 ,200 per man
per year.

• After 200 hours of programed instruction, students have shown an
average gain in educational level of 1.4 grades. When this instruction was
coupled with a reading improvement course, the gain was 2.5 grades,

• In the hope of getting a certificate of high school equivalency, 178
students have taken the State's General Education Development tests, and
all except nine have passed the tests and received the certificate. These
students typically had been at the seventh grade level or even below in at
least some of their subjects when they enrolled. (Incidentally, evidence of
the cash value of a high school education was turned up by a 1 967 study
of one group of former inmates. Men who had earned a certificate of high
school equivalency during their imprisonment were making an average of

374

$349,30 a month; the others were making £279. As compared with
pre-irnprisonment earnings, the high school equivalency group had added
more than $140 a month; the others, less than $40.)

• Thirteen students were accepted by colleges while they were still
inmates, and entered the colleges upon parole. Three of these have
graduated; the others are still in college. Most of the students have
received help from a scholarship loan fund called PACE (Program for
Achievement of a College Education), which the Foundation administers.
The fund was started in 1962 when the inmates who had taken part in one
of the early progiamed instruction experiments, for which they were
paid, donated their earnings to establish it. Since then it has been
suppoited by gifts from outside. Loans to college students, graduates of
the Diaper program, have ranged from $40 to over $2,000. The fund has
had to turn down some requests and meet others only in small part.

To disseminate information about its accomplishments and methods,
the Foundation has held four conferences—in Montgomery, Houston
(with the University of Houston as cosponsor), New York City (with the
Staten Island Mental Health Center as cosponsor), and Berkeley, Calif,
(with the University of California as cosponsor). Each was attended by
more than 100 persons from correctional institutions, pardon and parole
boards, State employment services, State manpower development and
training programs, and vocational rehabilitation agencies. More than 2,000
persons have visited the project for first-hand information. A manual on
Draper's training methods has been prepared.

Correctional institutions in a number of States— including California,
Florida, Georgia, Illinois, Indiana, Nevada, Oklahoma, South Carolina, and
Tennessee-have adopted some part of the Draper program, and others
have expressed interest in doing so. For lack of funds, Doctor McKee
reports, Alabama has not yet been able to extend the program beyond
Draper.

Programed Instruction

At the basis of the program is learning theory, or reinforcement theory,
which says that if desirable behavior is rewarded and undesirable behavior
ignored or punished, a person's behavior will eventually be modified in the
desired direction.

"There is no substitute in this world for competence," says Doctor
McKee. "And that is what these trainees have lacked all their lives— a
feeling of competence. They have failed in every major task they have
undertaken, perhaps with good reason: the family may have failed them,
or the school. As the warden says, they have even failed in crime, or they
wouldn't be here."

Programed instruction, or P.I., is intended to substitute success for
failure. As defined by Doctor McKee, it is "an attempt to systematically

375

01 learning theory to an individual learning
situation." Subject matter is said to be programed when:

1. Its training objectives are behavioral: that is, the learne-r will be able
to do something specific.

2. Material is presented in a logical sequence of steps-called frames or
exercises-each small enough to be easily learned.

3. The student is requited to respond actively to each step; that is, he is
asked to apply information or exercise skills immediately after they are
introduced.

4. The learner receives immediate feedback; that is, he is told at once
that he is right or wiong. In good P.I. he is almost always right. TJife
explains the behavior-shaping power of P.I., says Doctor McKee, because
the knowledge that a response is right is reinforcing to the student: he is
encouraged to go ahead. If he is wrong, he is told why and may be
referred to a section that helps him over the difficult part. He is asked to
conect his answer; then he is moved forward in the program,

A course on sentence structure, for instance, begins by defining subject
and predicate and giving examples. Then it offers a few simple sentences
and asks the student to pick out the subject and predicates. It tells him
the answers. If he has been right, he goes on to more complex sentences; if
wrong, he reviews the definitions and corrects his mistakes before going
ahead.

One series of lessons teaches personal hygiene as an important factor in
getting and holding a job. Other lessons try to teach desirable work habits
and customer-relations attitudes. Here is one unit, or frame, from the
customer-relations course for automobile station workers;

Make the first move count.- J. Q. Customer drives into your shop,
This is his first visit, and he feels a bit wary. The longer he is left
waiting, the more doubtful he becomes. He may even decide that
you're not interested in his business and leave.

* * * But no! You're on your toes! You see that no one has made
a move to help him. At that particular moment you cannot leave
your work, so you call out, "Good morning, sir! Someone will be
with you in a moment." He at least knows now that someone is
aware of his presence and is hurrying to assist him. The sooner you
can approach him, the better.

Put a check mark by the following statements that are true;

\ , Most customers don't mind waiting, • —

2. If the customer is kept waiting too long, he may leave. — — •

3. You should approach the customer as soon as possible, , .

4. Keep the customer waiting for a while to show him how hard

everyone is working.

5. If it is not possible to approach the customer immediately, tell

him that someone will be with him as soon as possible.

You scored 100 percent if you checked numbers 2,3, and 5, You
realized that most people do not appreciate being kept waiting. Time
is important to all of us, and no one likes his time to be wasted. Too,

376

a customer will be more impressed with your attention to his needs
than he will be with your efforts to fix someone else's car!

If you checked a wrong answer or failed to check the correct
answer, go back and correct your answer before continuing.

In the Draper project, a learning area replaces the usual classroom. It is
a room with rows of stalls or cubicles, like those found in college library
stacks, so that each student has his own, compartmentalized, study space.
A dozen or 15 students work in the same area under the eye of a learning
manager or supervisor, but no two students are likely to be engaged on the
same subject. The supervisor answers questions, offers encouragement,
and keeps records.

Programed instruction makes it easy to woik with a person at several
levels, depending on his need. One student, for example, is relatively far
advanced in English, so he is studying it at the eighth grade level. Since he
is extremely weak in arithmetic, howevei, he studies that subject at the
fifth grade level.

P.I. materials come either in text form or for use in teaching machines.
The machines have advantages. They can automatically skip ovei material
that the student's responses show he already knows, and they can present
extra practice sequences when the response indicates these are needed.
Further, they can control cheating and can tabulate the number of correct
and incorrect responses, thus easing the supervisor's job. The Draper
project uses them to some extent but, largely in the interest of economy,
has relied mainly on texts, or "software," and feels the results have been
good.

In the beginning, the available P.I. courses covered only a few subjects
and often lacked what Doctor McICee considered a necessity;
self-administered tests that evaluated the subject's knowledge at the start,
at various points along the way, and at the end. So the project set to work
to develop materials of its own. Some 30 Draper-tested courses covering
aspects of such basic subjects as grammar and mathematics, such
vocational subjects as electronics, welding, auto mechanics, and
bricklaying, and such social relations areas as applying for a job, good job
habits, and table manners have resulted; they are on sale through the
University of Alabama.

The development of its own courses went hand in hand with the use of
existing ones. One of these, used purely because it did exist, was Russian
language for beginners. But could this possibly have any value to prison
inmates? "Yes indeed, it could and did." answers McKee. "The fellows
who took it went back and spouted off a few words in Russian to one
another. And when the others asked 'What's that?1, our boys said, 'If you
went to The School, you'd find out.' Their new knowledge gave them
prestige, and that's important."

Some of the students began programing on their own. One day, a few
months after the project began, two inmates— call them Ted Jones and Bill
Smith— came to McKec and said, "Hello, Doc, take a look at this."
McKee looked and saw a sheaf of handwritten instructions and diagrams
entitled "A Programed Course in Fingering Movements on the Trumpet.

377

By Ted Jones and Bill Smith." It had 1,164 frames. Leafing through if,
McKee remarked, "Gee, I believe I could take this course and learn to play
the trumpet. It even teaches you how to blow through the trumpet. You
blow as if you were going to spit a piece of tobacco off your tongue/'

"Well," Ted Jones answered, "if you are as smart as the four men we've
already put through the program to test it, and at the end could play
those three tunes-yeah, if you are as smart as they are, you could Ictnn to
play those tunes, too."

"Another year and a day for you," McKee joked back, and went on to
encourage Ted and Bill to continue their work, which they did, writing
lessons for the clarinet and saxophone and a basic course in music
notation (One of the musicians is now in college; the other has been
returned to Draper for a parole violation.)

Over the years other inmates have developed courses in personal
grooming, etiquette, parts of speech, improving the memory, orientation
for new prisoners, and other subjects. One man who was studying
technical writing signed up for a half-year course in welding just so lie
could program lessons in the theory of it. The course in technical writing,
incidentally, was diopped after a few years because employers shied away
from hiring its graduates. But the welding lessons, like most of the other
courses developed by inmates, have become part of the Draper
curriculum.

Meanwhile the commercial production of P.T. materials has boomed to
such an extent that hundreds of courses are available. The problem fit
Draper is no longer to develop new courses but to select from the most
promising ones at hand, One new program, though, is definitely planned.
It will be written with the aid of offenders-in particular, successful iiml
unsuccessful parolees-and will teach how to succeed on parole.

Reinforcement

Although programed instruction has its own built-in reinforce ment-Ehe
lift one gets from finding that an answer is correct (or, if it isn't correct,
from learning what was wrong and from being encouraged to go
ahead)-the staff at Draper found that the typical inmate got bored after a
few weeks. So it has employed a number of tangible rewards to create a
motivating environment. For instance, students who complete a course arc
given points, and enough points win them a Certificate of Achievement,
Along with written recommendations, these certificates provide a stiong
basis for favorable parole consideration. Many former inmates have
reported that the certificates also helped them to get jobs.

Now undergoing trial is a "contingency management" plan under which
a student contracts at the end of each day's classroom work to complete ii
certain number of frames the next day. A frame comprises one question
and one answer in a programed instruction course, A contract is an
agreement, recorded on a form, between the student and the supervisor.
The student is free to specify the number of frames he will do provided

378

this is at least equal to his accomplishment during the baseline period at the
beginning of the course. The student also agrees to take the tests assigned
by the supervisor at the appropriate points in the course. And student and
supervisor agree that upon completion of a certain number of frames or of
a test the student will be free to take a 15-minute break. {The amount of
work to be completed before a break depends upon the total amount for
the day as specified in the contract.) The agreement makes it possible for
each student to average one break an hour.

The breaks are taken in a room- the "reinfoicing event" area-fitted
out with a coffeemaker and supplies, newspapers, books, magazines,
letter-writing materials, shoe-shining equipment, crossword puzzles,
playing cards, checker and chess sets, and a radio.

An experiment that preceded the adoption of this plan involved 16
students, First they worked through a 3-week baseline period having two
scheduled breaks per day. Then came 4 weeks managed by the
experimenter. During the first two of these weeks he set the performance
level at 20 percent higher than the baseline average; during the last 2
weeks, he raised it another 20 percent. Most students achieved these
increases.

Then came 2 weeks of self-management, with the students themselves
at the end of each day setting the amount of work they would do the
following day. In almost every case they agreed to work and did work
faster than during the second phase, when the experimenter was calling
the pace. Frames completed per hour averaged 61 during the baseline
period, 101 during the experimenter-managed period, and 125 during the
self-management period.

These increases in work per hour were accompanied by a decrease in
the number of hours worked. During the second and third periods, the
students were permitted to leave the study area when they had completed
the stipulated or agreed-upon number of frames and the appropriate tests.
They could go back to the dormitory and rest, work out in the gyrn, talk
to friends, or engage in other free-time activities. Consequently, the
number of hours worked per day dropped from slightly more than 5 in
the baseline period, to 4.35 in the experimenter-managed period, to 3.4 in
the self -management phase. But the total work completed rose from 320
frames in the baseline period, to 405, to 435. The increase in frames
completed did not adversely affect test performance.

In sum, the experiment indicates:

1 . The knowledge that a reward is waiting spurs students on. This is so
whether the work to be done is set by the experimenter or by the student.
But when it is set by the student, productivity seems to increase still
further.

2. In many cases, permission to leave the experimental area at the
completion of the performance contract for the day acted as a more
powerful reinforcer than the immediate reward-one or more of the
activities available during the 15 -minute-break periods. After students
were told, part way through the experiment, that they might continue to

379

study rather than take an eained break, the gioup as a whole chose
continued work as often as it chose immediate relaxation.

In addition to the built-in leinforcement of programed instruction ami
the extrinsic, tangible rewards offered at Draper, something else is al
work. In the process of learning to be a student, Doctor McKee obseives, a
man learns to enjoy certain intellectual pursuits. He reads the newspapers,
he picks up a book, he listens to the news on TV, he carries on iin
intellectual conversation— his whole world is changing and opening up, lie
is going from such reinforcers as a party, free time, and points towaid a
certificate to find intrinsic rewards in the subject matter itself and in tliu
process of his development. Tlu's is theory, but you see it working in the
lives of all of us.

The attitude adopted by the staff is also a motivating factor, Doctor
McKee believes. "We attempt to be firm but not punitive," he says,
"helping but not totally permissive, and flexible but not vacillating. Jty
employing positive rather than negative reinforcement whenever possible,
an atmosphere had been created that is much different from the hostile,
punitive one expected by most inmates."

As an example, he cites his own behavior when he helps administei a
test for a high school equivalency certificate. It's a long test, dragging on
for several hours. Since the typical inmate's span of attention is short, and
since McKee knows from experience that the inmate may say "The -with
it" and walk off, the director keeps coming back and showing interest.
"How are you domg-any pioblems?" he'll ask. "I don't tell him any
answers," McKee says, "but I get him to say, 'Well, yeah, this part I just
took,' or something like that. And I say, 'Well, let me see,' and look at the
part and say, 'Yes, you've got some tough questions there. Better go hack
over what you've been reading and writing to make sure you've got the
light answers. 1*11 be back.' " What I'm trying to do is to maintain alert
behavior-keep them sticking with the job and not bugging out.

"That's what they've been doing all their Jives-bugging out, escaping.
And strangely enough, this escape behavior has been reinforced. Rob a
house, forge a check, steal a car-and run away. Girl friend squeezing you
in? The job getting you down? Run away.

"Even people in correctional work have been reinforcing such behavior.
An inmate will get tired of the work he's been assigned to. Or he'll get
tired of the person he's working with. So he starts manipulating a change.
He'll tell the classification officer: 'I'm bugging out of this. I need a
different job. My supervisor's down on me-I don't know why.' And hell
keep trying and after a while he'll get his change."

From staff interviews with prospective students, McKee suspects that a
large proportion of the inmates who volunteered for school were simply
running away once again. But he accepted them because in his project
there is nowhere to run to, except back to the job they escaped from.

380

A "Second Language"

Doctor McKee emphasizes that with prisoneis-and probably any other
population, for that matter-programed instruction has to be
supplemented by discussion groups and sometimes by individual
counseling or teacher-student conferences. The inescapable inadequacy of
many P.I. courses, standing by themselves, was brought home to him one
day a few years ago when he congratulated a trainee on having made the
extraordinarily high grade of 96 in a high-school-level grammar course.
The trainee gave a deprecatory smile in response and said: 'This here
English, Doc, don't give me no trouble nohow." Like many other
students, he had learned the rules but wasn't applying them.

The project then tried the seminar approach-trainees meeting in groups
with a teacher and working together at speech modification. This was
much more successful. As the trainees listened to one another, they
learned to detect errors, which they called to the group's attention.

Now the staff is experimenting with what it calls the "second language"
approach; that is, teaching standard English by the oral methods used in
teaching a foreign language.

In this approach a man's present speech, no matter how poor by
conventional standards, is never criticized. It is the one he has picked up
as a child, and it is usually quite effective in meeting his needs to
communicate with his usual associates. The trainees are simply told that
the school wants to give them the ability to use another kind of speech on
appropriate occasions, such as when applying for a job or meeting a girl
friend's parents. "We all have different levels of speech," points out Sally
Roy, a former public school teacher who is now a research associate with
the Diaper project. "A child of five will speak to his mother one way, to
his friends another, and to his teacher a third. We speak one way to the
dog and another to the Mayor. The hope is that the trainees will use this
different level we are giving them more and more often and that
eventually it will become their usual level."

The project got the idea from St. Mary's Dominican College Business
School in New Orleans, a basic education center for the Job Corps. In
training Negro, Cajun, and other disadvantaged girls to be secretaries, staff
members from Draper found, Dominican College was trying not to correct
their language but to give them a new one for use on the job. It called the
new language "business speech."

In its second-language experiment, the staff at Draper gives extensive
drill work on the most common errors noticed during talks with the
trainees— errors exemplified by such sentences as "John and me want to
go to town," "He don't know no better," and "They is ready to fight,"
The trainees hear over and over again, on tape, the correctly spoken
version of these and hundreds of other examples. They also listen to
themselves at frequent intervals. And there is group work, including mock
job interviews, some of it recorded by a videotape machine.

Has any employer said, "I can't take this man because he doesn't talk
well enough?"

381

, . -_ ~. itvw uiieans," answers Doctor McKee. "People (lie«
said, l\Ve can't take tliis girl; she doesn't speak right; she'd give us ab:l
image/ We don't know whether or not it has happened here. But i! SLYJI>
reasonable to suppose that an employer hires a man on the basis of tto
total Image he casts, including his ability and his self-confidence, and ilui
his speech is part of this image. Even an employer who himself snys Ik
don't' cannot help being impressed by a man who speaks wcff
Rehabilitation is a cluster of things, including attitude changes. We tliinl
that proper speech is part of the clustei. It is not merely that the man wlio
learns what we are calling a second language has acquired a new skill but
that he is enhanced by having done so. People look twice at him urni think
that he is worth more. He himself has a greater feeling of worth ami a
sense of belonging."

For the Future

Aided by a new grant from the Department of Labor, Doctor McKw
expects to continue the program of education and vocational training urnl
to make more extensive and rigorous followups. He hopes to answer thesu
questions, among others:

» What is the most effective type of training program, as measured by
work adjustment and recidivism? Can P.I, simulate real work situations,
beyond that of a job interview? Can realistic work-stress conditions be
simulated and the means of handling them be transferred to nclunt job
conditions following release?

• What specific barriers, particularly those related to employe*
attitudes, make it more difficult for the offender to find and hold it job?

* Do released convicts who receive "labor mobility" funds (siiwH
grants of Federal money to tide them over until the first paycheck conies
in) or Federal bonding assistance, or both, have a better post-release
record than those who do not?

» How can an inmate's behavior, not merely during the hours given to
the training program but throughout the day, be shaped to conform to
desired standards? Involved are (a) deciding which kinds of behavior arc
desirable, because they make for a successful transition from the
institution to a job in the community, and which kinds of behavior are
undesirable, (bj determining the frequency with which each kind of
behavior occurs m the sample of men to be studied, and (c) selecting,
applying, and measuring the effectiveness of measures-am ong them,
probably, certain changes in the attitudes of the custodial staff-to
encourage one kind of behavior and discourage another. This, then, is n
major, long-term experiment in behavior modification and one that is dear
to Doctor McKee's heart.

382

Research Grant. MH 14990

Date of Interview. December 1968

References

Clements, C. B., and McKee, J. M. Programed instruction for institutionalized

offenders' contingency management and performance contracts. Psychological

Reports, 22:957-964, 1968.
McKee, J. M, Adult basic education for the disadvantaged: procedures used to raise the

basic educational level. Presented to the American Vocational Association, 1966.
McKee, J . M. Programed instruction in the correctional process. Presented at

Conference in Manpower Training for Offenders in the Correctional Process, 1968.

383

Investigator.

James F. Short, Jr., Ph.D.*
University of Chicago
Chicago, III.

Prepared by

Herbert Yahraes

Members of juvenile delinquent gangs, according to a theory atlviinc
m an important new study, suffer from a condition described us ,mo
disability. Much more so than other boys, they grow to school ii
without the skills needed to meet new situations. Thus Uiey ;i
handicapped both in getting along with teachers and classmates and
earning, and their school experiences in turn make them additional
Handicapped for conventional activities later on, including work. In t
effort to meet the universal need for relationships with other peopif, tin
drift into gang membership. it*

t0 3 co?Pleme"tary theory advanced in the same slmh
/ 3 gf"gs delmcluent ac«vities may be likened to playing
* Siight reward bllt j» which also a MOM,
may be exacted« If »c«<>n is coming u

is termed the ,/wory ,,

presses d gang memb°'-s'

the or r t,prresses r d gang memb°'-s' ™IMS

the course of a study o Ch icl lPf °ty mks "aVe been ^wtopcd I.
with support from the Si f ; t°riier groups' undertakeiUl1 IW
practically finished ""' Ins"tute of Ment^ Health and iio«

^ Dr; J— P. Short, J,, and Dr.
sociologist, is now dean of the graduate

*Wow at the Department of <:„ • i
Washington. P ^^ °f *"****. Washington State University, Pullman,

384

school at Washington State University, but during the study's fieldwork
served at the University of Chicago as a visiting member of the faculty. Dr.
Strodtbeck holds appointments at the University of Chicago in the
clepartments of sociology and psychology and is director of the Social
Psychology Laboratory. Another major contributor has been Dr. Des-
niond S. Cartwright, now. at the Univeisity of Colorado.

The investigators believe their study to be the largest attempt yet made
to get information useful in understanding and dealing with teenage
delinquent gangs. It grew out of a new piogram under which college
graduates working for the Y.M.C.A. of Metropolitan Chicago hang around
^vith gang membeis-listening to theii talk, going along with them to
parties and to encounters with other gangs, encouraging them to take up
sports, getting them admitted to ball games, seeing that they are treated
for venereal disease, talking them out of stolen cars, putting them in touch
"With employment programs, and so on. These Y employees, assigned one
to a gang and known as detached workers, came to be widely accepted
and counted upon.

Through weekly interviews with the detached workers, frequent
observation of gang activities, talks with the boys themselves, and the
administration of various psychological tests, the researchers studied 16
jgangs in considerable detail. These gangs, of which 1 1 were Negro and five
White, had 598 members.

For comparison, the researchers studied several hundred boys who lived
in the same areas as the gang members but did not belong to gangs. They
studied also two Negro and two white groups of middle-class boys, all of
them members of Y.M.C.A. Hi-Y clubs; no middle-class gang could be
located.

The principal concern was with group motivations and processes— with
what makes gangs tick.

The project's major findings are summarized in the numbered
paragraphs that follow:

1 . Gang members and other lower-class boys express as high a regard for
middle-class standards as do middle-class boys,

This finding came as a surprise, Dr. Strodtbeck reports, because it ran
contrary to the widely accepted theory that juvenile gang delinquency is a
reaction against middle-class standards.

The investigators arrived at the finding by using a semantic differential
to study values held by the boys. A semantic differential is a set of paired
•words, opposite in meaning, for learning how people regard a given set of
goals or standards. Among the pairs in this case were good-bad,
pleasant-unpleasant, and smart-sucker. Each word in a pair was one end of
a seven-point scale. If a boy considered something wholly good, for
example, it was given a value of one; wholly bad, a value of seven; just
rnidway between good and bad, a value of four. Among the images rated
by the scales were these:

Representative of middle-class standards;

Someone who works for good grades at school.
Someone who likes to read good books.

385

someone who saves his money.
Representative of lower-class standards:

Someone who has a steady job washing and greasing cars.
Someone who likes to spend his spare time hanging on the come

with his friends.

Someone who shares his money with his friends.
Descriptive of many boys in delinquent gangs-

Someone who is a good fighter with a tough reputation.
Someone who has good connections to avoid trouble with Lhc law.
Someone who makes easy money by pimping and other illegu

hustles.

All boys studied-gang, nongang, and middle-class-evaluated highly the
images representing salient features of middle-class style of life. No Oilier
images were evaluated more highly.

But lower-class images did get a higher score from lower-class boys than
from middle-class boys. And criminal images, though rated low, received
higher scores from gang membeis than from nongang members and higher
scores from the latter than from middle-class boys. Further, some
evidence was found that gang boys more than others might be inclined to
judge participation in illegal activities from the standpoint of whether or
not it was smart rather than whether or not it was good.

The boys' ratings of themselves, as they are and as they would like to
be, showed that regardless of the group to which a boy belonged he would
like to be, on the average, both smarter (meaning less of a sucker) niul

better.

2. None of the gangs studied can be characterized as strictly criminal or
as being part of a criminal subculture. However, illegal activities were
characteristic to some extent of all the gangs.

The activities of gang members, as reported by the detached workers,
were combined into 37 items of behavior. Factor analysis then disclosed
the presence of five groups of closely associated activities, some of the
groups being more characteristic of certain gangs than of others. The five
groups, or factors follow:

(a) Conflict. Main items: individual and group fighting, the carrying
of concealed weapons, and assault. Though robbery, theft,
becoming a public nuisance, and statutory rape also appear in
this factor, they are part of other factors as well.

(b) Stable corner activities. Characterized by sports, social activities,
and gambling. No serious delinquent item is prominent.

(c)A cluster of behaviors difficult to sum up but finally called
stable sex pattern. Principal items include sexual intercourse,
statutory rape, petting, and the buying, selling, and use of
alcohol. Holding a job is fairly prominent, too-the only factor in
which it is prominent.

frfjCalled retreatist because the buying, selling, and use of narcotics
and marihuana ranked highest among the activities. Only one
retreatist gang was found.

386

(e) Authority protest. Principal activities: stealing autos, driving
without a license, becoming a public nuisance, drinking, running
away from home, playing truant, joyriding.

From the individual scores, the investigators deteumned the average
score of each gang on each factor. Gangs varied most greatly on (a), least
on (e),

The ranking of a gang on any one factor was found to correlate
positively with its ranking on all the other factors except (b). The highei a
gang's scoie on the conflict factor* for instance, the higher its score on all
the others except the one built around sports and social activities.

The six gangs found to engage in the most conflict were all Negro.
Three of the four gangs that engaged in the least conflict were also Negro,
White gangs ranked significantly higher than Negro only on the "authority
protest" factor. The gangs ranking highest in "authority protest" were
those engaging least in sports and other activities of(b).

3. The white community is more concerned than the Negro with the
excesses of its young people, and in general controls its young people
more effectively. Among the many and complex reasons are the
greater economic stability of white communities and the existence of
indigenous institutions of established leadership.

From a statistical study of the relationships among various types of
behavior, the investigators find that Negro gang members tend to make no
clear distinction between delinquent and conventional behavior. The
Negro gang boys participated in such conventionally approved activities as
doing the household chores and taking part in organized sports; they also
participated in such conventionally disapproved activities as fighting, illicit
sexual relations, drug use, and auto theft. In contrast, white gang boys
were less inclined to participate in approved activities and were more
openly at odds with the adult community, particularly concerning
rowdyism, drinking (almost universal), drug use (rare, compared with the
situation in Negro gangs), and sexual delinquency. In shoit, the
Investigators suggest, delinquency among Negro gang members is part of
the total life pattern, and the community appears to accept it as such.

Life in the lower-class Negro areas was found to be organized around
sxich institutions as "quarter parties" (sometimes called "rent parties"),
pool halls, and taverns to a greater extent than in comparable white areas.
One Negro poolroom that was a hangout for prostitutes and thieves was
also a hangout for gang boys. Among the persons present on one occasion
was a man who had just escaped from the police, stilt wearing one
handcuff. He was being kidded.

In the white areas, the investigators report, life generally revolved
around more conventional institutions-the Catholic Church in particular,
ethnic groups, unions, bowling leagues, other organized recreational
patterns, and political and "improvement" associations: these last
organized in large measure to keep the Negroes out.

Economically and socially, "no white area [was found to be] as
disadvantaged as the least disadvantaged Negro area." The Negro gangs
came from neighborhoods where the median family income ranged from

387

S3, 200 to S5 100: the white gangs came from neighborhoods when;
ranged between S 6, 200 and S6,500. Unemployment was 10.6 portent
the Negro areas; 4.8 percent in the white areas, In the Negro areas, *
peicent of the dwelling units had more than one person per room; in tl
white aieas, 14 percent.

4. Rating themselves on friendliness, cleanness, smartness,
helpfulness, and goodness, white gang boys produced
higher scores than Negro gang boys. With the two
groups, though, (he situation was reversed.

if the Negro gang boys reaily are so fundamentally disadvantjigccl as tt
scores indicate, the investigators believe that "a major modification of tl
social system" will be needed to change "this negative self-image," (I
middle-class Negro boy rates himself highly, the study suggests, pnssib.
because he may see himself at the top, looking down, whereas Hie micHIl
class white boy may still see steps ahead.

The investigators reached their findings by having the boys in the sUitl
compare themselves and their associates with other neighborhood hoys u
the basis of 42 adjectives. The ratings— -more, the same, (ess— were givo
numerical values and subjected to statistical analysis. By various check
the researchers satisfied themselves that in making the ratings Um ho>
were actually describing themselves.

Some other findings:

Among Negroes, the higher the socioeconomc status (ranked 11
follows: middle-class, lower-class, nongang, gang), the highev the boy
rated themselves on masculinity, Among whites the opposite was Inn
What is involved, the investigators suggest, is a cultural difference in \\\
degree to which being manly is a focal concern.

The gang members who saw themselves as polite, loyal, helpful, swan
i-iean, obedient, and /-^(bus-rather than troublesome, mean, tough, am
coo! -were also the boys who reported engaging in more cleHmjucn
activity than the others. Explanation? Perhaps the boys with 111-
characteristics expressed by those adjectives are the ones who, more thin
others, hold the group together, the investigators suggest, and f
are thrown mto roles and situations leading to delinquent acts,

5. Gang members more than the other boys studied seem to su

in interpersonal retoltafis.

markedly higher than their

more likely a boy was to rate iSVTL ATT ^ Sodal Icvc!>

uccter tnan his associates,

388

Among lower-class, particularly gang boys, the investigators infer, there
exists an underlying dissatisfaction with their peers. Beneath the surface
solidarity of the gang, there are signs of mutual frustration and lack of
trust. Mental health depends strongly on accepting oneself, and the ability
to accept oneself, the investigatois note, seems strongly related to the
ability to accept others.

The inference about gang boys is apparently bolstered by a personality
assessment, which suggests that these boys are less self-assertive than the
other boys studied, more reactive to false signals, slightly more neurotic
and anxious, less gregarious, and more narcissistic. In sum, they are
lacking in qualities that make for helpful relations with others,

As an example of gang boys' insecurity, the study notes that they were
much more sensitive than nongang boys to how others were answering
questions and performing various tasks; they were more anxious about
their own performance as compared with others. Yet gang experience,
with its constant challenge to prove such qualities as toughness, smartness,
and adeptness with the girls, probably does little to alleviate such
insecurity except in the gang itself.

6, The roots of social disability appear to lie in early family life.

In considering the causes of delinquency, sociologists and psychologists
often emphasize status deprivation. But this is conceived of mainly in
occupational and economic terms, the present investigators point out,
whereas feelings of self-worth determined by factors not directly
connected with job levels and income rates may prove equally relevant.

Observations at an experimental nursery school for lower-class Negro
children-conducted at the University of Chicago by Dr. Stiodtbeck under
an Office of Education grant-suggest that at the age of 41/2 years these
children are less able than those of middle-class homes to associate with
other children without fighting. They seem to have been harshly brought
up and to have been cautioned frequently that the world around them is
threatening. Further, they are retarded in cognitive development and
verbal skills, partly because, the grantee finds, they have lacked
stimulation from reading materials and constructive play and partly
because they have not been exposed to an atmospheie that rewards good
use of language.

Gang boys, the researchers have observed, are relatively ignorant of how
to dress for a given occasion, eat in public, or carry on a polite
conversation. Their limited social skills are attributed to their narrow
social experience not only within the gang but also, and first, within the
family.

The family does not equip the child to meet the demands of school, the
investigators theorize, and unsatisfactory experiences in school further
narrow his opportunities to play roles that will prepare him for getting
along with employers and fellow workers and, in general, for dealing with
new situations. In contrast, middle-class parents early help their children
develop the ability to adjust as the situation requires. "Company" is
different from "family," and entertaining the boss teaches approved

389

means of relating to authority as well as something about the require
merits of dress and manners.

The study emphasizes that since the family life of the boys WHS no
investigated, there is as yet no proof of the suspected differoncc*
Neveitheless, the lower-class nongang boys did give evidence of having hai
more opportunities to learn how to get along with people, Roughly 7'.
percent of them had successfully adjusted to school, as compared witl
about 50 percent of the gang boys. Also, more nongang boys than ganj
boys reported having had contact with adults of relatively high status
such as the administrators of youth agencies, clergymen, and employers

As further evidence that the lower-class nongang boys had stiongc
family or institutional ties, these boys spent less time in places, inchidtni
street corners and poolrooms, where there was a high risk of becomiui
involved in delinquency.

Differences in intelligence, too, are reported as contributing to Uu
social disability of gang boys. The following table shows 1Q estimates
based on scores on so-called "culture-free" tests-tests that try tc
minimize the effects of previous tiaining. As the investigators put it, UK
school and other social institutions reward the bright boy-arid I he #m£
boys are handicapped in respect to brightness.

1Q Estimates

Negroes . .

Gang
69

Nongang
74

Middle class
Ofi S

Whites . .

85

91.5

JXl * J

in

7. One factor in the formation of Chicago 's juvenile gangs Is Die- It'ii.w,
fear/ul atmosphere of lower-class areas

^'°,IiVe '"."V3™ area v"y long, Dr. Strodtbeck comment*,
Ru' H ? fd? t0vvard one another "'at >^pccl properly
"le tt », r "iT" l'^'1 Negr° fam'Iy has moved th"« tillies

"

.V e- history'" the Srantee continues, «h.«

no

« d ,, yfbeeir°re distinctly Jerked
ft ™f ^ nfuP "elWOrks' Here in Wc^o, Negroos live in

™ H fei Jnoto^T' Ma^a tenement door has lhreo lwlts
trm We cornel l°m ^ that a neiSh^ may mean Harm but also Hint if

n7r tr4e" ™ --y°U ^i-311 ^ '° Wm> Y°U Wi" bC SUCktt'
ways of X-ee™gpeoT a San eTd ^ Can"0t be shsred' y°" fi>K'
•pu-tment door It does .in^^?™"1 communlty sto"s nt »c
extended kin." take m the buildi"g. «°r the block, nor

n te
too, if he can. S ° aSSUme that every «»»« else will got his,

agency representatives" the studv r" 7 St'f '°rs' police' and other

r

which other potentiany legitirna s± ™f ' "^ ™ atmosP|lere ""

Numate sources of support, as well as neighbors,

390

defined as part of a hostile outgroup. The result is that, for the
a liaison which is as informal as standing on the comer with
boys comes to be sought particularly because it requires so little
commitment of exposure."

By engaging in group delinquency, gang members help to satisfy
common adolescent needs both for working with and depending
upon others and for achieving status.

are likely to bring' to the gang a suspicion of the dependability of
relationships, the study notes, and the gang contributes to this
Status within the gang is subject to constant challenge, and
to status may disrupt even close friendships. But the gang does
a status system in which boys can succeed after they have failed in
~*Qre conventional settings.

Because adolescence is a period of emancipation from childhood, most
find it difficult to admit and to gratify their dependency needs,
is particularly the case among gang members, who are undei pressure,
n a culture that values toughness, to shy from anything likely to be
interpreted as a sign of personal weakness. But planning and canying out
^ang forays and talking about them afterwards— with opportunities to
Ldrnit to fright and to express appreciation for help—create a strong
Hough temporary bond of loyalty, the study reports, that probably serves
o meet dependency needs for a time.

The investigators point out that a gang boy's early decisions to do
something illegal are not necessarily a deliberate attempt to develop bonds
tvith other boys: Easy material and aggressive exploits are attractive in
tHemselves. The satisfaction that comes from working with a group may
be at the start only a bonus.

9. Short-run hedonism, or a search for pleasure without regard for the
consequences, has often been advanced as an explanation of why
gang members participate in seriously delinquent activities. A more
valid explanation may be a process that balances the immediate
outcome of an action, in terms of the gain in status if one joins the
action and the loss if one holds back, against the remote possibility
of punishment by the larger society if something goes seriously
wrong.

In putting forth this theory of aleatory risks, or risks strongly affected
t>y a probability process, the investigators conclude that the disposition of
gang boys to join the action cannot be explained either by the values they
Inold or by neurotic Or irrational tendencies. Further, the researchers
argue, short-run hedonism is not a good explanation because the term
implies, wrongly, that an atmosphere of abandon surrounds gang decisions
and activities.

As Drs. Strodtbeck and Short see it, in such a loosely organized group
as a juvenile gang there tends to be a continuous leakage of status and,
tlierefore, a need for continuously working status-maintaining
mechanisms. The boy who participates in a delinquent action wins a slight
reward; his status is maintained or even heightened.

391

To be sure-particularly in the case of fights, since guns and knives are
common in lower-class neighborhoods— there is a chance that the action
will lead to seiious consequences, with some of the participants being
hurt, killed, or arrested. But the investigators estimate that not more than
a fifth of the cases do have serious consequences to gang members, and
they point out that not more than a fifth of the offenses actually
committed by gangs studied were known to the police. Hence the risks are
likely to appear well worth taking. A boy can avoid all risk by staying
aloof, but in that case he suffers a loss of status.

1 0. Members of delinquent gangs do not pursue the middle-class goals
they espouse because, in part, (a) they don't know how and (b)
gang life, which they value, emphasizes behavior incompatible
with such pursuit.

Gang boys can be told how to reach middle-class standards, but the
information rarely penetrates to the point where it arouses action. In
this respect, Dr. Strodtbeck adds, they are like an inveterate cigarette
smoker who is told how he can improve his health.

The investigators believe that the boys mean what they say, both alone
and in groups, about such values as getting a job, saving money, getting
married. Alone, the boys late these values highly; in groups, they deride
them. In this respect they are like the medical school students who, in
another study, were found to speak idealistically about their chosen
profession only when alone with the investigator. Contradictory value
systems, the study notes, are a mark of modern society, particularly
among adolescents.

Gang life, as well as medical school, may be seen as a phase in a person's
career. The study explains that no gang boy expects this phase to last
forever. The trouble is that involvement in gang life hampers the
achievement of values held individually with respect to future phases.

Sex is one example. Among gang boys, as among boys generally, it was
found to be a matter of much concern and some anxiety. The pressure of
the gang compounds the problem, the investigators explain, because it is
harder for gang than for nongang boys to withdraw from sexual
competition and excel in some other endeavor. Among the gangs studied*
marriage was by no means taboo but "making out" was more highly
valued, and the boys who were the gang leaders always had finesse in
sexual matters.

Becoming the parent of an illegitimate child apparently does not affect
a boy's status one way or the other, and it may actually raise a girl's
status. Citing other studies, the investigators observe that the unmarried
lower-class Negro girl who becomes pregnant may suppose she has some
chance of landing the boy as a husband; in any event, she gets gratifying
attention from her female relatives and the social worker. And, Dr.
Strodtbeck adds, the financial help she receives from the Aid to
Dependent Children program may enable her to pull out and live in
slightly more savory surroundings that those at home.

"Both boys and girls," the report concludes, "are caught in a cycle of
limited social abilities and other skills. Their disabilities contribute to their

392

worries about status and in this way contribute to involvement in
delinquency."

As with sex, so with work. Individually the boys begged the detached
workers to help them get jobs, but on the street they bragged about
"hustles" of great variety. During one 3-year period, every member of one
of the gangs held at least one job, for a time, at the end of the period, not
one of the members was employed. The investigatois explain that the kind
of work such boys can get is generally neither sufficiently well paid nor
sufficiently challenging to serve as an acceptable status alternative to the
gang. Further, participation in gang activities is not conducive to a good
work record, and a good work record is not conducive to participation in
gang activities.

One gang member did fall in love with his job as messenger-in part,
because it gave him a uniform and a chance to talk to people. He married
his girl and began buying furniture and a car. Eventually he was promoted,
but the extra money was not enough to meet the credit payments. In
addition, the new job kept him from circulating and threatened to expose
his limited reading ability. He dropped out of legitimate work and is now
a well-paid employee of a group in the numbers racket.

1 1. Gang life offers few if any constructive tasks because, in part (a)
the leaders must choose activities at which most of the members
are proficient; (b) boys who have found fellowship and other re-
wards in delinquent activities may find the level of reward in
conventional activities too low; (c) the leaders know that the
easiest way to meet a threat to their status is to engage in de-
linquent activity.

Team sports might seem a promising basis for group solidarity, the
study notes, but gang boys tend to be aggressively poor losers. The
Y.M.C.A. program attempts to get around this handicap by seeing that
every team in a tournament gets a trophy —for participation or sportsman-
ship, if not for winning. The program also appoints influential gang mem-
bers as "field consultants*' at a small salary. Hence, at sports events,
consultants from rival gangs know one another and feel bound to some
degree to try to keep peace. Without such devices, the study reports, team
contests would often erupt into gang warfare.

V

12. The fewer the opportunities to which a group is exposed in its
own area— for example, recreational centers, counseling agencies,
churches, schools, chances to work— the more the group is in
trouble with the law,

Based on interviews concerning "the area where your group hangs out,"
the study developed opportunity scores for each boy studied and for each
group. These scores— which reflected the boys' awareness as well as the
actual existence of the opportunities-were then compared with the

393

average number of offenses, per group member, known to the police. Tiie
table gives the results.

Opportunity Scores versus Delinquent Behavior

Offenses, per

Oppoitumty boy, known to
Group score police

Negro gangs

9.0

3.14

White gangs

9.3

2.73

Negro lower class ,

11.0

.47

White tower class

13.7

.31

Negro middle class

15.6

.06

White middle class

20.2

.02

Possibly the most powerful influence against delinquency, Dr.
Strodtbeck reports, is the opportunity for a youngster to step into a job as
soon as he leaves school. "In Europe," he observes, "when you get out of
school, at any level, you can get a job that pays in accordance with the
contribution you can make to the production process. You have more
trouble doing that here— and more delinquency."

For the Future

The detached worker program, Di. Strodtbeck reports, has been
successfully used to prevent gang fights and is now being experimented
with as a gateway to what sociologists call "the opportunity structure," If a
boy can become interested in employment, perhaps he can also become
interested in going back to school or in getting more education and
training some other way. And if the road can be cleared to steady
employment for one boy, perhaps he can become a channel through
which other boys can be reached. But there are many forces pressing on a
gang boy to remain pessimistic about his chances in the larger world.

Research Grant: MH 3301
Reference'

Short, James F., Jr., and Strodtbeck, Fred L. Group Process and Gang Delinquency.
University of Chicago Press, Chicago, 1 965.

Investigators.

Malcolm W. Klein, Ph. D.

Helen E.Shimota, Ph. D.

University of Southern California

Los Angeles, Calif.

Prepared by:

Gay Luce

Juvenile delinquents, according to court records, account for less than a
million offenders in the country. But their impact ramifies, and the social
costs of their destructive behavior cannot be evaluated by a head count. A
high rate of illegitimacy and a new generation of underprivileged children
are among these costs. Because roughly four-fifths of all juvenile arrests
are boys, and because their gang activities have been most conspicuous and
amenable to study, the preponderance of our information about
delinquent gang behavior and background almost omits the part played by
girls, an information gap that should be filled if our remedial action
programs are to take root.

What is the nature of the delinquent girl gang? How does it influence
the boys' gang? Who joins, and how does she differ from her nongang
counterparts?

Some answers to these questions have emerged from a study in south
central Los Angeles. The neighborhood, 75 percent Negro, was a
residential region covering many square miles. Incomes ranged from under
$1,000 a year to $10— $15,000. Unlike the congested eastern slum, this
was an environment of small houses, duplexes, and apartments, spaced by
grass and trees. Less than 2 percent of the adolescent population there
belonged to gangs, and the gang membership was often dispersed over a
square mile. In the course of an extensive study of gangs and delinquency,
the investigators singled out a number of boys' gangs with a membership
totaling about 600, and a half dozen or so sister gangs, whose membership
amounted to about 1 50. These were Negroes ranging in age from 1 2 to 25.
Their gangs were in constant metamorphosis, shifting in membership,
disbanding, reforming. A rounded picture of these youngsters, their
backgrounds, psychometric evaluation, their influence upon one another
and opinion of each other was obtained with the help of a variety of Los
Angeles agencies, the schools, the police^ the juvenile court, and
particularly with the aid of special group workers assigned to the gangs by
the County Probation Department.

The procedure of collecting information utilized these sources at three
levels. In order to search out the interaction and influence of girl and boy
gangs upon each other and to determine the nature of these groups, a
system of personal interviews was established in which the gang members,

395

detained for a delinquent offense, weie asked to describe the role they
played, the roles of their associates and of the girls, and to give their own
peiceptions of the event. These were compared with police records, and
with the diiect observations of group workers and the study staff.
Interestingly enough, the youngsters cooperated well and offered candid
descriptions that jibe with the reports of police and group workers.

The half dozen or so girl gangs under scrutiny varied considerably in
size, cohesiveness, and activity. One of the larger groups had never met as
an organized unit. Another gang concentrated upon making life miserable
for schoolmates and fighting giils who were not gang membeis. The most
extreme gioup indulged m violence, hitchhiking and mugging drivers,
robbery, and vandalism. This gang dispersed when many of its members
became piegnant, and was succeeded by another gang of 20 younger girls
who may be carrying on the tradition.

In the course of the study it became clear that gangs of girls were not as
long-lived as boys' gangs. They were indeed dependent units, and with one
exception, they all began in relation to a boys' gang. They were
characterized by greater turnover in membership, and tended to disband if
their brother gang dispersed.

Within the literature on delinquency it has been speculated that sister
gangs often play the devil's advocate, inciting violence, starting fights,
encouraging greater and grander illegal plans. The findings within the
Negro gangs in south cential Los Angeles would indicate an opposite
effect, By questioning boy-gang members in this area after each incident,
it became apparent that girls rarely participated in the planning or action,
Frequently, the girl's role was that of an observer on the fringes of theft
or assault. When asked how. a girl's presence on the scene would influence
plans or action, almost half of the boys replied that the accidental
presence of girls would have postponed or prevented the event, especially
if the plan were a theft. In good measure, the boys' responses were a
reflection of a need for esteem. Among the adolescent boys, notably, the
girls' opinion weighed heavily, and few of the boys thought they would
win respect for attempting an offense and getting caught. In sum, the boy
who valued the girls' opinion of him said he would be prevented or
delayed in his illegal plan, It seemed that the boys could take chances of
being apprehended without worrying about their image, but if a girl
actually witnessed the act there would be no way of denying or
disclaiming the behavior.

Although the actual pillage and violence perpetrated by these juvenile
gangs is probably overdramatized, these are youngsters who disrupt their
neighborhoods and schools, use drugs, and destroy more than they con-
struct. They are part of a minority group in an extended ghetto, low in
income and high in unemployment. They are a small part of a major social
problem under study, a group whom remedial classes and camps, and
special resocializing schools can barely hope to help utilize their potential.
Perhaps the major burden of their delinquent pattern is borne by the
generation beyond their own, for the social cost of their many illegitimate
children outweighs the robberies and assaults they conduct today. An

396

understanding of female gangs may indeed make it possible to counteract
this trend.

Research Grant: MH 7993

Date of Interview February 25, 1965

397

Investigator'

Barbara Fish, M.D.

Mew York University Medical Center

New York, N.Y.

Prepared by:
Herbert Yahraes

An improved method of evaluating the effect of psychiatiic drugs on
emotionally disturbed children has been developed by the Childien's
Psychopharmacology Unit of the New York University School of
Medicine, Department of Psychiatry, and is being applied to the study of a
number of compounds.

The Unit, established in 1961 with the aid of a special grant from the
National Institute of Mental Health, is directed by Dr. Barbara Fish,
psychiatrist-in-charge of the Childien's Service of Bellevuc Hospital's
psychiatric division,

Dr. Fish and her fellow workers have confirmed that a child's response
to a psychiatric drug may be quite different from an adult's. For example,
triflu opera zine (Stelazine), a tranquilizer having less sedative action than
chloipromazine, was found to be a stimulating agent when given to
retarded schizophrenic children and to be capable, on the basis of
preliminary findings, of bringing moderate improvement. The hope is that
even more potent agents will be discovered.

To get a trustworthy measure of a drug's effect, the Unit has found it
necessary to divide emotionally disturbed children into four general
groups and to observe the results of medication in each group. These
groups or types, constituting a clinical typology, are as follows:

I. Autistic-dysjunctive: severely impahed schizophrenic children.

II. Immature-labile: a borderline group, less severely impaired til
present than the children of group I but more severely impaired at present
than the children in groups III and IV.

III. Anxious-neurotic: children with picdominantly neurotic manifesta-
tions. These children show anxiety and feelings of helpless dependence
and inadequacy.

IV. Sociopathic-paranoid: children with predominantly sociopathic or
paranoid features. These children tend to deny personal responsibility
for their feelings and acts; their behavior may be antisocial; they may be
overly suspicious.

The children in the first two groups are the most severely disturbed and
those in the last two, the least disturbed. The classification is done on the
basis of psychiatric interviews. Specially developed rating scales are used

398

to determine the severity of each child's condition before and after treat-
ment.

In one pilot study of childien on the psychiatric ward of Bellevue Hos-
pital, the Unit has investigated the effect of chlorpromazinc and
diphenhydramme on each of the four groups. Chlorpiomazine (thorazine)
is a widely used tranquilizer; diphenhydramine (benadryl) is an
antihistamine that makes some adults sleepy but acts in young children as
a mild tranquilizer.

Of the more severly disturbed children—those in groups I and II— 80
percent improved on chlorpromazine and 50 percent on diphen-
hydramine. None improved when given only a placebo.

Howevei, among the less seveiely impaired children— those in groups III
and IV— the percentage of those who impioved on chlorpiomazine did not
differ significantly from the percentage of those who improved though
they received nothing but a placebo. The improvement in several of the
chlorpromazine-treated children, however, was more marked. This
indicates that a large proportion of the children in groups III and IV get
better or worse for reasons having nothing to do with drug action.

To sum up, improvement in the fiist two groups of children seems to
depend primarily upon the effectiveness of the drug being studied and in
the second two groups upon such factors as hospitalization, psycho-
therapy, and special education. If the children in this study had not been
classified, chlorpromazine would have appeared effective—that is, better
than placebo treatment— for the entire group. In reality it was significantly
better only among the severely impaired children.

•|! * $ H i!

When drugs are evaluated, the investigators report, children must be
matched by age as well as by severity of illness, because some types of
treatment are more effective with certain age groups than with others. For
example, all the children who improved with diphenhydramine were
under 81/2 years of age. Older children were not helped by this drug. Again,
all those children who improved when receiving only a placebo were more
than ten yeais old. On the othei hand, there were children in all age
groups who improved with chloipromazine.

Dr. Fish points that whether or not a particular kind of behavior is
abnormal often depends on the age of the child, since behavior ap-
propriate to one age may be immature or markedly deviate if it occurs in
an older child. "The simple fact that the child is a growing organism," she
observes, "complicates every aspect of the evaluation of psychiatric treat-
ment. . . . One must detect changes produced by therapy in an organism
that is already in the process of change."

In a second pilot study, the Children's Psychopharmacology Unit is
investigating the effects of trifluoperazine on a number of group I children
raning in age from 2 to 6. Trifluoperazine is a tranquilizer used for chronic
adult patients. Dr. Fish found several years ago that it increased the alert-
ness and motor drive— and sometimes even the responsiveness and
language ability-of severely apathetic, withdrawn, schizophrenic children.

399

The children selected for the current study showed gross withdrawal
and greatly impaired speech. With each child the dose of trifluoperazine
was gradually built up to the level at which the drug was doing the most
good without causing side effects, With some children the best dose was
six times as large as with others. Dr. Fish's team also noticed considerable
differences among individuals in the doses of chlorpromazine and diphcn-
hydramine required for best results.

So far only 12 children have been followed in the trifluoperazine study,
Four of these have improved on the drug as compared with one in the
control group. They show increased alertness and improved language and
social ratings. These results, however, are not yet statistically significanl.

The investigators are now-

-drawing more children into the study, to make a total of 24;

-trying other potentially stimulating drugs on the childien who have

not improved on trifluoperazine;

—developing ways to measure changes in alertness, in order to facilitate

the rapid screening of compounds.

Hi * t * $

Dr. Fish estimates that at least 60 peicent of severely disturbed children
can be moderately improved by presently available drugs. Among children
in an outpatient population who were too distuibed to benefit from
psychotherapy, drugs enabled one-fourth to go to regular schools and
another one-half to participate in group activities and special classes.

Special screening of drugs potentially valuable for disturbed children is
essential, the New York University Unit reports, because a drug's action
may depend upon the stage of the child's development. This is true in the
case of diphenhydiamine, but a better known example is phenobarbilal,
which calms adults but may excite young children. Quite possibly, drugs
that are ineffective in adult animals or human beings may prove valuable
for children.

The investigators point out that dosages of psychopharmacologic drugs
must be determined separately for children and not merely adjusted from
adult dosages according to body weight. A dosage that has been scaled
down in this manner may be so high for a child as to cause toxic symp-
toms, or ft may be so low as to be ineffective.

Research Grant MH 4665
Published References

Fish, Barbara. Drug Therapy in Child Therapy, Psychological Aspects. Comprehensive

Psychiatry, 1:1, February 1960.
Fish, Barbara, The Influence of Maturation and Abnormal Development on the

Responses of Disturbed Childien to Dntgs, Reprinted from proceedings of Thntl

World Congress of Psychiatry, June 1961.
Fish, Barbara. Evaluation of Psychiatric Therapies in Children. Delivered at the

meeting of the American Psychopathologicai Association, February 24, 1 962.
Fish, Barbara Progress Report to NIMH, February 1963.
Fish, Barbara and Shapiro, Theodore, .4 Typology of Children's Psychiatric Disorders:

Its Application to a Controlled Evaluation of Treatment. Presented at the annual

scientific meeting of the Academy of Child Psychiatry, September 1963.

400

Investigator.

Leon Eisenberg, M.D.*
Johns Hopkins University School of Medicine

Baltimore, Md.

Prepared by.

Herbert Yahraes

Significant findings in the ticatment of emotionally disturbed children
are reported by a clinical research group at the Johns Hopkins University
School of Medicine. The most important result of the work to date, in the
view of the group's leader, Dr. Leon Eisenberg, professor of child
psychiatry, is its demonstration that effective treatment depends upon
accurate diagnosis. The group reports that:

1. Brief psychotherapy-half a dozen sessions-worked a marked
improvement in the behavior of disturbed children diagnosed as neurotic.

2. Disturbed children diagnosed as hyperkinetic, or overactive, showed
little response to psychotherapy but improved to a significant extent
when treated with a stimulating agent. (Such agents have been used for
some years with hyperkinetic children; the Johns Hopkins work verifies
their effectiveness in a carefully controlled study.)

3. The same stimulating agent markedly improved the behavior of
delinquent boys in a training school.

The work suggests that the country's lesources for helping disturbed
children before they become disturbed adults can be stretched through
the wider use in appropriate cases of a short-term course of treatment
instead of long-term, intensive psychotherapy.

The research also suggests that certain psychoactive drugs may be useful
in the management of institutionalized delinquents. There is at least the
possibility that they could make institutionalized youngsters more
amenable to training and education and, therefore, less likely to return to
antisocial activities.

In current work Dr. Eisenberg and his associates are trying to: (a) Find
out how psychoactive drugs work-that is, what functions they affect; (b)
find more objective ways of diagnosing disturbed children and measuring
the effects of therapy; and (c) provide more information on some of the
defects in mental functioning that result in low IQ scores.

*Now at the Department of Psychiatry, Massachusetts General Hospital, Boston,
Massachusetts

401

All these studies should lead to improvements in therapy for disturbed
children. The investigators hope that the third study will also lead to
stionger efforts to prevent reasoning disabilities in children from deprived
homes. The iccently completed and continuing research of the Hopkins
group is presented in more detail in the following sections.

Neurotics and Hyperkmetics

In a pioject now nearing completion, Dr. Eisenberg undertook to
expand the results of earlier work by his group in the Children's
Psychiatric Clinic of the Johns Hopkins Hospital. This had shown that
disturbed children given a placebo and psychotherapy for 7 weeks were at
least as likely to register maiked improvement as disturbed children given
a tranquilizer and psychotherapy. But the improvement was closely
related to the type of disoider. Two-thirds of the neurotic child: en, bill
only one-third of the hyperkinetic children, showed substantial gains.
Later studies indicated that most hyperkinetics did respond to either of
two stimulating agents, dextroamphetamine and methylphenidate.

The investigator describes the neurotic child as shy, introverted, afraid
of new situations, inhibited, anxious. Typically the parents complain that
he is feaiful, won't go out and make friends, can't even be dragged to
parties. The hyperkinetic child, in contrast, is distractible and forever on
the go. In school he pays attention to every disturbance rather than to the
main activities of the class. He does not follow directions. He is often
accused of being aggressive because he doesn't keep his hands to himself.
His parents complain that they cannot manage him and are afraid ho is
going to get into serious trouble.

The new project called foi both a psychotherapy study with neurotics
and a psychopharmacology study with hyperkinetics. Eighty children
were selected for each. Ruled out were children whose IQ was less than
80, those who had severe neurological defects, those who were in trouble
with the law, or those who were so sick they required inslitutionalization.
"I don't mean you can't treat such patients," Dr. Eisenberg explains. "I
mean they were different from the ones in whom we were interested for
this particular project. We wanted to define our population carefully/1

The children ranged in age from 5 to 13 years. On the basis of parent!) 1
education, they came from three social classes-middle, lower middle, and
lower-but mainly fiom the first and second of these. The hyperkinetics
tended to come from lower social levels than the neurotics and to be
younger. Most of the children had been referred to the clinic by the
schools and other community agencies; some, by private physicians. In a
few cases mothers had called the clinic themselves.

The Value of Brief Psychotherapy

The problem of setting up a control group-untreated persons against
whom to compare the treated ones-for the psychotherapy study was

particularly difficult. That's because the process of including a person in a
study of psychotherapy may in itself be therapeutic. "The moment you
talk to a disturbed child and his parents," Dr. Eisenbcrg points out,
"potentially you change them— and if you don't talk to them, you don't
know what the child was like at the beginning of the study. Mere
questioning may amount to interpretation even when the doctor has not
intended to give advice. For example, if the mother says that the child is
disobedient and you then ask how she disciplines him, you aie implying a
connection between his misbehavior and her methods of discipline. You're
already giving treatment, so to speak."

The control gioup in this case was a consultation-only group. After the
intake process— three sessions of history taking and psychological
testing— the parents were told that the child should do well without treat-
ment if certain recommendations were followed. The recommendations
were tailored to the case. A mother who was exceptionally harsh or
punitive might be advised to let up on the discipline; a mother who was
too lenient might be advised to show more firmness. Parents might be
given suggestions for improving relations with each other.

The point is that the consultation was limited to one 30-minute period,
at the end of which the parents were assured that the child's condition
would be checked again after 2 months. (The families in the consultation
group were offered treatment after the 8-week period of the study if they
wanted it and the doctors thought it desirable.)

In contrast, the experimental group was given brief psychotherapy,
defined as five additional interviews lasting from 45 minutes to an hour.
During these periods the child was seen by a psychiatrist, and one or both
parents by a social worker.

After 8 weeks, most of the children had improved, but those in the
psychotherapy group significantly more so than the others. This finding
was based partly on ratings by the psychiatrists who were treating the
children, and such ratings, the grantee points out, may be unintentionally
biased. However, the psychiatrists' ratings were confirmed by an in-
dependent set of ratings made by the children's teacheis, who did not
know to which group a child had been assigned. Further, on measures of
friendliness and aggresiveness, derived from the mothers* descriptions of
the children on a rating instrument known as the Clyde Mood Scale, the
childien in the psychotherapy group showed a greater change for the
better than the others.

The children in the earlier studies who showed improvement were
found a year later to be maintaining it. The children in the new study will
be followed up, too.

The results of this part of the investigation strengthen Dr. Eisenberg's
conviction that psychiatric clinics and psychiatrists in private practice
should place considerably more emphasis, in treating disturbed children of
the neurotic type, on brief psychotherapy. For one thing, it works. For
another, a given clinic or individual psychiatrist can reach more people
with it. Further, says the investigator, "Brief psychotherapy makes much
more sense to the parents, who in general are grateful for the statement

403

that you will see their child so and so many times instead of the vague,
'Well, it may take a long time.' With brief psychotherapy, there are fewer
dropouts."

The results of this form of treatment have not been compared with
those of long-term, intensive psychoteraphy, the investigator observes.
"But if I were the director of a community agency, with the task of
setting up a program for disturbed children that would bring the greatest
benefit to the community on a service level," he says, "I would certainly
put much moie emphasis on brief psychotherapy than has been tradition-
al."

Medicine for Hyperkinetics

In the drug study, 40 of the hyperkinetic children were given cither
dextroamphetamine or methylphenidate, commonly used with adult
patients as stimulants, and the rest were given a placebo. At the end of the
8-week study period, those who had been receiving a drug were rated both
by clinic personnel and by teachers as significantly more improved than
the others. As viewed by the mothers, they still scored liigh in aggressive-
ness but less than before.

On several objective tests, too, the drug-treated group surpassed the
other. One was a discrimination task, in which the children were shown
pairs of pictures flashed on a scieen in rapid succession and asked to tell
which picture in each set was the larger. When the time between the
pictures in a set was more than 2 seconds, no change occurred in either
group during the course of the study. But at a spacing of 2 seconds the
drug-treated group made appreciably fewer errois at the end of the study
while the other group showed no change.

The investigatois consider a 2-second spacing in this test rather stressful
for a child. So the drug takes hold, they think, at the point where per-
formance begins to break down under pressure.

Dr. Eisenberg was especially impressed by the results of the For tens
Maze Test, which calls for the subject to trace his way, with a pencil,
through a series of mazes of increasing difficulty. When a child enters a
blind alley, he is scored as having failed on that particular maze even

^i^i!16 baCkS °Ut and continucs in the right direction. An impulsive
child, then, is likely to end up with a low score. Both groups of hy-
perkinetic children made poor scores at the beginning of the study. Eight
«'^ks later the placebo group showed no improvement, but the scores of
Irug-treated group shot up about 15 points.

t;"gly, the children with the lowest IQ's-all of which were
"-ma! range-showed the greatest improvement. "Apparently
is functioning well/' Dr. Eisenberg explains, "isn't going to
the drug to levels of superfunction. But children with
Q's may have a better potential than the IQ scores indicate.
>e using it because of their inability to pay attention and to

control their motor activity. These are the childicn who may be helped
most, at least in the skills demanded by the maze test."

This reasoning may also explain some of the results of earlier work with
emotionally disturbed children in two institutions. In each case, the in-
vestigator found a strong trend for the childien who had been given a
stimulating drug to make fewer mistakes than the others in learning a
standard laboratory test. (The children in this preliminary woik were not
differentiated either by the type of disorder or by IQ scores.)

To test the idea that a stimulating agent improves the learning ability of
certain disturbed children by improving their ability to pay attention, Dr.
Eisenberg's associate. Dr. C. Keith Conners, has now worked out a way of
measuring the latter. The child being tested watches a pattern of lights on
a panel and presses a particular light whenever a new pattern appeals. He
has to keep watching the panel in order to press the right light at the right
time. The idea is to learn how many mistakes he will make over a
10-mimite period. Once the researchers have baseline rates for disturbed
children, they propose to test a variety of psychiatric drugs to learn if the
attention span can be favorably altered. If the plan works out, the group
will have a way of telling beforehand which kind of drug will affect a
given type of child most favorably.

Why should a stimulating agent work with hyperkinetics— children who
appear to be already overstimulated? Before answering, Dr. Eisenberg
poses a companion question: Why should phenobarbital, a sedative, tend
to overexcite children and, often, the elderly? The usual explanation, he
goes on, is that interfering with the functions of certain structures of the
brain may produce different effects at different stages of development.
That is, a given drug has a certain, unchanging pharmacologic effect, but
the physiologic expression of this effect depends on the balance among
the brain centcis, and this balance changes as the brain develops. In the
case of stimulating agents, one has to suppose that the drug in some way is
improving the inhibitory or control centers of the brains of hypeikinetic
children. If scientists can learn how this is done, the search for more
effective drugs will be greatly advanced.

Treating Delinquents

Twice within recent years the Hopkins team has gone into a training
school and tested the effect of psychoactive drugs on delinquent boys
ranging in age from 1 1 to 17. The first time the team used perphenazine, a
tranquilizer. Some of the boys in the study were given the tranquilizer,
others a placebo, and the rest nothing. Those who got medicine, whether
it was the active drug or the placebo, responded with a substantial
improvement in behavior.

"We had altered the social environment," Dr. Eisenberg explains. "All
the youngsteis wanted to get out, and the medication was something that
was supposed to make them better and help them get out sooner. To be in

405

the tieatment cottage, where some of the boys were on perphenazine and
some on placebo, became a matter of piestige."

The tranquilizer, in other words, had a placebo effect: the boys
expected the medicine to make them better and they did get better. In
this respect they were just like the boys taking the placebo itself. Failure
to allow for such an effect has clouded the results of many a test of new
medicines.

When the team returned to the institution, they tested dextro-
amphetamine, one of the stimulating agents used successfully with hy-
perkinetics. The ding was given to some of the boys in each of two
cottages and this time the results were quite diffeient. Only the boys
receiving the medicine showed a decided improvement in behavior, as
rated by houseparents, teachers, and cottage mates. However, there was
some improvement among their associates, both those who were untreated
and those who received a placebo, and this change may have resulted, the
investigators suggest, from a more harmonious atmosphere brought about
by the change in the drug-treated boys.

Though the improvement lasted no longer than the treatment, the
Johns Hopkins group holds that further research with delinquents along
the line it has pioneered is "a compelling social necessity."

"I don't think you can change the symptoms of delinquency merely by
medication," Dr. Eisenberg declares. "But if you can diminish a
youngster's anger and hostility and aggressiveness so that instead of fight-
ing everything he may be willing to listen to what's said to him, then the
ordinary treatment procedures might be more effective."

Putting it another way, medication may lead to improved behavior, and
improved behavior may lead to bettei relationships with houseparenls,
teachers, and other personnel in a training school, and these better
relationships should make possible a more constructive outcome of a
boy's training school experience.

So far as Dr. Eisenberg knows, this lead has not been followed up. He
points out that most of the facilities for delinquents are not in medical
hands and do not have much medical support. Beyond this, when staffs
remain underpaid and undertrained, and are hard put, consequently, to
work toward the professed goal of reeducation, medicine that might be
effective in other circumstances is not likely to do much good.

The investigator doubts that drugs can help the delinquent who remains
in his decaying neighborhood and troubled home. Without change in the
social circumstances that breed delinquency, he says, the behavior will
almost certainly continue. Social rehabilitation is the key to delinquency
control; drugs, counseling, and other measures are adjuncts or aids that
may be useful and essential-because patterns of antisocial behavior tend
to persist— once the environment has been altered.

Tranquilizers

The use of stimulating drugs for hyperkinetics is bound to increase as
the lesult of the recent findings, Dr. Eisenberg believes, while the use of

406

tranquilizing drugs for disturbed children in geneial should be consider-
ably reduced. He judges that the tranquilizers have been prescribed far too
freely in the treatment of children.

No drug is free of hazards, he holds, and no child should be placed on
medication unless there is a clear need for it and clear evidence that the
chosen medicine is likely to be beneficial. None of his group's three out-
patient studies with disturbed childten produced evidence of beneficial
effects that could be ascribed to the tranquilizers, and he finds little or no
evidence in the literature that tranquilizers are of any benefit for the usual
child who conies into an outpatient clinic. With the schizophrenic child
the story is different.

Diagnosing the Disturbed Child

Because of considerable overlap in types of psychiatrically disturbed
children, the John Hopkins group is looking for more objective diagnostic
means. Among the numerous measures being studied arc the answers to a
symptom questionnaire filled out by the parents of the 400 disturbed
children treated by the clinic since 1959, when the NIMH began support-
ing its research. The questionnaire lists 70 symptoms, or types of
behavior, and the parent indicates to what extent each of these applies to
his child.

Preliminary analysis shows that certain clusters of these symptoms arc
more characteristic of one type of disturbed child than of the other.
Neurotic children register higher on the clusters, or factors, tentatively
labeled inhibited, shy, and psychosomatic. Hyperkinctics register higher
on the factors labeled hyperactive, tantrum behavior, aggressive acting
out, and sibling rivalry. The symptoms grouped under the labels anxious,
stubborn, and school problem seem to apply as much to one type of child
as to the other.

In work now going on, the investigators hope to leain if factor scores
provide a good diagnostic and prognostic tool and if improvement is
related to changes in the scores on particular factors.

Several hundred parents of normal children, approached through PTA's,
have filled out the same questionnaire. The researchers wanted to learn if
such parents see as many things wrong with their children as the parents
who have sought help from the clinic. The answer is that they don't:
about 40 of the 70 symptoms were found to have been checked much
more frequently for the psychiatric population. Through factor analysis
the reseachers expect to come up with a detailed description of how a
disturbed child differs from a normal child in terms of behavior as viewed
by the parents.

The Thinking Process in Children

The Johns Hopkins group has also undertaken basic research to learn
more about the reasoning processes of the developing child and thus to aid

407

in understanding and treating defects in these processes. As Dr. Eisenberg
explains it, such work should make it possible in many cases to determine
the factors responsible for a low IQ scoie and do something about them.
This phase of the program is the particular interest of Dr. Sonia F. Osier, a
psychologist.

Dr. Osier began by testing the ability of normal children to form an
idea 01 concept and by studying the process involved. In the simplest test,
the child was shown two pictures at a time, one of a bird and one of
something else. The fiist pair, for example, might show a bluebird and an
automobile; the second, a robin and a dog; the third, an airplane and a
canary. There was a lever associated with each picture, and the child had
to press one of the levers each time. If he pressed the one under the bird
picture, he was rewarded with a marble. Soonei or later, most childien got
the idea that the common denominator or unifying concept was birdam\
pressed the correct levci each time.

In her first study, Dr. Osier presented three concepts-bird, animal, and
living thing-to groups of public school children aged 6, 10, and 14 years.
Half of the children in each group were of average IQ and half were above
average.

Most of the results were predictable. The bird concept proved the
easiest to grasp; the living thing concept, the hardest. Older children did
better than younger children, and brighter children did better than aveiage
children. Interestingly, however, the average and the bright children used
different approaches. The average child tended to follow a hit-and-miss
process. He would make a random number of mistakes, then begin getting
the right answer more frequently, and finally grasp the idea. The brighter
children, on the other hand, tended to go along making mistakes and then
jump all at once from the random to the perfect level of performance.

The results suggested that while the average child was proceeding
blindly on a trial-and-error basis, the brighter one was making guesses or
hypotheses and at some point saying, "Ha! Now I see-that's what the
answer is!"

To check on this, Dr. Osier devised a test calculated to confuse the
hypothesis makers by providing a good deal of irrelevant information. The
problem was to get the concept of two. One card of a pair might show a
dog and a ball; the other, two birds and an automobile. In another pair,
one card might show a shoe and a baseball bat; the other, three fire trucks
and a cat. In this situation, brightness proved no asset. The higher IQ
children, presumably because they had seen and been obliged to discard so
many hypotheses, ended up scoring no better than the others.

Next the investigator changed the rules of the game so that right
answers were rewarded only one time out of two. This situation bothered
both the older children and the brighter children more than the younger
children and those of average intelligence-a finding reached by comparing
the scores of each group with the scores made when the right answer was
rewarded every time. In fact, in some cases age and intelligence proved to
be absolute disadvantages; college professors, for example, were stumped
by some of the problems handled by the average 6-year-olds. In a test like

408

this, Dr. Eisenberg observes, a person of intelligence and sophistication
keeps making hypotheses and discarding them, since none of them
rewards him consistently; a 6-year-old thinks less and performs better. The
investigator points out that even animals can learn certain types of
responses without being rewarded every time.

These results again are taken to indicate that normal children use two
approaches in solving problems of the kind presented in these tests. Young
children make the responses that a trial-and-error approach leads them to
think will pay off at least some of the time. As children grow older, they
tend to look for some hypothesis— some rule— that will give them the
answer every time. Bright children show the same tendency.

Dr. Osier has these projects under way:

1. An analysis of children's information-processing ability as shown by
their responses to a new series of tests in which the amount of irrelevant
information is systematically varied. The investigator is learning how
much information, of the kinds presented in this research, can be handled
by normal children of a given age. She is also learning just where the
concept-forming process breaks down when a failure occurs.

2. Comparisons between normal and retarded children and between
normal and disturbed children.

3. Comparisons between the normal children studied so far, all of them
from good neighborhoods, and children from slum areas. Since other
investigators have found that deprived children lag in intellectual develop-
ment, the Hopkins group expects that the slum-area children will have
considerably more difficulty with the tests. This would be an indication
either that they had less native ability to form concepts or draw logical
inferences, or that for some reason they were hampered in using that
ability. The research group believes the second explanation to be the true
one. Specifically, they believe that difficulty with the tests could be
traced to backgrounds so impoverished that the children had had little
experience with materials like those used in the tcsts-not only in the
variety of pictures but also in objects of different sizes, shapes, and colors.
If the comparison gives the expected results, the investigators will test
their line of reasoning by training some of the slum-area children before-
hand to recognize these materials.

Through this phase of the work, in sum, the investigators are looking
for evidence that a lag in a child's reasoning ability may be caused by a
deficiency in experience instead of in intellect, and can therefore be
prevented or corrected.

Service and Research

In discussing the implications of recent findings by his group and
others, Dr. Eisenberg stresses two compelling mental health needs; to act
more effectively on the basis of what we already know and to make
greater efforts to get the answers to what we still do not know.

409

One thing we already know is that distuibed children are more likely
than other children to become disturbed adults; hence it is good
preventive psychiatiy to reach these childien. And child psychiatrists
could reach many more of them, the investigator observes, by following
the implications of two of the findings reported earlier, as to the value of
brief psychotherapy for neurotic disorders in childhood and of psy-
chiatrically supervised treatment for delinquents in an institution.

A clinical study by Dr. Eisenberg a few years ago when he was a
consultant to Baltimore's welfare department suggests a third way of
making our psychiatiic resources go father. At that time he talked with a
number of disturbed children from foster homes, discussed each child
with the social worker in charge of the case, and made recommendations.
These had to do with such matters as advising the foster parents to modify
their treatment of the child in some respect, finding a different school, or
getting the child interested in such an organization as the Boy Scouts.
They did not include psychotherapy, because there was none to be had, or
medication.

When the records of the welfare department were checked a year later,
those children for whom the recommendations had been cairied out were
found to have made a much better adjustment than the others. (In the
other cases the recommendations had not been carried out for such
leasons, generally, as that the caseworkei had become overloaded or had
been transferred, or that a certain facility was not available in the child's
neighborhood.) This study was not so rigorous as one might like, Dr.
Eisenberg observes, but it does offer evidence that if psychiatrists are used
to enhance the effectiveness of other mental health workers, a grcatci
number of disturbed children can be helped.

Dr. Eisenberg points out that while measures to reach disturbed
children are essential, preventive psychiatiy will be most effective if it
helps correct the conditions that lead to disturbances in children. To this
end he holds that psychiatrists must go beyond the clinic and the consult-
ing room and work with other professional people to broaden the
availability, or improve the effectiveness, or both, of family planning pro-
grams, good health care, decent housing, training for displaced woikcrs,
casework services to minimize family breakdown, substitute-care for
homeless children, enriched school programs, and the like. Some deprived
children manage to grow into functioning adults, he observes, but far too
many contribute to statistics on delinquency and disease; "They become
premature and inadequate parents themselves, fated to repeat for a
succeeding generation the cycle of deprivation."

In spite of research achievements in the last decade or so, Dr. Eisenberg
emphasizes, our knowledge about pieventing and treating psychiatric
illness has serious limitations. For example, we don't know how
'"•IPS work, so we cannot see clearly how to get better ones,
"'-Jtely acceptable system for classifying children's
-an impediment both in treatment and in research,
sons-including the placebo effect, the lack of fully
ir measuring changes in attitudes and personalities,

and the possibility that even m double-blind studies the investigator will
detect the children who have been receiving the medication-it is
extremely difficult to gauge scientifically the worth of a given form of
treatment.

Psychiatrists are charged with relieving suffering, Dr. Eisenberg notes,
but if they aie to do this most effectively psychiatry must acquire ad-
ditional knowledge and new skills. Hence, though there is an urgent need
for more psychiatric service, theie is also an urgent need for a heavier
investment of psychiatric peisonncl in research.

Reseat ch Giant: MH 2583

Date of Interview Apr. 14-15, 1965

References:

Conners, C. K., Eisenberg, L., & Sharpc, L. Effects of rnethylplicnidate (Ritalin) on
paired-associate learning and Porteus Maze performance in emotionally disturbed
children./. Consult. Psychol, 1964, 28, 1

Cytryn, L > Gilbert, Anita, & Eisenberg, L. The effectiveness of tranquilizmg drugs
phis supportive psychotherapy in treating behavior disorders of children: a double-
blind study of eighty outpatients. A met: L Orthopsychiat., 1960, 30, 1.

Eisenberg, L. The Strategic Deployment of the Child Psychiatrist in Preventive Psy-
chiatry Presented at the World Congress of Psychiatry, Montreal, 1961.

Eisenberg, L. The sins of the fathers: urban decay and social pathology. Amer. J.
Orthopsychiat., 1962, 32, 1

Eisenberg, L, Role of drugs in treating disturbed children. Children, 1964, 11,5.

Eisenberg, L., Conners, C. K., & Sharpe, L. A Controlled Study of the Differential
Application of Outpatient Psychiatric Treatment for Children. Presented at the
Sixth International Congress of Pscy ho therapy, London, 1964.

Eisenberg, L , Gilbert, Anita, Cytryn, L., & Moiling, P. A. The effectiveness of psy-
chotherapy alone and in conjunction with perphenazine or placebo in the treatment
of neurotic and hyperkmetic children. Amer. J. Psychiat., 1961, 1 17, 12.

Eisenbeig, L.t Lachman, R., Moiling, P. A,, Lockner, A., Mizelle, J. D,, & Conners, C.
K. A psychopharmacologic experiment in a training school for delinquent boys.
Amei J.Oithopsychfat,, 1963,33,3.

Osier, Sonia F., & Shapiro, Sandra L. Studies in concept attainment: IV. The role of
paitial reinforcement as a function of age and intelligence. Child Develptn., 1964,
35.

411

Investigator1

Marian K. DeMyer, M.D.

Indiana University School of Medicine

Indianapolis, Ind.

Prepared by.

Herbert Yahraes

In a bioad-front research attack on early childhood schizophrenia, the
Indiana University School of Medicine is searching for causes, collecting
information to facilitate diagnosis, and experimenting with promising new
ways of treatment.

The piogram is headed by Dr. Marian K. DeMyer, a children's psy-
chiatrist and director of the medical school's Clinical Research Center for
Early Childhood Schizophrenia, housed in the La Rue D. Carter Memorial
Hospital, Indianapolis. This hospital contains the State's residential treat-
ment institution foi disturbed children.

The children most intensively studied are inpatients at the Center.
When admitted they are between 2 and 5 yars old. In general, they cannot
get along with adults or other children, fly easily into rage, either do not
speak at all or else use language in odd and incomprehensible ways, do not
listen to directions, find little pleasure in play, and show a remarkably
narrow range of inteiests. Some of them twill or rock themselves for long
periods; others bite or hit themselves. They often seem unsure of who or
what they are.

Some of these children have been diagnosed as autistic, a type of child-
hood schizophrenia maiked by a withdrawal from other people; others us
symbiotic, a type in which the child physically clings to his mother.
Because some of the children display, at different times, each of these
characteristics, Dr. DeMyei classes them as symbiotic-autistic. Still other
patients suffer from what Dr. DeMyer labels chronic undifferentiated
schizophrenia. Their behavior is less abysmally abnormal than that of the
autistic and symbiotic children; they have some conversational ability and
usually show some conformance to social amenities.

Investigations under way are directed toward:

• presenting detailed descriptions of the behavior of various types of
schizophrenic children and of other children with psychiatric clis-
ordeis;

• learning how, if at all, the preschool schizophrenic child differs
biologically from normal children and whether or not the differences
are responsible for his illness;

412

« learning whether or not certain influences in the home, notably the
personalities of the parents and the way the parents treat the child,
contribute to the onset of childhood schizophrenia,
• determining the value of several new approaches to treatment.
So far the Center's research piogram has led to two results of major
importance.

First, out of a group of 149 young psychiatric patients, 51 percent
were found to have clearly abnormal electroencephalograms and 15
percent had experienced at least one epileptic-like seizure.
In each instance the proportion would have been highei had borderline
cases been counted. The abnormalities would be found in less than 1
percent of any randomly selected noimal population.

The patients included neurotics, childhood schizophienics of all types,
and children with a severe but nonpsychotic behavior clisordei, marked by
such activities as stealing, setting fires, attacking other children and adults,
and refusing to get along in school. A high propoition of abnormal EEC's
was found among all the patients except the neurotics.

The results are important because they indicate that many children who
are mentally ill have a physical impaiiment of the brain. The possibility
that the impairment signals an effect of the illness rather than a cause, Dr.
DeMyer observes, cannot yet be ruled out. Current studies of schizo-
phrenic and nonschizophrenic children may turn up clues to the origin of
the impairment.

Befoie the EEC study, Dr. DeMyer had leaned to the psychogenic
explanation of most psychiatric illness and had given only superficial
consideration to organic factors. "As the result of this study, done in a
careful way and using the severest ciiteria for EEC dysrhythmias," she
notes, "I knew that our research program would have to include a careful
study of the central nervous system and other biological matters."

Second, conditioning principles similar to those used in training
animals-and to those often used, unknowingly, by parents in train-
ing their children -have been successfully applied to improving the
behavior of severely schizophrenic children and broadening the
activities of which they are capable.

Under the Center's new treatment program, called the semester system,
the child goes through cycles of 5 months at the center and 7 months at
home, and the staff works first of all to make him easier to live with. It
tries to correct those behaviors the parents have found most disruptive of
family life and it instructs1 the parents how to handle him. In accordance
with conditioning theory, desired behavior is rewarded, both with food
and with a "social reinforcer" such as praise, and undesiied behavior is
ignored or is associated with an "aversion stimulus" such as physical
restraint.

Proposed by Dr. Don Churchill, children's psychiatrist and the Center's
assistant director, the semester system was introduced in the fall of 1965
with the admission of four new patients. In all these cases, after the
children had gone home for a 7-month stay, the parents reported they
were having much less, if any, difficxilty with the particular kinds of

413

behavior they had identified as the most troublesome. The system is now
being used with all the Center's inpatients-1 1 at present.

How many times a child will return for the 5-month hospital cycle
lemains uncertain. But Dr. Churchill observes that he is not inteiestcd
merely in impiovmg behavior and has not thrown overboard the psycho-
analytic concepts in which he was trained; he just doesn't find these
concepts very useful in working with extiemely sick children. According
to psychoanalytic theory, the therapist establishes a relationship with a
mentally ill person and uses this relationship to induce normal behavior,
But with most children like those at the Center, Dr. Churchill notes, it has
been extremely difficult both to establish a relationship and to effect
much improvement. If a certain amount of normal behavior is first trained
into such severely ill children, the investigator suggests, they may show
more interest in the adults about them, and the therapeutic relationship
will follow and be fruitful. In any event, the trained-in normal activities
may crowd out at least some of the customary abnormal ones.

As one promising way of training in normal behavior, the Center has
been experimenting with a technique for inducing a schizophrenic
youngster to imitate specific actions of the therapist. The fiist results arc
exciting: two of the most severely sick children have been trained to
imitate several hundred simple activities and to say a few dozen words. To
Dr. Joseph N. Hingtgen, the psychologist in charge of this work, the
preliminary outcome suggests that the schizophrenic process is not
completely irreversible-that perhaps even terribly withdrawn children can
be brought up to something approaching the normal level.

Dr. Hingtgen emphasizes that he is speaking on the basis of only a few
cases-two of his own, and several reported by another psychologist, Dr.
0. Ivar Lovaas, who works at the University of California, Los Angeles,
and whose ideas have influenced the Indianapolis group. Further, iho
technique requires such an immense amount of work that only a lelatively
few children could be helped if it had to be administered professionally,
But Dr. Hingtgen finds that it can easily be taught to parents. The basic
idea is that normal children, to a great extent, learn by imitating adults.
Gravely psychotic children, however, pay no attention to adulls and,
therefore, fail to imitate. They have to be trained to imitate so that they
can learn to behave normally. The training can be accomplished through
many long and laborious sessions in which the child is rewarded for doing
what the therapist-or the parent-does, beginning with such simple
activities as holding up a finger. Eventually the child begins to imitate the
adult spontaneously, and even to do what the adult requests.

five learning technique fits in nicely with the ideas and

he new semester system and, depending on the results of

nay eventually be tried with all the children at the Center.

ition about the findings noted, together with a brief report
»ther research of the Indianapolis group, follows.

Abnormal EEC's in Young Psychiatric Patients

The EEC study was pait of the research seeking to describe the
anatomical, biochemical, and physiological makeup of the preschool
schizophrenic child. Among the children tested were 58 diagnosed as
autistic, symbiotic, or symbiotic-autistic, 44 with chronic undifferentiated
schizophrenia, 37 with a nonpsychotic behavior disorder, 10 as neurotics,
and 13 with normal controls. They were from 4 to 10 years old. Since
most of the psychiatric patients fought against the test, the EEC's in
virtually all the cases were taken after the children had been calmed by a
tianquilizer, promazine— given in too small a dose, Dr. DeMyer reports, to
affect the pattern of electrical discharges.

As noted earlier, 51 percent of the psychiatric patients were found to
have abnormal EEC's, a proportion that would have been slightly higher
had the neurotics, only one of whom showed an abnormality, been
omitted. All of the normal children had normal EEC's. Among the
children with nonpsychotic behavior disorders of the acting-out type and
among those with the various types of schizophrenia, there were no
significant differences in the proportion of abnormalities; also, the types
of abnormalities were the same and were present to about the same
extent. Most chaiacteristic was the type of activity described as
paroxysmal spike and wave (PSW), which consists of a mixture of
synchronous spikes and slow waves, This type appeared in almost two-
thirds of the children with abnormal EEC's.

The EEC study, which was directed by two neurologists— Drs. Philip T.
White and William DeMyer— in cooperation with the investigator, opens
several important questions, including;

1. What relationship has the central nervous system disorder, demon-
strated in half of the psychiatric patients, to the abnormal behavior of
these children?

On the basis of this and previous studies, the investigators answer, it
seems established that at least the EEC abnormality labeled PSW has more
than a chance relationship to disturbed behavior. This means that in
seeking the causes of childhood schizophrenia and other psychiatric
disorders, the role of cerebral dysfunction cannot be ignored. But the
exact relationship between abnormality and behavior remains to be
discovered. One puzzling fact is that most children with epilepsy, who
have EEC abnormalities similar to those found in this study, have normal
social and intellectual skills and behave normally except during a seizure,
In the children with schizophrenia and behavior disorders the EEC ab-
normalities presumably could be the result of biochemical changes caused
by the stress associated with psychiatric illness. But they could also be one
aspect of an organic disturbance, congenital or acquired very early in life,
directly responsible for the illness.

2. Why did only about half of the psychiatric patients show abnormal
brain waves?

One possible answer, says the investigator, has to do with the electro-
encephalographic technique itself, which cannot pick up all abnormalities,

415

Another possibility has to do with the way abnormalities are defined. The
Indianapolis study counted only those variations in the EEC that weic
distinctly different from normal; now the records are being reexamined to
determine the effect of taking borderline changes into account. Pievious
EEC studies of psych uitrically disturbed children. Dr. DeMyer recalls,
reported abnormalities in as low as 2 percent of the patients, and in as
high as 90 percent. She thinks that biases both in psychiatric diagnosis ami
in EEC diagnosis explain part of this wide range.

As still anothei possibility, the sampling may have been inadequate,
meaning that many of the records which were normal for the particular
hour when they were made might have been abnormal if made at some
other time. The investigator points out that even peisons with epilepsy
often have normal records except when seizures are occurring; perhaps,
she reasons, the iccords of most psychiatric patients would show ab-
normalities if taken at times of intense anxiety or shortly afterward. A
study in which biain waves would be picked up and telemeteied to a
recording station, by an appaiatus worn by the child, throughout a day of
usual activity is being planned.

There is the possibility, too, that a numbei of the children with normal
records will show abnormal ones a little later in life.

The Center is preparing a laboratory to get further information about
the brain's electrical activity in psychiatric patients by recording cerebral-
evoked potentials, meaning the brain's electrical responses to external
stimuli-a light, a touch, a sound. The subjects will be schizophrenic
children whose EEC's are normal. Normal and retarded children will be
tested the same way.

3. Do the families of patients who show no EEC abnormalities differ
significantly from the families of those who do?

In the EEC study, no attempt was made to evaluate the families.
However, a study nearing completion, in which families with a schizo-
phrenic child are being compared with one another and with matched
families whose children are normal, should provide an answer. The subject
bears upon the genesis of childhood schizophrenia. Another investigator,
working with somewhat older children, found that children showing no
organic impairment came from more deeply disturbed families than the
others, the implication being that in some cases an organic impairment
contributes to the onset of schizophrenia, but in other cascs-whcre the
family is seriously disturbed-schizophrenia can develop whether or not
something is organically wrong.

Dr. DeMyer and her associates have also used the EEG to study the
sleep and dream patterns of seven schizophrenic children hospitalized at
the Center. Their sleep habits at home, the parents reported, had been
extremely erratic. At the Center, though, the staff considered their sleep
to be remarkably regular, and the records of brain waves and eye move-
ments during sleep showed that these children spent the same proportion
of time in the dreaming stage as normal children. This study, then, finds
no evidence to back up the suggestion that patients may hallucinate in the

416

waking state because they have been unable to hallucinate as much as
other persons during sleep.

Earlier Experiments in Conditioning

The Center's use of conditioning techniques-with piomising results-
follows a series of experiments to find ways of getting very sick
youngsters to come out of their shells and act moie noimally toward
people and things.

In one of the earlier experiments, directed by Dr. C. B. Ferster, an
autistic child would spend an hour or two, every day, in a room contain-
ing such fascinating equipment as a pinball machine, an electric train set, a
phonograph, a picture viewer, a telephone set, a trained monkey in a cage,
and vending machines. Each device could be operated by the child if he
put a coin in the proper slot (in the case of the monkey, the coin tinned
on a light in the cage and the animal went into his act); and the child
could get the coins by pressing a key that operated a coin-dispensing
machine.

Rats and cats will learn to keep pressing a bar, and pigeons to keep
pecking at a disc, if such activity occasionally brings them food. This
learning process is known as operant conditioning because it gets the
subject to work, or operate, in a desired manner. The same principle held
good with the schizophrenic children. They learned that by pressing the
key often enough, they would get a coin good for any of the devices, and
soon they spent most of their time working for coins and spending them.
The most popular devices by far were the candy vending machines.

Dr. Ferster's next question was whether or not the same technique
could be used to help the child understand more difficult situations and
engage in more complex bchavior-an important question because one of
the most marked characteristics of schizophrenic children is their
extremely narrow range of activity.

In one of the new situations, coins were delivered only when a panel on
the dispensing machine was lighted. In another, a device could be operated
only if a coin was placed in a slot when the slot was lighted. A more
difficult task required the child, if he were to get a coin, to compare two
colors or figures or pictures and touch the one that matched a third,
which he had been previously trained to touch. Dr. Ferster's question was
answered in the affirmative, even the autistic children came to understand
and master the new situations, though more slowly than normal children.

To sum up, the study found that children who rarely had taken any
account of their environment could soon begin, through the use of
conditioning techniques, not only to notice the environment but also— by
pressing keys, matching pictures, and dropping coins— to manipulate it.
True, the environment was artificial and the behavior needed to control it
a little out of the ordinary. But the results were taken to indicate, in these
schizophrenic children, "at least the existence of normal processes at a

417

very basic level," with no suggestion of an underlying deficit except in tl
rate at which the children learned.

The next question was whether 01 not a similar laboratoiy situatio
could be used to teach socially adaptive behavior~specifically,interactio
with other children. Normal youngsters in a preschool group are almo:
constantly interacting: by playing together, foi example, grabbing eac
other's toys, and fighting. But schizophrenic children hold thcmselvt
aloof.

In research directed by Dr. Hingtgen, the Center tried to devclo
cooperative behavior in some of its children. Only two machines wcr
used— the coin dispenser, set to deliver one coin for every 1 5 presses of th
key, and a vending machine offering candy, crackers, and cereal. Si
children participated. After each had been trained to opeiate th
machines, they were paired, and for the rest of the experiment bo I
members of a pair went to the laboratory room together.

For a while, each member of a pair, A and B, was free to operate th
coin lever at any time. When it was operated by A, a panel behind th
lever showed red; when it was operated by B the panel showed green.

Step 2 required a little cooperation because A could operate the leve
only when the panel showed red, and B only when it showed green. Aftc
one child received a coin, the other child's light was presented. Tin
children learned by trial and error that they had to take turns.

In step 3, the children had to use a two-key panel in another part of thi
room. One key could be illuminated by a red light; the other by a green
When the red light was on and A pressed the corresponding key, the con
lever showed red; then A could go to work to get a coin.

In the final step, when A pressed the red-light key, the coin Icvei
showed green, and B could operate the machine; when B pressed the
green-light key, the light on the coin machine went red for 4. Thus A and
B had to work for each other by providing the appropriate coin-lever light,

Starting with step 2, the children learned to work for each other in an
average of 23 sessions.

To the investigators' surprise, in view of earlier observations of schizo-
phrenic children at play, the youngsters in this particular situation
frequently made physical contact with each other. One would lead the
other to the correct lever, or one would operate the correct lever by
manipulating the other youngster's hand, or one would pull the other
away from the coin lever or from the vending machine.

In one session a 4-year-old boy was working at a very low rate. He had
to press the coin lever 15 times in order to be rewarded, but he often
stopped and began twirling. Since his partner, a 6-year-old girl, could not
go to work for her com until he received his, she began pulling him back to
the lever whenever he wandered off. Once she slapped him, and ills

XTtoT.T up', AnotheVime' a slap set hta *> «y"i "nd *

ht hnln I ™<*me "* a11 After a long period' she wen' over to
him, hugged him around the neck and led him back to the lever- his

went up< In later sessions> she slapped him a

418

All the children appeared to be communicating with their partners
through vocal and facial expressions, and two of the less seriously
disturbed used words-"That's the red light." "That's enough," "Get the
coin."

This social interaction, however, did not carry over from the ex-
perimental room to life outside. Evidently young schizophrenic children
were capable of modifying their usual behavior, but how could the
modifications be made to stick?

The next experiment tested the effect of directly rewarding the
children for making physical contact with each other. The mechanical
devices were not used: an observer simply handed the children a small
piece of candy or a bit of a cookie or cracker whenever they performed
the desired behavior. In the first stage the childien were rewarded when-
ever one touched the other. The required behavior was made gradually
more complex until, in the end, the reward was given only when the child
touched the other with both hands and said something, whether com-
prehensible or not.

This time there was a little more carryover. Parents and ward personnel
reported that the children would come up to them once in a while, touch
them, and make a sound or two. But this social behavior died out in a few
weeks, presumably because it no longer brought the kind of reward that
had been used in developing it.

The research group then tackled one phase of an old and basic problem:
why the severely schizophrenic child does not respond like a normal child
to what he hears and sees. Clinicians are satisfied that he can hear and see,
in spite of the many indications he may give to the contrary, particularly
in the matter of hearing. But sometimes, apparently, he shuts out his
perceptions, and other times, apparently, he fails to organize them— at
least he does not act upon them in normal fashion. Part of the trouble,
some investigators have thought, may be an inability to discriminate
among stimuli.

Dr. Hingtgen used an auditory discrimination test with six of the
children. First a child would learn that by pressing a lever often enough he
would get a coin, which he could use to obtain candy from a vending
machine. Then the experimenter fixed the dispenser so that it would drop
coins only when music was playing. Like rats, pigeons, and normal
children in earlier experiments, the schizophrenics soon learned to work at
the dispenser only during periods of music; in other words, they learned
to discriminate between music and silence.

The children also showed they could discriminate between speech (the
voice of someone reading a story) and silence, and between speech and
music. Then the investigators tried single words, in pairs. When a voice
said chalk, for example, the machine was operating; when the voice said
ball, it wasn't. Of the six children being tested, all except one learned to
distinguish even between wordSj like mama and apple, that are rather
similar phonetically. Again, though, the learning took longer than with
normal children.

419

Learning by Imitation

If schizophrenic children have this power of auditory discrimination,
why don't they use it all the time? Maybe, Dr. Hingtgen ventures, it's a
matter of paying attention, and perhaps they pay attention only when
they are going to get something-and when that something meets such a
basic need as for Food.

The researchers were discussing ways of testing this and other
possibilities, and of putting the findings to practical use, when Dr. Lovaas,
the University of California psychologist mentioned earlier, came to the
school of medicine and described some of his work with young schizo-
phrenics. He, too, was using reward s-and punishments, also-but in a
nonautomated situation, and with just one adult and a child working
together, intensively. He theorized that a schizophrenic child fails to learn
because he fails to imitate, and he fails to imitate because he doesn't
associate paying attention with reward. Once the child has been taught to
pay attention and to imitate, he can start learning on his own.

What was new here was not the general learning-by-imitation concept
but its systematic and intensive application to schizophrenic cliildien. One
of the aspects that impressed the Indianapolis group was the comparative
simplicity of the technique. You present the child with a model, or
example, of the type of tiling you want him to do. He imitates you. You
reward his imitation. This is easier than waiting for the child to perform,
more or less accidentally, the behavior you have in mind, then rewarding
him for it and going on to shape that behavior by rewarding him for
increasingly complex variations. Getting two children to the point where
they weie touching one another with both hands and saying something
required as many as 30 daily sessions of at least an hour each.

But the most hopeful aspect of the learning-by-imitation concept was
the possibility that eventually the schizophrenic child would conic to
imitate spontaneously. As Dr. Churchill puts it: "To teach a child
separately every little thing he has to learn in the course of his growing up
is patently impossible. But if we can teach a child to imitate others, there
is the hope that the learning process will become more automatic-will not
stop at the end of the learning session, but continue through the day."

The Indianapolis investigators decided to test the new technique-
modified, though, to omit punishment- with some of the children at the
Center.

Tommy was first. He was six-and-a-half, a mute child with a very
narrow range of behavior. He had been subjected to every available type
of therapy, but in 3 years at the Center had shown no significant improve-
ment.

He was taken off the ward, given his own room, and kept there for 20
days, Every morning about 8 o'clock Dr. Hingtgen and Mrs. Susan
Coulter, his research assistant, would come in when Tommy got up, and
every evening about 8 o'clock they would put him to bed. They were
the*e at least eight of the intervening hours and-so that he would come to

420

associate food and drink with the adults who were asking him to imitate
them— gave him all his meals.

The investigators aimed to get three types of imitation. One was imita-
tion in use of body parts-touching two fingers together, for example.
Another was imitation in use of objects-from so simple an activity as
picking up a pencil to so complicated a one as using scissors to cut out a
picture. The third, since Tommy was mute, was imitation of sounds and
speech.

Like other seriously schizophrenic children, Tommy generally avoided
looking at adults, so the first sessions were rather wearing. The procedure
went like this. One of the teachers— Dr. Hingtgen and Mrs. Coulter
alternated-would seat Tommy in a chair directly in front of him and say,
"Tommy, look here. Clap your hands." And the teacher would clap Ms
own hands. At the beginning, Tommy would not clap his hands. So the
teacher would clap his hands for him, to demonstrate what was wanted,
and give him a piece of candy or sugar-coated cereal or some other food.
The idea was that Tommy could not get out of his chair— he was held in
by the teacher's legs if necessary-until he had done what his teacher
wanted him to do. Fussing and crying did no good. After 20 minutes 01
so, Tommy clapped his hands.

When the boy had learned 15 or 20 imitative responses through such a
grueling process, he began imitating immediately. For example, Dr.
Hingtgen folded his hands in a rather complex fashion, and Tommy,
though he had never before seen him do this, did it himself right away.

Sometimes the boy would try doing the same thing over and over, only
to find that repetitive responses were not rewarded; so he would stop
making them. For the sake of the reward, he learned to watch his teachers
very carefully,

Tommy's use of objects had been limited and bizarre. He would busy
himself for hours with ritualistic-like activities, such as moving a piece of
string back and forth in front of his eyes or tearing paper to bits. Con-
sequently, getting him to imitate the investigators' use of objects was
especially difficult. He got around to picking up a pencil and drawing a
straight line on a piece of paper only after 3 hours of work-spread over
several days, for the sake of everyone's nervous system. But once he drew
the straight line, it was easier to get him to draw X's, squares, and circles.

Learning to fold a piece of paper once and then to fold it again took 5
hours.

Getting him to use scissors was a traumatic experience, too. He watched
Mrs. Coulter use them and he wanted to hold them but had no idea how.
Eventually she used tape to help keep them in place, and in 2 hours she
got him to make one little cut in paper. After a number of additional
sessions, he was cutting out circles and squares.

Working like that 7 days a week, the investigators would go home "a bit
edgy," Dr. Hingtgen recalls, but once the behavior started coming, a
session with Tommy was rewarding as well as exhausting.

Getting this mute youngster to make sounds and say words proved to
be, as expected, the most difficult part of the process. Since Tommy'had

421

no idea what to do with his mouth in order to voice a word, the teachers
began by having him imitate the position of the mouth necessary foi
forming a given sound. In the case of the "ah" sound, for example,
Tommy was rewarded at first merely for pointing to his chin and opening
his mouth; later he had to make the sound.

To teach him the "P" sound, the teachers brought in a harmonica,
which Tommy knew how to use. As he blew it, they would take it away
and reward him for continuing to blow. After a long session, he finally
came to voice the required sound every time the investigators offered it.

At the end of the 3 weeks, Tommy would say 18 words, imitativcly.
Some woids weie very clew— baby, mama, daddy, puppy. Some wore
appioximations— feer instead of finger. He also had more than 150
imitative responses, including standing, running, and jumping, thai
required him to make use of his body. And he had more than 100 uses of
objects.

He was doing all these things only imitatively. But he was imitating
consistently and paying consistent attention to the two adults closest to
him. At this point, Tommy was returned to the ward, and Dr. Hmglgcn
and Mrs. Coulter began teaching the learning-by-imitation technique to
the nurses and other peisonnel and to Tommy's parents. The emphasis
now was not merely on increasing his imitative repertory, but also on
getting him to attach meaning to his vaiious responses, particularly (he
vocal ones. After 2 more months of morning and afternoon sessions in
the hospital 5 days a week, at home on weekends-Tommy had about 30
words to which he attached meaning. For example, when a teacher
pointed to her nose Tommy said nose\ when she held up a picture ol*£i
puppy, he said puppy,

Whatever the original cause of early childhood schizophrenia, Dr.
Hingtgen believes a youngster persists in his withdrawn attitude and
bizarre behavior because he lias learned that he can get by without doing
anything his parents want. "At home," the investigator points out, "the
parents have never been able to command as much from the child as we
have because they have not been able to invest so much time. Also, they
never think of spending 3 hours trying to get him to hold a pair of
scissors, for example, because no normal child requires that long.

"Some authorities have thought that these children really cannot learn,
but our idea is that they are very adept at learning how not to learn -how
not to have to do something." At the beginning of Tommy's training, for
example, the boy would giggle or laugh instead of trying to do what had
been requested. Then he'd act as if he were going to cry. Then he'd try to
fight his way out of the chair, and he might break into real tears. But the
investigators would not go away, and they would not cease making their
demand, so eventually Tommy would come out with the behavior they
wanted. Since Tommy's avoidance behavior, as psychologists call it, was
never successful, within 3 weeks he stopped using it.

The same technique has brought similar results in another autistic child,

5-year-old Peggy, who had been even more withdrawn at the beginning
than Tnrnrnw

Don't the children get tired of the rewards— become literally fed up?
Tommy and Peggy did not, although they weie eating almost con-
tinuously dm ing the 3 weeks of intensive treatment, and each gained
about a pound a week. But the rewards were very small; one potato chip
would be broken into six or eight pieces.

Then, too, the food reward was coupled with two other reinforcement
measmes. One was the easing of physical lestraint if such restraint had
been necessary, as at first it generally had been. The other was social
reinforcement Whenever Tommy made the appropriate response, the
investigator would say, "Good boy, Tommy "-or hug him, or give him a
ride around the room, or just touch him. In an earlier study, the
Indianapolis team had tiied using social reinforcers alone to improve the
behavior of children like Tommy, and the gains had been neither very
large nor consistent. But when Tommy and Peggy had finished their
intensive treatment, the investigators found it possible to reduce the
frequency of the food rewards and count more and mote on the social
reinforcers alone. "One theory of learning," Dr. Hingtgen observes, "is
that all social reinforcers gain their strength by having been initially paired
with food and warmth. Hopefully, at some point in the training of these
children, social reinforcers— paired only occasionally with food— will be
sufficient."

Because only certain behaviors are rewardcd-thoae the adults request
and demonstrate— the children do not become automatons, doing every-
thing they see their teachers do. Eventually, it is hoped, the habits they
are learning, of paying attention and of doing what is asked and demon-
strated, will take hold to such an extent that the children will sponta-
neously imitate behavior-including speech, play activities, use of house-
hold equipment—that normal children imitate. Tommy and Peggy do a
little of this now. They also obey some simple instructions even when no
demonstration is given.

Dr. Hingtgen throws in a cautioner. Both Tommy and Peggy have been
significantly improved but are still a long way from where they should be,
considering their ages. So the question is: How far can you take such
children through the use of this technique? On the basis of Dr. Lovaas's
work, Dr. Hingtgen notes, there is evidence that at least some severely
schizophrenic children can be brought into a classroom situation. Whether
or not they will ever be normal, no one cay say.

Further, this investigator emphasizes, the technique has been used with
probably not moie than half a dozen children. "So we don't know yet
whether this is a valid technique for all early child schizophrenics," he
observes, "or whether those six just happened to be ones that it worked
with."

The Indianapolis group, though, has made headway in answering some
other questions about the technique. For one thing, its application can be
made less strenuous for the teacher than Tommy's case may suggest.
Tommy was isolated for 3 weeks and all the shaping of his behavior was
done by two persons. Peggy, too, was isolated, but half a dozen persons
guided the conditioning process. Dr. Hingtgen now suspects, and will try

423

to confirm, that the same results can be obtained if the child continues to
live on the ward instead of being isolated when the conditioning process
begins. The investigator also plans to test his idea that the technique, after
the intensive preliminary treatment, can be used with two or three
children at a time. As he sees it, hospital care might then consist mainly
of a series of classes in which the children would learn to imitate various
kinds of responses. Sometimes the child would be with a small group;
sometiines alone with a therapist or another adult.

In many cases, this investigator believes, it will be good for the parents
to come light into the hospital and work with the child on a regular basis*
as Tommy's and Peggy's mothers have been doing. "This is a very simple
technique," he points out, "and can be taught to anybody: simply
rewarding the child at the appropriate time and withholding reward at the
appropriate time."

Among the basic questions still to be answered is how many models of
behavior a schizophrenic child must be trained to imitate before he begins
imitating on his own.

The Semester Program

As noted earlier, the Center now uses conditioning techniques with all
its patients in an effort to correct those behaviors in a child that have
most upset the family. Though the application is less intensive than the
process used with Tommy and Peggy, the techniques are basically the
same. While the children are in the hospital, the parents, individually and
in groups, are instructed how to take care of them when they come home,
as they do for 7 months of the year.

"The predominate experience everywhere in trying to help a young,
severely ill child," Dr. Churchill remarks in explaining these rather
revolutionary innovations, "has been of working hard and for a long time
and then, more often than not, of seeing him transferred to a State insti-
tution for the mentally ill or the retarded."

After a child has been a patient at the Center or some comparable place
for 2 or 3 years, the psychiatrist continues, the parents "have sort of
fallen out of love with him." It is then extremely difficult to get him back
into the family again, and this is partly because not very much has been
done with either the child or the parents to enable him to live in the
home.

"In nther words," Dr. Churchill observes, "the emphasis has been on
an all-nr-nni-hing proposition— on curing." A psycho-
Mmself, he goes on, "has been trained
•'tient and to consider, in line with
aing on inside the patient, and why,
Dssible to establish this relationship,
lis may take a couple of years— and
ent. But generally the child still is

At the Centei, where the experience has been typical, he believes, the
instances in which these young children have improved sufficiently to
return home and to enter a public school have been so few and far
between that "we'ie not sure whether they improved to this extent
because of what we did or in spite of what we did."

Now, under the new semester system, the Center has a dual objective:
to cure the children, "if they can be cured/' and meanwhile to make them
easier to live with and to help the families get along with them.

"Family morbidity in these cases is very high," Dr. Churchill notes.
"These are families who haven't gone to a restauiant or even out on a
family picnic for several years because they cannot manage the sick child.
The parents may never go out together because babysitters won't come in.
Often the house is stripped, with everything of value kept behind locked
doors, and the family is living within bare walls."

In working with the parents, the therapists focus not so much on the
emotional conflicts that may have led or contributed to the child's illness
but on specific methods of dealing with his most disruptive types of
behaviors -methods used by the Center itself while the child is an
inpatient. For example, if the child is sitting at the table and throwing
food, the parents are instiucted to give him a warning, If he throws food
again, that's the end of the meal for him and he doesn't eat until the next
meal,

This, too, is operant conditioning. It pairs an undesirable behavior,
throwing food, with a unpleasant result, removal of food. In the case of a
child who begins to wreck a room or a grocery store, the unpleasant result
may be simply physical restraint.

Many times, though, the best way to handle unwanted behavior is
simply to ignore it. It was failure to do this when the behavior first
appeared, Dr. Churchill believes, that may have helped perpetuate it. In
other words, even if disturbed behavior has a physiological basis, the first
adult response to it may have had a rewarding and thus a strengthening
effect.

Sometimes, notably in the case of temper tantrums, there is more than
one way of ignoring a behavior, and choosing the effective way requires
the parent to consider why the child is acting as he is. For instance, if the
child has been using tantrums as a means of compelling an adult to come
and do something for him that he should be doing for himself, the
effective form of ignoring the tantrum is to stay away. But if the child has
been using tantrums to hold off an adult and prevent necessary ministra-
tions, the effective form of ignoring the tantrum is to plow ahead and do
what is needed.

Usually it is extremely difficult or impossible to learn whether the
abnormal behavior of these young children had its roots in organic trouble
or in childhood experiences or, as the psychiatrists and psychologists at the
Center are inclined to think, in both.

As an example of the difficulty in tracing the origin of abnormal
behaviors, Dr. Churchill tells about S-year-old Jimmy who socks himself
on the head and jaw until he is bruised and his knuckles have calluses.

425

During some 20-minute periods he has been observed to hit himself more
than 2,000 times.

After studying Jimmy's history, the psychiatrist first thought that the
behavior probably had been learned, in the sense that Jimmy had received
a lot of attention for it. Adults would step in and hold his hands. Jimmy
would hit himself while eating, too, so his parents would hold his hands
and feed him. Things got to the point where he was being held most of the
time and was refusing to feed himself.

Under treatment at the Center, everything that might be considered to
reinforce or reward the hitting behavior was withdrawn, but instead of
dying out, as learning theory says should be the case, the behavior
continued. Theie was some improvement, in that Jimmy began feeding
himself, but Dr. Churchill noticed a strange correlation: the more the boy
ate, the more he also used himself as a punching bag.

Recently the staff tried a different approach— the administration of
painful but harmless electiic shock, paired with the word shock, whenever
Jimmy began hitting himself. Within 3 days his hitting behavior, which
had been essentially unaltered for 3 years, practically stopped. Further,
the investigator reports, this behavior can now be controlled by the word
shock alone. The boy has become more approachable and is playing with
toys, which he had ignored for months.

Some physiological factor is believed to be at least partly involved in
Jimmy's tiouble,

Sally, another child at the Center, typically withdraws to a corner of
the playroom, sits on the floor, and rocks endlessly. All children do a little
rocking, Dr. Churchill notes: they find pleasure in the rhythm. Why some
schizophrenic children keep it up for hours is not known.

In this connection, Dr. Gerald D. Alpern, a psychologist* calls iU-
tention to blindisms, the term applied to the repetitive behavior—such as
rocking, swaying, and passing the hands back and forth in front of the
eyes— seen in many blind children. With much of their external slimuhi-
tion cut off, these children have turned to themselves for stimulation.
Blind youngsters are not mentally ill, but they do obviously have im
organic defect: they cannot see. There is a possibility, then, that schizo-
phrenic children who engage in behavior similar to blindisms also have un
organic defect. There is also a possibility that blind children who engage
most strongly in such behavior have an emotional disturbance as well as itn
organic defect. This second possibility is now being studied at the
University of Indiana Medical School as part of a different research
program.

Some of the Center's Other Research Projects

To help pin down the factors contributing to early childhood psychosis,
Dr. DeMyer and her associates are comparing in great detail 30 schizo-
phrenic children and their families with 30 normal children and their
families. The children are given physical and neurological examinations

426

and are tested psychologically. Samples of blood and urine are analyzed
for the presence of abnormal metabolic products. Information about their
behavior and growth from birth onward— including eating, sleeping,
physical and mental development, and social skills— and about the
attitudes, personalities, and life histories of their parents is obtained from
an exhaustive series of interviews, covering 13 schedules, with the fathers
and mothers.

The investigator points out that the study, like other research on the
causes of schizophrenia, looks backward and therefore has to rely for
much of its information on what the parents remember and choose to tell.
However, it carefully checks the account of one parent against that of the
other and it carefully compaies all the data about a family in which a
child is schizophrenic with all the data about a family that is similar to it
except for the absence of schizophrenia.

Besides having to make its observations after the fact, research on the
causes of schizophrenia is frequently hampered also by the fallibility of
clinical judgment. Drawing upon her own experience, Dr. DeMyer says an
investigator is likely to get quite definite impressions about the first
families he sees in a study like this and then, unless he is very careful, he
tends to see the rest of the families in the light of his early impressions.
Dr. DeMyer, trying to be very careful, found that the more schizophrenic
children she studied, the wider the lange of characteristics she saw in their
families.

To minimize the problem of clinical judgment and to permit hundreds
of facts about each family to be considered, all data in the current
research are given numerical ratings, and the comparisons will be made by
a computer.

Tying in with this woik on the antecedents of childhood schizophrenia
is a recently completed survey of the families of 99 schizophrenic children
and 146 disturbed but not psychotic children. These children had been
seen at the La Rue Carter Hospital between 1955 and 1 963, and the data
about their families came from the files. The parents of schizophrenic
children, no matter what the type of schizophrenia, were found to be
significantly better educated than the other parents (who had about the
same amount of schooling as the average Indiana adult) and the fathers
held much better jobs.

These findings (a) support the earlier observation by another investigator
that parents of autistic children tend to be well educated and successful,
and (b) extend this observation to include the parents of schizophrenic
children in general. Another interesting and unexplained finding: among
the disturbed but nonpsychotic children, broken homes were common;
among the schizophrenic children, they were not.

Other research underway by the Indianapolis group includes:
• Observations to test Dr. Churchill's hunch that autistic children at
times pay considerably more attention to adults when the adults
seem to be paying no attention to the children. In terms of condi-
tioning theory, this would mean that the attention of adults had

427

become negatively reinforcing, and it would have important implica-
tions for therapy.

A variety of studies intended to learn more about, and to describe
systematically, the behavior of schizophrenic children. In one of
these, a moving picture is made of each child's "behavior clay" by
filming him for 10 seconds eveiy 5 minutes, from the time he wakes
up to the time he goes to sleep. The films will be used to compare
the dominant behavioral characteristics of the different types of
schizophrenic children and also to study changes in the children from
one year to the next. Another study of behavior is concerned with
the way autistic children use toys. The first findings show that
autistic children, as compared with groups of normal and retimltxl
children, (a) make fewer different uses of a given toy, and (b) spend
less time using one toy in combination with another.
An attempt to develop a quantitative means of measuring Uie
abilities of young schizophrenic children who cannot be tested on
standard intelligence tests. In one test being used by the investigator,
Dr. Alpern, the youngster is scored on the basis of his failure to curry
out certain activities of which he is known to be capable. The reason-
ing here is that it may be less important to know what a schi/.o-
phrenic child can do than to know what he won't do. The variability
in his "failure score" over time, it is hoped, may be a useful indicator
of how well he will respond to treatment.

Research Grant: MH 5154
Date of Interview: Apr. 26, 1966

References.

DeMyer, Marian K., & Ferster, C. B. Teaching new social behavior to schizophrenic

children./. Child Psychiat., 1963, 1, 3
Ferster, C. B., & DeMyer, Marian K. A method for the experimental analysis of (he

behavior of autistic children, Amer. J. Orthopsychiat., 1962, 32, 1
Hingtgen, J. N., Sanders, Beverly J., & DeMyer, Marian K. Shaping cooperative

responses in early childhood schizophrenics. Presented at annual meeting of the

Amer. Psychol, Association, 1963.
Lowe, Lois H. Families of children with early childhood schizophrenia: scluclcd

demographic information. Arch. Gen. Psyclnal., 1966, 14.
Onheiber, Phyllis, White, P. T., DeMyer, Marian K., & Ottinger, D. R. Sleep and drcmu

patterns of child schizophrenics. Arch. Gen, Psychiat., 1965, 12.
Tilton, J. R., & Ottinger, D. R. Comparison of the toy play behavior of autistic,

retarded, and normal children. Psychol Rep,, 1964, 15.
White, P. T., DeMyer, W., & DeMyer, Marian K. EEC abnormalities in early childhood

schizophrenia: a double-blind study of psychiatrically disturbed and normal

children during promazine sedation. Amer. J. Psychiat., 1964, 120, 10.

Investigator

Douglas A. Sargent, M.D.

The Merrill-Palmer Institute

Detroit, Mich.

Prepared by:

Clarissa Wittenberg

Through a special piogram of foster home placement, a number of
seriously disturbed boys who seemed headed foi a lifelong series of
troubles that would burden the community as well as themselves appear
to have been helped toward a relatively normal adolescence.

This program, known as the Detroit Foster Homes Project, so far
appeals to have been successful even though the boys, neglected by their
parents, had previously undergone at least two and as many as nine un-
successful foster home placements. The community agencies that referred
them to the project considered that further placement in the usual type of
foster home would be difficult if not impossible.

Persons acquainted with the neglected childien in our communities find
it hard to be optimistic about their future. But the Detroit program
suggests that some of the most troubled and troublesome of these children
may be put on the path to a normal life.

The projects which was sponsored and staffed by the Merrill-Palmer
Institute of Human Development and Family Life, grew out of concern
for the boys who were being returned to the care of the Juvenile Court
because they were deemed no longer suitable for foster care. These
children weie either so aggressive or disturbed as to be difficult to manage,
yet foster care was considered the treatment of choice. Until these boys
liad had the stabilizing opportunity to live in one place with people who
cared about them, it was felt, all other forms of assistance would be
ineffectual.

Fifteen boys between the ages of 7 and 13 were selected to test two
ideas: (1) That the "holding power" of foster homes could be increased to
the point where it could sustain the impact of such problem children and
thus keep them from detention or reform facilities, and (2) that place-
ment of troubled children in this age group would make it possible for
them to adjust and stabilize before the turbulence of adolescence. These
boys, 1 1 of whom were Negro and 4 white, had histories that included
chronic anxiety, school failure as well as such behavior as stealing,
vandalism, truancy, setting fires, and fighting.

429

Children were not accepted if they showed signs of serious physical
illness, brain damage, psychosis, or retardation. One of the boys accepted,
however, was suspected of being psychotic and this was later confirmed.

Only a small number of the boys had had a few as two past placements,
the number established as a minimum for acceptance by the project. The
average number of "official" placements was four. Often a child had also
been in numerous unrecorded, "temporary" placements, Besides the
placements, many of the children early in life had been shifted from
relative to relative. In short, the relationships and living situations of all
the boys had been continually disrupted.

One boy, for example, had been deseited by his mother, cared for by
his grandmother, and been placed in two temporary homes, all by the
time he was a year old. He remained in his next home for 5 years. It was
then declared unfit and he was moved. His consequent unhappincss and
disorientation made themselves appaient in school, so he was moved
again, this time to let him "benefit" from a better, more suppoitive school
program, but he then began to picsent difficulties in his foster homo as
well, so he was taken to a community shelter for children. There he wiis
described by the staff psychiatrist as the most disturbed child he had ever
seen. He was moved briefly to the Youth Home of the Juvenile Court
because his behavior was uncontrollable at the shelter. From there he was
admitted to a small children's unit of a psychiatric hospital. After rcniiiiii-
ing there half a year, he was referred to the project. During his first visit lo
the office for evaluation, he hid under benches and lockeis and refused to
speak. He attempted to frustrate an attempt at psychological testing, by
trying to stick his hand into the blades of a fan. Incidents or even minor
disappointments often set off tantrums of unreasoning rage which lasted
as long as an houi. During these he was very destructive and it sometimes
took two adults to restrain him. Later on he killed and maimed pets En Die
foster home.

Another boy came to the attention of the Juvenile Court after three

siblings were hospitalized for tuberculosis and a man living in the house

with the children and their mother was found to have an active cast) of I he

disease. When the juvenile authorities came to investigate, the mother

could not be found. Subsequently this boy maintained that if his mother

had known they were going to come that day, she would not have lei Ihe

police take him and that she would try to get him back as soon us site

could. This he maintained despite the fact that she rarely visited him tnuE

made no attempt to remove him from the detention home. While lie was

detained there he fought constantly. Obscene or critical comments about

mothers weie especially likely to trigger his rage. He was academically

retarded. He couldn't read. His school record showed that for many years

HP hnH rtfto.i ™«p to school so tired and hungry he could not study. He

:ause he felt it better to be considered bad than stupid.

stories were similarly dreary, varying only in detail. In

/orkers had tried to help the family and avoid tlic

of these attempts were ineffectual. The Juvenile Court

. to take custody of the child. In no case was a return to

the parents a possible alternative, as their lives were too chaotic. Most of
the parents had chopped all contact with the boys. One mother was in a
mental hospital and seveial others had histories of hospitalization for
mental illness or alcoholism. All of the families were broken by divorce,
separation, death, or the fact that the mother had never been married.
One father was in prison. None of the fathers visited their sons. Several
mothers themselves had been'court wards as children. The occasional visits
between these children and their relatives usually were followed by in-
creased anxiety or distuibed behavior in the children.

The extent of family breakdown and its attendant problems was in-
creased by the fact that the boys often came from large families. Ten of
these families had from 7 to 14 children. Each child with siblings had
siblings in foster care. Rarely had any been placed together. Only six of
the boys ever had been able to maintain contact with their siblings.

The Foster Parents

In spite of an intensive campaign and the full cooperation of other
agencies, the project found it difficult to find foster parents who met even
Its rathei permissive standards. These required that the foster patents be
self-supporting, without obvious mental disorder, be able to read and to
value education. Further they were to have some emotional flexibility and
the capacity to cooperate with professional workers. (This last point
turned out to be of paramount importance.) As it turned out, applicants
were eliminated only for such serious reasons as psychosis or marginal
financial 01 occupational status. Only 5 percent of applicants were final-
ly accepted.

In the end, the project reports, it had an acceptable group of foster
parents, seriously interested for the most part in what it was trying to do
but more like the average foster parent than the ideal— and better so, since
this was a demonstration program. A few of the foster parents turned out
to have rather serious emotional problems which only became apparent as
involvement with the child deepened.

The foster parents ranged in age from 30 to 64, with the average father
being 49 and the average mother 44. Ten couples had had only one
marriage. Eight had been married 21 years or more. Ten had children of
their own, and eight of these couples had children still living at home. Ten
of the foster parents came from the South, the others from the urban
North, Four had been foster children themselves.

Ten of the families were buying their homes, and two owned their
btisinesses. One father was a professional engineer, and the rest had had
long, steady employment in various Detroit factories. Several of the
women had had training in practical nursing, beauty parlor work, office
machine operation. Several had held long-term domestic positions. One
had worked in a factory. Family incomes for the most part were in the
vicinity of $6,000.

431

Most of the foster patents had had considerably less than a high school
education, a finding that agrees with the results of a preliminary study, by
the project directors, showing a low state of literacy to be common among
foster parents in general. Because of this low-educational level> project
workers used little written material and routinely offered assistance wlieu-
ever forms had to be filled out. This lack of education became a more
obvious handicap when the parents tried to help the child with school-
work or to communicate with his teacher.

A high degree of community responsibility existed among the foster
parents. Several were active in their churches, unions, or neighborhood
groups. Most of them were quite involved with their relatives and spent a
great deal of time with them.

These people gave a variety of reasons for wanting to be foster parents.
Several childless couples said that with a foster child they could feel they
had a family. A few other couples said they had worked hard, had
achieved substantial income and comfort, and now wanted to do some-
thing for an unfortunate child. Several wanted companions for their own
children. Some had previously been foster parents and had found it a
satisfying experience. Several foster mothers hoped to repeat the highly
gratifying experience they had had in caring for their own children. In itt
least one case the foster mother was lonely and wanted company. At leiist
one couple hoped ultimately to adopt the foster child. Subsequently it
became evident that in addition to these overt expectations, there were
other needs which led these couples into foster care and not all of these
needs were compatible with the job, as it turned out.

Working With the Foster Parents: the First Steps

Through frequent, sometimes daily discussions with individual pjirunts
or couples and through group discussions, both before and after the hoys
had been placed, the staff tried to help the parents understand and cope
with the boys' behavior.

Major attention was given to widening the foster parents' range of
disciplinary techniques. Because many of the boys had suffered physical
abuse and been emotionally hurt by it, spanking was specifically
prohibited and help given the parents in developing alternative ways of
dealing with undesirable behavior, Particularly encouraged were (ho
anticipation of explosive situations, the avoidance of overstimulation, and
the use of words rather than blows as a means of communication.

Many of the foster parents had had an abiding faith in the efficacy of
spanking in raising their own children. The project's rule against it created
and was not especially successful. Spanking occurred, the project
Tn school, too, spanking or slapping was sometimes resorted lo in
-3 project's plea. They conclude that, for a variety of reasons,
sonditions of the project, the occasional use of physical punish-
levitable.

The project attempted to place the children so that each would be the
youngest child in the family to lessen the competition with the family's
own children.

Before a boy was actually placed, a caseworker attempted to build a
good relationship both with him and with the foster family. Great pains
were taken to pace the transition from detention home, etc., to the new
foster home according to the needs of the child and foster family. Then,
for many days after placement, the caseworker was available almost
continually. This was important, the project found, because the parents
and the child tended to have a great many questions and worries about
each other and to be reluctant to deal with them. Many of these worries
were acted out directly by the children, and the casewoiker's help was
needed to cope with these reactions.

In some of the first cases, the parents had parties for the children when
they arrived. But some foster children, it developed, were under so much
strain that they could not stand this welcoming ceremony and ran away.
So in the latei cases, on the advice of the staff, the family's response to the
arrival of the foster child was modified and played down.

In general, the best beginnings were in those families that deviated as
little as possible from the noimal routine and had other children to help
ease the strain of fitting into the new neighborhood.

All the boys tested the parents to see if and when they would be
rejected. The hardest kind of behavior for some of the parents to deal
with was withdrawal. Some of the boys became almost mute, and
physically unavailable to their new parents. In other cases, oaths and
obscenities were used to test foster parents or to keep them at a distance.

Some of the foster parents asked that the boy be removed from their
care. These parents, in general, felt less well prepared for the child and
looked upon the caseworker as an investigator checking up on them. The
other parents tended to see him as a helper and themselves as part of a
team.

The parents who gave up the children seemed to have been most upset
by aggression directed against themselves or other family members. They
mentioned, for example, "sassing me," "fighting with my son/*
"defiance," and "cursing and temper."

The parents who kept children, on the othei hand, seemed to be more
worded about the child's well-being. Typically their concerns focused on
such activities as "staying out all night," "backing car into the street
(7-year-old)," "walks to swimming pool by himself.'* It may be that the
two groups of parents were not exposed to the same kinds of behavior.
Temper tantrums were reported by most of the parents who gave up
children but by only a few of the other parents.

Approaches and Techniques

In spite of the insulting and sometimes assaultive behavior exhibited by
many of the boys, the project staff made the continuing stability of the

433

Most of the foster patents had had considerably less than a high school
education, a finding that agrees with the results of a preliminary study, by
the project directors, showing a low state of literacy to be common among
foster parents in general. Because of this low-educational level, project
workers used little written material and routinely offered assistance when-
ever forms had to be filled out. This lack of education became a moie
obvious handicap when the parents tried to help the child with school-
work or to communicate with his teacher.

A high degiee of community responsibility existed among the foster
parents. Several were active in their churches, unions, or neighborhood
groups. Most of them were quite involved with their relatives and spent a
great deal of time with them.

These people gave a variety of reasons for wanting to be foster parents.
Several childless couples said that with a foster child they could feel they
had a family. A few other couples said they had worked hard, luui
achieved substantial income and comfort, and now wanted to do some-
thing for an unfortunate child. Several wanted companions for their own
children. Some had previously been foster parents and had found it a
satisfying experience. Several foster mothers hoped to repeat the highly
gratifying experience they had had in caring for their own children. In ut
least one case the foster mother was lonely and wanted company. At least
one couple hoped ultimately to adopt the foster child. Subsequently it
became evident that in addition to these overt expectations, there were
other needs which led these couples into foster care and not all of these
needs were compatible with the job, as it turned out.

Working With the Foster Parents: the First Steps

Through frequent, sometimes daily discussions with individual pnrcnts
or couples and through group discussions, both before and after the boys
had been placed, the staff tried to help the parents understand and cope
with the boys' behavior.

Major attention was given to widening the foster parents' range of
disciplinary techniques. Because many of the boys had suffered physical
abuse and been emotionally hurt by it, spanking was specifically
prohibited and help given the parents in developing alternative ways of
dealing with undesirable behavior. Particularly encouraged were the
anticipation of explosive situations, the avoidance of overstimnlation, njul
the use of words rather than blows as a means of communication.

Many of the foster parents had had an abiding faith in the efficacy of

""""uing in raising their own children. The project's rule against it crcntcd

^ and was not especially successful. Spanking occurred, the project

chool, too, spanking or slapping was sometimes resorted lo In

project's plea. They conclude that, for a variety of reasons,

jnditions of the project, the occasional use of physical punish-

evitable.

The project attempted to place the children so that each would be the
youngest child in the family to lessen the competition with the family's
own children.

Before a boy was actually placed, a caseworker attempted to build a
good relationship both with him and with the foster family. Great pains
were taken to pace the transition from detention home, etc., to the new
foster home according to the needs of the child and foster family. Then,
for many days after placement, the caseworker was available almost
continually. This was important, the project found, because the parents
and the child tended to have a great many questions and worries about
each othei and to be reluctant to deal with them. Many of these worries
were acted out directly by the children, and the caseworker's help was
needed to cope with these reactions.

In some of the fiist cases, the parents had parties for the children when
they arrived. But some foster children, it developed, were under so much
strain that they could not stand this welcoming ceremony and ran away.
So in the later cases, on the advice of the staff, the family's response to the
anival of the foster child was modified and played down.

In general, the best beginnings were in those families that deviated as
little as possible from the noimal routine and had other children to help
ease the strain of fitting into the new neighborhood.

All the boys tested the parents to see if and when they would be
rejected. The hardest kind of behavior for some of the parents to deal
with was withdrawal. Some of the boys became almost mute, and
physically unavailable to their new parents. In other cases, oaths and
obscenities were used to test foster parents or to keep them at a distance.

Some of the foster parents asked that the boy be removed from their
care. These patents, in general, felt less well prepared for the child and
looked upon the caseworker as an investigator checking up on them. The
other parents tended to see him as a helper and themselves as part of a
team,

The parents who gave up the children seemed to have been most upset
by aggression directed against themselves or other family members. They
mentioned, for example, "sassing me," "fighting with my son,"
"defiance,** and "cursing and temper."

The parents who kept children, on the other hand, seemed to be more
worried about the child's well-being. Typically their concerns focused on
such activities as "staying out all night,** "backing car into the street
(7-year-old)," "walks to swimming pool by himself." It may be that the
two groups of parents were not exposed to the same kinds of behavior.
Temper tantrums were reported by most of the parents who gave up
children but by only a few of the other parents.

Approaches and Techniques

In spite of the insulting and sometimes assaultive behavior exhibited by
many of the boys, the project staff made the continuing stability of the

433

foster home one of its key objectives. Trying to hold to this objective
became very difficult at times. For example, one child's behavior was
occasionally so disturbed-it reached a peak when he killed a neighbor-
hood puppy-that maintaining the placement was, to some degree, a
threat to the community.

The easiest action in this case, the project reports, would have been to
remove the child. This was avoided, the crisis passed, and the placement
has been sustained. There is no certainty that he will not eventually be
hospitalized anyway; as the fostei mother put it, however, "If he doesn't
make it here, he'll never make it."

A foster home placement, the staff points out, demands a great deal of
work and devotion from the foster patents and a great deal from the
foster child as well. The child, however, cannot begin attempting to
modify his own behavior until he has some certainty that people aic going
to support him and to continue to be concerned about him. Boys like
those in the pioject cannot be expected to increase their trust until they
become involved with people who aie tiustworthy.

The staff iccognized, and tried to get foster paients and teacheis to
recognize, that the boys had suffered real deprivation. The aim was to
repair the effects of this, to meet the child's needs at all levels, and at tlie
same time to make appropriate demands on the child.

Each foster home had a variety of cases— among them illness, death,
financial difficulty, the mother's going out to work, and the threatened
separation of the parents— that endangered its stability. But the crises were
often temporaiy or, in the judgment of the project, not so harmful to the
child as moving him would have been. The policy was to scrutinize the
health of the home and its membeis continually but not to move a child
without exceptional reasons. There is some evidence to suggest that the
involvement of the foster child in coping with these crises was beneficial
to him.

In its work with the children and their foster parents, the project used
numerous approaches and techniques, the main ones summarized below.

Casework in the home itself was emphasized. -To answer questions us
they came up, and to give all the assistance possible, the caseworker
visited the foster parents and the child at least once a week— more
frequently if the situation wairanted. Even when the parents icporlcc)
having no particular problems or nothing to discuss, the visits were made
and-often these were the most fruitful ones.

Originally it had been planned to have the foster parents meet regularly

as a group, and this was to have become the key educational medium.

However, most parents showed little interest in the planned programs anil

attended only sporadically. The only programs that were well attended

were those planned to recognize the work of the parents, such as riu:

annual Hinnpr Nnr did a group session turn out to be a place to shait

arents, by and large, considered these difficulties tc

-nd were often reluctant to discuss them, even witl

Psychiatric and other professional help was always available.— Periodic
psychiatric interviews were held with the boys. Psychiatric consultation
was always available to the caseworkers, whenever an emergency seemed
to warrant it. Other specialists in appropriate fields, such as education and
psychology, were project consultants and were called upon for help as
needed. Beyond this, the actual progress of the project was kept under
continued surveillance and modification by the administrative staff and
consultants.

Group work services were used for the boys, the general objective of
which was to help a boy fit into his new home and neighborhood. Specific
aims were to give him healthy experiences with other children, help him
attain social skills, especially to Icain how to play.

In the beginning boys were added to the group as they were accepted
by the project, but the group became unwieldy. To accomplish the
therapeutic aims of the group, these boys needed extensive assistance and
new caiefully selected groups were formed of thiee or four boys each.

Where theie was another young child in the foster home, he was included
in the group. Neighborhood friends weie invited to share in the group
activities which included games, trips, and other recreational activities.
Many of the boys, it turned out, had never been to even a store, a bank, or
a post office. Several had never been to the city park, Belle Isle, to which
the entire city seems to flock in hot weather. Such trips were also used to
observe and modify the boy's behavioral troubles.

Special emphasis was given to schooling,— Because parents were not able
to keep a child who failed continually in school, and because these
children were likely to do so, the project staff—along with the foster
parents, if possible-met with the school's administratois and teacheis and
discussed the boy's problems before a boy was enrolled in a school. It was
found that this procedure made for a better relationship between school
and paients and helped to save the foster parent from being criticized for
the child's behavior or conditions which had developed before he came to
their home. Discussions about the boy's needs and the project's ideas
about discipline were held with each teacher.

The staff kept in close touch with each school and was ready to help
whenever difficulties arose. When a school called for assistance, members
of the staff responded at once. Consultants were brought in when
advisable. Personal history data and diagnoses of the children were shared
with each teacher whenever the project judged this information to be
useful. Most of the time this confidence was justified. Sometimes it
backfired.

The staff found that the boys tended to do better in highly structured
situations, such as a class in arithmetic, than in looser, more informal
groupings such as music classes and library and gym sessions. Consequent-
ly, when a boy needed a shorter school day in order to survive the strain,
these latter were the first classes eliminated.

435

Efforts were made to have the boys spend as much time as possible
with those teachers who were found to be making outstanding contri-
butions. Letters of appreciation were written to teachers and administra-
tors. With the project's coopeiation, several of the teachers who wert
attending university classes wrote reports of their work with the boys.

Tutoring became an important part of the program.-^ commor
problem-arid one contributing to behavior difficulties in school-wa:
reading retardation. Even in the third and fourth grades, some of the boyi
were virtually nonreaders. For all the children having trouble in reading
the pioject established a therapeutic tutoring program.

The tutors set out to reach the boys by using material familiar to them
including the songs of the Beatles and books of comics. After such ,11
introduction, it was easier to get on with conventional reading material
The sessions were planned to use to the utmost the child's ability t<
concentrate and comprehend yet keep fatigue to a minimum. Food nm
other treats were a valuable entree into the confidence of these deprive*
boys.

Reading became a very important area of achievement, the projcc
reports. Often the boys would want to demonstrate their new road in
ability to the caseworker and other adults.

Incidentally, the tutoring sessions were considered a valuable means c
helping to meet the difficulties brought on by the summertime closiu
of the schools. Foi some of these boys, June meant still anolhc
sepaiation from people who had become important to them, and face
them with the rather aimless and formless-and therefore difficult time c
summer. Of course, some of the boys simply were glad to get away froi
school.

Efforts were made to strengthen the boy's sense of identity, -T\\u ii
formation available about his background was given to the foster parent
and the healthy ties remaining from his past were nurtured. Altonipi
were made to fill in the voids in the child's knowledge of his own past.

The project staff initiated contact with the natural parents when the
could be located and, in the beginning, in some cases, arranged for thej
to visit the boys. These visits, it was found, always caused serioi
difficulty for the boys and in some cases jeopardized the foster hon
placements. Since it was impossible for these boys to return to the
natural parents, and since the foster homes were considered essential
their healthy development, parental visits came to be virtually eliminalc
By and large, the staff found, the situation that had led to parental ncglc
or abuse, and to court action on behalf of these boys had tended
deteriorate rather than improve.

wnc a relative who visited and maintained
)oy. Visits in such cases were encourage
ilped a boy to visit a past foster parent.
sd routinely by the psychiatrist, who us
ing technique. The boys were fascinated 1
;s-sometimes they were taken alone a
their foster family. The pictures becat

valued possessions. The pictures were also used as tangible evidence of the
project's concern for the child, a concrete demonstration that his person
was valued (the photos decorated the project's walls). It also seemed to
enhance the child's self-concept.

Anothei identity strengthening technique was used in the tutoring
sessions, though primarily tp provide the child with material he would like
to read. The tutor would help the child compose a story about himself
and then would have it typed for him.

The Outcome of the Program

The project repoits the following results:

1. All the boys have received good medical and dental care and are
physically well.

2. Thirteen out of fifteen boys remain either in foster homes or have
been adopted. Some are enjoying a high degree of security for the first
time in their lives.

3. Enuresis, existing in one-half of the boys, has ceased in every case.

4. They are less depressed than they were.

5. Their relationships with people are becoming deeper and more ap-
propriate, although there is still a tendency to try to keep from getting
involved.

6. Most are behaving more nearly at a level appropriate to their age. In
the beginning, many had a tendency to play with children much younger
than themselves- frequently to the despair of the foster parents. One
father, for instance, who had hoped for a companionable youngster,
found that the boy liked most of all to ride a tricycle or play with the
neighborhood small fry. In some cases there are still remnants of such
behavior.

7. Fears and anxiety, both conscious and as revealed in the dreams and
fantasies of the boys, are less. Those remaining are more realistic. In the
beginning, in many cases, violent scenes— peopled with monsters, dis-
membered peisons, and lost children-predominated. Fears of personal or
family destruction were common,

8. The boys' attitude toward food has improved. It is less frequently
used to assuage anxiety or as a substitute for love. Early in the project,
one boy who was offered a piece of cake ate the whole cake, and children
offered snacks during a group program or a casework session frequently
demanded an entire meal. Another boy, fearful that his foster parents did
not love him, feared that he could not have a second helping; another
took literally his foster mother's remark, "We don't have a thing in the
house to eat," and slipped out to a neighbor's. Still another ate voraciously
and became very fat. Tills boy is now secure enough to be able to diet, has
lost 30 pounds, and is proud of the change in his appearance.

9. Antisocial behavior has decreased. From diffuse outbreaks against
the community, such as throwing bricks through a school window, they
have shifted to more localized acts with a more immediate bearing upon

437

themselves, such a truancy. One boy has stopped stealing even though he
had been taught to steal by his mother.

10. In school, some of the boys are still behind the grade level ap-
propriate for their age but all have made progress. One boy had to be sent
to a special school but the others are in regular classes. The nonreadcrs
now read.

11. Perhaps the most marked gain has been in verbal facility and
spontaneity. This improvement is attributed mainly to the increase in
personal security, the lessening of hostility, encouragement of the ability
to use words lather than action to express themselves, and the
opportunity to talk with friendly people.

12. The boys are less tense and better able to concentrate than when
they first came into the project. Then they were inclined to be reckless
and physically overactive. They often avoided eye contact or met the eyes
defiantly. They either slumped in chairs or were unable to lemain seated.
Regular interview situations were difficult foi them.

13. Contrary to some commonly voiced fears, the boys have not
"infected" other children -either those in the foster family or those in tlic
neighborhood. In the case of the boy who killed a puppy, his sadistic
behavior with animals has been of concern to the neighbors, but Ihcii
children, instead of following his lead, tend to reject this behavior and
thus to exert a helpful influence upon him. In only one case was there a
serious problem between the foster child and the natural child: they
fought to such an extent that the mother asked to have the foster child
removed. She was concerned also that this boy's stealing would set a bad
example for her son. (This is the same parent who wanted a foster child us
a companion for her son. This placement was an error in judgment, ac-
cording to the project.)

Still piesent in each of these children are the tremendous aftereffects of
early deprivation and of repeated separation from parental figures. The
children in varying degrees still show signs of hostility, distrust, ap-
prehension, and sadness. Nor has the program been successful in every
case. One boy, for example, was sent to a mental hospital. (Luler
authorities returned him to his natural mother, though she was barely able
to care for herself.) One is in a boy's home described as "rather pleasant";
a third is in a treatment-oriented detention facility described as "relatively
good,"

Nevertheless, most of these "impossible to place" youngsters have

become far better able to manage their day-to-day living in a foster home.

-am i-.nn ^»ionstrated, its directors submit, that many of Ihc

rp JK a means of helping such children can be

of comprehensive services, It then

tment with the advantages of a stable

ome from living in a community. Thus

ly to have required institutional care

y and needed no halfway house to ac-

,he end of their treatment,

Research Grant' Rll-MH-1551
Date of Interview , Apr. 6, 1966

References:

Ambinder, W J. The extent of successive placements among boys in fostei homes.

Child Welf., July 1965,397-398.
Ambinder, W. J., & Falik, H. The behaviorally disturbed foster child in school— a

preliminary report. Unpublished paper. Detroit Foster Homes Project, 1965.
Ambinder, W. J. & Falik, H The social acceptance of the behaviorally disturbed boy in

his classroom group. Unpublished paper. J, Sch. Psychol
James, Adnenne Differences between two groups of foster parents. Paper presented at

Second Annual Conf. on Foster Care for Emotionally Disturbed Children, Detroit

Foster Homes Project, February, 1965.
James, Adnenne. Casework with emotionally disturbed children in foster care.

Presented at NCSW, Chicago, 1966.
Mahaffey, Maryann. Progress report of an experiment in foster care for disturbed boys,

Paper read at Ohio Welf. Conf., Cincinnati, October, 1 965.
Redl, F., & Wineman, D, The aggressive child. Glencoe, 111.1 Fiee Press, 1957.
Sargent, D. A., & Ambinder, W. J. Foster parents techniques of management of

preadolescent boys' deviant behavior. Child Welfare, February, 1965, 90—94.
Sargent, D. A., Ambinder, W. J., & Fireman, Laura. Verbal abilities and literacy levels.

required of foster parents. Child Welfate, December 1 963, 502-503,
Sargent, D. A , Ambinder, W. J,, Fireman, Laura, & Wineman, D. Role phenomena and

foster care for disturbed children. Amer. J. Orthopsychiat., 1962, 32, 1,
Shiefman, Emma The Beatles? Yeah! Yeah! Yeah! The Reading Teacher, October,

1965.
Shiefman, Emma. A school teacher in a child welfare agency. Children,. May— June

1966, 116-118.

439

Investigator.

Abraham WiklerM.D.*
Addiction Research Center, NIMH
Lexington, Ky.

Prepared by

Antoinette A. Gatozzi

The postadclict patient leaves the Lexington hospital free of drugs. He
has had the best tieatment that medicine knows how to provide. Whether
he entered as a Federal prisoner or a voluntary patient, he was carefully
and humanely withdrawn from his drug, usually heroin, and received
expert aftercaie and rehabilitation services. But the grim fact is that the
discharged postaddict cannot be given much chance to lemain free of
drugs. One of three patients is readmitted because he has relapsed to the
use of narcotics.

This calculation is based on patient records of the two U.S. Public
Health Service hospitals, at Fort Worth, Tex., and Lexington, Ky., that arc
charged by law to care for narcotic addicts in the United States. Patients
are Federal prisoners addicted to narcotic drugs and narcotic addicls who
voluntarily enter for treatment. Because of the inherent selectivity of this
group, the readmission late would seem to represent a lower instance of
relapse after cure than that which prevails among the postaddict popula-
tion as a whole. A certain number of relapsed postaddict patients defy
accountability as such.

Tragically, some shoot themselves with a fatal overdose of opiiilcs,
unknowingly or perhaps suicidally oblivious of their lost tolerance to
dosages formerly self-administered. Doubtless there are others who lake
up using opiates again, stay clear of the law, and do not choose to be
withdrawn. Many postaddicts turn to abuse of different drugs, notably
barbiturates. A study made 2 years ago of 1 ,000 consecutive admissions to
Lexington found one fourth of the patients physically dependent on
barbiturates as well as addicted to a narcotic drug. Although it is difficult
to discover the true number of postaddicts who relapse after cure, all
observers agree that the rate is high and the problem pressing.

At the Addiction Research Center, a facility of the intramural program
of the National Institute of Mental Health, the question of relapse is one
of many that researchers are investigating. Understanding why postiulditHs
relapse after cure depends upon insight into the factors involved in the

*Now at the school of Medicine, University of Kentucky Research Foundation,
Lexington, Kentucky.

initial use and abuse of drugs. More information must be gained about the
mechanisms of tolerance, dependence, and the stereotyped withdrawal
syndrome that follows abrupt discontinuation of addicting compounds
befoie the question of relapse can be answered definitively.

The classic drugs of addiction are the opiates. Of these, people most
often hear of morphine and heroin. Morphine is of great value in the
practice of medicine; its prescriptive use by physicians is controlled by
law. Heroin is almost identical to morphine in biological activity. Grain
for gram, it is two to three times as potent as morphine, which may
represent a convenient economy to those who deal criminally in opiates
and thereby explain why heroin is the opiate of the majoiity of narcotic
addicts. All heroin is illegal in the United States. An opiate interacts with
the cells of an individual's somatic and autonomic neivous systems,
profoundly alters the functioning of vital organs and glands, and affects
his mood and motivation. Although there is evidence that the
phenomenon begins with the first dose, with prolonged use the individual
becomes physiologically tolerant to the drug's actions, and the original
dose no longei induces the original effects. A very small proportion of
those who experience the effects of an opiate over a period of time
become dependent on the drug. They feel they must continue using it in
order to maintain a sense of well-being, both physical and mental.

Why Do They Relapse?

Addicts call morphine "God's own medicine." Many patients with
severe pain who have been given the drug by their physicians would agree
with that assessment because the sine qua noti of an opiate is the capacity
to relieve pain. No other compound approaches it in this respect although
the search for an equally effective analgesic with lower abuse potentiality
has been long and intense. Opiates also possess narcotic, or sleep-
producing properties, and for this reason opium derivatives and their
synthetic analogs are called narcotic analgesics. Medical scientists do not
fully undeistand how analgesics (from aspirin to morphine) relieve pain.
There is still much to learn about pain itself and the influence of
emotional factors on the threshold level at which noxious physical stimuli
are perceived as pain. The outstanding capacity of the opiates to relieve
physical pain and its associative psychic components is probably involved
in the postaddict's proclivity to relapse.

Once withdrawn from the drug, the postaddict faces again the personal
and social difficulties that formed at least part of the reason why he
plunged into drug abuse initially. The narcotic addict's personal problems
are thought to center around the expression of aggression and the primary
drives of sexual gratification and avoidance of pain. These normal human
drives provoke in him sharp, sometimes intolerable anxieties, and he may
have abused the drug originally because it led him to a state of dreamy
indifference to reality. The social factors that dispose to relapse stem from
such handicaps as a truncated education, little or no job experience, racial

441

minority-group status, and a criminal record. These are common features
of the backdrop of addiction. All aie obstacles to the postaddict's
attempts to become productive and responsible.

Other factois in relapse may relate to the physical changes wrought by
prolonged use of a narcotic. For months or years the addict sm rendered
himself many times a day to the pharmacologic thralldom of his drug. It is
possible, indeed it now seems likely, that he does not regain his foimer
physical status after the 7-10 days required to detoxify him and lite few
months in a duig-free environment needed to free him of all obvious signs
of addiction. Dr. William R. Martin, director of the Addiction Research
Center, is now in the midst of a detailed study of the effects and after-
effects of chronic narcotic addiction in man. The classic study done in (he
mid-1 930's by C. K. Himmelsbach, the fiist director of the Cenlcr,
showed that as long as 6 months after withdrawal the postaddict is hyper-
reactive to some physical stimuli.

Then there is the habit itself, a pattern of behavior that becomes a way
of life for the addict, pivoting on the compulsive acquistion and use oi' the
duig. The narcotic habit is embedded in fundamentals of character, social
backgiound, and neurophysiological functioning, and breaking it may
prove insupeiable for many postaddicts. To account in part for this aspect
of relapse after cure, a two-factor learning theory has been proposed by
Dr. Abraham Wikler. Dr. Wikler is professoi of psychiatry at the Univei-
sity of Kentucky College of Medicine and consultant to the Addiction
Research Center. For many ycais he served as chief of the Center's section
on expeiimental neuropsychiatiy.

Postaddicts have desciibed experiencing physical symptoms very like
those of acute drug abstinence when they return to their home environ-
ments many weeks or months after withdrawal. These reports suggested to
Dr. Wiklei that physical dependence on the narcotic drug might become
conditioned dining the course of addiction to certain situations in I he
addict's enviionment This conditioning would follow the Pavloviim
model, in which a dog came to salivate at the sound of a bell that had rung
many times before in temporal contiguity with the presentation of meal
powder.

The investigator reasoned that Pavlovian conditioning could occur when
the addict failed to get his drug in time to avert the emergence of" with-
drawal symptoms, which might begin within 3 or 4 hours after the last
dose. A conditioned response could be engendered through repealed
temporal associations between such sporadic, accidental abstinences iiml
specific environments, for example, a street corner, room, or bar, Thus,

O1'° '""' '" had been withdrawn from the drug for some time, when

HA cellular logic for abstinence symptoms, Uic
•»f past abstinence might trigger in him
' mimicked real abstinence.
• hypothesizes in relapse is antilogous
ierant conditioning. In this model uu
sously (he "operates" on his cnviron-
led by him as positive or negative. If

the consequence is positive and rewarding, the theory holds, it will
reinforce the act that proceeded it. That is, the act will be repeated if it
culminates in reward often enough. The addict's frequent intermittent
bouts of abstinence are totally and dramatically relieved when he does
succeed in acquiring and using the drug. Thus, his drug-acquisitory
behavior is reinforced again and again by the reward of relief from
abstinence symptoms.

Both kinds of conditioned learning may influence behavior in the same
individual. Dr. Wilder further postulates that the permanence of this leatn-
ing'is directly related to the amount of effort put into the acquisition of
drugs. Here again the investigatoi took his lead from the reports of former
addicts, who often brag among themselves about how hard they "hustled"
(almost always ciiminally) and how good they were at getting their supply
when they were "on the street." Dr. Wikler suggests that the status
accorded to a successful hustler by addict society is a potent secondary
reinforcer of drug-acquisitory behavior. Scheduling of reinforcement and
stimulus generalization are other concepts of conditioned learning theory
that, Dr. Wikler believes, also are peitinent to the role of learning in
relapse after cure.

Addiction and Relapse in Rats

Experiments to test the Pavlovian and operant learning hypotheses were
done with rats. The studies involved these basic design elements; Rats that
weie physically dependent on morphine, a suitable reinforcing agent,
methods for conditioning the animals, and testing procedures to deter-
mine whether conditioning had occurred.

Rats were made physically dependent on morphine by injecting them
with the drug once a day at about 8 a.m. Doses were gradually increased
over a period of 6 weeks until the desired high-dose level was reached. The
animals then were maintained on this dosage. They wore addicted. Control
rats weie treated similarly, except that their morning injection was an
innocuous saline solution. By evening the addict rats' doses were wearing
off and they were exhibiting signs of abstinence including "wet-dog
shakes." The phrase aptly conveys the rat's movement: A shaking and
twitching of trunk and limbs that resemble a wet dog shaking water from
itself. Normal rats briefly Behave this way after handling. Earlier work by
Dr. Wikler and his colleagues had shown that elevated wet-dog frequencies
were reliable signs of early abstinence in the chronically addicted rat,
roughly paralleling such other abstinence symptoms as loss of weight and
low body temperature, increased activity and elimination. Wet-dog shakes
may be the murine equivalent of a human abstinence symptom called, in
another graphic phrase, "cold turkey." Addicts use it to refer to wave
after chilling wave of gooseflesh (medically termed piloerection) and to
describe the whole range of symptoms that occurs on abrupt termination
of narcotic use.

443

The reinforcing agent had to be an opiate or any other chemical
compound that exerted opiate-like effects. Morphine itself was the logical
choice. It proved to be unsatisfactory, however, presumably because the
quantities required for reinforcing purposes and the methods of delivery
to the rats were aversive to the animals. Then the investigators learned of a
synthetic drug called etonitazene, which was just being described,
Etonitazene produced morphine-like effects in animals. It was shown to be
1,000 times as potent as moiphine as an analgesic in the rat. The re-
searchers undertook a series of experiments with this interesting com-
pound to deteimine if it could serve as the reinforcing agent in condition-
ing studies.

Their pieliminary observations were encouraging. First, they found that
rats would drink a dilute water solution of etonitazene without being
forced to, that is, without prior water deprivation. Water deprivation had
been a complicating but necessary procedure when morphine was tried as
the reinforcing agent. If etonitazene were used, water deprivation could be
eliminated. Second, drug-treated animals that were abstinent 24-48 hours
drank much more water if it contained etonitazene than if it were plain,
and abstinent rats consuming etonitazene solution showed lower wet-clog
frequencies than those drinking water. It looked as if the grealci
consumption of etonitazene solution compared to water might have
resulted from the drug's power to suppress abstinence discomfort. Further
study demonstrated that this was indeed so.

In normal, control rats, fluid consumption was not affected when a
solution containing 5 or 10 micrograms of etonitazene per milliliter of
water was substituted for plain water. Nor was fluid consumption altered
when the substitution was made for animals in a state of morphine
intoxication, a period lasting about 7 hours from the time of the single
daily drug injection. However, experimental animals showing morphine-
abstinence symptoms, which began about the 13th hour after injection
and continued until the next morning injection, drank much greater
volumes when their fluid was etonitazene solution than when it was water.
Moreover, the usual primary abstinence symptoms did not appear when
experimental animals in a condition of emerging abstinence were allowed
to drink drug solution. The investigators reasoned that etonitazene
solution was consumed in greater quantities than water by morphine-
abstinent rats because for them-and only for them-it had reinforcing
properties, specifically, relief from abstinence discomforts. In sum,
etonitazene's various attributes made it an excellent choice for the role of
reinforcing agent.

The procedures devised to condition animals were carried out in in-
dividual linear mazes. Mazes were constructed so that an animal's access to
one or the other end could be blocked. On certain days of the week, from
the time of morning injection until 2 p.m., all rats had free run of their
mazes and could drink plain water available at either end. From 2 p.m.
until the next morning they were allowed access to only one end where,
again, they found water to drink at will. The evening routine was altered
on other days of the week: From about 8 p.m. until the next morning all

444

rats were confined to the opposite maze end. The fluid available for
drinking here was different for each of four subgroups. In every maze, this
end also was cued to help the animal discriminate it from the other end.
Anise flavoi tagged the drinking fluid in the first study; in a replicate
study, visual and tactile cues were used instead.

• Drinking water available to one group of morphine-addicted rats
contained 10 micrograms of etonitazene in every milliliter of water, a
concentration sufficient to relieve abstinence.

• A second group of morphine-addicted rats had no etonitazene in
their drinking water. Only the discriminative cues marked this end as
different from the other end for them.

• One control group had available for drinking a solution containing
half the amount of etonitazene placed in the first addict group's water.
The smaller potency reduced the chances of these animals becoming
tolerant to, or dependent on the drug.

« A second control group was placed in the same situation as that of
the second addict group.

Two basic strategies were behind these rather complicated procedures.
First, both groups of addict rats suffered withdrawal symptoms nightly in
the mazes; they were conditioned to the mazes as an environment of
abstinence. Second, one addict group was conditioned on alternate nights
to associate relief from abstinence with the behavior of drinking water
(containing etonitazene), tagged with certain cues, while the second addict
group was never allowed to relieve abstinence and therefore made no
association between relief and the behavior of drinking water (not
containing etonitazene), tagged with the same cues. The first group was
trained in the operant conditioning model and the second was not.

Conditioning procedures were carried out for 6 consecutive weeks.
Animals remained in mazes the following week but were not allowed
access to the cued end-compartments. This was a precautionary step to
equalize the degrees of addiction of the two experimental subgroups; in
case the one that had consumed etonitazene solution every other night
during conditioning had acquired a stronger drug-dependence than the
subgroup that received only the morphine injections.

Then all injections were halted and animals were returned to their home
cages. All drug-treated rats underwent abrupt withdrawal and showed the
typical acute abstinence symptoms. Beginning about a week after
injections were stopped and continuing at intervals up to several months
afterward, the animals were observed in a battery of procedures and
relapse tests to determine the effect of training in relation to previous
treatment.

Dr. Wikler and his coworker, Frank T. Pescor, a biologist, conducted
two complete conditioning studies, retracing each step of treatment, train-
ing, and testing. The second study differed somewhat in details from the
first, which is the one described above, but the principles of each were
identical. The studies yielded comparable results.

The investigators found that wet-dog frequencies of postaddict rats
were higher in the mazes than in the home cages long after ail other signs

445

of primary abstinence had subsided and postaddict wet-dog frequencies in
home cages had returned to the level shown by controls. In some relapse
tests, wet-dog fiequencies of control rats changed in the same direction as
those of postaddict rats, but the amount of change was smaller for
controls. Thus, the experiments demonstrated the Pavlovian condition-
ability of this abstinence phenomenon in the rat. Considered along witii
the suggestive clinical reports of former addicts, the experimental evidence
supports the hypothesis that abstinence phenomena may be similarly
conditioned in man.

Evidence concerning the role of reinforced ope rant behavior in relapse
could not be obtained. During the test period of the study, rats were
exposed to both forced drinking of dilute etonitazene water and free-
choice drinking of either plain or doped water. According to the
hypothesis, those postaddict rats which had learned during training thnl
the doped water alleviated their discomfort (abstinence symptoms) would
choose to drink more doped water after cure than would postaddict rats
which had not been so trained. The investigators found that, although
both postaddict groups drank significantly more doped water than did
control rats, there was no difference between the two postaddict groups in
the amount of doped water consumed by free choice after withdrawn!.
Both relapsed after cure, but the investigators could not find any dif-
ference between them in the nature of their relapse. Dr. Wikler believes
that a test with greater discriminative power than free-choice drinking will
have to be devised to expose differences in relapse liability that might
exist between the two groups, and thereby to isolate the role of reinforced
operant behavior.

The experiment did demonstrate that, regardless of the roles of
Pavlovian and operant conditioning, previous addiction itself disposed rats
to consume etonitazene solution long after primary withdrawal symptoms
had subsided. Dr. Wikler suggests that postaddict rats' greater free-choice
consumption of doped water, compared with the amount never-addicted
rats chose to drink, may have been due to prolonged aftereffects of ad-
diction. The blatant, well-recognized symptoms of primary abstinence in
the chronically addicted rat run their course by the third day after abrupt
withdrawal, but earlier studies by these investigators had demonstrated
so-called secondary abstinence phenomena in the chronically addicted nit,
persisting 4-6 months after withdrawal. Moreover, they had found that
even 6 months after withdrawal there were minor differences between
postaddict and control rats. In the conditioning study, the persistence of
elevated wet-dog frequencies long after withdrawal may be regarded as i\
manifestation of secondary abstinence. The evidence suggests that a long-
-g, subtle homeostatic imbalance in postaddict rats is an important
i generating relapse after cure. Does protracted abstinence occur in
/hen Dr. Martin and his colleagues complete the work mentioned
we shall be able to answer this question with considerably more
ice than is now possible.

Procedures To Break the Habit

The etiology of relapse after cure is certain to involve interactions of
both physiological factois, including conditioned learning to a greater or
lesser extent. Every promising lead must be followed; until we understand
why postaddicts relapse after cure we cannot be sure that the prevailing
methods of treating narcotic addiction are truly curative. If postaddict
patients are found to experience long-enduring residual aftereffects, for
instance, the exact duration and physiological nature of the effects will
need to be determined and methods of specific treatment devised. To the
extent that conditioning factors are shown to contribute to relapse after
cure, extinction procedures would become part of the treatment legimen.
Dr. Wikler has speculated about the general outlines of such a piogram. He
points out that extinction of conditioned abstinence phenomena and of
reinforced drug-acquisitory behavior require different procedures and
should be done separately, beginning with the latter.

Two new drugs would be needed. One would be used with incoming
addicted patients during the detoxification phase of treatment. Its
qualities would be such that it could be substituted for the opiate patients
had been using. On withdiawal, it would not cause the emergence of
characteristic heroin or morphine withdrawal phenomena. Then the new
drug would be abruptly withdrawn, and the result would be a piolonged
but not severe abstinence syndrome. At this point the second new chug
would be used. This drug should possess some rewarding properties so that
patients would work for it, but it should not be effective in relieving
abstinence. In this manner, the usual reward associated with drug-
acquisitory behavior would not be forthcoming, and the old conditioned
stimulus-response pattern could be steadily weakened.

Dr. Wiklei proposes that rewards to condition socially acceptable
operant behavior be provided in separate training. Postaddicts might earn
money for useful work, which they would be allowed to spend for things
they want other than drugs.

Extinction of conditioned physical dependence would begin after this
hospital program. It would be carried out in the postaddict's home en-
vironment. Surveillance, nonnarcotic drug therapy, and psychotherapy
would be used to bolster the postaddict and forestall relapse should
conditioned abstinence occur. In this way, the classical (Pavlovian) condi-
tioning factor would gradually be extinguished.

Recent studies made at the Addiction Research Center have provided
another possible means of extinguishing conditioned abstinence and drug-
seeking behavior. A new synthetic substance— cyclazocine-seems to have
especially desirable characteristics. Dr. Wikler thinks that maintenance of
postaddict patients on cyclazocine, as proposed by the research group
headed by Dr. Martin,1 would lead to expeiimental extinction of the

'Martin, W. R., Gorodetzky, C. W., and McClane, T. K. Treatment of Narcotic
Addicts With Cyclazocine, Clin. Pharmacol., Therap,, 1966, 7,455-465.

447

conditioned abstinence symptoms and drug-acquisitive behavior he
postulates are factors in relapse.

Cyclazodne is a narcotic antagonist, a compound that blocks many of
an opiate's effects. Like several of the narcotic antagonists known,
cyclazocine has analgesic properties; indeed, it possesses many times the
analgesic potency of morphine. (It has been found to produce hallucina-
tions in man frequently, however, and this disturbing side effect makes it
unacceptable as an analgesic for general clinical use.) Another significant
property of many of these agents is that chronic usage does not produce
tolerance to their opiate counteractions. This is true of cyclazocine.

The Martin group found that subjects (abstinent postaddict patients)
who had been chronically intoxicated and then withdrawn from
cyclazocine experienced only mild abstinence symptoms and did not crave
more of the substance. Further work demonstrated that, when subjects
were taking cyclazocine chronically, even very large doses of morphine
failed to produce such customary effects as severe respiratory depression
and euphoria. Under these circumstances of cyclazocine premedication,
the patients were also protected fiom developing physical dependence
when morphine was chronically administered.

The researchers outlined how cyclazocine might be employed wi!h
abstinent narcotic addicts trying to stay free of opiates. The physically
dependent addict first must be withdrawn from his drug because cycln-
zocine precipitates a violent abstinence syndrome in the morphine-
dependent patient. Then cyclazocine may be given orally, twice a day, in
small doses that are gradually increased to the desired level, The
postaddict may be maintained at this level until he is judged to luivc
gained all he can from the treatment; then the agent may be gradually
withdrawn. The investigators advised that withdrawal from narcotics mul
stabilization on cyclazocine be done in a controlled hospital setting. Treat-
ment may continue in an ambulatory setting once the patient has reached
the stabilization dose level. Other types of care, provided by psycho-
therapists and social workers, may then help him build a new life for
himself.

While the postaddict patient is maintained on cyclazocine, he would be

protected from the euphoria of even large doses of morphine or heroin

taken, perhaps, while on a spree-and from developing physical

dependence on opiates again. Cyclazocine maintenance thus would avert

two of the pharmacologic actions of opiates believed to be imortanl in

narcotic addiction. "There may be other benefits," the Martin group

Dr. Wikler's work, they concluded; "It is possible Unit

™«+ to r^dict themselves while receiving a narcotic

there may be extinction of conditioned

iking behavior."2 Clinical trials of cycla-

addicts are currently underway in New

Intramural- NIMH

Date of Interview: August, 1966

References'

Martin, W. R., Wilder, A., Eades, C. G., and Pescor, F. T Tolerance to and physical
dependence on morphine in rats. Psycho pharmacologia, 1963, 4, 247-260.

Martin, W. R., Gorodetzky, C. W., and McClane, T K. Treatment of narcotic addicts
with cyclazocine. CUn. Pharmacol. Therap., 1966, 7, 455-465.

Wikler, A. Opiates and opiate antagonists. Public Health Monograph No. 52 U.S
Department of Health, Education, and Welfare. Washington, D.C., !95S.

Wikler, A. Conditioning factors in opiate addiction and relapse. In Kassenbaum, G. G.
and Wilner, D. M. (eds.), Narcotics. New York. McGraw-Hill, 1965.

Wilder, A., and Pescor, F. T. Factois disposing to "relapse" in rats previously addicted
to morphine (abs.). The Pharmacologist* 1965, 7, 171.

Wikler, A., Martin, W. R., Pescor, F. T., and Eades, C. G. Factors regulating oral
consumption of an opioid (etonitazene) by morphine-addicted rats. Psycho-
pharmacologta, 1963,5,55-76.

Wikler, A., and Pescor, F. T. Classical conditioning of a morphine abstinence
phenomenon, reinforcement of opiod-drinking behavior and "relapse" in morphine-
addicted rais.Psychopharmacologia, 1967, 10, 3,255-284.

449

Investigators

Constantine J. Falliers, M.D.

Kenneth Purcell, Ph. D.

William W. Hahn,Ph. D,

Children's Asthma Research Institute and Hospital

Denver, Colo.

Prepared by:

Gay Luce

The Children's Asthma Institute and Hospital is widely known for Us
multidisciplinary research on asthma, and for outstanding success in treat-
ing severely ill children. Partly aided by NIMH, investigators there arc
studying the manner in which emotions and physiology interact to induce
asthmatic episodes. Dr. William W. Hahn has found that asthmatic
children tend to withdraw from criticism and to respond to stress with
distinctive respiratory and pulse changes, Currently, he is monitoring the
sleep of asthmatic children in order to determine what physiological
and/or psychological states in sleep are correlated with night attacks. Dr.
Kenneth Purcell has devised a system of FM monitoring which will enable
the staff to review a child's social reactions in an effort to sec the
emotional antecedents of acute attacks. Although asthmatic attacks muy
be triggered by some combination of emotional and allergic circum-
stances, a new approach has been to scan for biological rhythms that niny
hold clues to susceptibility. A recent study by Dr. Falliers indicates that
theie is a 24-hour rhythm in the peak expiratory flow rate—a measure of
lung function— in asthmatic children. Furthermore, the timing of hormone
therapy alters the timing of this peak.

Innovations in long-distance monitoring by FM radio, and perhaps later
by biotelemetry, are opening up new possibilities in understanding how
emotions, external conditions, and perhaps body rhythms conspire to
produce the crippling symptoms of asthma and how timing may become
an effective element in future treatment of the illness.

Background

pling and dreadful illnesses, most people

This is odd, because asthma or related

out of every 10 Americans. Attacks arc

st most people do not consider asthma a

ailment. In point of fact, it causes about

9,000 deaths a year, deaths that are particularly horrible since they arise
from strangulation. About 3 million American children suffer from
asthma at one time or another; 300,000 have chronic asthma, and many
are so severely ill that they do not improve under conventional treatment.
They miss from 1 to 3 months of school each year, cannot join in the
normal physical activities of their peers, and are retarded in physical
growth. Psychologically, these children and their families must learn to
cope with a frightful life-and- death emergency that can suddenly strike;
A child in an acute attack may gasp for breath and in the failuic of his
respiratoiy system, begin to turn blue-sometimes requiring the oxygen
equipment of a police department to save his life.

The Children's Asthma Research Institute and Hospital treats about
150 of these youngsters from all over the country. The medical program
began in 1939, with the Jewish National Home for Asthmatic Children,
and as the need for multidisciplinary reseat ch became evident, CARIH was
created in 1959, and combined residential and hospital facilities with
research. The campus resembles a private boarding school, strewn with
small residence quarters, playing fields, and recreational buildings. Incon-
spicuous among the buildings is a modern hospital and a separate research
building. The critically ill children who come to CARIH, for periods of
18-24 months, live with their peers in groups, attend school, and receive
an unusual regimen of psychological and medical treatment, and various
studies. When they leave, about 80 percent require much less medication
than before to help them lead normal lives, and the remainder can lead
normal lives with the help of drugs judiciously employed.

The success of CARIH, indeed, points up the very mysterious nature of
the illness itself.

In 1966, a group of boys, who had been too ill to walk to school before
they came to CARIH, began to play football against teams of healthy
children. They not only played: they competed and won. Bediidden chil-
dren had been transformed into sports competitors. The staff were the
first to admit that this transformation was uncanny, and it underscored
their problem in understanding the ailment. Were the boys enabled to play
football by treatments that caused physiological development? Were they
liberated by psychological therapy and the emotional change of being
away from home? Was there something generally desirable about the
Denver environment? "If we knew for sure," remarked one doctor, "we'd
soon have asthma licked."

Psychological Role of the Family

Inevitably, when a child arrives at the institute he enters a totally
changed world. He may leave behind a city like New York, with its partic-
ular climate and air pollution. There are changes in pollen characteristics,
foods, and factors such as the molds that may grow in his basement at
home, or household pets, or neighborhood pets. The child is removed
from many of the usual physical attributes of his environment that may

451

trigger or contribute to his respiratory troubles. Moreover, he leaves what-
ever emotional conflicts are engendered by his family. About a third of
the children who have come to CARIH have improved so fast, with so
little aid from medication, that they were at first called "mircale children/1
Their improvement suggested that emotional factors may have been sing-
ularly important in the acuteness of their illness.

Asthma has long been considered a psychosomatic disease by many
investigators. One old and prevailing theory held that asthmatic attacks
were the bursting forth of "suppressed crying," and the asthmatic child
was supposed to have been unable to express certain emotions. Dr.
Kenneth Purcell surveyed the interview responses of CARIH children
about the causes of their attacks. A number of them felt that crying
precipitated an attack rather than relieving it.

Others thought that laughing or coughing brought on an attack. Thus it
appeared that general excitation of the respiratory system was singled out,
and it seemed highly likely that these children learned to inhibit crying >
coughing, or laughing— precisely in order to avoid an attack. It was the
respiratory excitation, not the emotional expression, that they seemed to
want to avoid. Suppressed crying might indeed be a learned response,
Similarly, excessive mother-child dependency has been cited as a cause of
asthma: Does the mother's undue attention elicit the asthma, or is her
overprotectiveness inspired by the plight of her asthmatic child? In survey-
ing the literature and evaluating the CARIH patients, Dr. Purcell has been
struck with the fact that asthma is by no means homogeneous. Pollens,
dust, colds, foods, or climate changes precipitated attacks. Many children
have unexplained attacks in sleep, and physical exertion or weather
changes were common piecipitants. Still, there did seem to be some dis-
tinction between the so-called "miracle children" who promptly remitted
almost on arrival at CARIH and needed no medication— the intermediate
group that could respond to therapy without relying on adrenocorticoid
hormones-and the group that absolutely depended for survival upon
some hormone therapy. Dr. Purcell postulated that the drug-dependeni
children may have been more influenced by infection and allergy than the
others. Perhaps the rapid remitters were children in whom symptoms had
a predominantly emotional basis. On responses to questionnaires, parents
of these children seemed to indicate that they had more punitive and
authoritarian attitudes than did the parents of the other youngsters. Was
family conflict at the bottom of their allergic disease?

One youngster, who had improved remarkably on arrival at CARIH,
suffered a terrible attack, suddenly, 6 months later. His family had come
for their first visit. However, if the staff had been tempted to conclude
that this was a prime example of emotional triggering- they discovered
that the family cat had been brought along, and the child was highly
allergic to the beast, Thus, the interrelationship of environmental factors
had made it difficult to pin down the specific triggers. An Ingenious plan,
involving 2 years of work and considerable patience, has been devised in
order to study the role of the family in asthmatic illness. It was designed

452

to hold environment constant, so that the asthmatic child experienced
only one change— the absence 01 the presence of his family.

Children suffering from chronic asthma in the Denver area were
selected for this study. These children are outpatients of CARIH, who
have come to the institute for daily tests. After several weeks of
evaluation, the families were offered a 2-week paid vacation in a hotel
away from home. During this time, the child would be cared for and
studied by a surrogate mother from the CARIH staff. This staff member
would live at home with the child, with the daily visits to CARIH con-
tinuing throughout this period and after the family's return home.

There have been 22 children studied so far. Some have been selected on
the conjecture that they would show improvement when their families
departed, while others were not suspected of so many emotional factors
underlying their illness.

About 70 percent of the youngsters in the emotional group did actually
improve when their families were away and showed a resumption of
respiratory symptoms when they returned. Only 10 percent of the
children whose ailment was judged as less overtly emotional showed
change that could be correlated with the presence or absence of the
family. Thus, it seems that there are at least two kinds of asthmatic
children; those in whom family emotional factors play a role, and those in
whom these factors do not seem important. These two groups of children
are now being compared on tests of autonomic responses, allergic history >
and a variety of other measures. It is, however, too early to remark on the
characteristics that distinguish the two groups.

Behavioral Antecedents of the Attack

If there arc emotional interactions that precipitate asthmatic attacks,
no clear correlation has been found among studies of individual patients.
Teams of allergists, psychologists, and pediatricians have concentrated on
a few youngsters, obtaining information from the children's school-
teachers, house-parents, physician, and psychotherapist. By converging on
all the various aspects of the child's life following an asthma attack, they
had hoped to single out the piecipitating factors. Invariably there were
complications, such as changes in weather or social environment, that
made it impossible to construct a convincing link between the attack and
the situation that preceded it. Moreover, there was no comprehensive
observation of the child's social interactions, excepting when he was in the
presence of an adult, and knew it.

Dr. Purcell and his associates decided that a more natural and inclusive
mode of observation might be developed by having the children wear FM
transmitters. All of their vocalizations, as well as breathing sounds, could
then be recorded, and in the event of an asthmatic episode, the antecedent
behavior could be heard on tape recordings. This kind of field observation
without observers is possible with a miniature radio transmitter, small as a
cigarette package, that can be worn on an ordinary belt. It will transmit to

453

receives in a 200-foot range, all verbalization, whispers, ami
piratoiy sounds to signal the audible signs of asthma lie attacks. If a cliifd,
so observed, were to have 20 attacks, it would be possible to examine the
portions of the recoiding just preceding, and to compare llwsc wift
samples taken on other days at the same time. Later, the investijuiluii
might be able to re-create situations that led to attacks, substiinluiliim the
con-elation between behavior and respiratory trouble in the tohonitoiy,
One question, however, demanded an answer before such a slmly could Iw
planned. Would the children become self-conscious and dislorl their lv-
navior if they knew they were being bugged?

A pilot study now seems to have answered that question. Lust your. 16
boys and 1 0 gnls were selected from the adolescent groups as bcinn |urlie-
ulaily self-conscious and therefore good to study and test. Half ol Hie
youngsters were given live transmitters; the other half were ftivvti dum-
mies. hey wore these packages for an hour each evening for 1 0 ihiys, .nij
each child knew whether or not he was really bugged , but had been avkcd
not to tell anybody. At the end of the test, the houscparcnls xveu- nsked
D Chlldre? Wore live transmitter* and which wore dummy

D

f ™ lr P™'mably> if the bugged children w^ behaving iinion-nlly
we inS^r H T1?' ^ WOUld fac apparent to the houacimwnis. who
end Whad been asked to fil1 OL't bchovOr sc.il« on

e been asked to fil1 OL't bchovOr sc.il« on

test children dunng different stages in the 1 0 clays.

n^ Hve transmitte''s 8«ve every evidence »r

bvf. fief|m°nito/ln8 a«d of having become oblivious In

* °rf

-.wwu. itJui^uvci, til tnfi pnn nf i-Im f-.:,.i ^i . *

than chance hleue^nPwhM/^ tna1' the houseparenis did no Dodcr
radio tiansmitters Beh»v i-ff "dren W0re dummies and which wine live
been very striking 'indeed ^,1-1?°^ lf there WCre any' couUl not ljin'°
were far more sensitive tM cmidren s own descriptions of (heir Jtvlin^s

c .

lost any effect on their interac?fons ton8ua««. the monllorlim MIOII

™..Ci«:aii wation Fnr chp*? inrvestiSators have contracted with
Astern. Beidnninp i» ^leTOtopl^ «f a campus-wide tck-

Sleep

Mysteriously, out of an apparently untroubled sleep, the asthmatic
patient sometimes awakens gasping foi air, feeling himself on the verge of
suffocation. In some children the frequency of nocturnal attacks is 5-10
times as great as those occurring during the day. The night of sleep has
been one phase of the daily round in which CARIH children have been
mostly unobserved, but perhaps this span of quiet can reveal how an
attack develops and why it occurs when it does. Dr. William W. Hahn has
established a sleep-monitoring laboratoiy, within the research building,
with a lecording room in which childien now sleep through many con-
secutive nights, while their brain waves, pulse, and respiration are continu-
ously recorded.

Sleep has been found to consist of several stages, each characterized by
particular brain wave patterns and physiological changes. A person passes
down and up through these different levels of consciousness in a relatively
predictable manner throughout the night. Four or five times in the course
of a night the progression is repeated, and at each level the sleeper's
muscle tone, his arousability, and his psychological experiences appear to
be different. At approximately 90-minute intervals the eyes begin moving
rapidly, the pulse changes, respiration grows irregular, and if awakened
now, most people remember vivid dreams. This rapid-eye-movement stage
of sleep has been associated with the gastric secretions of ulcer patients,
with attacks of nocturnal angina, and some investigators have wondered
whether it may also be associated with asthma. Dr. Hahn has been looking
for a possible link between the asthmatic attack and a particular stage of
sleep. So far, five children have been studied for a total of 12 nights and
some identifiable patterns are beginning to appear.

On nights when a child in the laboratory is experiencing fairly constant
wheezing he will be likely to spend very little time in deeper stages of
sleep and proportionately more time in the lighter stages. It isn't possible
to tell, at this point, whether the child is initially restless, tense, or
uncomfortable— and his resphatory distress is a result of this psychophysi-
ological tension. Or does the distress of mild asthma and an impending
attack elicit various psychological and physical stresses in the child? Dur-
ing the night it appears that the awareness of discomfort, to the point of
waking up and doing something about it, may vary with the stage of sleep.
In the laboratory a child with very mild asthma symptoms has awakened
from a light stage of dreaming sleep to request a nebulizer (spray medica-
tion), yet the same child has continued to sleep for an hour or more,
wheezing and showing signs of greater respiratory distress during the other
stages of sleep.

Sleep study has a unique advantage for tracking the development of an
asthmatic attack. This is the one long period of time when a child will stay
still without complaint.

455

Psychophysiological Reactivity in Asthmatic and Normal Children

Two years ago Dr. Hahn and his associates began to find a suggestive
difference between asthmatic and normal children in their reactions to
mental arithmetic, interference while they were solving problems, and a
situation in which they anticipated a mild electric shock. The asthmatic
youngsters showed a notably higher heart rate. Careful checking indicated
that this was not caused by acclimatization to the Denver altitude, that it
could not be attributed to lack of adaptation, nor to medication. The
unusual incidence of tachycardia among asthmatic youngsters suggested a
possible malfunction within the autonomic nervous system. The invcsti*
gator speculated that this might be a premonitory signal, the first sign of
an abnormal response which might develop into an asthmatic attack were
the stress continued.

In a recent study, normal and asthmatic children aged 10—14 were
compared as they underwent more severe stresses, The 18 CARIH young-
sters were very much improved, had shown virtually no symptoms since
shortly after their arrival at the institute, and were taking no regular
medication. Their counterparts were 21 healthy youngsters who were
Denver residents participating in a summer recreation program. At the
outset, each boy was asked to check a list of words that approximated the
way he felt most of the time: a quick scanning for attitudes and emotions.
Then the youngster was wired for physiological recordings of heart iatc,
finger and face temperature, finger pulse volume, skin resistance, ami
respiration. As they lay in the recording room they received four kinds of
stimuli-tones, shocks, and two sets of arithmetic problems geared for
their age level.

The problem was played to the child by tape recorder, and he was
asked to solve it mentally and announce the answer. After each answer, «n
voice would make some remark which, unknown to the child, was pre-
recorded. Following responses to the first set of problems, the rcrmirk;
were neutral or encouraging. However, during the second batch of prob-
lems the child heard increasingly severe criticism. "I wonder if you'rt
really trying-most children your age don't have any trouble with these
problems."

After the session the boy was questioned. Did he realize that the voice
was a tape recording? Was he aware that ciritcism was part of the oxpcri
ment? Did he realize that his answers actually had been correct? The
questions began with an exploration of his feelings, his reactions to UK
criticism. Once the child learned that the criticism was a tape-recorclee
dupe he was usually relieved and more accurate about describing his feel
ings. In general, the asthmatic children seemed to have been more limit
and self-reproachful when they were criticized. They were apt to say thn
they had gotten angry at themselves, that they felt discouraged aiu
wondered why they were giving wrong answers, The normal youngster:
more typically believed that the voice had to be kidding or putting then
on. They didn't doubt their ability to compute the problems and tcnclcc
to be angry at the invisible critic.

456

A close look at their physiological records indicated a curious differ-
ence between the asthmatic and normal boys. The asthmatics showed a
generally higher heait rate, but their respiration exhibited a curious pat-
tern as the stiess mounted. The normal children showed an elevated respi-
ration rate, but the asthmatics reacted by breathing faster up to a point
and then, as if this were a kind of ceiling, or as if they exerted control,
their lespiration became no faster. Slowed expiration is one of the
symptoms of an asthmatic attack. It is interesting that the physiological
response to criticism resembled, in this manner, the initial stages of the
asthmatic attack.

The psychological after re act ions of the asthmatic children, as judged by
the checklist of adjectives of mood, showed a preponderance of negative
answers when once again they were asked to mark the words that de-
scribed their usual state of mind. Psychologically, as physiologically, they
gave a different lesponse to the stress of adverse criticism in comparison
with nonasthmatic peers,

In future studies the investigator will attempt to discover what this
difference means. One important possibility is that the emotional reac-
tions of inward anger and n on aggressive self-reproach are stress reactions
functionally related to the asthma attack, and not merely a byproduct of
the illness. "Although we cannot document a functional or causal relation-
ship, data from the present study illustrate that a complex psychological
stimulus initiates emotional and physiological responses in the asthmatic
child which differ from those aroused in normal children exposed to the
same conditions or stimulation."

Rhythmicity in Physiological Responses and Drug Effects

CARIH children, with all the study, monitoring, and therapy they re-
ceive, might seem to lead extremely abnormal lives. Yet for every one of
these children, life had never before been so normal. Medically hopeless,
always in jeopardy of strangling attacks, these children had lived a more
disabling life at home. There, a slight wheezing might be sufficient to
bring tension and fear into the whole family. It is clear why the particular
symptoms of this disease incur strong psychological repercussions.

At home a child might have to be rescued by the police, or taken to the
hospital in an acute attack, stricken, gasping, and turning blue. These
episodes sometimes last a few hours, and sometimes linger for as long as 4
days, Special equipment is needed, for the victim's chest would expand,
trapping air, his lungs filling with thick mucus. At the CARIH a child in
such trouble would immediately receive aid from a pressure-breathing
instrument, from chemical sprays to open his bronchial tubes and lungs,
and from an instrument designed to remove plugs of mucus from the
lungs. He would be attended by calm competent nurses and doctors, at
once soothing him and also studying the nature of his attack. To a child
who had experienced the emergency measures that all of the children
knew, the gamut of laboratory research tests was not particulary imposing.

457

Each of the children was studied along many dimensions throughout hi
stay. Within the laboratory, the children were exposed to changes ii
temperature, humidity, barometric pressure, and psychological provoc;
tion-in attempts to single out the triggering factor in their allergy. Th
general finding was never a single factor, but a combination of simultimi
ous changes that might be very slight. Each child regularly takes tests c
pulmonary function during laboratory exercise and controlled climat
changes. Analyses are made of bacteriological samples from the respiratoi
tracts. Tests of pituitary-adrenal function and immunological factors ai
continued, along with tests of particular nose spiays used to dilate tl
bionchial tubes, and studies of the manner in which these dilators nu
affect the nervous system and catecholamine metabolism. For the child
CAREH, life as a continued object of research is still more normal thi
was the disabled life of a youngster always on the precipice.

In their efforts to uncover the origins of asthma, and hopefully sor
unique factor which might be medically countered, the CARIH researc
ers have taken a new and unusual path. They have been studying o
dimension that is usually overlooked in medicine: the periodicity of
ternal functions and of physiological responses. An almost 24-hou
circadian— fluctuation has been found to be the endogenous charactci is
of many organs, cells, and subcellular activity in animals and man. Abe
every 24 hours, body temperature rises to a peak and falls to its nac
repeating this fluctuation day after day in the normal person. The level
certain adrenal hormones in the blood, or excreted in urine, also show
roughly 24-hour rhythm. At almost any level, from large body syste
down to cell division, rhythm seems to be an important property of i
body's organization. The circadian, or almost daily rhythm, is nearly
prominent among certain physiological functions as is the mcnslr
rhythm. In a manner remotely analogous to the winding of a clock
person's environment, his regimen of sleep and waking, of meals and so
activity, seem to synchronize his multitudinous internal rhythms.

Although new in application to medicine, the relevance of circadiim i
other biological rhythms has been amply demonstrated by a numbei
scientists, notably by more than a decade of research on the part of
Franz Halberg and his associates at the University of Minnesota, Not o
must the body deliver the right substances to the right places in
correct amounts-but it must do it at the right time. In the case of
victim of Cushing's disease, for instance, a clue in diagnosis is the rise
fall of circulating adrenal steroids, for it is the inverse of a norma I rhy
even when the levels remain within the so-called normal range. PI
relationships among body rhythms can also provide meaningful chic
pathology.

Two years ago, Dr. Constantino J. Falliers and Dr. Halberg took a
step in evaluating circadian rhythms in asthmatic and normal child
They compared these two groups of children on certain major body I
tuations as they occurred around the clock.

In January 1966, 19 boys at CARIH-all of them youngsters who s
free of symptoms and who required no medication-began to be the f<

458

of special and continuing interest. During the daytime, every 4 hours, they
were briefly interrupted by a staff member, and were asked to bieathe
into a tube as haid and rapidly as they could. This flowmeter measured
the number of liters of air they could expel from their lungs each minute.
The peak rate of expiratory flow is an indicator of pulmonary function.
An asthmatic person, whose respiratory passages are often blocked, may
expire too little, too slowly. During stress or an oncoming attack, expira-
tion rate diminishes and air collects in the lungs. Thus a high peak expira-
tory flow rate is a welcome sign. As these studies were to demonstrate, the
peak is not the same at different times of day, even within an individual
who is noimal. One question that interested the CARIH staff was whether
normal and asthmatic children would show their peaks at the same points
on the daily round.

In April, and again in May, the boys underwent more than their daily
tests. Now, around the clock for 48 hours, they were tested on several
physiological changes. Seventeen normal boys of the same age, orphans in
Denver institutions, were brought to the institute for scrutiny and com-
parison. At each age level, the boys lived the same schedule of mealtimes,
bedtime, and waking. For a little more than 2 days they were housed in a
rescaich wing of the hospital on campus. There they lived in scaled rooms
under the constant supervision of nurses. They could not even go to the
bathroom unsupervised, since all urine was being collected in 4-hour
samples for analysis of 17-hydroxycortiscosteroids.

Beginning at midnight, and every 4 hours thereafter, the boys were
briefly interrupted in their sleep or activities for a measure of blood
pressure, pulse, a urine sample, and rectal temperature. They were asked
to breathe into the expiratoiy flowmeter, and to estimate when 2
minutes had passed by counting to 120. This ubiquitous little test indi-
cates approximately whether a peison senses the passage of time as slow,
fast, or indeed as the stopwatch reads. In a relatively predictable fashion,
all of these physiological variables and responses cliange around the 24
hours.

It is, of course, far easier to describe in writing a 48-hour study of 3
dozen youngsters than to confront the actual doing. When measures must
be taken at equally spaced intervals, around the clock, and when precision
is crucial, staff coordination and exertion is intense. Many staff members
volunteered to work on the 2-day marathons, around the clock; an ex-
hausting procedure for all.

Throughout the longitudinal studies as well, it was necessary to chase
the boys down promptly every 4 hours, whether they were out playing,
roller skating, or swimming. A staff member would have to find them and
make them blow into the flowmeter.

The very arduousness of such a study prescribes a very cautious experi-
mental design. There are limits to the endurance of even a mildly sick
child, and limits to the energies of the staff. Thus, it is necessary to
ascertain beforehand how many samples, at what intervals, are needed to
determine the peak of a circadian rhythm, and by what analysis the fre-
quency and peak phase will become clear. A longitudinal study of certain

459

functions could be run for several months with the asthmatic children aj
one normal child, but measures could only be taken during daytim
Interruption of sleep over several months would not only be likely to alt
the rhythms under study, but also to exacerbate the children's symptoi
and subject the staff to extreme exhaustion. Thus, within the framewo
of an unevenly paced longitudinal study containing nightly gaps, she
intensive studies were conducted around the clock. By suitable analys
performed by computer, these could fill in some of the gaps. As I
Falliers said of his collaboration, the study could not have been feasible
a small institution without the sophisticated statistical help and the co
puter analysis offered by the Minnesota laboratory.

In comparing the normal and asthmatic children, the researchers ch<
to look for the highest daily values-trie peak expiratory flow rate, th
horns in which adrenal steroids were excreted most, the point of high
pulse or blood pressure, etc. These crests were then plotted on a 24-h<
compass, and it was easy to see how the two groups differed or coincid
The normal children, for instance, were close to one another in the dis
bution of their highest body temperatures, which occurred during af
noon and early evening. The asthmatic children were even more hoi
geneous, showing their peaks exclusively in the afternoon. Their fas'
pulse, however, was more distributed over the day than was the peak h<
rate of the normal youngsters. Again, in the excretion of adrenal stero
the healthy boys showed their peaks between 8 a.m. and 1 p.m., whi
few asthmatic children exhibited peaks very early in the morning o
early afternoon. Although a harmonic analysis of the data did indi<
some differences between the groups, in the crest phase of their phj/
logical functions, no further conclusions could be drawn from sue
preliminary study, Perhaps the asthmatic children as a group showt
little more variability in circadian crests.

On the other hand, the circadian rhythm of peak lung functioning t
seem to differ considerably among asthmatic and normal groups. In in
sive 48-hour studies, and in longitudinal studies, it has become appa
that there are circadian fluctuations in the peak expiratory flow ]
Healthy children showed crests between 1 and 4 p.m., while the ave
peak of the asthmatic child occurred around 2 p.m. Here, the hea
children were more heterogeneous and exhibited more individual v
tion.

Further data, still in analysis, were obtained from May until the en
September 1966. Every other week, the 17 boys and one normal gi
this study underwent 48 hours of round-the-clock study.

Circadian Rhythms in Drug Effects

Adrenal hormones play an important role in the therapy of asf
sometimes in severe cases cortisone and its analogs are the only me
tions that make life possible. Judged by available techniques, asthi
youngsters do not seem to suffer from adrenal insufficiency, an<

460

steroids they take— not for correction of a deficiency but for their potent
anti-inflammatory action— have many side effects. One of these appears
to be a dampening effect upon growth. These youngsters are usually be-
hind their peers in growth and maturation, although tests at CARIH indi-
cate that they have the normal levels of growth hormone. Some of them
have been given anabolic hormones to stimulate growth. Still, the side
effects of their lifegiving hormone treatments persist at high dosages. Al-
though the corticosteroid drugs have been available for more than 1 5
years, no definitive recommendation can yet be made regarding the best
time schedule for treatment (e.g., once a day, once every 48 hours, di-
vided doses every 4-6 hours, etc.).

Could side effects be diminished by giving smaller doses at appropriate
times? Animal experiments have demonstrated circadian susceptibility
rhythms for bacteria, drugs, X-ray, and other stimuli. The treatment that
kills an animal at one point on his daily cycle will merely make him sick at
another phase. Thus, there is reason to believe that the changing state of
the body might enhance or reduce a drug's potency, according to the time
it is administered. In June of 1966, Drs, Falliers and Halberg began to
study a time factor in the efficacy of hormone treatment. Between June
and the end of September, 28 CARIH children were medicated on specific
schedules and systematically evaluated for the crest time of their expira-
tory flow rate. All of the children were sick enough to require hormone
medication daily. They were divided into four groups matched as closely
as possible for age, sex, home locale, allergies, the amount of hormone
they needed, and their expiratory peak flow rates. For 4 months they
were maintained on prednisone. One group received it at 1 a.m., another
at 7 a.m., another at 1 p.m., and another at 7 p.m. At the end of 4 months
it was clear that the timing of the hormone affected the time at which the
child now showed his peak expiratory flow rate. These asthmatic children.,
who had been very similar in their crest phases before the study, now
diverged significantly .

Youngsters who received their drug at 1 a.m. began to show their crests
almost 2 hours earlier than they had before. Those who received it at 7
a.m. were little changed, but the children who took their drug at 1 p.m.
now showed their crests about 6 hours later than they had before. The
time of expiratory crest shifted according to the time they received the
hormone,

Although this novel finding was not immediately applicable to treat-
ment, it did suggest future possibilities in the scheduling of drugs. The
medical staff had the impression that children receiving their medication
at 1 a.m. or 7 a.m. were clinically better off than the others. As a result of
the study, Dr. Falliers felt that the dosage might possibly be reduced if it
were administered in two segments. This procedure is now being tried.
However, it is still too early to discover whether the timing of prednisone
had any influence on side effects, for these sometimes appear only after
many months or even after a drug is discontinued. The youngsters will be
monitored for side effects relative to drug timing throughout their stay at

461

CARIH. The feasibility of such long-term surveillance makes the institute
an excellent focus for such studies.

A clinical study of about 2-y ears' duiation will be needed before the
clinical value of drug scheduling can be determined. This may be a next
step, although there are some inherent limitations in the endurance of the
children, who are ill, away from home, and undergoing all the trials of
growth and of being young. An intermediate study may ascertain whether
these kids can function on smaller amounts of steroids, taken in two daily
doses.

The implications of this study extended far beyond the treatment ol
asthma, for steroid hormones are staple medication for many illnesses
whether they be chronic ailments such as arthritis or ulcerative colitis, 01
an emergency crisis such as pneumonia. The administration of steroids is ;
delicate matter, as illustrated by the kinds of serious consequences tha
can follow-the development of diabetes in predisposed persons, or th
so-callecl steroid psychosis. The role of timing in the regulatory action o
hormones makes the medical substitution— by regulation from without-
far more complicated manipulation than supplying an artificial limb to a
amputee. The exploration of this factor thus has ramifications for hclpin
the victims of a great many illnesses.

The importance of circadian rhythmicity in asthma, and in undcrslam
ing asthma, may be determined better when round-the-clock studies (
physiological functions can be made without disturbing the children
through bio telemetry . In Dr. Purcell's study of vocal and respiratory b
havior via FM transmitters worn by the children, there may be a fir
indication of the value of continuous monitoring. So far, at least, a fir
look has shown that the asthmatic child may have different timing c
specific functions than the normal child. This, in itself, is a clue and ;
encouragement to continue examining time structure.

Research Grants. MH 10385, 8415, 3269, Also: A 5963 and Ch 5523
Date of Interview: Nov. 23, 1966

References:

FalHers, C. J. Corticosteroids and anabolic hormones for childhood asthma, C\

Pediat,, 1965,4,441,
Falliers, C. J., McCann, W. P., Chai, H., Ellis, E., and Yazdi, N. Controlled study

iodotherapy for childhood asthma./. Allergy, 1966, 38, 183.
Halberg, F., and Falliers, C. J, Variability of physiologic circadian crests in group*

children studied "transversely."/. Pediat., 1966, 68,741.
Hahn, W. W. Autonomic responses of asthmatic children, Psychosom, Med., 1 9(i6,

4.

Hahn, W. W., and Clark, J. A. Psychophysiological reactivity of asthmatic child

1966 preprint.
Jorgensen, J. R., and Falliers, C. J, A rational approach to corticosteroid thcrapj

asthma in children. J.A.M.A., 1966,198,773.
Purcell, K. Critical appraisal of psychosomatic studies. JV. Y. State J, Med,, \ 965

16.
Purcell, K,, and Brady, K. Adaptation to the invasion of privacy; monitoring bolii

with a miniature radio transmitter. Merrill-Palmer Quart, Behav. and Dcv. 1

12,3.

462

investigator-
Margaret F. Gutelius, M.D.
Director

Child Health Center

Comprehensive Health Care Program

Children's Hospital

Washington, D. C.

Prepared by-

Clarissa Wittenberg

Increasingly it has been realized that helping a child "in" must begin

ry early. People involved with Head Start, a federally sponsored pre-

100! program for disadvantaged children, came quickly to the realization

at they were not beginning nearly early enough. Some suggested expand-

g the schools and preschool programs to even younger groups while others

It very strongly that the young child should be assisted while still in his

jme. One research program at Children's Hospital in Washington, D.C.,

is attempted to reach young mothers to help them learn ways of caring

jr their children and stimulating their intellectual growth so that they will

3 able to meet with success when they enter school. This program, work-

ig with black, unmarried, teenage mothers has shown that many can be

elpecl to provide successfully both well-baby care and the sort of interne-

ion that helps a baby "learn to learn." Results show that babies who have

.ad this combination of weli-baby care and infant stimulation have higher

cvclopmental I.Q.'s at 6 months of age than babies who have not had this

u-ogram. This difference continues to show up and to become more pro-

lounced when the babies are compared again at 1 year and 2 years of age.

Almost all of the mothers in the program had ambitions for their chil-

Iren to "go further" than they themselves had done. They wanted their

children to be successful in school and do well for themselves in life. This

project attempted to help them realize the important first steps in making

these dreams come true for their babies. Although in many cases the

mothers were quite immature and themselves locked in cycles of poverty

and lack of education, most were cooperative and eager to participate in

the program.

MOBILE COACH PROJECT

A total of 140 young women expecting their first child were found at
city prenatal clinics. The girls selected were black, and between 1 5 and 1 9

463

years of age and lived in a specific area surrounding the hospital. The
were in their seventh month of pregnancy, of normal intelligence, withoi
chronic physical disease, and without gross emotional pathology. Tl
group was then divided and one section received counseling from t(
research team and particularly from the public health nurse; the other, t!
control group, received prenatal and baby care given by city clinics ai
city health department nurses. The deliveries took place in city hospifc
or in private hospitals under contract to the city. After the babies we
born, a second screening was done, and only normal babies weighing
least 5 pounds 8 ounces remained in the study. This assured that t
experimental and control babies had essentially the same characteristi

Of the original 140, 95 pairs of mothers and children remained in I
final study group. Twelve babies were born prematurely. A typical rate
prematurity for babies born to ghetto mothers is about 12 percent. Til
were other pregnancies that miscarried or resulted in stillborn babies. 1
odds are against these babies even in cases such as these where the motr
had obtained prenatal care. After delivery many mothers moved away i
a few became uncooperative. Those who moved within a half hour's
by car were retained. In Washington, this encompasses almost the en
city so that only those who moved to the suburbs or out of the city v
lost for this reason. This left 47 mothers and children in the experimc
group and 48 in the control group.

A mobile coach brought well-baby care to the door for those in
experimental group. A public health nurse worked with each mo the
dividually to teach her about her baby's development. The mothers :
encouraged at each phase of development to respond to their ba
particular needs and to supply a stimulating and protective envirorm
The mobile coach made possible a very active reaching out and facilit
well baby care. Those in the control group were visited by the city R.
By and large, the city nurses only followed the babies for two or t
months. However, if there were problems it was up to the mother t>
that the child got to a well-baby clinic for check-ups. All mothers it
program used Children's Hospital clinics and facilities in emergencies
at times of serious illness.

The orientation of the stimulation program was frankly middle-

and its major goal was success in school for each child. This was a pnu

is schools remain the main way in which children can learn the t

•>eed to know to be successful in our society and to be able I

1 good jobs. The methods taught the mothers were built upo

-*• — *_„ ting and expanding the child's curiosity and develi

n was to work with the mothers until the child rei
lien to assist the families in placing their child
Start programs. However, Head Start programs
3 3-year-olds in this neighborhood and there fort
i have been placed in preschools. Some success i
'ed, however, as one child is attending a Monl
e been considered eligible for scholarships to a

private school in the city. However, it is almost impossible for these
mothers to provide transportation and to pay even incidental expenses for
private preschool education The loss of the planned Head Start programs
is serious, as it has been demonstrated that it is necessary to keep up a
level of stimulation if a child is going to continue to progress and not fall
behind.

The Mothers

All of the mothers were between 15 and 19 years of age, unmarried,
and pregnant for the first time. They all lived in a 15 -census-tract area
around Children's Hospital. This area is basically a low-income, black
ghetto. It is an area of old rowhouses, old and crowded apartment build-
ings, a high-crime rate, and many social problems. It was an area highly
involved in the riot in 1968, All of the girls had normal intelligence,
between 70 and 115 on the Peabody Picture Vocabulary Test, and were
without evidence of chronic physical disease or major mental pathology as
determined by a short interview with the psychiatrist. They were all wil-
ling and able to participate in the program. Almost 70 percent were born
and raised in Washington. All of them had finished elementary school, and
about 20 peicent had finished high school. Most had received average
grades in the regular academic curriculum and wished to finish school.
Many wanted to take additional training and learn some skill or trade.
Very few showed anxiety or depression. About 90 percent were pleased
about or at least accepting of the pregnancy. Almost half had known the
father for one to three years and more than half expressed love and
confidence in him. Most wanted to get married and have one to three
more children. None saw adoption as a desirable course for their baby.
Most had the assurance of help from the grandmother or other relatives.

The girls lived in crowded households, About a quarter of the girls lived
with both parents, but for another half of the group their mother was the
head of the house. Over half of both groups had significant contact with
their father or another man who had helped to raise them. Sixty-eight
percent had an. adult male living in the household. Their families' income
tended to be at the level for unskilled laborers. However, only three of the
girls came from families then on public assistance. Although none of these
girls came from the lowest economic level or from the most disorganized
families, they were still definitely poor and lived in a marginal way. Al-
though all of the families said they served meat daily, it was found that
they lived at a crisis level of food expenditure, spending less than the
Department of Agriculture 1962 guidelines for emergency minimum ex-
pense. Although a large number of families had life insurance, almost none
had -health insurance or savings. Most of the girls came from homes that
were characterized by good housekeeping, although there were many
basically substandard conditions, such as shared kitchens and deteriorated
structures. The families, by and large, were struggling to do the best they

465

could. In all of these respects, the mothers and their backgrounds wet
essentially the same in both the experimental and control groups,

A survey was done piior to the program to determine the child -rear in
attitudes of these girls. This covered the entire group. Questions wei
asked to assess their attitudes about child tearing and to see what thi
understood of child development.

The first question dealt with the handling of aggression toward tl
parent. The girls weie asked what they would do if at dusk they had
biing a child in from play who then became angry and kicked and \
them. It was found that even mild aggression towards the parents is n
tolerated and is usually punished. About 60 percent of the mothers wov
respond punitively and about half were in favor of spanking in such a ca
Some favored explaining why and then spanking.

Another question dealt with the use of punishment for immature 1
havior during toilet training and was concerned with the acceptance
slow development toward maturity. For a 2-year-old who had a lapse
toilet training most would scold or spank, with scolding preferred. Oi
one girl would ignore such an incident, whereas most middle-class molh
would do so.

With regard to diet, most felt a rather active controlling role was
quiied rather than a passive one. Almost all of the girls felt that (1
would coax or make a child eat vegetables rather than simply offer
them and then ignoring it.

In the crucial area of verbal development, only 25 percent of the g
would encourage a child to talk when adults were sitting around visit
Many would permit an answer if the child were spoken to, but a tl
would piefer the child to be absolutely quiet. Here the girls with
higher I.Q.'s gave significantly different and more positive answers t
those with lower scores. This encouragement of speech and the acquisi
of early and basic verbal skills, as well as having pleasant associations >
speech, is extremely important in the development of comnuinicn
skills. The investigators noted here that initially the mothers were pie;
with the sounds of their first child and responded naturally with co
and other sounds. It was found that with help the mothers continued
healthy trend even though it became more difficult when their chil
became active physically and, in addition, began to bombard them
questions.

When asked what they would do if a child was misbehaving all day
- ** Question designed to tap the girl's perception of underlying cans

demonstrated love and devotion for their infants during the newborn
period. One question with regaid to what to do when a child was aftaid of
the dark elicited interesting answers. About 73 percent would leave a
night light on, stay with the child, or make some other sympathetic
gesture. Even girls who were generally harsh about discipline seemed lo be
understanding in this situation.

When asked as expectant mothers what they thought they could do in
the first five years of the child's life to help him do well in school, many
replied with answeis such as "teaching a child obedience," "teaching him
honesty," "name and addiess," "ABC's." Only about a third said "read to
him" or other such types of stimulation.

It must be remembered that both experimental and control groups took
part in this study, and that the two had essentially the same attitudes and
basically were the same in approach to their children.

Prenatal Nursing Activities

The public health nurse associated with the research visited the girls in
the experimental group at home and discussed with them many topics.
Among these were the issues of diet, their pregnancy, the development of
the baby and other things of importance to the girls. Visits were made
approximately every two weeks prior to the ninth month of gestation and
then each week during the final month. The nurse, Mrs. Marion Brooks, is
black as weie the girls, and she felt that this helped her to reach them and
be accepted by them. She felt that good lapport was developed with all of
the girls that she visited, and that most were eager to learn and com-
municate with her. Her general approach was to be quite definite about
methods of child rearing and very supportive of the mother's desire to do
well with her baby. Her visits prior to the birth of the baby lasted about
one-and-a-half hours each time.

Prior to their deliveries, the mothers were encouraged to prepare for the
birth and to plan their care for the child. They were helped to collect
bright, colorful pictures for the walls of the baby's room, and the nurse
stressed to them the importance of visual, auditory, and tactile stimula-
tion to the young baby.

Well Baby Care

After the birth of the baby, a comprehensive program of well-baby care
was begun which included the infant stimulation program. Well-baby care
was offered in a dramatic way by taking a mobile coach and a pediatric
team directly to the home. One visit was made each month for the first six
months and then less frequently until the child reached 3 years. Between
each visit the nurse visited to carry on the infant education program.
Although there were some technical problems with the use of the coach,
parking, etc., it was considered successful, as it demonstrated to the

467

mothers the concern that was felt that their babies receive regular i
tion. A rather large coach was used with two examining rooms £
toilet, and it was felt that perhaps a smaller coach would have been j
useful. The coach allowed the home examinations to go on in a p
sional setting and a clean environment and also assured that the staff
pleasant setting. The use of the coach has now been discontinued as
of the children have now reached 2 years of age.

Medical well-child supervision was planned and administered accc
to the guidelines of the American Academy of Pediatrics. The progr
infant stimulation was developed after an examination of the lite
and a study of Cald well's inventory of Home Stimulation.

After the coach was discontinued a doctor and nurse made honu
via a car. The visits totaled 22 during the three-year period. The ch
in the comparison group were followed in the regular District of Col
health clinics, and it was up to mothers to keep appointments and
that the child received well baby care.

The research families were counseled at each visit of the docU
nurse on feeding, hygiene, sleep habits and normal growth. Toilet trr
feeding problems, temper tantrums, thumb sucking, discipline, and
developmental problems were anticipated and received attention pi
and during the period of their occurrence.

The infants were provided iron and vitamin medication in proph
doses from the newborn period through the first three years of HI
mothers were given the home phone numbers of the pecliatricii
nurse, and the director of the project, and were urged to call for
whenever it was needed. The mothers called often in the early mo
their child's life and more infrequently later. When a child was
mother was directed to take him to Children's Hospital, althoi
project staff treated minor illnesses, such as colds, diarrhea, impetij

Group meetings for mothers were held occasionally on Sundn;
noons with the project psychiatrist, nurse, director, and occasiona
to discuss problems in child rearing. It was observed that thclntct
of ideas was reassuring to these young mothers and stimulate
interest in the development of their children. In addition to these s-
social events were arranged. One such outing was to a flower sh
another was a swimming party. In addition, there were parties held
staff member's home. Some of these events took place for mo the
and some included children. These were designed to provide c
evidence of the research team's interest in the mothers as well
babies.

Some of the mothers had a feeling of being singled out and the.
why they were selected while pregnant friends had not been in
join the program, They were told that it had to do with the timing
pregnancy, and that the research had to be with young womoi
seventh month.

468

Infant Stimulation Program

Aftei the babies were born, the mothers were visited by the public
health nurse and a carefully designed stimulation program was carried out.
It was felt that the very best teaching of the young infant has to be done
by the mother who presumably has the strongest relationship, and who
can achieve that blend of intimacy, affection, and stimulation that is the
key to a responsive baby who is actively able to learn. Theie is an assump-
tion here that these mothers are less able to do these things than are
middle-class mothers. There is also the assumption that these children will
be disadvantaged when in school with middle-class children if they are not
helped. There is concern on the part of the investigators that these
mothers, often overwhelmed as they are, will not see the importance of
the type of interaction that produces an open and curious child who is
able to learn successfully. The stimulation program was particularly de-
signed to help build the type of verbal-cognitive background that typically
occurs in the middle-class home. It is thought that particularly the acquisi-
tion of verbal skills depends upon the type of relationships fostered in the
early days of a child's life. Intelligence in this study is conceived of as
being made up of many abilities, many attributes, and built on many
expeiiences and supported by good physical health.

This program was designed to give the ghetto mother the maximum
suppoit and assistance. The public health nurse helped, but did not replace
nor displace the mother in her role with her infant. Every effort was made
to communicate the value that the research team placed upon her and her
infant. Efforts were made to reduce the feeling that only perfunctory care
would be available. Many of these girls had so diminished an expectation
of assistance that they were slightly suspicious of the program and the
quality of care that it offered. The survey on characteristics of the
mothers and their attitudes showed a basically positive approach to their
children, but one that reflects the pressures of their lives and also their
own lack of positive experience in many learning situations. Although
middle-class families do not always reflect the most sophisticated ideas
about learning and development, they tend at least to have a better idea
about these factors and to be more in a position to afford the type of toys
and to provide the type of environment where a child can at least explore
ways of learning. At this point in time the research personnel were content
to try and achieve a little more of the middle-class attitudes towards child
rearing.

Throughout the well-baby program and specifically in the stimulation
contacts, another message was given indirectly to each mother. It was
hoped that by reaching out to her and helping her care for her infant the
importance of her task and the value of it would become more obvious to
her and that her own early sense of being glad about becoming a mother,
and her desire to take good care of her child would be supported and to
some degree protected.

469

In both the well-baby and the stimulation piograms, as the n
visited, she observed the mother's behavioi with her child and alsoij
duced and discussed various issues. Records were kept via checklist
each visit in the first three years. If the mother was not home, the
taker, often the grandmother, was seen.

An emphasis was placed on all forms of visual, auditory, tactile
motor stimulation. The encouragement of exploration, the use of p
the understanding of normal developmental stages were all impoi
Avoidance of overstimulation was discussed and protection of the
from accidents without inhibiting his curiosity. An emphasis was p
on the alleviation of superstition and on assistance to the child in har
his own fears. Normal child development was discussed continuouslyH

An effort was made to see the mother on each visit, but since ma
these mothers either went back to school or were forced to go to '
often the baby was in the care of a relative. When the caretakers we
grandmothers, they weie, in general, less receptive to the prograir
were the mothers. They had already reared children and tended 1
back on ways they had found successful. Also, Mrs. Brooks fouri
some were quite superstitious, for instance, fearing to let a baby loo
a mirror as they believed it caused trouble with teething. It was esti
that on the average a mother was at home with her baby until 1
about 1 5 months of age. Four of the mothers remained with their
until they were 3 years old. In many cases a good day care center
have given the baby more stimulating care and a better opportm
have the type of experiences that would help him cope more succe
with school and have a better chance of breaking a cycle of poverty.

The First Year - Stimulation Program

Ten visits were made by the nurse. Each one lasted about 1 ho
nurse gave the baby toys appropriate to his age. A mobile, a rattle,
squeeze toy, and plastic cookies on a chain were given to the ncv
Later, at 3 months, a cradle gym, a terry cloth toy; and at 6 me
cloth book. At 7 months, a plastic hammer and a plastic milk botl
colorful objects inside were given. A stack toy was chosen for 9 i
and at 1 year, a more advanced cloth book. The nurse in each <
plained why the toy was appropriate and, if necessary, showed the
how to use it. The toys were provided rather than suggested as the
gators knew it was difficult for the mothers to afford extras, and

helped in obseiving how the baby begins to speak with throaty sounds and
then progresses. The mother would be encouraged to respond to the baby
and be told how important that was. Trips out of doors and to new
places were encouraged. The development of tactile sense, important all
throughout life, was explained. Cuddling, stroking, holding, and kissing
weie encouraged and the mothers were warned not to hamper a child's
movements by more furniture or clothing than what was necessary.

As the need for discipline grew, the mothers were helped to use con-
sistent and patient methods with an emphasis on praise and the under-
standing that a child's curiosity is valuable and not "bad" behavioi. A
frequent problem was that the mothers tended to confine or restrict their
babies rather than safeguard the environment. The mothers were discour-
aged from slapping and from continually repeating "no" to their babies.

The Second Year

About 8 hour-long visits were made and special emphasis was put on
the mother's talking with her child and encouraging him to speak. The use
of books and bedtime stories was stressed, The families were advised to
help the child get large muscle exercise and to let him explore, jump,
climb, and otherwise be active. Again, as in the well-baby visits, the
subject of accident prevention was raised.

The children were given more complicated toys, many of which stressc-d
fine motoi coordination, such as wooden beads to string or snap beads.
Balls were suggested as a good toy for helping with gross motor develop-
ment. The nurse also helped the mothers to develop homemade toys and
to see that some household items make good playthings.

Although the nurse encouraged water play and mud pies, etc., generally
the mothers were uncomfortable and felt such activities were too messy.
Generally, the mothers did not want their children to get dirty or to put
things the mothers considered dirty in their mouths. The nurse interpreted
the need for the child to be free from unnecessary concern about tidiness
in order to explore textures and to play in an imaginative and creative
way.

The Third Year

Here, there was an increasing emphasis on trips outside the home, on
visual motor coordination, and language development. The use of pencils
and crayons was encouraged. Discussions with the caretaker or mother
emphasized the need to satisfy the child's curiosity, to praise, and reward
him frequently, and to admire his achievements. The mothers were en-
couraged to answer the interminable questions asked and were helped to
see that questioning was important to the child and the basis for later
intellectual growth. The mothers were told to be honest with their chil-
dren and to admit it when they didn't know the answers. Consistency in
discipline was again stressed.

471

The nurse visited about six times during the third yeai and each visit
lasted about one-and-one-half hours. Dental hygiene was promoted as m
the overall well-baby progiam, and a toothbrush was given the child if he
didn't already have one. A small blackboard and chalk and eraser were
given at 28 months and a small hardback book at 32 months. A puzzle was
given as a farewell present at 3 years.

What Happened to the Mothers

Of the 47 mothers in the experimental group, 17 have married. Out of
these 17, five have already had separations occur. In the remaining intact
marriages, most of the husbands work at semiskilled jobs, such as barber-
ing, electrical work, etc. It must be remembered that many of the mothers
were still dependent upon their families' support and that those who
worked generally had low-level jobs. Six of the mothers are receiving
public assistance. In one case, the money goes to a grandmother who is
caring for the child. Of the experimental mothers, 1 1 finished high school
before delivery and 9 finished after the birth of their child. Twenty-one of
the mothers remain ungraduated. Four of these took some further training
after delivery and four are still in school. All were encouraged by the
research personnel to finish high school,

Nineteen of the 47 experimental mothers have had other children.
Fourteen of these had one more child and five had two more. All of these
mothers have been given access to birth-control information and devices as
part of their maternity care provided by the city. Of the mothers who
have had other children, 10 have obtained consistent well-baby care for
their additional children.

About one-fourth of the research mothers received both more than $50
a month and fairly frequent visits from the alleged father. Another one-
fourth had visits but not money, and the rest either sent small amounts of
money and didn't visit or else ignored the entire situation. Not nil of the
mothers who married wed the father of their first child.

Most of the mothers had enough help from friends or relatives to leave
the baby occasionally. A very few were terribly confined. The degree lo
which they left their babies seemed to depend upon the wishes of the
girls. Some were more socially active and felt more hindered by the baby
than others. Most were seen to be caring well for their babies and demon-
strated love and affection.

Observations of the mothers and babies were made by a psychologist
during the physical examinations at 6 months of age. The observer was un-
aware of the groupings. Results showed that the mothers who had partic-
ipated in the infant stimulation program were markedly more verbal with
their babies than the control mothers. This has not been completely tested
out, but appears to be the case as the children grow older.

Most of the mothers seen by the research team were positive about the
program and made contacts continually for information, in addition to
regular appointments. For the most part, they were receptive to the ideas

472

suggested and eager to learn. By and large, they were able to communicate
well with the research personnel.

Early Results - The Babies at 6 Months

Although all the babies started out well at the newborn stage and both
groups were comparable, there were differences in the Bayley test in
developmental I.Q.'s at 6 months, and at 1 year, and even more pro-
nounced differences at 2 years of age when the experimental children had
significantly higher scores than did the controls. In addition, there was
significantly less anemia in the experimental group. This was undoubtedly
due to the provision of iron and the attention given to the diet of the
children during the "out reach" well -baby care. The experimental chil-
dren, it was found, had also received more meat and had been taken out
of doors more often than the control children. Both practices had been
encouraged in well-baby checks and by the public health nurse in her
stimulation visits. Babies can develop anemia very easily if kept on a
basically milk diet since milk has little iron. Their bodies need this sub-
stance for growth and development, and unless supplemental iron is given
or large amounts of iron rich foods are added, anemia often develops.

At 6 months, there were only four children in the experimental group
with hemoglobin levels below 10.0 gms. per 100 ml. Levels below 10.0
gms. are considered anemic, and this is considered one index of general
health. There were 14 control children with levels below 10.0 gms. This
was a significant difference. In addition, the experimental babies had less
skin trouble-primarily diaper rash-than did the others. They had better
appetites and less thumb sucking, too. Although it might be predicted that
babies in the experimental group would have fewer illnesses, this did not
turn out to be the case and patterns in this respect did not differ. Growth
patterns were not significantly different in any way cither.

The children were also studied intensively physically, anthropomorphl-
cally, and in terms of medical history. Again, in most respects, they began
the same and continued to be about the same.

In terms of the mothers and their family situations, again many factors
were studied, such as crowding, the number of people in the homes,
incomes, illnesses, financial resources, diet, and education. The two groups
were basically the same, all were overburdened.

More Recent Results

The Stanford-Binet test is used when the children reach 3 years of age,
and to date a trend has emerged that gives the experimental children the
advantage. This is a statistically significant figure, but not all the children
have reached this age. It appears to offer some evidence of the value of the
research intervention.

473

Research at this Point

A number of questions are being examined now. A large number of
variables have been recorded and a statistical analysis is being conducted
with consultation from Arthur Kirsch, Ph.D., who is an Associate Profes-
sor of Statistics at George Washington University. To date most variables
explored have either been unproductive or both groups have remained
very similar, except in terms of the health and development of the infants.
A close examination of success in school will be made as these children
begin school. No attempt is being made to modify their school experience
in any way. The relationship between the mothers and their children will
be observed and to a certain extent the relationship of the mothers to
subsequent children. It will be important, if positive results are achieved,
to find what aspect of the program can be held responsible and to know
which aspects really work.

One major problem in this area is the subtlety of the results that arc
being measured. All of these children, both control and experimental,,
have received well-baby care, in fact, the D. C. health department R.N.'s
who do this also have a certain amount of infant stimulation education
that they include in their routine contacts with new mothers. The differ-
ence then is one of degree. All the mothers live in crowded homes, all have
low incomes, and so on. This similarity in samples, combined with the
difficulty in finding sensitive evaluative instruments, means that differ-
ences may not be easily seen or documented, even if they exist. There is
considerable conviction on the part of the staff that differences exist.
Some differences, they feel, are related to the relationship between the
mothers and the staff and the enhancement of the mother's self esteem
due to her participation in the program. There is a growing awareness Unit
the large social problems that weigh down these families need large mid
powerful answers to go along with the improvement of infant cure in
order for differences to be seen. One idea behind working with the
mothers, rather than the staff working directly with the child, is to give
benefit to subsequent children as well as the child in this research pro-
gram. Here, again, tremendous pressures make themselves felt. Mothers
who participated well and followed directions, who seemed to understand
the value of well-baby care and took advantage of it when offered by the
mobile coach team still do not always take subsequent children to well-
baby clinics in a consistent manner. There is also some concern Unit
despite pressing social problems and the availability of birth-control infor
mation, these young women seem to be on their way to targe families
However, it is possible that many want two or three children rclalivcl;
close together and that having achieved this, they will then exercise ai
option to limit their families. It is perhaps too early to tell if the Idea o
birth control has been successfully communicated.

474

Conclusions

• Early results indicate that young children can be helped to an ade-
quate level of verbal ability through work done with their mothers.

« Relatively uneducated mothers can be assisted to provide a stimu-
lating environment for their children even in overcrowded ghetto homes.

• It is important to work with the mother who is a key figure for the
baby, and she may be able to use the same training with subsequent
children,

• Well-baby care will be better utilized if taken directly to the home.

• Infant anemia can be wiped out.

• There is a dire lack of preschool facilities.

• There is a desperate need for good day-care centers for infants.

• A nurse who has been given only a small amount of special orienta-
tion can be effective in an infant stimulation program.

• Infant stimulation programs should be incorporated into well-baby
care.

• Infant stimulation programs and good well-baby care alone may not
alter the future of a baby born to a young, poor, black mother living in a
poverty area.

This program is in a sense a minimal one -relatively inexpensive, using
limited personnel, and the extensive use of a nurse to provide services. It is
not a radical departure from the care now generally believed to be desir-
able. The teaching of the mother makes possible the continuation of
the spirit of the program and hopefully will benefit the entire family. It
appears that marked changes can be achieved in children through this type
of stimulation, and that this is a fruitful approach.

Research Grant: MH 09215

Dates of Interviews; April and September 1970

References:

Brooks, Marion. A stimulation program for young children performed by a Public
Health nurse as part of well baby care. Paper presented at the American Nursing
Assocition meeting, 1970, and accepted for publication in the American Journal of
Nursing.

Caldwell, Bettye. Descriptive evaluations of child development and of developmental
settings. Pediatrics, Vol. 40, July 1967, No. 1 , 46-54.

Gutelius, Margaret, M. Child rearing attitudes of teen-age Negro girls, American Journal
of Public Health, Vol. 60, No. 1, January 1970, New York, 93-104,

Levenstein, Phyllis, Cognitive growth in preschoolers through stimulation of verbal
interaction with mothers. Paper presented at the 46th annual meeting of the
American Orthopsychiatric Association, New York, New York, April 1969.

The me nobody knows, Children's voices from the ghetto. Edited by Stephen M.
Joseph, Discus Book, Avon Books, 1969, New York, p. 36.

475

Indeed, what is there that does not appear marvelous when ft comes to
our attention for the first time? How many things, too, are looked upon
as quite impossible untff they have been effected?

-Pliny the Eider

Investigator

Jerome S. Bruner, Ph.D.

Director, The Center for Cognitive Studies

Harvard University

Cambridge, Massachusetts

Prepared by:

Maya Pines

For the past four years, Dr. Jerome S. Bruner has been concerned
primarily with the cognitive development of babies: How does the human
infant - so helpless and limited at birth — learn to control his environment
and himself? How does he grow up intelligent?

Although human babies at first appear more stupid than chimpanzees
of the same age, by the age of 2 or 3 the normal child has achieved one of
the most difficult intellectual feats he may ever perform: he has re-
invented the rules of grammar, all by himself, and he has learned to speak.
He has also constructed a fairly complex mental model of the world,
which allows him to manipulate various aspects of the world in his
thoughts and fantasies. And he has learned to mobilize various skill pat-
terns whenever he needs them. Under the guidance of Dr. Burner, the
Center for Cognitive Studies at Harvard is trying to unravel the sources of
these formidable achievements.

There are various theories about how children acquire language. In his
book, Verbal Behavior (1967), Dr. B. F, Skinner claimed that children
learn to speak as a result of stimulus-response conditioning. All compli-
cated behavior is learned, he argued; one learns to behave in ordered ways
because there is order in the environment. Linguists such as Dr. Noam
Chomsky disagreed, suggesting that human beings have an innate compe-
tence for language which sets them apart from all other creatures. This
innate competence is what allows babies to learn language on the basis of
relatively few encounters with words and sentences, they claimed, credit-
ing babies - and mankind — with far more "mind" than most scientists
were willing to accept at the time.

Innate Competence

Dr. Bruner and a number of other psychologists in the U.S. and abroad
have now gone one step further than Chomsky in emphasizing the im-
portance and activity of the infant's own nervous system. By studying

479

babies well before they learn to speak, these researchers have come to the
conclusion that language competence is just one example of an even more
significant ability with which infants enter the world: the basic ability to
pick up logical rules from mere fragments of evidence, and then use these
rules in a variety of combinations. There are programs of action in the
human mind right after birth, they believe, not only for language but also
for the intelligent use of hands, eyes and tools.

"It's a very different view of man," says Dr. Brunei, "and it's just
beginning. People are staiting to see that skids of this wide-ranging type
couldn't possibly be learned element by element. There must be some
kind of predisposition in man to allow babies to pick up so quickly rule?
that go for such a large number of situations.

Only a few years ago, it was generally believed that newborns could no
see more than the differences between light and dark; that during thei
first three months of life they were so absorbed by their insides that the
could hardly react to the outside world; and consequently, that the
physical environment had little impact on them - all that mattered was ti
provision of food and comfort. It now turns out, however, that even c
the day of their birth, infants can track a triangle with their eyes. By 0
time they are 1 month old, they can spot the identity of objects and knc
when something has been changed. Furthermore, they actively invc
rules of theories to explain what they perceive. Even at 3 weeks of age,
infant will have faiify complex hypotheses about the world he has ji
been born into - and if he is proved wrong, he may burst into tears.

That was exactly what happened at the Center two years ago, during
experiment conducted by a Radcliffe undergraduate, Shelley Ros
bloom. Researchers there wondered whether babies of 3 to 8 weeks ret
understood that a person's voice should come from the spot where
person stood - whether babies had the idea of a locus. If so, did 11
babies also grasp the fact that when the person moved, the sound of
voice should travel along the same path? To what extent had they orj
ized their experience at that age? To find out, it was decided to use sl<
speakers that could separate the sound of a voice from its origin
response to ads in the Harvard Crimson, there is always a proccssio
babies ~ mostly the offspring of graduate students - to the Center's
where they are made comfortable, given toys, and usually offered st
thine interesting to see or do as part of a psychological experiment. \

increasingly differentiated. But there is some notion of it right from the
start."

The Study of Infancy

The Center's study of infancy has focused on five issues:

1) How the infant achieves voluntary control of behavior in a fashion
governed by prediction and anticipation;

2) How visually guided and intelligent manipulative behavior emerges,
with emphasis on the transfer ability and generativeness of skills;

3) How the infant progresses from being a "one-track" enterprise to
being able to carry out several activities simultaneously and under the
control of an over-rule;

4) How attention develops and its control shifts from external con-
straints (novelty) to internal constraints (problem-solving); and

5) How pre-linguistic codes develop, particularly in the interaction of
the mother and infant,

O'Our underlying assumption is that the codes of language, while they
may indeed reflect innated patterns, are first primed by a great deal of
interactive code-learning of a nonsyntactic type,'* writes Dr. Bruner.
"When certain of these pre-linguistic manifestations are understood, per-
haps light will be shed on the deeper question of the nature of language as
such.")

In babies' hands, Dr. Bruner believes, lie clues to much of their later
development, and he particularly wants to find out how babies learn the
value of two-handed ness. Nobody, teaches infants this skill, just as no-
body teaches them to talk. Yet around the age of 1, a baby will master the
"two-handed obstacle box," a simple puzzle devised by the Center to
study this process. Seated on his mother's lap, he will suddenly use one
hand to push and hold a transparent cover, while the other hand reaches
inside the box for a toy.

To Dr. Bruner this is extraordinary, for it shows that the baby has
learned to distinguish between the two kinds of grip - the power or "hold-
ing" grip, which stabilizes an object, and the precision or "operating'* grip,
which does the work. Monkeys and apes have developed a precision grip,
Dr. Bruner says, but "it is not until one comes to man with his asymmetry
that the power grip migrates to one hand (usually the left) and the preci-
sion to the other.'* From then on, he emphasizes, many routines can be
devised for holding an object with one hand while working it with the
other, leading to the distinctively human use of tools and tool-making.

The experiments at the Center are essentially very simple, but their
interpretations are not. Some of these interpretations parallel Noam
Chomsky's "transformation" approach to linguistics, which reduces lan-
guage to basic kernel sentences, each one made up of a noun phrase and a
verb phrase. Early in childhood every human being learns the logical rules
which allow him to transform these kernels into any possible sentence. Dr.
Bruner speculates that when a baby learns to differentiate between the

481

two kinds of manual grip, this foreshadows "the development of to
comment in human language" - the basic sentence form of ;
predicate, which may be found in all laguages, with no exception
ever, and which a baby expresses when he combines a holophrase I
word or a very short phrase that is used as one word) with anotVu
Thus, man may be uniquely predisposed, at birth, to reinvent the
grammar, to process information, and to develop "clever hands
born with a highly complex programing system, the result of mi1
years of evolution.

What about disadvantaged children, then — why should they t
ent, if they are born with the same programing system? "Mind )
can ruin a child's inheritance, too," warns Dr. Bruner, "with an
ment where he acquires helplessness. You can also be traine
stupid."

Before man's marvellous programing system can be activated
guage, for instance, a baby must learn a series of primitive cot
these require interaction with an adult. "What seems to get es
very quickly between infant and parent is some sort of code o
expectancy," says Dr. Bruner, "when the adult responds to an
on the part of the child, thus converting some feature of tl
spontaneous behavior into a signal." Right from the start, pi
infant are busy communicating through eye-to-eye contact, si
sounds. As early as 4 months of age, an infant will smile more
that smiles back than to one that does not respond; and if the :
then stops smiling back, the infant will look away. In some case
even struggle bodily to look away. A child's other attempts at lej
similarly be brought to a halt when his expectancy is thwarted, i
stop making sense.

The Development of Strategies

The prolonged infancy of man has definite functional impor
Bruner concludes: During that time, the infant is basically deve
strategies that will later be combined for intelligent action - f
and language, as well as for the manipulation of tools.

One such strategy is "place-holding." The earliest evidence i
be seen in infants' sucking behavior. As everybody knows, a p
calm a baby. But why? Earlier research had shown that sucki
hunger pangs and relieves muscle tension. "Well, but putting ele
the temples of babies as they were watching a movie here, we'\
find out what a pacifier really does," says Dr. Bruner. "One c
ipal effects is to cut down scanning eye movements, which cut
baby's information intake." At birth, and for a few days therea
can't cope with more than one activity at a time. When they w
they close their eyes tightly, to avoid taking in informatio
outside. When their eyes are open, they stop sucking. By the a
weeks, however, they can suck with their eyes open - but as s

482

become really interested in something, the sucking stops. Finally, between
the ages of 2 and 4 months, a new strategy appears. Whenever something
catches their attention while they are sucking, they stop their usual suc-
tioning and shift to a sort of mouthing which keeps the nipple active,
though at a reduced rate. This allows them to pick up where they left off
with great ease, once their curiosity has been satisfied- A neat solution to
an early problem, "place-holding" of this sort leads to many later skills,
both manual and linguistic.

"As I got more into this work on skilled behavior, it became increas-
ingly evident to what extent intention and hypothesis are central to the
organization of knowledge and to the filtering of input," declares Dr.
Bruner. In his most recent study, "Studies in the Growth of Manual
Intelligence in Infancy," which he did with Karlen Lyons and Kenneth
Kaye, Dr. Bruner emphasizes the importance of the infants' own programs
of action. "When one observes the early behavior of infants — say at the
onset of visually guided reaching at around 4 months of age — one is struck
by the fact that arousal of intention is the initial reaction to an 'appropri-
ate' stimulus," he writes, "The earliest overt expression of activated inten-
sion is not 'trial and error,' but an awkward but recognizable instrumental
act that expresses a preadapted progiam of action."

His movie, "The Intention to Take - The Infancy of Object Capture,"
illustrates how babies begin with an intention, act out its intended results
(or an approximation thereof) and then work backward to the compo-
nents that will in fact make such results possible. "First, they look at the
object," says Dr. Bruner, describing the movie. "They want to take it. It's
an intense gazing. Then, as the child's intention gets organized, his lips
come forward in what we call an 'A-frame mouth.' Later on, when he
takes hold of the object, it will go into his mouth; but already, his whole
system is activated, his mouth works. Then his arms come up in an anti-
gravitational movement, and up comes that fist." The infants1 actions are
not yet in the right ordei for success. "Six weeks later, these actions will
seem so well regulated that we'll forget the complexity of even so simple a
task. Then they will leap forward to a fully orchestrated act. But the
preparation is slow and demanding."

The infants' own intentions, then, are crucial. Of course, some goals can
be imposed from outside, and babies can be taught, for instance, to re-
spond to a buzzer in certain ways. Thus, Dr. Hanus Papousek, a Czech
psychologist who is now spending a year at the Center, has conditioned
newborns to turn their heads sharply to the side at the sound of a buzzer,
in order to get milk from a buzzer. "It can be done," says Dr, Bruner,
"but it's endless. The babies show so much aversion to this. They're so
slow at learning it, you have to present the stimulus hundreds of times,"

By contrast, when the infant uses his own initiative, learning often
comes with lighning speed. In the Center's lab, a medium-sized room
which might be called a baby theater, babies are placed in a well-padded
seat facing a blank wall which serves as a screen. Then, with a pacifier in
their mouth, they are shown a movie. "We didn't want to condition them
to respond to a stimulus," Dr. Bruner explains. "Instead, we wanted to

483

choose something the child does and give it some consequence. Then he is
at the controls. So we chose sucking. Would they learn to suck at different
speeds in order to produce changes in their environment'? And, lo and
behold, these little 4-, 5- and 6-week-old infants do learn to suck in longer
bursts to produce a clear focus. Or else, if you reverse the conditions so
that sucking blurs the pictuie, they learn to desist from sucking on this
pacifier. They respond immediately, during the very first session, to
changes produced by their own acts."

The movie that the babies watched so eagerly showed an Eskimo
mother playing with her child. "It was shot in winter, indoors, and she
was constantly involved in little games with him — string games and so on,1
explains Dr. Bruner. The experiment was devised by a graduate student
Mrs. Ilze Kalnins. When the babies discovered that sucking made th
pictures clearer, they cut down their pauses between sucks, stopping onl
four seconds. On other visits to the lab, when they found that suckin
blurred the image, they lengthened the pauses to about eight seconds.

The babies' performance was all the more remarkable because of the
inexperience with "place-holding." To bring the picture into focus, tlu
had to suck in longer bursts without looking at the film, then take a quw
look before it blurred again.

Curiously, this experiment comes quite close to the kind of opera
conditioning pioneered by Dr. Skinner, in which rewards ate used
"shape" a child's activity. But Dr, Bruner interprets it quite different
seeing the babies' rapid learning as the effect of fulfilling their o
intentions. Sucking to produce a sharp focus involves quite camp
strategies to coordinate looking and sucking. Such strategies come IV
the inside out, from an innate preadaptation, Dr. Bruner believes. O
after their appearance has been evoked by events can trial and error i
reinforcements be of any use,

"What reinforcement is doing, in effect, is locking in that response
set of alternative responses which in fact works/' he writes. "It docs
bring into being new responses. For the most part, the children do
gradually improve their strategies, but rather increase the skill with w1
they perform old routines. , . . Two-handed efforts make their appears
abruptly, rather than by some gradual route, and seem to be 'ready
triggering.11

He points to an experiment performed at the Center two years ago
three groups of babies of different ages. The babies were seated in fro
a table on which a jingly toy was placed behind a small transparent so
open at one end, The youngest babies, only 7 months old, simply veil
for the toy with the nearest hand and bumped into the screen.
banging on and clawing at the screen for a while, they lost intcrcs
gave up. The next group, the 1-year-olds, began in the same fashioi
then let their hands follow the edge of the screen and reached bchint
a sort of backhand grasp until they got the toy. Only the 18-nion1
babies knew right away how to reach the toy efficiently, and did so
1 6 trials each, none of the babies ever changed his initial strategy ; th
the best he was capable of at that stage.

484

"Trial and error implies the capacity to hold an end constant while
varying means," notes Dr. Bruner. "The segments in which this is possible
are very short in duration for the child. What thwarts him is distraction,
not error." This is why he is so interested in the development of the
child's own intentions, and in the kind of "planning control" described by
the Russian psychologist, Dr. A. R. Luria, as being located in the frontal
lobes. He hopes to study the development of strategies and plans in pri-
mates and compare it to that of human infants, so as to gain further
insight into this issue.

Studies of Perception and Thinking

The Center for Cognitive Studies came into being in 1960. During its
first years of operation, it paid no attention to babies. For Dr. Brunner,
too, infant development is a comparatively recent interest. Unlike many
other psychologists, who study the same topic for their entire working
lives, he has ranged all over the field. And before calling attention to the
cognitive giowth of infants, he had helped to create interest in four major
movements: 1 ) the so-called "New Look" in perception in the late forties
and early fifties; 2) the study of cognitive processes, mostly in adults; 3)
educational reform, with emphasis on new curricula; and 4) the study of
children's cognitive development, Throughout, he always came back to
the same basic questions: How do human beings gather, categorize, store,
use and communicate knowledge?

"You can never get a direct test on reality," he says. "You must take
scraps and test them against your mental model of the world." In his work
on perception, he wanted to learn how people register information
through the filter of their own experience. He concluded that the same
objects — for example, coins — are perceived differently by different people,
in accordance with their values and needs. "Perceptions are highly regu-
lated entry ports," he notes. "An experienced eye will pick up so much
more!1' In contrast to work that considered perception to be strictly
passive, this approach was called "hot" perception, or "the New Look" in
perception- It led him to the boundary line between perception and
thinking.

Together with other members of the Harvard Cognition Project, he then
spent five years studying cognitive processes — "the means whereby organ-
isms achieve, retain and transform information." At the time, this was a
major departure from the accepted approach to psychological problems,
behaviorism. For roughly 30 years, most positions of prestige in American
psychology had gone to people who studied stimuli and responses, by-
passing anything that smacked of the "mental."

Spurred on by work in computer simulation and information theory, a
few psychologists were beginning to worry about the rnind again. Some-
times they called it "the black box." Clearly, the black box had to sort
out all the inputs and outputs; but how did it do it? The behaviorists did
not even attempt to answer this question, which they considered irrele-
vant. The members of the Harvard Cognition Project did, as described in

485

Dr. Bruner *s book, A Study of Thinking (1956). Specifically, they tried to
deal with what Dr. Bruner called "one of the simplest and most ubiquit-
ous phenomena of cognition: categorizing or conceptualizing. On closer
inspection, it is not so simple. The spirit of the inquiry is descriptive. We
have not sought 'explanation' in terms of learning theory, infonviiiliov
theory, or personality theory. We have sought to describe and in a smwl
measuie to explain what happens when an intelligent human being seek
to sort the environment into significant classes of events so that He mtv
end by treating discriminably different things as equivalents."

"There were some strategy theories I had picked up from John vo
Neumann," recalls Dr. Bruner. "I wanted to show how, in problcn
solving, as in perception, people use strategy for choosing the instanc
they want to think about. I was arguing that strategy and syslemat
search efforts are characteristic of all living systems - that there arc sin
lures and hypotheses in the mind, and that you're constantly testing ttu
against fragmentary evidence from the environment. You're locked
the most tragic — you're locked into the structures that are spccics-spuci
to you, because that's the way the human nervous system is. But over u
beyond that, there is a way in which, through the exercise of iniUalive
your part, you can turn on your own infoimation, reorder it, ami puna
hypotheses. The structures in men's minds are productive, generative,.1
as grammar makes it possible for men to emit any numbei of ultcraiiu<

The Impact of Piaget

The emphasis on strategy in Burner's work on peiception and Hio
caught the interest of the famous Swiss psychologist, Dr. Jean Piagel
was the last tiling he expected from an American psychologist,"
Bruner notes. "I guess I'm not a very typical American psychologist
least my colleagues don't think so. I think I'm right in the IradiUnr
started with William James, of pragmatism, and that they're very imi
the tradition of Ivan Pavlov! You know - 'we don't have to look insit
organism, there's no structure at all, all the order is outside > and till y>
is mirror it.* Well, I take a drastically different view." Drs, Bruiu
Piaget first met 16 years ago, when Dr. Piaget came to Boston lo
lecture. And Dr. Bruner was among the first Americans to nppreuta
importance of Dr, Piaget's work,

Piaget's monumental studies of child development had been i*
in the U.S. for several decades, until the cognitive movement awake
their value. In bold strokes, as well as painstaking detail, PmgcL I
scribed the growth of human intelligence, from the first day oT IH
adulthood. He had shown how children construct their own
models of the world in successive stages, following an invariant se<
though they may go through the stages at different rates. When u cl
experienced enough conflict between reality and his imago ol
changes this image to make it more accurate. Thus, at first a chikl
understand that when water is poured out of a full glass into ;i wit

486

which it fills only half way, the amount of water is unchanged, Being
"centered" on only one aspect of reality at a time, he sees that the glass is
half empty and says there is "less" water than before. Through a series of
experiments, Piaget explored how children develop what he calls "conser-
vation," the understanding that a quantity of water or clay will remain the
same, regardless of the shape it takes. As children realize that objects and
people have properties that do not depend on their immediate appearance,
they become able to deal with symbols. Intelligence consists of such leaps
into abstraction - but it depends on a large repertoire of images with
which one can visualize certain sequences of cause and effect.

Dr. Brunei- devotes considerable space to the contribxUions of "the
Geneva school" in his book, The Process of Education. Many of his own
papers show a strong Piagetian influence, particularly those in which he
discusses the stages in cognitive growth. But eventually he developed dif-
ferences of opinion with Piaget about how children acquire the notions of
conservation and - much more fundamentally — about what produces
intellectual growth.

''Mostly we argue about prefixedness," he says. "I found increasingly
with Piaget that his notions of interior order were much more prefigures,
prefixed than mine. I think this was the thing that caused something of an
intellectual rift between us, I think that he misunderstands me more than
I misunderstand him. He is too concerned with how the mind just proc-
esses things. I told him once, only half-jokingly, that his study of mollusks
(conducted when he was only 15) was characteristic of him. His idea was
that there was a mollusk, and no matter what that mollusk ate or what
that mollusk did, it always turned out to have the same prefigured shell.
Piaget's notion of intellectual development is a bit too much like his early
conception of the way in which a mollusk grows. As one of his colleagues
pointed out when he was here a few weeks ago: What does Piaget need a
theory of education for? Either the child hasn't reached the right stage,
and there's no point in trying to teach him anything; or he has already
reached that stage, and why bother to teach, as he'll learn anyway."

In Dr. Burner's view, evolution has given man a wide range of possibil-
ities — far wider than Piaget would allow — because man is a cultuer user,
and his growth depends largely on the kinds of tools he uses. "I don't
believe you can or should separate anthropology from psychology," he
declares.

The Center for Cognitive Studies

By I960, a number of converging trends made the study of cognition
seem particularly promising. Some central place was needed to stimulate
interdisciplinary research on the subject. With grants from the Carnegie
Corporation of New York and other foundations, Dr, Brunei then
founded the Center for Cognitive Studies, together with Dr. George A.
Miller, a psychologist who was known for his work on psycholinguistics —
the study of how cognition and language interact.

487

At first the Centei focused on four areas: psycholinguistics, h'
memory, perception, and the cognitive growth of children, Amoi
many research fellows and visitors could be found psychologists, p
ophers, physicians, linguists, anthropologists, sociologists and cyb
icists. Dr. Burner was most involved in research on the cognitive gro\
children, particularly those between the ages of 9 and 13. A Mobile J
atoiy helped him and his associates to do experiments on the develo
of perception, attention, and judgment in children under controlled
lions, right next to the children's schools.

In the meantime, he had become famous in another field - edui
This helped to make it respectable for psychologists to be concerne
the subject. His involvement began when he served as chairman oi
fercrtce of scientists, scholars, and educators at Woods Hole, Cape C
better ways to teach science. His resulting report, The Process of
tion (I960), was the clearest work on curriculum reform at the tir
won him instant fame. It has since been translated into 22 language
still being studied by teachers all over the world, particularly the
statement - which has been quoted over and over again - that "any
can be taught effectively in some intellectually honest form to ar
at any stage of development,"

Although many of Dr. Bruner's ideas have changed since then, h
by this famous statement, declaring that there is "absolutely no t
against it." Another dominant theme persists: physics (or math. It
or any other subject) is not something that one "knows aboi
something one "knows how to do." It is a way of thinking, rath*
series of facts. Thus, when Dr. Bruner devised a social studies cu
for the fifth grade, "Man: A Course of Study," he gave 10-year
raw materials with which to act like social scientists and three b?
tions to start them off: What is human about human beings'?
U\ev set that way? And how can they be made more so? The
include films on the life cycle of the salmon, on free-ranging babe
on the Netsilik Eskimos, the purest surviving example of traditu
mo culture - the kind of authentic records previously avanabl
college or graduate students. The course has now been adopted
than 1,500 schools. "Intellectual activity is the same wheth
frontiers of knowledge or in a third-grade classroom, Dr. Brune
Wto£l*tfBdu£tton. He still believes it passionately. And
involved in the creation of new curricula - right now, a new
adolescents on principles of child development.

At the college level, he proposes a dual curriculurn to take ad
young people's drive to control their environment: On Monday
days and Fridays, students would continue with the essential bs
sue'h » mathematics or language, in which one step must be ta
another; and on Tuesdays and Thursdays, they would be 1,
gove heir own learning in ways as experimental as possible.
httUide taking part in budget decisions, teacher evakiation .
TJ more than that, it would mean that they coulc

488

own problems to study. Preferably, these should be problems for which
no answers yet exist.

Students are usually exposed to only two types of problems, Dr. Bruner
points out those which require analytical thought — e.g., dealing with
abstract formulas - and those which require them to do some kind of
laboratory exercise. "Both are formulated by the instructor or the text or
the manual, and both are important in any science, art, or practical
sphere," he says. "But neither is much like problem-finding. This requires
the location of incompleteness, anomaly, trouble, inequity contradic-
tion "

The Growth Sciences

In the mid-sixties, as his studies on children's cognitive growth pro-
gressed, Dr. Bruner became increasingly dissatisfied with the age group he
had been working with. "We were left with a sharp sense of incomplete-
ness concerning the origins of what we had studied ,'* he noted. He saw
that by the age of 3, a repertory of skills is already well developed.
Therefore, he began studying younger and younger children.

By 1967, when Dr. George Miller left the Center, the transformation
was complete: nearly all the Center's research dealt with the cognitive
development of babies, including infants only a few weeks or a few
months old. Traditionally, this age period had been neglected because the
child seemed so inaccessible between his fifth day of life, when he left the
maternity hospital, and his entry into nursery school at 3, Dr. Bruner
urged his students to adopt the viewpoint of a naturalist exploring a new
species, rather than try to test specific hypotheses derived from a general
theory of infant development. "Assume that you are studying the great-
chested Jabberwocky," he advised.

In this way, he took a lead in the development of what he calls "the
growth sciences," a new composite discipline concentrating on the early
years of life. "Just as medical research was organized around concepts of
pathology, so today we would do well to organize our efforts anew
around the concept of growth," he declared. "Those sciences that can
help us understand and nurture human growth - biological, behavioral
and social sciences alike - should find ways of joining forces as the
growth sciences. Let them then make their knowledge relevant to those
who are practitioners of the nurturing of growth: parents, teachers,
counselors. It is bizzare that no such organization has yet emerged, though
it is plainly on its way."

Research Grant; MH 1324

Dates of Interview: September and October 1970

References:

Bruner, J. S.,Goodnow, J. J., and Austin, G. A., A Study of Thinking. Wiley, 1956.
Bruner, J. S., Going beyond the information given, in, H. Gruber et al, (Eds.) Con-
temporary Approaches to Cognition. Harvard University Press, 1957,

489

Brimer, J S., The Process of Education. Harvard University Press, 1960. (Paperback

edition, Random House Vintage Edition, 1963)

Burner, J. S., The course of cognitive growth. American Psychologist, 1964, 19, 1-15.
Brunei, J. S , Oliver, R. R., Greenfield, P. M., et al, Studies in Cognitive Growth, John

1 Wiley & Sons, 1966.
Bruner, J. S., Toward a Theory of Instruction. Harvaid University Press, 1966.

(Paperback edition published in 1968 by W. W. Norton.)
Bruner, J. S., Processes of Cognitive Growth' Infancy (Vol. Ill Heinz Werner Lecture

Series). Clark University Press with Barre Publishers, 1968.
Bruner, J. S. and Bruner, B. M., On voluntary action and its hierarchical structuio.in

A. Koestler and J R. Smythies (Eds), The Albach Symposium 1968, Beyond

Reductionism, New Perspectives in the life sciences The Hutchinson Publishing

Group, Ltd., London 1968
Bruner, J S. The growth and structure of skill, in K. J. Connolly (Ed.), Mo tor Skills in

Infancy, Ciba Conference, November 1968.

Bruner, J S., Lyons, K., The Growth of Human Manual Intelligence: I. Taking posses-
sion of objects, in preparation.
Bruner, J. S., Lyons, K. & Watkins, D , The Growth of Human Manual Intelligence: II.

Acquisition of complementary two-handedness, in preparation.
Bruner, J S., Kaye, K., and Lyons, K., The Growth of Human Manual Intelligence: 111.

The development of detour reaching, in preparation.
Bruner, J. S., The Relevance of Education, W. W, Norton, 1971, in press.

490

Investigator:

Roger Brown, Ph. D.

Harvard University

Cambridge, Mass.

Prepared by

Herbert Yahraes

A number of ingenious tests for showing how much a child knows
about language have been developed by an NIMH grantee and his associ-
ates and are being used by several investigators here and abroad in work
with aphasics, childien with retarded speech and speech disorders, and
children who are deaf.

The grantee is Dr. Roger Brown, professor of social psychology at
Harvard University and a leading authority on psycholinguistics, or, as Dr.
Brown piefers to call it, the psychology of language.

This investigator himself has collaborated in using the tests to help
determine how much language-processing ability has been retained by
persons with brain injuries. But his goal is quite different: to learn how
normal children acquire grammar. This is important, he points out,
because speech is the most characteristic human performance, yet little is
known of the mental processes involved in its development.

A child gets his first knowledge of language from the persons close to
him, and he tries to imitate what he hears. But by the time he puts words
together^ he has begun, apparently, to induce a sets of mles for their use.
Before the average child is 3, he has somehow developed a basic grammar.

The child's first grammar is not an adult's. It leads to mistakes like/
throwed the ball and to peculiarities of word usage like Kitty (til gone, and
it is far from complete. However, it does enable a child to put together a
very great variety of sentences, most of which he has never heard, and
thus to express himself on his own.

Dr. Brown's group is attempting to discover sets of rules that will
produce, in the sense that a program for a computer will produce, the
sentences that a young child produces. The aim is to get behind the words
and the sentences to the machinery that turns them out. Since the ap-
propriate rules would lead to the same product as the child's brain, they
might constitute a model of the kind of operations going on in his brain.

Ultimately, therefore, this research may add to knowledge of how the
brain works. By elucidating the natural language-learning process and by
demonstrating how many children know about language by the time they
go to school, the research may also lead to advances in the teaching both
of foreign language and of reading.

The investigators use two general techniques: application of grammar
tests, particularly in the case of children older than 3; and intensive study
of language usage by very young children.

491

Ill some of the tests for grammar comprehension, children are given
nonsense words and encouraged to use them in speech, with the investi-
gators noting what rules are applied in forming inflections, like the plurals
of nouns and the tenses of verbs.

In other tests, children are shown pictures and asked to distinguish
between, for example:

The dogs dig and The dog digs.

The cup is falling and The cup will fall,

The round dish on the table and The dish on the round table.

The boy is pushed by the girl and The girl is pushed by the boy. (Chil-
dren under 4, Dr. Brown reports, almost invariably go wrong on this one
because, evidently, their grammar contains no passive voice.)

A test may be a simple direction, such as Put the penny in the glass,
when the glass is standing upside down so that putting the penny on it
would be simpler.

Dr. Brown and his cowoikers have found they can use such tests down
to the age of 3 to learn at what point in a child's development the brain
processes small grammatical differences of various kinds. Below the age of
3 they have to depend mainly on obseivation.

In studying a very young child, the investigators visit him frequently
and record his and his mother's speech over a several-hour period of
normal activity, The transcription of the recording is then analyzed anil
rules drawn up that seem to match those the child himself has followed in
putting together two or more words.

The technique used for this procedure is distributional analysis, in
which the words spoken are grouped into syntactical classes (on the basis
of where and how frequently they have been used in the sentences and
what contexts they have shared with other words). The rules take the
form of a computer-like program for generating sentences by selecting
words from the groups in a particular order.

The investigators test this program by using it to tiy to duplicate what
the child has already said and by noting how well it predicts the child's
new utterances, during later visits, using the same vocabulary. Preliminary
work with such programs indicates they do indeed represent the grammar
used by a given child at a given stage of development.

The first children studied intensively have been a girl, Anna, who was
18 months old when this phase of the work began (in October 1962) and
a boy, Dale, who was then 24 months old. They have been visited every
other week.

Anna and Dale are the children of middle-class, well-educated people.
The next subjects for intensive study will be a boy and a girl from families
considerably lower on the socio-economic scale. Dr. Brown reports tlml
such children lag behind others in every index of speech development
because, he thinks, there has been less interplay between mother and
child.

492

In addition to collecting more information for a language-learning
model, Dr. Brown wants to test this hypothesis about lower-class families
and If it is substantiated, try remedial measures.

# # :ii # *

From the many sentences that a child hears, he somehow extracts the
latent structure of the language and for the rest of his life operates in
accordance with that structure. Dr. Brown and his associates are "trying
to discover the normal progression by which children attain to the rules of
English and also trying to understand the learning that is involved."

Some of the clearest evidence that children form construction rules, Dr.
Brown observes, lies in the errors children make. So long as a child speaks
correctly, it is possible that he says only what he has heard. But when a
child from an educated family says "I digged in the yard," or "I saw some
sheeps," or "Johnny hurt hisself," it is likely that, instead of imitating, he
is applying the rules he has induced. (If walked, why not digged? If dogs,
why not sheeps? If myself and herself, why not hisself?)

Other evidence comes from some of the tests the researchers have used.
For example, a child is shown a picture of a small animal and told, "This
is a wug." Then he is shown a picture of two animals and told, "Now

there are two of them. There are two " The child generally says

wugz. If the animal is a bik, he gives the plural as btks; If it is a niss, he
gives the plural as nissez. Thus, even with words he has never heard, he
follows the rules he has derived, correctly, for forming and pronouncing
plurals.

When a child begins to put words together, generally around the age of
18 months, his utterances are very short-'*Anna walk," for example, and
"Dale play car." Dr. Brown thinks that this is less a matter of limited
memory, since the child may know several hundred words and be able to
recall them when needed, than of limited programing ability. During the
second year of life, this usually prevents a child from planning sentences
of more than two or three words. (Of a dozen children studied by the
Harvard group, those about 2 years old had an average span of 2 words,
with a range of from 1 to 4; children about 3 years old had an average
span of 5 words, with a range of from 1 to 1 1.)

Even when a mother asks a child to "say what I say," the child reduces
the sentences to telegraphese. "Mommy is going to have her soup" be-
comes "Mommy soup" whether the child is speaking spontaneously (using
words heard before, of course) or in immediate Imitation of his mother.

In the beginning, Dr. Brown reports, the child probably does not know
that in reproducing an adult sentence he cuts it down to the most signifi-
cant words. More likely he selects these words, mostly nouns and verbs,
mainly because of the emphasis placed upon them by the speaker. The
same principle holds with polysyllabic words-when they are beyond the
chikTs grasp, he repeats only the stressed syllable.

On the basis of the work done under the NIMH grant and previously,
Dr. Brown describes a child's first speech as a systematic reduction of
adult speech marked by (1) a short programming limit and (2) selection in

493

favor of stressed elements. Features of the child's grammar can be predict
from the average length of his utterances. The investigators found,
instance, that children whose utterances averaged less than 3.5 woi
invariably omitted such modal auxiliaries as will and can and said "I
inside" and "I make a tower." Children who averaged less then 3.2 wo
per utteiance always omitted the foims of the verb to be in progress
constructions, saying not "I am going" but "I going."

In helping a child go from telegraphic English to grammatically mi
complete speech, the investigators note, interplay between mother a
child is highly important. For example, Anna's mother says to li
"Anna's going to have her lunch," and the girl, picking out the stres:
elements parrots: "Anna lunch." But some day the girl heiself ventui
"Anna lunch," and the mother then expands the utterance into the m
appropriate simple sentence. In one situation she will say something li
"Yes, Anna is going to have her lunch now;" in another situation, "\
Anna has had her lunch." Thiough many repetitions of this sort, the cl
learns to express time-and to develop rules applicable to other sentem

One way to teach a foreign language is to give explicit instruction in
rules of grammar. A newer method, Dr. Brown points out, is to treat
student like a child and have him repeat sentences over and over agt
Eventually he is expected to be able to produce new sentences-ones t
are somehow implied by those in the practice set. This is essentially w
the child does in learning to speak. Hence studies of how the child g
about making his new sentences may contribute to the improvement
foreign-language instruction.

The most important contributions of this study, however, may CQ
from the increased knowledge it piovides of normal language clevcl
ment. Arrested or abnormal language development are two of the m
prominent symptoms of emotional, mental, or neurological disorder:
childhood. Understanding the factors which contribute to normal
guage development may serve as a guideline for finding the factors wl'
contribute to the erratic speech habits of the emotionally disturbed cli

Research Grant: MH 7088
References:

Brown, R. The Acquisition of Language, Presented at 1962 meeting of the Associa

for Research m Nervous and Menial Diseases,
Brown, R. and Fraser, C. The Acquisition of Syntax. In C. Gofer and B. Musg

(Eds.), Verbal Behavior and Learning: Problems and Processes. McGraw Hill, 1

York, 1963.
Fraser, C., Bellugi, U., and Brown, R. Control of Grammar in Imitation, Comprc

sion, and Production, Journal of Verbal Hearing and Verbal Behavior, 2:121-1

1963,

494

Investigator:

Richard M. Held, Ph.D.

Massachusetts Institute of Technology

Cambridge, Mass.

Prepared by:

Herbert Yahraes

A number of authorities on 'child development have reported in recent
years that babies raised in a stimulating environment progress faster than
other babies. One of the first experimental tests of this observation was
recently made in connection with research to uncover basic mechanisms
involved in the development and maintenance of sensorimotor coordina-
tion.

Shaping this research program, directed by Richard M, Held, professor
of experimental psychology at the Massachusetts Institute of Technology,
is the idea that voluntary, self-produced movement, with its accompany-
ing feedback signals to the sensory centers of the brain, is highly im-
portant to the growth of the sensorimotor systems. This means, for
example, that we can reach for the telephone and, without fumbling, pick
it up because we have had a good deal of practice in coordinating messages
to the brain from the world around us, messages from the brain to our
muscles, and then more messages as the muscles act to change our rela-
tionship to the external world.

Dr. Held and his associates have been working in two ways to demon-
strate the importance of motor-sensory feedback: They have tampered
with its normal operation in adult human beings and in newborn animals,
and they have increased its occurrence in babies.

The work with adults has involved, typically, the wearing of prism
goggles that make objects a few feet away appear to be displaced 3 or 4
inches— right or left, up or down. People soon compensate for the dis-
placement, it has long been known, and after the goggles have been re-
moved the compensatory effect carries over for a time, now in the form of
error. But how is compensation achieved? In a series of experiments, the
MIT group has demonstrated that it occurs when what the individual sees
is correlated with what he does. If he views his surroundings as he is
pushed in a wheelchair instead of as he takes a walk on his own, for
instance, or if he holds his hand still as he looks at it, or if it is moved by
somebody else and not himself, his brain makes little or no adjustment to
the distortion of the visual signals. This work seems to show, then, that

495

favor of stressed elements. Features of the child's grammar can be predicts
from the aveiage length of his utterances. The investigators found, fi
instance, that children whose utteiances averaged less than 3.5 won
invariably omitted such modal auxiliaries as wilt and can and said "I ^
inside" and "I make a tower," Children who averaged less then 3.2 won
per utterance always omitted the forms of the verb to be in progress!
constructions, saying not "I am going" but "I going."

In helping a child go from telegraphic English to grammatically mo
complete speech, the investigators note, interplay between mother ai
child is highly important. Foi example, Anna's mother says to IK
"Anna's going to have hei lunch," and the girl, picking out the stressi
elements parrots: "Anna lunch." But some day the girl herself venture
"Anna lunch," and the mother then expands the utterance into the mo
appropriate simple sentence. In one situation she will say something lik
"Yes, Anna is going to have her lunch now;" in another situation, "Yt
Anna has had her lunch." Thiough many repetitions of this sort, the chi
learns to express time-and to develop rules applicable to other sentence

One way to teach a foreign language is to give explicit instruction in t
rules of grammar. A newer method, Dr. Brown points out, is to treat t
student like a child and have Win repeat sentences over and over agai
Eventually he is expected to be able to produce new sentences-ones tli
are somehow implied by those in the practice set. This is essentially wh
the child does in learning to speak. Hence studies of how the child gc
about making his new sentences may contribute to the improvement
foreign-language instiuction.

The most important contributions of this study, however, may cot
from the increased knowledge it provides of normal language devclc
ment. Arrested or abnoimal language development are two of the me
prominent symptoms of emotional, mental, or neurological disorders
childhood. Understanding the factors which contribute to normal h
guage development may serve as a guideline for finding the factors whi
contribute to the erratic speech habits of the emotionally disturbed chi

Research Grant: MH 7088
References'

Brown, R. The Acquisition of Language . Presented at 1962 meeting of the Associnti

for Research in Nervous and Mental Diseases,
Brown, R. and Fraser, C. The Acquisition of Syntax. In C. Cofer and B. Musgn

(Eds.), Verbal Behavior and Learning: Problems and Processes, McGraw 1 1111. N

York, 1963,
Fraser, C , Belhigi, U., and Brown, R. Control of Grammar in Imitation, Compreh

sion, and Production, Journal of Verbal Hearing and Verbal Behavior, 2:121-1'

1963,

494

Investigator:

Richard M. Held, Ph.D.

Massachusetts Institute of Technology

Cambridge, Mass.

Prepared by.

Herbert Yahraes

A number of authorities on child development have reported in recent
years that babies raised in a stimulating environment progress faster than
other babies. One of the first experimental tests of this observation was
recently made in connection with research to uncover basic mechanisms
involved in the development and maintenance of sensorimotor coordina-
tion.

Shaping this research program, directed by Richard M. Held, professor
of experimental psychology at the Massachusetts Institute of Technology,
is the idea that voluntary, self-produced movement, with its accompany-
ing feedback signals to the sensory centers of the brain, is highly im-
portant to the growth of the sensorimotor systems. This means, for
example, that we can reach for the telephone and, without fumbling, pick
it up because we have had a good deal of practice in coordinating messages
to the brain from the world around us, messages from the brain to our
muscles, and then more messages as the muscles act to change our rela-
tionship to the external world.

Dr. Held and his associates have been working in two ways to demon-
strate the importance of motor-sensory feedback: They have tampered
with its normal operation in adult human beings and in newborn animals,
and they have increased its occurrence in babies.

The work with adults has involved, typically, the wearing of prism
goggles that make objects a few feet away appear to be displaced 3 or 4
inches—right or left, up or down. People soon compensate for the dis-
placement, it has long been known, and after the goggles have been re-
moved the compensatory effect carries over for a time, now in the form of
error. But how is compensation achieved? In a series of experiments, the
MIT group has demonstrated that it occurs when what the individual sees
is correlated with what he does. If he views his surroundings as he is
pushed in a wheelchair instead of as he takes a walk on his own, for
instance, or if he holds his hand still as he looks at it, or if it is moved by
somebody else and not himself, his brain makes little or no adjustment to
the distortion of the visual signals. This work seems to show, then, that

compensation depends to an important extent upon the feedback that
accompanies voluntary movement.

Other conditions besides active movement together with its sensory
feedback will produce some change that may be called adaptation or
compensation. This is particularly the case when a person can gain infor-
mation about the errois he makes while wearing prisms. He can then note
his mistakes and very quickly correct them. The MIT researchers believe,
however, that this kind of correction involves a rather different sort of
process from the one they have been mainly concerned with, which seems
applicable to questions about the very early development of sensorimotor
control.

Though the experiments with prism goggles may superficially appear
rather strange and even bizarre, they are considered quite relevant to a
normally occurring process. During growth, Dr. Held explains, head size
increases until its linear dimensions become at least one and a half times as
large as at birth. As that happens, the distance between the eyes grows,
the distance between the ears grows, and other parts of the sensory system
change. Because of these anatomical changes, the adult and the young child
will have different sensoiy inputs for the same external conditions. The
fact that the adult can handle the same sensorimotor tasks as the child
indicates that some adjustment must have taken place. Further, the
optical properties of the eyes change, too, in the course of a lifetime,
and these changes may require compensation. "In our experiments the
changes are abrupt," says the investigator, "while in the natural course of
things they are very slow, but we think that some of the factors affecting
adaptation are probably the same."

In the work with babies, done under the immediate supervision of a
psychologist named Burton White who is particularly interested in child
development, the researchers studied several groups of infants who had
been born and were being reared in an institution, A control group got the
institution's usual care, which was adequate for general health but pro-
vided a minimum of handling and other stimulation. Other babies were
given 20 minutes' extra handling every day for a month, beginning when
they were 6 days old. Up to their fifth month, at least, these handled
babies were significantly more visually attentive-shifted their gaze more
often from one thing to another-than the controls. But there was no
other apparent change.

Another group of babies who had received the extra handling were then
given a better-than-usual opportunity to move in their cribs and to look at
things more interesting than the monotone ceiling and the white sheets
and crib liners. The mattresses were flattened so that the babies, instead of
lying in a trough all day, could roll a little and more readily move head,
arms, and trunk. Three times a day, for 15 minutes at a time, the babies
were placed on their stomachs and the crib liners removed, thus encourag-
ing the children to raise their heads and make other movements in order
to watch the ward activities. Interesting objects were hung where the
babies could see and swipe at them.

496

One big change occurred. At the median age of 3 months, these chil-
dren began to reach for and grasp objects effectively; the control group
needed 5 months. The results are what would have been predicted, the
investigators point out, from the theory that the development of sen-
sorimotor coordination depends to a great extent upon self-initiated
movement and the accompanying feedback to the senses. At least as far as
the ability to reach and grasp is concerned, a baby apparently develops
faster when he has an opportunity to move and visually interesting sur-
roundings to encourage him to move.

Experiments by Dr. Held and Dr. Alan Hein with cats and by Dr. Held
and Joseph A. Bauer, Jr., with monkeys tend to support the same theory,
Kittens fitted in infancy with ruffs that prevented them from seeing their
paws were unable a few weeks later, when the ruffs had been lemoved, to
guide their paws on the basis of vision. Time after time these kittens
would pounce on a slowly swinging ball and miss it. Much the same thing
happened with monkeys raised for some weeks in a "holder chair1' so
constructed that they could use their hands but could not see them. When
they were finally allowed to sec one hand, they could hardly keep their
eyes from it, and the experimenters had difficulty enticing them to reach
for food or toys. When the animals did reach, they usually missed. Only
when they had seen the hand a long time, from 10 to 20 hours, did they
bring the movements under control. Allowed then to see the other hand,
the monkeys were just as surprised— and just as unable to control its
movements to get what they wanted.

Dr. Held points out that persons who have been blind since birth do
achieve sensorimotor coordination. In this case the sensory feedback that
accompanies movement is delivered by the senses of touch and hearing.
Every once in a while a person long blind recovers his sight and then finds
considerable difficulty, for a time, in coordinating what he sees with what
he does. He reaches for a cup of coffee and, because he has had no
opportunity to maintain motor-visual coordination, upsets it.

Besides increasing our basic knowledge of how the central nervous
system operates, the MIT research may have results of immediately prac-
tical importance. The work with babies has already provided scientific
evidence, the investigators point out, that certain kinds of experience veiy
early in life lead to increased alertness and faster visual-motor develop-
ment. And authorities on child development believe that the growth of
sensorimotor capacities leads to the growth of intellectual capacity. The
findings and prospective findings may also throw some light on the devel-
opment of certain visual disturbances, notably strabismus and amblyopia.
The research may have a suggestion, too, for rehabilitation workers.
Patients long bedridden lose something of their ability to control the
movements of their arms and legs. This is less a matter of muscular weak-
ness, Dr. Held suspects, than a failure to have performed those movements
and thus to have obtained the feedback from them— conditions that seem
to be important to the maintenance of control.

Voluntary movement and the accompanying sensory feedback are im-
portant, the investigator sums up,

497

compensation depends to an important extent upon the feedback that
accompanies voluntary movement.

Other conditions besides active movement together with its sensory
feedback will produce some change that may be called adaptation or
compensation. This is particularly the case when a person can gain infor-
mation about the errors he makes while wearing prisms. He can then note
his mistakes and very quickly correct them. The MIT researches believe,
however, that this kind of correction involves a rather different sort of
process from the one they have been mainly concerned with, which seems
applicable to questions about the very early development of sensorimotor
control.

Though the experiments with prism goggles may superficially appear
rather strange and even bizane, they are considered quite relevant to a
normally occurring process. During growth, Di. Held explains, head size
increases until its linear dimensions become at least one and a half times as
large as at birth. As that happens, the distance between the eyes grows,
the distance between the ears grows, and other parts of the sensory system
change. Because of these anatomical changes, the adult and the young child
will have different sensory inputs for the same external conditions. The
fact that the adult can handle the same sensorimotor tasks as the child
indicates that some adjustment must have taken place. Further, the
optical properties of the eyes change, too, in the course of a lifetime,
and these changes may require compensation. "In our experiments the
changes are abrupt," says the investigator, "while in the natural course of
things they are very slow, but we think that some of the factois affecting
adaptation are probably the same."

In the work with babies, done under the immediate supervision of a
psychologist named Burton White who is particularly interested in child
development, the researchers studied several groups of infants who had
been bom and were being reared in an institution. A control group got the
institution's usual care, which was adequate for general health but pro-
vided a minimum of handling and other stimulation. Other babies were
given 20 minutes' extra handling every day for a month, beginning when
they were 6 days old. Up to their fifth month, at least, these handled
babies were significantly more visually attentive-shifted their gaze more
often from one thing to another— than the controls. But there was no
other apparent change.

Another group of babies who had received the extra handling were then
given a better-than-usual opportunity to move in their cribs and to look at
things more interesting than the monotone ceiling and the white sheets
and crib liners. The mattresses were flattened so that the babies, instead of
lying in a trough all day, could roll a little and more readily move head,
arms, and trunk. Three times a day, for 15 minutes at a time, the babies
were placed on their stomachs and the crib liners removed, thus encourag-
ing the children to raise their heads and make other movements In order
to watch the ward activities. Interesting objects were hung where the
babies could see and swipe at them.

496

One big change occurred. At the median age of 3 months, these chil-
dren began to reach for and grasp objects effectively; the control group
needed 5 months. The results are what would have been predicted, the
investigators point out, from the theory that the development of sen-
sorimotor coordination depends to a great extent upon self-initiated
movement and the accompanying feedback to the senses. At least as far as
the ability to reach and grasp is concerned, a baby apparently develops
faster when he has an opportunity to move and visually interesting sur-
loundings to encourage him to move.

Experiments by Dr. Held and Dr. Alan Hein with cats and by Dr. Held
and Joseph A. Bauer, Jr., with monkeys tend to support the same theory.
Kittens fitted in infancy with ruffs that prevented them from seeing their
paws were unable a few weeks later, when the ruffs had been removed, to
guide their paws on the basis of vision. Time after time these kittens
would pounce on a slowly swinging ball and miss it. Much the same thing
happened with monkeys raised for some weeks in a "holder chair" so
constructed that they could use their hands but could not see them. When
they were finally allowed to see one hand, they could hardly keep their
eyes from it, and the experimenters had difficulty enticing them to reach
for food or toys. When the animals did reach, they usually missed. Only
when they had seen the hand a long time, from 10 to 20 hours, did they
bring the movements under control. Allowed then to see the other hand,
the monkeys were just as surprised— and just as unable to control its
movements to get what they wanted.

Dr. Held points out that persons who have been blind since birth do
achieve sensorimotor coordination. In this case the sensory feedback that
accompanies movement is delivered by the senses of touch and hearing.
Every once in a while a person long blind recovers his sight and then finds
considerable difficulty, for a time, in coordinating what he sees with what
he does. He reaches for a cup of coffee and, because he has had no
opportunity to maintain motor-visual coordination, upsets it.

Besides increasing our basic knowledge of how the central nervous
system operates, the MIT research may have results of immediately prac-
tical importance. The work with babies has already provided scientific
evidence, the investigators point out, that certain kinds of experience very
early in life lead to increased alertness and faster visual-motor develop-
ment. And authorities on child development believe that the growth of
sensorimotor capacities leads to the growth of intellectual capacity. The
findings and prospective findings may also throw some light on the devel-
opment of certain visual disturbances, notably strabismus and amblyopia.
The research may have a suggestion, too, for rehabilitation workers.
Patients long bedridden lose something of their ability to control the
movements of their arms and legs. This is less a matter of muscular weak-
ness, Dr. Held suspects, than a failure to have performed those movements
and thus to have obtained the feedback from them-conditions that seem
to be important to the maintenance of control.

Voluntary movement and the accompanying sensory feedback are im-
portant, the investigator sums up,

497

—in helping infants develop sensonmotor coordination;

-in helping the central nervous system adjust to the growth of the

body; and
-in maintaining normal coordination even in a situation, such as may

occur in space, where the incoming signals are distorted.

Compensating for Displaced Sensory Signals

In one of the team's early experiments, subjects sat at a table ant
looked down at a mirror, which earned the reflection of a diagram. On tlu
table top, some inches below the mirror, lay a sheet of paper. The subject;
were asked to take a pencil and to mark with dots on the paper, whicr
they could not see, the positions of the four corners of the diagram. Thei
they put on prism goggles for a while and, with the mirror removed
looked at theii right hands. One group kept the hand motionless; anothei
moved it back and forth; a third group had it moved back and forth b)
the investigator. Then the prisms were removed and the mirror replaced
and the subjects were asked again to mark the apparent positions of tin
corners of the diagram. When people compensate for the distortion pio
duced by prisms, the effect lasts for a while after the prisms are removed
In this experiment, consequently, the subjects who had become adapte<
to the prisms should have marked the comers several inches to one side o
the positions originally marked. The only subjects who did mark them thi
way were those who had moved their hands while wearing the goggles.

The investigators then tried the effect of movement involving the entirt
body, not just the hands and arms. In a dark room a subject was asked tc
turn his chair until he was face to face with a target -a slit of light. Thii
showed his ability to judge the direction of a target in reference to himsel
alone. Then he put on prism goggles, went for a walk of an hour or more
came back, took off the goggles, and was tested again. Invariably he now sav
the target as being some distance to the side of where it actually was. Thi
was evidence that while wearing the goggles he had compensated for thci
distorting effect. But prism-wearing subjects who had taken a ride over tin
same path in a wheelchair, pushed by someone else, showed little or IK
adaptation.

Experiments in directional hearing brought analogous results. Normal];
a person locates the source of a sound because the waves reach one eai
the closer one, a split second before they reach the other. But in this worl
the subjects, wearing microphones, were exposed to noise from separate)
sources in a fashion that produced rapidly changing dichotic time diffei
ences; that is, differences in time of arrival at the two ears of correspond
ing acoustic signals. In effect, the apparent source of the sound signals-
like that of the visual signals in persons wearing prisms-had been dii
placed. Subjects who wore the microphones while walking up and down
busy corridor later showed temporary inaccuracies in locating the sourc

498

of a sound. While weaiing the microphones, apparently, they had compen-
sated for the aural distortion. The subjects who had not moved about
showed no evidence of such an effect.

These and other experiments seemed to support Dr. Held's idea that
adaptation to sensory distortion strongly depends upon the close correla-
tion between signals from the motor nervous system, producing the physi-
cal movement, and the consequent feedback signals to the sensory system,
showing the results of the movement. Seeking further evidence, he set up
an experiment in which the correlation was impaired. The apparatus was a
rotating prism arrangement that made a peison's hand appear to be mov-
ing back and forth even when it was held motionless. When the person
moved his hand, therefore, his sensory nervous system presumably could
not distinguish between the signals resulting from the actual movement
and those resulting from the apparent movement. This hypothesis seemed
to be borne out by a test of the subjects' accuracy in reaching for an
object after they had been looking through the rotating prisms. Subjects
who had not moved their hands (but had had them moved by the experi-
menter) showed no decline in their accustomed accuracy. But those who
had been moving their hands themselves showed much less accuracy than
usual: their eye-hand coordination had been temporarily impaired, pre-
sumably because the brain had been getting more signals than it could
manage. In engineeiing terms, there had been "noise" in the feedback
system.

The investigators have produced the same effect by introducing what
Dr. Held calls "inertial noise." In one experiment, for instance, they have
had a subject push with his arm against an apparatus that, in spite of his
efforts, moves his arm in a direction opposite to the one he wants it to
take. The normal relationship between signals to the muscles and the
consequent movement of the arm, with the accompanying feedback to the
brain, is upset. When wearing prisms in this situation, he compensates
significantly less, or not at all.

Further, a recent experiment shows, the compensatory process in com-
pletely blocked when there is a delay between the time a subject moves
Ills hand and the time he sees the result of this movement. Ordinarily, the
results of movement reach the retina with the speed of light. To introduce
a time delay, the investigators built a device enabling a subject to sit at a
table and move a control slick hidden from his sight under a mirror on the
table. Movement of the control stick actuated potentiometers and these in
turn modulated voltages. When these voltages were fed to an oscilloscope
they produced movements of the scope trace-a narrow bar l!/2 inches
long-just like those being made by the hand. The trace was projected
onto a ground glass screen, which the subject viewed in the mirror over his
hand. In sum, though the subject could not see his hand, he could watch
every movement it made. When the voltages were fed directly to the
oscilloscope, the image of the movement was perceived instantaneously.
In this situation, before-and-after tests showed that a person wearing
prism goggles adapted himself to the displacement caused by the goggles.
But the voltages could also be fed to magnetic tape and then given to the

499

oscilloscope after a delay. In this case the visual feedback was delayed; the
subject did not see the movement until after it had occurred. The investi-
gators found that even delays as small as 270 milliseconds completely
eliminated adaptation.

Though efforts to produce and study the effect of still smaller delays
are being made, Dr. Held is fairly well satisfied that unless motor-sensory
feedback is virtually simultaneous with movement, the brain cannot
handle it. Consequently, the nervous system will not adapt itself to the
apparent displacement, caused by prisms, of visual signals. This is quite as
it should be, the investigator notes. If the feedback signals did not have to
be closely coupled with the output signals, the system would sometimes
be in chaos trying to decide which of several moving hands was its own.

Dr. Held's work is related to other ongoing research at MIT under the
direction of Dr. Hans-Lukas Teuber, chairman of the department of
psychology, dealing with the "corollary discharge" and the supposed role
of the brain's frontal lobes in controlling it.1 The corollary discharge
theory holds that when the brain calls for movement, some of the signals
go not to the muscles concerned but to the appropriate sensory systems.
In this way the senses are prepared for the results of the movement, For
example, if a person turns his head while looking at an object, the image
of this object moves across the retina of his eye just as it would if the
object itself were moving. But he perceives that his eyes are moving and
not the object; the retinal change has been discounted. And the discount-
ing occurs, Dr. Held postulates, because the sensory feedback (occurring
as the result of movement) is matched to the corollary discharge (occur-
ring as the result of the order to move). When the feedback is not matched
to the discharge, the movement is perceived in the world outside.

Several investigators have reported recently that adaptation, as mea-
sured by the accuracy with which hand and arm reach toward a target
after prisms have been worn, may result from a change in the felt position
of the parts of the body. Under this view, a person misjudges the location
of the target because he no longer accurately feels the position of his
reaching arm. But after a series of experiments involving, among other
things, the ability of subjects to locate with the unseen right hand various
positions taken by the left hand, out of sight under the desk top, the MIT
group offers another view.

Before wearing prisms, the subjects, blindfolded, could mark the posi-
tions of the left hand fairly accurately; after wearing prisms, they made
mistakes of the kind to be expected -marking locations to the right of tlie
actual location, for example, if the prisms had made objects seem dis-
placed to the left. The investigators ask why a person should misreach for
his hand following prism adaptation when he doesn't need vision to reach
for it accurately -when he can feel his way to it, so to speak.

Dr. Held offers the following hypothesis: In this case, there are actually
two different but interacting modes of reaching. One involves the felt

'"Exploring the Brain's Functions." In Research Project Summaries No, 2, Na-
tional Institute of Mental Health: Public Health Service Publication No. 1208-2, 1965.

500

positions of the two arms, but there is no reason for either of the two felt
positions to be changed by adaptation, and they probably do not change.
The change that occurs lies in the other mode of reaching, which is based
upon the matching of arm movements with the orientation of the head. It
is reasonable to suppose that adaptation changes the direction of reaching
with respect to the orientation of head to target. But in the case consid-
ered, perhaps the effect of the change is modified by the influence of the
other mode of reaching. Hence a person who has just taken off prisms
should make smaller mistakes in reaching for his unseen hand than in
reaching for a visible target. The experimental evidence seems to point in
this direction.

The Development of Visual Control

A few years ago a research group at the University of California, River-
side, found that when kittens were restrained from walking, they devel-
oped marked deficiencies in the visual control of activity. Why? Probably,
thought the California group, because the kittens had lacked a variety of
visual stimulation. More likely, thought the MIT researchers, because they
had lacked visual stimulation correlated with movement.

To get additional facts, Dr. Held and Dr. Alan Hein ran an experiment
that has been much admired for its ingenuity. The apparatus was a sort of
merry-go-round, with a center post supporting a rotating beam. Attached
to one arm of the beam was a kitten, which was able to walk around the
post, to jump up, and to turn toward the right and the left. Attached to
the other arm was a little box carrying another kitten. Any movements
made by the first kitten were transmitted to the box, so that the second
kitten made them, too, but passively instead of on its own. The center
post and the surrounding circular wall of the apparatus displayed a pattern
of stripes, so each kitten received essentially the same visual stimulation.

The pairs of kittens were raised in darkenss till they were 1 0 weeks old;
then they were brought out 3 hours a day for experience in the apparatus.
Ten days later the active kittens responded normally to several tests of
visual responses. They put out their forepaws to ward off collision when
gently lowered toward a surface; blinked at an approaching object; and,
when walking on a sheet of glass beneath which appeared a shallow drop
to one side and a steep drop to the other, avoided the steep one. The
kittens that had been passively moved did not show these responses.

To learn more precisely the conditions influencing visual-motor coordi-
nation, Drs. Hein and Held have continued their studies .of kittens
(financed largely by the National Science Foundation but part of the same
general research program supported by NIMH) and recently made a puz-
zling discovery. Kittens were put through the merry-go-round procedure
under both active and passive conditions of movement. But when a kitten
moved itself, only its right eye could see, the other being covered by an
opaque contact lens, and when the kitten was passively moved, only its
left eye could see. The result was an animal with one "active" and one

501

"passive" eye. And the investigators found that when the active eye was
open, the kitten could perform normally in the tests of visually controlled
behavior; when the passive eye was open, it could not. There was no
transfer of information; one eye was disassociated from the other, as if
this had been a split-brain study and the connections had actually been
severed.

Additional work supports this finding. Kittens were raised so that they
could see their forepaws with one eye but not the other. This second eye,
though, could see just about everything else, and the animals were free to
move. Then the experimenters tested the kittens' ability, first with one
eye open and then the other, to reach for and hit an object. This ability
was found to be present only when the eye that had always been able to
watch the forepaws was open.

"Here is a failure of information to pass with an intact biain," the
investigators comment. "It is very surprising." As yet they have no ex-
planation, but they point out that the findings are consistent with the
central idea guiding the group's research-that movement and feedback are
essential to the development of eye-hand coordination in infants and an
important source of the information enabling an adult to correct for
sensory distortion.

In work with animals, the investigators are now trying to study what
actually happens in the brain under conditions like those used in the
experiments. This phase of the research may or may not clear up the
puzzle of the disassociated eyes, but at least, Dr. Held believes, it should
help identify those parts of the brain essential to the development and
maintenance of the ability to direct a visually guided reach and grasp. This
is one of the primary abilities enabling man to orient himself to, and get
along in, the world around him.

Research Grant: MH 7642
Date of Interview; August, 1966

References:

Efstathiou, Aglaia, Bauer, J., Greene, Martha, and Held, R. Altered reaching following
adaptation to optical displacement of the hand. /. Exp. Psychol., 1966, in press.

Held, R. Plasticity in sensory-motor systems. Sclent. American, 1965, 21 3, 5.

Held, R. Plasticity in sensorimotor coordination. Paper for XVIII International Con-
gress of Psychology, 1966.

Held, R , Efstathiou, Aglaia, and Greene, Martha. Adaptation to displaced and delayed
visual feedback from the hand./. Exp. Psychol., 1966,72, 6, 887-891.

Held, R., and Freedman, S. S. Plasticity in human sensorimolor control. Science,
1963,142,3591.

Held, R., and Hein, A. Movement-produced stimulation in the development of visually
guided behavior./ Comp. Physiol. Psychol., 1963, 56, 5.

White, B. L., and Held, R. Plasticity of sensorimotor development in the human infant.
In Judy F. Rosenblith and W. Allinsmith (Eds.), The causes of behavior: readings in
child development and educational psychology. Boston: Allyn & Bacon, 1966,

502

Investigator.

Benson E. Ginsburg, Ph.D.

University of Chicago

Chicago, III.

Prepared by

Herbert Yahraes

A person's experiences during infancy and childhood work powei fully
to shape his later attitudes and personality. Indeed, according to psycho-
analytic theory, they largely determine the state of his mental health.
Looking for scientific proof of this concept, a number of investigatois
have reported evidence linking mental illness with stressful situations in
early life. Animal studies, too, have stressed the importance of early ex-
perience. Now comes a geneticist asking that research into the causes of
behavior consider more rigorously the effects of heredity. Early experi-
ences may indeed affect later behavior tremendously, reports Benson E.
Ginsburg, professor of biology at the University of Chicago. Or, depending
upon the genetic situation, they may not affect it at all.

As part of his evidence, this investigator and his students subjected
carefully bred mice to certain types of stress during infancy. Some of
them, as adults, became significantly more aggressive than usual; others
became less aggressive; others showed no effect. The results depended
upon the strains to which the mice belonged: that is, upon heredity. In
most cases the results depended also upon the time when the stress had
been encoimtered-during all the first 4 weeks of life, or only the first 2,
or only the second 2. "All mice," Dr. Ginsbuig notes, "are not created
equal.'1

Because work by himself and other researchers shows a strong relation-
ship between an animal's inheritance and certain facets of its behavior,
including some abnormal behavior under stress, the investigator believes
the same relationship may well hold true in people. If one can select for
aggressive strains in rabbits, guinea pigs, dogs, fighting cocks, and bulls,
and if equivalent early experience can have dlffeiential effects in mice,
depending on the genetic background, he asks, why should it be supposed
that human beings all respond to a given situation in the same way,
regardless of their inheritance? He agrees with psychotherapists that
psychological factors are important, but he holds that these must act on n
biological substratum— on inherited tendencies and mechanisms.

In recent work with mice, Dr. Ginsburg believes that his research group
has uncovered the inherited mechanism responsible for a certain kind of
abnormal behavior. Apparently the mechanism lies in a particular cell

503

layer in one structure of the brain, and appaiently its manifestations can
be treated chemically in ceitain cases. In some other mice, the same kind
of behavior seems to have a different cause and is not affected by the
same chemotherapy.

Dr. Ginsburg's work is supported by a grant from the National Institute
of Mental Health. His main laboratory animals are highly inbred mice
(every individual in a strain essentially an identical twin of every other
individual), wolves, coyotes, and dogs.

Stress Mechanism in Mice

The investigator and his group have set out to study differences in
behavior among closely related animals, to learn what laws of genetics are
followed in the inheritance of these differences, and to determine what
body or brain tissues are affected by the differences in heredity and,
therefore, give rise to the differences in behavior.

In mice the principal characteristic under study is susceptibility to
audiogenic seizures-epilepticlike convulsions induced by the ringing of a
bell. The grantee chose this characteristic because; (a) it leads to behavior
of an extreme kind, relatively easy, therefore, to identify; (£) it is, like
some types of abnormal behavior in human beings, triggered by stress— in
this case, sound; (f) it appears only in certain strains of mice and conse-
quently must be under genetic control— a deduction Dr. Ginsburg and an
associate, Dr. Starbuck Miller, have demonstrated, through breeding ex-
periments, to be correct.

Some highlights of the findings to date:

1 . Three strains of mice highly susceptible to seizures show a character-
istic difference in one tiny area of the brain. It appears to be caused
by a difference in either the amount or the activity of one or more
closely related enzymes.

The difference came to light in the course of a meticulous, area-by-area
study of activity involving adenosine triphosphate, or ATP. This is the
compound supplying most of the energy for biochemical reactions, such
as those concerned with moving the muscles, building proteins, and flash-
ing nerve signals. The energy is made available when ATP is broken down
in an almost instantaneous reaction involving the enzyme ATP-ase. The
investigation centered on this reaction because of an earlier report, which
has not yet been verified, that the brains of seizure-prone mice contained
less of the enzyme than the brains of other mice.

The Ginsburg group incubated slices of brain tissue in a medium con-
taining ATP or a related compound and then examined them under the
microscope for a brown stain produced by one of the chemicals resulting
from the breakdown of that compound, The darker the stain, the greater
the amount of ATP that had been broken down and, presumably, the
greater the amount of ATP-ase (or the greater its activity) in the area of
the brain under examination,

504

Of all the areas studied, a difference between seizure-prone and seizure-
resistant mice was found in only one-a thin layer of cells in the hippo-
campus (specifically, the granular cell layer of the dentate fascia), which is
one of the oldest parts of the cortex. In the mice susceptible to seizures,
this layer stained consistently darker than in the other mice. It also
stained darker when triphosphates other than ATP were used in the incu-
bating medium, suggesting that the activity may be caused by other
enzymes that break down this class of compounds.

Breeding and other experiments showed that the staining differences
were inborn, with two pairs of genes apparently being involved, and did
not arise from the seizures themselves. The experiments seemed to show
also that whatever led to the staining difference led as well to the seizuies,
a finding that was given additional support when structural diffeiences
were discovered in the cell layer.

In sum, the research group reports, these findings indicate that the
epilepticlike seizures of some strains of mice are indeed traceable to an
inherited difference most probably involving one small part of the brain.
In the test tube, at least, this difference apparently involves a reaction
related to energy release. It is conjectured that in the brains of these mice,
a certain kind of stress leads to abnormal energy release due to a genetic
deficiency in a localized biochemical mechanism.

Whether or not these findings relating to audiogenic seizures in mice
can be applied to epilepsy in human beings remains to be determined.
Until recently, Dr. Ginsburg notes, little was known of the relationship
between seizures in mice and those in people. But electroencephalographic
studies by other investigators now indicate that the seizures induced in
mice are quite similar to those found in some forms of epilepsy.

2. The biochemical mechanisms involved in the type of abnormal be-
havior under study differ from animal to animal, depending upon
what genes have been inherited

In one strain of seizure-prone mice, the staining reaction, instead of
being typical, resembles that found in mice resistant to seizures. (This
strain arose, by natural mutation, from a seizure-resistant strain,) Even
strains showing the typical ATP-ase reaction differ in other characteristics
—for example, the age at which they arc most susceptible to seizures and
the effect of repeated exposure to noise, which in some strains increases
susceptibility.

"A behavior peculiarity," Dr. Ginsburg comments, "can be a kind of
stew with a lot of ingredients in it." As further evidence he cites experi-
ments to learn whether or not seizure-prone mice could be chemically
protected. Some of them could be. The chemical used was monosoclium
glutamate, which is closely related to glutamic acid, a compound that
appears in the brain and has been widely investigated because of reported
findings that it can improve learning ability. Following injections of
rnonosodium glutamate (which also has a common use, to accent the
flavor of food), mice of certain strains had fewer and lighter seizures. Mice
of other seizure-prone strains, however, were not affected.

505

3. Mice that have seizures when exposed to a ringing bell also respond
abnormally to a stressful learning situation.

An ordinary mouse that has learned to run through simple mazes and
then is confronted with a maze a little too complex for his learning ability
will keep on trying foi a while. But a seizure-pione mouse goes to pieces,
as evidenced by squeaking and by frantic jumping. An injection, before-
hand, of monosodium glutamate will lead to normal behavior, but only if
the mouse is from a stiain in which the compound will reduce seizures.

4. Mice become less aggressive if they have had a seizure while young.
Along with the other findings, this is believed to suggest a connection
between the seizure mechanistn and various forms of behavior under
stress,

In some human beings with mental disorders, therapy involving convul-
sive seizures-induced by shock treatment-is followed by a return to
normal behavior. Dr. Ginsburg wondered whether or not seizures would
affect the later behavior of mice. As the trait to be studied, he and his
co-workeis chose aggressiveness, measured by the proportion of times two
mice with the same backgrounds attempt to fight when given the oppor-
tunity.

The mice came from two strains highly susceptible to seizures, Members
of both these strains, as well as those of a third seizure-prone strain, hart
been shown to be considerably more aggressive than members of three
seizure-resistant strains.

Seizures were induced when the animals were 28 days old. Each mouse
was then housed alone until its 75th day, when the fighting trials started.
Results; In one strain the mice that had had seizures tried to fight on only
37 percent of the occasions as against 72 percent for the control mice, In
the other stiain the figuie was 56 percent as against 80.

5 . The effects of stressful situations during infancy are governed at least
in part by heredity.

In so-called "handling" experiments, a young mouse is lifted from its
home cage every day and placed for a few minutes by itself in an un-
familiar place, or it is briefly exposed to cold or pain. When mice that
have been subjected to such stress reach adulthood, their behavior is com-
pared with that of mice reared without the stress. A common criterion is
aggressiveness.

As noted earlier in this report, Dr. Ginsburg's group has found that in
most cases studied, the results-a significant increase in aggressiveness, a
significant decrease, or no change-depend both upon the strain of mice
and the period of stress. For instance, one strain becomes more aggressive
if handled during its second 2 weeks of life but not if handled only during
its first 2. Another becomes less aggressive if handled during its first 2
weeks but not if handled only during its second 2,

506

The grantee compares the implication of these findings with the in-
terpretation given to the results of earlier handling experiments, princi-
pally with rats. The older results had been taken to show that early experi-
ence profoundly affected various kinds of behavior, including aggressive-
ness, in a stereotyped -way. Thus, the grantee observes, animal experimen-
tation had seemed to bear out Freudian theory.

But the findings of the older work were in most cases the averages of
the results obtained from experimenting with mixed-or average- popula-
tions. "When you break down a population into a number of compo-
nents by using pure strains," Dr. Ginsburg continues, "you find that early
experience in one genetic situation has a very different effect on later
measures of behavior from early experience in another genetic situation.
One individual may be so well buffered that early stressful experiences
show no effect later on. But they may decidedly change another indi-
vidual in either one direction or the other.

"One mouse is not the same as another mouse, and certainly one person
is not the same as another person. This is true in respect not only to the
experiences an individual has had but also to his potentiality forieacting
to such experiences."

Consequently, in Dr. Ginsburg's opinion, the way drugs and other
forms of therapy affect behavior can be expected to show considerable
variation.

Inherited Aggressiveness

A tendency to engage in certain kinds of aggressive behavior, the
grantee notes, is definitely inherited. For example, rabbits of one partic-
ular strain, first found at the Jackson Memorial Laboratories, Bar Haibor,
Maine, will almost invariably fight an intruder. A person who pokes a
hand toward them or tries to pick them up will be clawed. This happens
no matter how the animals have been raised. Dr. Ginsburg has given young
ones away as Easter bunnies to see what would happen if they were
brought up with tender, loving care. They turned out mean.

Terriers have been selected for aggressiveness -and so successfully that
investigators at Bar Harbor have found it almost impossible to raise a litter
of four or more wirehaired fox terrier pups under natural conditions. If
the litter is kept together after weaning and one puppy goes down in a
fight, the others gang up on him. One puppy can defend itself against two
of its littermates, but not three. Eventually it is mutilated or killed. On
the other hand, when wirehaireds are raised in isolation from mother and
littermates and placed together only after having been weaned from the
bottle* no fights develop. The innate genetic aggressiveness is not e

circumstances, is bad. By genetic selection the program now produces a
high proportion of German shepherds that have only the desirable trait.

Eventually, the University of Chicago geneticists hope to learn the
physiological basis foi aggressiveness in rabbits, for the diffeienccs in
aggressiveness of strains of mice exposed to stress, and for differences in
the aggressiveness of a wild animal, the wolf, and a related domestic
animal, the dog,

The Socialization of Wolves

Preliminary work with wolves has been underway for several years. It
aims at noting marked behavioral diffeiences between wolves and some of
their relatives, principally coyotes and dogs, and then eventually at learn-
ing the physiological bases for these differences. The wolves come from
the Chicago Zoological Park and are wild, in the sense that they have not
been handled by human beings. (They were first housed in pens that had
been occupied by the university's watchdogs. The watchdogs never
learned to work the latches; the wolves learned easily, and the first wolf
out would open all the other pens. The researches had to install locks.)

So far the investigatois have learned, among other things, that woives
can be socialized at any age and that tranquilizing drugs have no perm-
anent effect on the process by which this is done. The first part of this
finding, Dr. Ginsburg repoits, demands a reinterpretation of the widely
held hypothesis of a critical period for socialization-a limited time early
in life during which, and only during which, an animal can learn to accept
human beings. That hypothesis is based mainly on work with dogs and
birds.

Socializing an adult wolf is a heroic job. As the grantee outlines it, you
begin by entering the pen. The wolf, becoming extremely upset, trembles,
moves away, defecates, tries to get out. You have to accustom it to your
presence by going to the pen time and again and just sitting there, This
phase is known as the aversive, emotional stage.

As the animal gets over its extreme fear, Dr. Ginsburg continues, it
enters the stage of the slightly aggressive approach. It takes cautious steps
to investigate you. Now is a critical time because you want to establish
contact, yet any move may provoke either fear or aggression. So you
watch for the signs that tell how far you may go. Curled lips, raised hair,
growls, and, especially, bared canine teeth are all warnings that the wolf is
in a fighting humor. In addition to these signs, if the tail droops and the
ears are back, you know that the animal is still quite frightened and that
you may approach it -very slowly, in full view, and without jerky move-
ments.

If the tail is held high, however, and the ears forward, the wolf means
to attack. Stand your ground, and you'll be snapped at and nosed, Re-
treat, and the wolf will be at you every time you enter the pen. Is there an
odor of fear? "We don't think so," answers Dr. Ginsburg, "because we've all
been as frightened as we could be. But the animal doesn't know that. How

508

it reacts to us is not a matter of how we feel but of what we do"
Apparently what the experiments weais has a hearing, too Gkms and
other protective clothing aie not worn because in the beginning they
appear to frighten the animal and then to make it more aggressive.

Next comes the investigative stage, duiing which the animal rubs
against you, tugs at your clothing and may mark you with urine.

Attempts to domesticate the animal during the second or third phase
will bring back the fear response and retard the socialization process

Finally you are accepted. The wolf greets you by wagging its tail, and it
comes up to be patted and scratched and to mouth your chin. Then,
having accepted you, it soon comes to accept other human beings It has
been socialized.

Once socialized, the grantee reports, adult wolves remain friendly with
human beings— for a considerable time, anyway-even after taking up life
again with their brethren. Eighteen months after a socialized adult has
been returned to the zoo, where it ran with wild wolves and was not
handled by the keepers, the Ginsburg group took it back. It was still easy
to handle.

In contrast, when a thoroughly socialized young wolf was left alone
with an unsocialized littermate at the age of 4'/i months, it regressed.
Eighteen months later it was as wild as the unhandled wolf and had to be
socialized all over again.

Tranquihzers, over the long run, do not speed the process. Under any
of three quite different compounds-chlorpromazine, reserpine,
chlordiazepoxide-adult wolves can be socialized in a matter of days
rather than months, but when the drugs are withdrawn, the socialized
behavior drops away and the animal is back where it started from. In
other words, tranquihzers can help a wolf overcome fear and aggressive
tendencies but only during the period of tranquilization. The answer to
the important question, Does learning carry over to the undntgged
state?, is-in this situation-wo.

With dogs, the story appears to be different. When tranquihzers arc
given to dogs that have been reared ta isolation and are, in consequence,
extremely frightened, the animals adjust to the world of man muct
rapidly than untranquilized dogs, and they do not regress when the
dropped.

hand. They comprise a ritual, each element of which elicits an appro p
ate, built-in response. This symbolic behavior serves to establish don
nance and to defend or claim territory-usually without the need of figli
ing. Other symbolic behavior governs courtship.

The grantee finds an analogy here to the behavior of human beings. I
points out that psychoanalysis is largely based on the proposition that
mental event, as in a dream or a fantasy, can be a symbol of who;
meaning the individual is not consciously aware. Animals, too, deal i
symbols, Dr. Ginsburg observes, and the level at which these may hav
meaning is analogous to, if not identical with, a level generally consideie
to exist "only in our own psyche."

Dogs have no dependable series of signals; the grantee says domestic!
tion has cost them part of their inheritance. Until the experimenter gets ti
know the individual animal, he cannot tell whether or not a given postur
has the same meaning it would have in a wolf.

To get back to tranquilizers: The investigators reported during tin
early part of this work that chlordiazepoxide seemed to eliminate a wolf:
fear and bring out its aggressiveness. Much the same effect was bciii(
observed with human beings, a psychiatrist told the grantee, Ccrlaii
patients taking that tranquilizer tended to attack people, and the patieiils
it developed, were usually those with records of assault and other violence,
Since wolves, too, have violence in their background, the psychiatrist
wondered if the situations might not be analogous.

Dr. Ginsburg doesn't know what conclusions the psychiatrist eventual-
ly reached. His own continued observations, however, showed that the
tranquilizer was not releasing a wolf's aggressiveness; instead, it was
greatly shortening all the stages of socialization, including the first one, in
which the animal is afraid, and the second, in which it is aggressive. As
noted earlier, the drug's effect does not hold.

Tranquilizers did have one lasting effect. Experimenters who thought
they were working with tranquilized wolves-as sometimes they were and
sometimes they were not-handled the animals more confidently and got
better results.

Work To Be Done

The grantee and his group will continue to search for the underlying
biochemical reasons for different behavior. One project aims at exploring
other genetic pathways (besides the one that apparently involves the
breakdown of ATP or related compounds) involved in audiogenic seizures,
Another hopes to learn why seizure-prone mice can differ from one
another in such matters as the age of greatest susceptibility and the cffecl
of early, stressful experiences upon later aggressiveness. Studies of the
adrenal gland, the catecholamines, the acetylcholine-cholinesterase system
in the hipocampal area, and the metabolism of phenylalanine-all of which
are known or believed to influence brain function -are underway in one or
another of the strains showing behavior differences.

510

In trying to establish a connection between a difference in behavior
and a difference in some biochemical characteristic, Dr. Ginsbnrg observes,
research has relied mainly on statistical methods. The grantee prefers to
emphasize breeding experiments designed to show whether or not the
behavioral difference appears only in animals showing the biochemical
difference. He believes that genetically controlled strains must be used if
the physiological reasons for variant behaviors are to be demonstrated
clearly.

Research Grant. MH 3361

References:

Giiisburg, Benson E. Genetics as a tool in the study of behavior. Perspectives in
Biology and Medicine, 1(4), 1958.

Ginsburg, Benson E. Causal mechanisms in audiogenic seizures, In: Psycho physi-
ologic, Neitropharmacologte, et Biochemie de la Crise Audiogene, 312:227-240,
1963. (Colloques Internationaux du Centre National de la Recherche Scientifique,
Pans, 1963.)

Ginsburg, Benson E. Genetics and personality. In: Wepman, Joseph M. and Heine,
Ralph W., eds. Concepts of personality Chicago, Aldine Press, 1963, p. 63-78.

Ginsburg, Benson E. and Slatis, Herman. The use of pure-bred dogs in problems.
Proceedings of the A mmal Care Panel, 1 2(4): 1 5 1 - 1 56, 1 962 .

Ginsburg, Benson E. and Miller, D. Starbuck. Genetic factors in audiogenic seizures.
In: Psychophysiologie, Neuropharmacologie, et Biochemie de la Crise Audiogene.
(Colloques Internationaux du Centre National de la Recherche Scientifique, Paris,
1963.)

511

Investigator.

Steven G. Vandenberg, Ph. D.
University of Louisville School of Medicine
Louisville, Ky.

prepared by:

Herbert Yahraes

Information about the influence of heredity on the behavior of hum,
beings is being obtained by an Institute grantee through a long-time slut
of twins, one of the most comprehensive investigations of the kind ov
attempted.

Coupled with the findings of other investigators, the results of II
study are expected to answer with greater assurance than is now possili
such questions as:

• Just wheie, in the bioad fields of intelligence and personality, ilui
heredity have its principal influence?

• Should emphasis on family-child relationships as a major 0:111x0 <
mental illness give way in part to a strengthened emphasis on heicditur
factors?

• What hereditary characteristics seem to be involved in the ilcvelo]
ment of mental illness?

• In predicting whether or not a child will later deviate from noun:
behavior, which tests of his intelligence and personality and which nspocl
of his early life are most useful?

Tlie grantee is Dr. Steven G. Vandenberg, research associate profcsso
of child development at the University of Louisville and director o
psychological research for the university's Louisville twin study. In oiu
part of this project, all pairs of twins of the same sex born in Louisvilk';
eight hospitals are studied during infancy and the preschool years, h
another part of the project, like-sexed twins already in the public mui
parochial school systems are picked up as they become high school fresh
men and, once a year until they leave school, given a battery of psycho
logical tests. Under plans for expanding the project, information about [hi
home environment of these older twins will be obtained through visits iiml
questionnaires, and the twins may be followed after they leave school,

Psychologists and geneticists have turned to twin studies many times in
the past for light on the roles played by heredity and environment, but
the current project promises to be especially valuable because (or) il lists
more advanced statistical techniques to estimate the strength of hereditary
factors and to select tests that most clearly bring out hereditary influ-
ences; (6) it uses a relatively large sample of twins (about 1 20 pairs in Ihc
baby study and 250 pairs in the high school study-numbers that may he

512

considerably increased if plans foi expansion go through), thus strength-
ening the validity of the findings; and (c) unlike most eailier studies, it
uses a reliable technique for distinguishing between fraternal and identical
twins, another factor adding to the validity of the results.

Obstetricians and patents often make mistakes in identifying types of
twins, but analysis of blood groups -the technique used in this study-is
reliable at least 90 percent of the time. This means that in possibly 1 pair
of fiaternal twins out of 10, both twins will have the same blood factors
and will appear on this basis to be identical. Such cases can virtually
always be classified correctly, however, by differences in the structure of
the heads, hands, and teeth, in the shape of the ears, or in the color and
shape of the eyes. (In a doubtful case, absolute proof could be offered
only by the skin-graft test.)

The study of baby twins was instituted by Frank Falkuer, M.D., now
head of the Department of Pediatiics in the University's School of Medi-
cine, to obtain fuller information on how heredity affects physical devel-
opment. Dr. Falkner brought in Dr. Vandenberg, who added the study of
high school twins. In addition to NIMH research grants to Dr. Vandenberg
and his appointment as a research career development investigator, the
work as a whole has been supported by other NIH and USPHS grants' and
by a small National Science Foundation grant. Under a new grant from
the Institute of Child Health and Human Development, Dr. Vandenberg
hopes eventually to establish a Center for Human Behavior Genetics,
where additional studies would be undertaken.

Before going to Louisville in 1961, Dr. Vandenberg had already done
considerable work on heredity and behavior, principally as director of
psychological studies for the hereditary abilities study of the University of
Michigan, which NIMH and the McGregor Fund of Detroit supported.
That project, too, studied twins-82 pairs, mainly from high schools. The
grantee's findings discussed in this report are based in part on preliminary
results from the Louisville study of high school twins, but in greater part
on recently published reports of the Michigan project and on new analyses
by Dr. Vandenberg of data from a review of other twin studies.

The findings may be summed up as follows:

1. Heredity influences human behavior in many of the aspects com-
monly measured by psychological tests. Its influence shows up strongly
even in some types of behavior, such as performance in arithmetic and
spelling, in which people have had a good deal of training.

2, Four of the six factors found by the late Dr. L. L. Thurstone to be
essential aspects of intelligence and included in the widely used Primary
Mental Abilities test are under strong genetic control. These are the
factors making for the abilities described as numerical, verbal (under-
standing of words), spatial (the ability to perceive patterns and structures
and to move them around in the mind, keeping the relationships intact),
and word fluency. Two other abilities included in the test, reasoning and

1 RG 5527, A 4847, H 7233, HD 482.

513

memory, do not show a strong hereditary influence-quite possibly be-
cause the test does not deal with enough of the factors comprising them.
Memory for names, as an example, is included in the test, but it has been
found unrelated to memory for faces, which is not. Further, a person's
willingness to memorize the items during the testing session is highly
influenced by his attitude towards the testing, so the test measures his
motivation as much as his memory.

3, Motor skills decidedly affected by heredity include dexterity in
using tweezers to place small pins into holes, the ability to soit cards fast
and accurately, and the ability to walk a beam without losing balance. But
tests of hand steadiness and body sway give insignificant results.

4. In the area of personality, one group of tests shows a high degree of
heritability for the factors of activity, vigor, impulsiveness, and sociabil-
ity—but not for dominance, stability, or reflectiveness. Other tests indicate
a relatively strong hereditary influence upon neuroticism, the display of
nervous tension, and control of the will; and there is some evidence that
the degree to which a person is an extrovert or an introvert is partially
determined by heredity. However, most measures of personality traits fail
to disclose any significant genetic role for them -evidence, Dr. Vanden-
berg believes, not that personality is little affected by heredity but that
present tests are measuring traits primarily affected by environment. He
points out that such tests are right and good for most applications but not
for studies of hereditary factors.

5. Tests of musical ability and preference have been even less successful
than those of personality in pointing to a hereditary factor. Perhaps only
the exceptional talent of great composers and musicians has such a factor,
the investigator suggests. Perhaps, in the case of the musical and certain
other abilities, heredity studies might make greater headway if they
looked for innate defects in basic skills rather than for innate profi-
ciencies.

6, Some of the ways in which the body reacts to stressful situations
appear to be influenced by heredity. In a laboratory situation, most of the
twins in the Michigan study responded to mild stress— induced by a sud-
den flash of light and the fall of a hammer-by changes in (a) the rate of
the heartbeat, (b) the rate of breathing, and (c) the galvanic skin resist-
ance. The changes in the heartbeat and breathing rates, but not in the
galvanic skin resistance, were much more alike in the identical twins than
in the fraternals, indicating heritability factors at work in the first two
responses. The grantee would like to see more research along this line
because he believes that the evidence for a genetic factor in some types of
responses by the autonomic nervous system may also be partial evidence
for a genetic factor in psychosomatic illness. The person who naturally
responds to stress by an above-normal increase in his heart beat, Dr,
Vandenberg asks-may he not be the one most likely to develop a psycho-
somatic heart ailment? Other investigators have suggested that persons
with different personalities develop different kinds of psychosomatic com-
plaints. Hence, Dr. Vandenberg believes, research might well find a close

514

relationship among illness and two factors controlled in part by heredity-
personality and the activity of the autonomic nervous system.

Fraternal and Identical Twins: Analyzing the Differences

Since a pair of identical twins have developed from the same feitHized
egg, they have the same heredity, and any difference between them must
have been caused by environment alone. But a pair of fraternal twins,
having developed from two fertilized eggs, have different heredities; hence
the differences between them have been caused by both heredity and
environment. Through statistical analysis of these differences, investiga-
tors are able to say whether a given aspect of intelligence (for example,
facility in using numbers) appears to be strongly or weakly controlled by
heredity.

Like earlier researchers, the Louisville gtoup assumes that the differen-
ces in the environment of a pah- of twins are roughly the same foi identi-
cal as for like-sexed fraternal. The investigators know that this is not quite
true. For one thing, the difference in the prenatal environments of identi-
cal twins, taken all together, is probably greater because of the greater
possibility of an imbalance in placental circulation. For another, identical
twins are more likely than fraternals to be treated alike by their families.
The Louisville studies may eventually help to show whether or not these
differences are significant. At the most, Dr. Vandenberg believes, they
make for either a slight overestimate or a slight underestimate of hered-
ity's role.

In analyzing the differences between the two types of twins, the grant-
ee uses an index based on a statistical concept (the F-ratio) developed by
the British statistician, R. A. Fisher. For each test, he subtracts Johnnie's
score from Jimmy's, Ronald's from Donald's, Ellen's from Helen's, and so
on, Then, for each kind of twins, he adds the squared differences and uses
the results to get an index of the strength of the hereditary factor. The
index, or F-ratio, is obtained by dividing each sum by the number of
pairs and then dividing the figure for the fraternal twins by the figure for
the identical twins. The result is almost invariably greater than 1, indi-
cating that the wi thin-pair variance of the fraternal twins is greater than
that of the identical twins. If the ratio is statistically significant-meaning
that the probability it could have occurred by chance is very slight-the
particular ability being tested probably has a strong hereditary element.

By this technique the contribution of environment to the expression of
a particular ability is subtracted, loosely speaking, from the contribution
of heredity. The method is most reliable when fairly large numbers of
twins are studied. The findings for a group, since they give th

the chaiacteristic is expressed, if at all, may depend in large part upon the
environment. He has found that tests commonly used for measuring basic
mental abilities give much the same results whether administered to
American college students or to young people who, though going to col-
lege in the United States, have been brought up in quite different cultural
backgrounds-South American and Chinese. This finding is considered by
Dr. Vandenberg to be evidence that the abilities measured by the tests are
cross-cultural and perhaps innate.

In spite of their value, Dr. Vandenberg describes twin studies as only
"an economical first step in human behavior genetics," They are only a
fiist step because they can be used to deduce what traits are inherited but
not how. Eventually, when enough reliable tests for assessing the influence
of heredity are available, researches will have to look for the how by
undertaking family studies. They will have to search for the appearance,
over generations, of behavioial characteristics related to psychological
functioning much as medical geneticists have searched for evidence of
hereditary factors in phenylketonuria, hemophilia, diabetes, and other
physical disorders.

"If we remember that a rather clear-cut and, one would think, rather
noticeable defect such as color blindness has been known to science for
little more than a century," Dr. Vandenberg observes, "it does not seem
unreasonable to expect that we are overlooking other anomalies,11 As a
possible case in point, he notes the suggestion by a Swedish investigator
that severe reading difficulties may stem from a hereditary factor con-
u oiled by a single gene.

If an inherited characteristic is harmful, perhaps we can learn how to
prevent its full expression, as we have learned how to prevent phenylketo-
nuria from leading still fuither to permanent mental retardation.
Emotional illness, however, is a far more complex problem than
phenylketonuria, partly because many genes may be involved instead of
only one, and partly because the biochemical mechanism can be fully
activated, perhaps, only by certain stressful environmental conditions.

Of the present psychological and biochemical measures, the giantec
believes we shall probably find that no single one is clearly related to such
complex conditions as schizophrenia and obsessive compulsive neurosis.
More likely > statistical analyses of the results of many tests given to large
numbers of patients will eventually succeed in isolating a variety of factors
involved. Efficient tests of these factors can then be developed.

If the influence on behavior of heredity, environment, and the two of
them acting together is to be clearly understood, Dr. Vandenberg suggests
that we may need at least a nationwide system of storing information
about individuals and families. But we can start on a smaller scale. For
example, we can go after proof of the theory that schizophrenia has a
rather strong hereditary component. The grantee thinks that a carefully
planned 10-year study of several thousand patients and their family
backgrounds-a research project in connection with an existing service
project-would come up with a definite answer. It might even supply an
inkling of the genetic mechanism.

516

Tests of Intelligence

It is widely agreed that heredity plays a huge role in intelligence, but
intelligence is many things. In fact, one authority attending the recent
Conference on Research in Human Behavior Genetics (supported by the
Institute through a grant to Dr. Vandenberg) reported that he had pre-
liminary evidence of two entirely different kinds of intelligence, one de-
pending mainly on heredity and the other on culture. Dr. Vandenberg
prefers to say that certain of the factors 01 abilities comprising intelligence
can be developed through training to a much greater extent than otheis.

Psychologists do not agree on the number of diffeient abilities contri-
buting to intelligence-more than 100, by one estimate-oi on the exact
nature of many of them. However, five of the factors in the frequently
used commercial test referred to earlier, the Primaiy Mental Abilities test
(PMA), are generally considered to represent in fact distinct chatacter-
istics. This is because, on the average, the scores made in tests of any one
of these five abilities-numerical, verbal, spatial, word fluency, and
memory— do not conelate highly with the scores made in tests of any of
the others. As previously noted, Dr. Vandenberg has found that the first
four of these abilities are not only independent but also strongly influ-
enced by heredity.

The grantee began by administering the standard version of the PMA as
an important part of his tests, but this year he is using an expanded version
drawn from Dr. Thurstone's original material. The new version covers
perhaps 14 separate abilities. The complete Louisville battery includes a
number of other "intelligence" tests, and several tests of personality—
about three dozen tests in all, some widely accepted, otheis expeiimental.
The tests are made up of about 1 25 measures, 01 subtests,

Through a method similar to factor analysis, the investigator can infer
whether or not the differences between fraternal twins on a number of the
tests can be attributed to the same hereditary factor. He can also estimate
what percentage of the variation within all the pairs of fraternal twins on,
say, the verbal ability factor, is accounted for by a given test. Thus he can
combine and weigh-and discard and add to-tests until hehas put together
the battery that makes for the best possible separation of fraternal and
identical twins. In sum, when he is considering many factors of intelli-
gence and personality, he can say which ones appear to be under strong
genetic control, and, when he is considering many tests, which ones most
clearly differentiate between the two types of twins, and therefore are of
most value to studies of the heritability of psychological characteristics.

A number of provocative findings about psychological tests as indi-
cators of inherited characteristics were turned up by the University of
Michigan study with which the grantee was connected.

Close to half of all the psychological measures used in that project, but
less than a third of those in the area of personality, Dr. Vandenberg has
found, gave statistically significant evidence that the characteristics being
measured had a hereditary component. Sometimes the heritability esti-
mate was strongly influenced by the type of measure being used. For

517

instance, one vocabulaiy test calling upon the subject himself to supply
the definitions yielded results showing a decided hereditary influence,
while two othei vocabulary tests, one giving a choice of answers and one
asking the students to match words with pictures, produced opposite
icsults. The Michigan group found half a dozen instances of this sort, and
conceivably would have found more if each ability or trait had been
measured by two or more tests. It also found some characteristics that
showed a stiong hereditary component but have little or no practical
application-so far as anyone knows now. One of these is the ability to
trace a pattern while watching one's work in the mirror. Dr. Vandenberg
cites these findings as evidence of the need for wider experimentation in
psychological testing if the hereditary elements of behavior are to be
pinned down. Tests that work well in predicting success in certain studies
or professions, he suggests, are not necessarily the best ones for research in
genetics. It may well be that some inherited psychological differences are
lying undiscovered because they have no applied usefulness or because we
have not been measuring them.

Of paiticular interest to the grantee right now are the spatial tests
because he thinks spatial visualization, an important ability in engineers,
may prove to have a stronger hereditaiy element— and therefore depend
less on training— than most other facets of intelligence. (Present tests of
this ability include putting a puzzle together in the head, imagining what a
given object would look like if cut in half, finding a given pattern among
other patterns, and saying where the holes will occur in a piece of paper
when it has been folded several times and punched.) A test of vocational
preferences also seems to get at tilings having a hereditary basis-a finding
that probably can be traced to the influence on occupational choice of
tempeiament and ability, which are to some extent genetically governed.

Another aspect of intelligence, or perhaps of both personality and intel-
ligence, seems to be closure, an ability to round out or complete things,
Some types of brain-damaged persons lack it. In the closure tests used by
Dr. Vandenberg, developed by the Canadian, C. M. Mooney, the twins are
presented with incomplete drawings of faces and asked to identify each
face as male or female and say in which direction it is looking. Though
closure is at present not known to be related to any other ability, some
people do better than others on such a test. Closure items have been
included in intelligence measures, Dr, Vandenberg explains, mainly be-
cause they do not seem to be influenced by verbal ability, a factor
strongly affected by environment and generally difficult to control for.
Dr. Vandenberg guesses that high scores on closure tests perhaps indicate
some artistic ability, but he may have to wait a long time befoie lie
knows.

In experimental use, as well, is a test of the long-held theory that some
people react more to color and others more to form, the difference being
traceable to a deep-seated personality difference. Preliminary findings by
Dr. Vandenberg indicate that scores on the form-color test, developed by
Thurstone, show a marked hereditary influence.

518

Tests developed by J. P. Guilford to measure various aspects of creativ-
ity are also used. (One asks the student to suggest, among other items,
new uses for old newspapers and two improvements each to various social
institutions, such as marriage.) Whether or not the tests really do measure
creativity in the common sense of the word is in question. Pieliminary
Louisville findings indicate that whatever they measure is not under strong
hereditary control.

Among the measures is one developed by the grantee himself while
working with the Schizophrenia and Psychopharmacology Project of the
University of Michigan and Ypsilanti State Hospital.2 In this test a person
looks at pictures of faces and chooses a woid that best describes their
expressions. One of the major characteiistics of schizophrenia, Di.
Vandenberg explains, is a loss of social sensitivity of social perception,
marked by a withdrawal from other persons and a lessened ability to
communicate. As used in the Michigan schizophrenia study, the test ap-
parently provided a measure of this loss; schizophrenic patients performed
significantly worse than nonschizophrenic mental patients, while the non-
schizophrenic patients did almost as well as normal persons. Perhaps the
schizophrenic cannot make the kind of social judgment demanded by the
test because he cannot bring himself to look at the pictures long enough;
perhaps some other disability is involved. In any event, Dr. Vandenberg
thinks that the inability of schizophrenics to make such a judgment is
related to their tendency toward withdrawal. Preliminary findings of the
Louisville twin study indicate that social sensitivity as measured by this
test is more influenced by environment than heredity.

Heredity and Handedness

Dr. Vandenberg is much interested in handedness, a psychophysiologi-
cal characteristic that may well be controlled by only a few genes. If just
two alternative genes are involved, for right-handedness and for left-
handedness, a person who received two doses of the one would be right
handed; two doses of the other, left handed; one close of each, either right
or left handed, depending mostly on the training received from his
parents. A number of right-handed persons, the investigator believes, are
perhaps naturally left-handed or ambidextrous~but don't know it.

The biggest difficulty in studying the heredity of this trait is the lack of
a good test for natural handedness. A simple test for brain dominance
might be the answer, because a person whose left hemisphere was shown
to be dominant might be expected to be naturally right-handed. But +hth™

illumination reaches one of the eyes through a filter The shorter axis
the apparent ellipse is greater when the filter is held in front of the m<

dominant eye.

The grantee is concerned with handedness, he explains, because infr
mation about its genetic mechanism would increase interest Jn-ai
provide another model for-research in the heritability of more "psych
logical" characteristics. Besides, it would be useful in answering that o
question about the effects of forcing a naturally left-handed person ]
become right-handed. In tests involving motor skills, incidentally, pc
formance with the right hand generally shows a stronger hereditary infli
ence than performance with the left.

Genetic Control of Physical Characteristics

The Michigan study, like the Louisville project, to some extent wa:
concerned with the degree of hereditary control over physical as welt a(
psychological development. Recently Dr. Vandenberg compared the
Michigan findings in this respect with those of five other twin studies since
1926 and found a high degree of heritability for virtually all the measures
that had been included. Among these were height, arm length, middle
finger length, leg length, foot length, chest girth, neck girth, head length
and breadth and girth, eye spacing, nose height and breadth, face length,
and ear breadth. Measures of length generally had higher heritability
values than measures of width-perhaps, the grantee notes, because tlie
latter are more affected by "such environmental excesses" as obesity and
malnutrition. Perhaps for the same reason, the studies did not agree as to
the influence of heredity upon the waist measurement and upon weight,

When twin differences in body measurements were related to twin dif-
ferences on certain intelligence tests, the Michigan group found that the
larger the head, the higher the test score. This relationship was statistically
significant among the identical twins but weaker and not significant
among the fraternals. Among the general population, Dr. Vandenberg
points out, the relationship is obscured by other factors and cannot be
observed. It is only when differences in age, sex, background, and educa-
tion are controlled, as they can be in comparing twins, that the relation-
ship shows up.

Among the fraternal twins studied in the Michigan project, the one who
had been born first was found usually to have made the higher score, aiuE
there was a tendency among both types of twins for the one who had
been born first to have the larger head.

Some Research Needs

Dr. Vandenberg considers research looking toward the following goals*
some of them indicated earlier, to be highly important:

• A better understanding of the basic elements making up intelligence
and personality.

520

» Better information about the characteristics measured by present
psychological tests.

• New tests or combinations of tests that bring out clearly the heredi-
tary elements in our abilities and personalities.

• Better tests of motor skills in children, particulaily those under six.
(At least during babyhood, the giantee observes, motor development and
mental development are related, and he points to work by another NIMH
grantee indicating that disturbances of motor development in infancy may
be followed by psychological disorders, including schizophrenia.)

• Better evidence than we now have as to the role, if any, played by
hereditary factors in schizophrenia and other mental disorders.

• Better information on how learning or training interacts with hered-
ity in the development of a given characteristic. (As an example of how
this might be obtained, the grantee suggests that one twin of a 3-year-old
pair of identical twins might be given intensive training in reading, the
twins then being followed up to learn whether or not any lasting differ-
ence resulted.)

Dr. Vandenberg remarks that while he himself is primarily interested in
the contribution of heredity, complete understanding of how human be-
havior is controlled by genes will be reached only through the work of
scientists in many fields. Studies of the role of the environment are just as
important as studies of heteditary factors,

Research Grants: MH 6203, MH 7033, MH 7708, MH 7880, K3-MH 18,382
References:

Clark, Philip J., Vandenberg, Steven G., and Proctoi, Charles H. On the relationship of

scores on certain psychological tests with a number of anthropometric characters

and birth order in twins. Human Biology, 33(2), 1961.
Cohen, William, Vandenberg, Steven G., and Falkner, Frank. Aims of Louisville Twin

Study. Report No. 2, Child Development Unit, University of Louisville, 1962.
Fish, Barbara. The study of motor development in infancy and its relationship to

psychological functioning, A merican Journal of Psychiatry, 1 17(12), 1961.
Button, H. Bldon, Vandenberg, Steven G,, and Clark, Philip J. The hereditary abilities

study: selection of twins, diagnosis of xygosity and program measurements.

American Journal of Human Genetics, 14(1), 1962,
Vandenberg, Steven G. Behavioral methods for assessing neuroses and psychoses. In:

Uhr and Miller, eds. Drugs and Behavior, New York, Wiley, I960.
Vandenberg, Steven G. The hereditary abilities study: hereditary components in a

psychological test battery. American Journal of Human Genetics, 14(2), 1962.
Vandenberg, Steven G. How "stable" are heritability estimates? A comparison of

herittibility estimates from six anthropometric studies. American Journal of

Physical Anthropology, 20(3), 1962.
Vandenberg, Steven G. Innate Abilities, One or Many? A New Method and Some

Results. Report No. 3, Child Development Unit, University of Louisville, 1963.
Vandenberg, Steven G. Contributions of Twin Research to Child Development. Report

No. 5, Child Development Unit, University of Louisville, 1964. Draft of chapter

from Lipsitt, L, P., and Spiker, C. C., eds. Advances in Child Development and

Behavior II, New York, Academic Press.
Vandenberg, Steven G. Multivariate analysis of twin differences. Methods and Goals in

Human Behavior Genetics, New York, Academic Press, 1964,
Vandenberg, Steven G, The Primary Mental Abilities of South American Students, A

Comparative Study of the Stability of a Factor Structure. (In press.)

521

Vandenberg, Steven G , and Mattsson, Eira. The interpretation of facial expressions 1

schizophrenics, other mental patients, normal adults and children. Acta Psvch

logica, 19,1961
Vandenberg, Steven G., Clark, Philip J., and Samuels, Ina. Psychophysiologicai Rea

tions of Twins- Hentabihty Estimates of Galvanic Skin Resistance, Heartbeat at.

Breathing Rates. (In press.)

Investigator:

Irwin Feinberg, M.D.*

New York Downstate Medical Center

Brooklyn, N.Y.

Co -contributors:

Howard Roffwarg, M.D.,

Montefiore Hospital, New York City

H, W. Agnew, Jr.,

University of Florida

Anthony Kales, M.D.,

University of California at Los Angeles

D. R, Hawkins, M.D.,

University of Virginia

Arthur Parmelee, M.D.,

University of California at Los Angeles

Prcprared by:

Gay Luce

Background

From the darkness of the womb the newborn child only slowly
emerges: As if incapable of plunging into consciousness for too long at
once, he sleeps and awakens in biief alternations. At the other end of life,
as if reluctant to let go of consciousness for too long at one time, the aged
brain shows a sleep that is punctuated with awakenings. The depth of
infant sleep and the alterations brought by youth, adulthood, and age
reveal, like indirect mirrors, changes that inevitably take place in the
nervous system during the span of a lifetime.

At birth the nervous system is incompletely developed, The brain
triples in weight after birth, reaching almost adult proportions by about
age 6. During these preadolescent years the brain has its greatest plasticity
and presumably the greatest capacity for learning. The plasticity of the
brain can be inferred by the ability of very young brains to transfer such
functions as speech after significant damage or surgery. During adoles-
cence this plasticity diminishes, and hormonal events make their great
impact on the sexually maturing human being. Later, after the plateau of
maturity, the middle-aged hormonal system again begins to change. An

*Now at Veteran's Hospital, San Francisco, California

523

individual may then begin to notice in himself a trend that has actu;
transpired throughout his adulthood: It has become more difficult for J
to change his habits, to learn a language, to acquire the skill of a n
sport, or adopt a new profession. Nor is he sleeping as soundly as
youth. In old age, finally, this trend accelerates, and the individual beg
to find that he can no longer practice his old skills and no longer
members what once he knew. His sleep, punctuated by wakeful momeii
knits night and day together in a prolongation of consciousness.

The relationships between the underlying brain physiology and lifetii
behavioral development are now being explored by using the electi
encephalogram (EEC) of sleep to depict the sleep patterns of each a|
from birth to senescence.

The overall picture of lifetime sleep, as compiled by Dr. Feinberg ar
his associates, also includes research conducted by other NIMH grantee
who have concentrated upon particular age groups. In correlating Hfetiir
behavioral changes with sleep patterns, the researchers expect that the;
sleep-behavioral relationships may offer clues about the biological functio
of sleep and its various stages. Knowledge of sleep patterns is, moreovei
of clinical importance. Dr. Marvin Schultz of UCLA has demonstratei
that sleep patterns can now be used to confirm very early diagnoses o
illness such as hypo thy roidism or of retardation in infants, A knowledge
of sleep patterns is only beginning to reveal why elderly people complaii
of insomnia, and what, indeed, this insomnia looks like and augurs. At
understanding of nighttime insomnia and delirious wanderings in the
senile may finally lead to the development of rational therapy, rather thar
the current practice of liberal and unconsidered drug use for insomnia and
of institutionalization for the senile. Base line studies of sleep, at different
ages, now offer a yardstick against which researchers are looking at the
effects of hypnotics and other drugs, and from which they are judging the
impressions the various illnesses mark upon nighttime sleep. The base line
of sleep patterns— for life-as compiled by these investigators will un-
doubtedly be refined again in the near future. It is nonetheless the founda-
tion, the bedrock, on which many clinical studies of sleep will increasingly
rely.

The formulation presented here he is not solely the substance of one
laboratory, but includes the data of other research teams, and offers a
reformulation of the age-sleep picture.

The Sleeping Population

For half a dozen years, many infants born in the bright new hospital at
UCLA have been subjects for Dr, Arthur Parmelee and his associates,
while other infants who were born at Columbia-Presbyterian Hospital in
New York were observed by Dr, Howard Roffwarg and his associates. In
one of the laboratories at UCLA, infants are being followed inio the
preschool and early school years, with sleep records and testing at regular
intervals. The data contributing to the base line for childhood have been

524

collected for a variety of reasons For instance, Di . Anthony Kales and his
associates compared normal and somnambulistic childien, while Dr. David
Hawkins, then at the University of North Carolina, and Drs, Chester Pierce
and Roy Whitman of Cincinnati were interested in enuresis. Dr, David
Foulkes has studied sleep in young school children, moved by particular
interest in their dreams. In the course of buigeonmg sleep research, adoles-
cents, young adults, and middle-aged people have become the subjects of
diverse studies. Dr. Wilse Webb and H. W. Agnew, Jr., at the University of
Florida, and Dr. Anthony Kales and his associates at UCLA have studied
normal sleep in some elderly subjects for the purpose of obtaining a norm.
Dr. Charles Fisher at Mt. Sinai, and Dr. Ismet Karacan now of the Univer-
sity of Florida, were correlating the occurrence of erections with rapid-
eye-inovement sleep throughout the lifetime, and were thus moved to
examine the sleep of quite elderly men.

This laboratory conducted its own studies of young adults, normal
older people, and deteriorated elderly people out of an interest in the
relation between sleep patterns, intellectual function, and age. As there
began to be data of disparate sorts from laboratories all around the
country and also in Europe, the investigators were challenged by the need
for a summary picture, a lifeline of sleep. They added to their own data a
study of sleep in children who were 5 to 10 years old. This filled in the
picture of sleep at a time of life when the brain is physically almost
mature, yet at its most plastic.

It is difficult to obtain normative data on human sleep, and while a
lifetime scale of sleep patterns promises to become an invaluable diag-
nostic yardstick, it will be hard to acquire. The difficulty in amassing
such a yardstick can be seen from the human investment in a single study.
Dr. Feinberg and his associates have given the better part of 4 years to
their study. Yet they studied only 38 normal people and 15 abnormal
elderly persons. The age span was 5-96. It would be hard to make these 38
people evenly represent those 91 years. Ideally, a graph of changing sleep
patterns would evolve from records taken from several normal people, one
group of subjects for each year in the lifespan. Discounting the time it
takes to find and acclimatize subjects, discounting the tediousness of con-
secutive nights of sleep recording, it would probably take 20 man-years to
run the ideal study of 400-odd people-five for each year under study.
Some researchers are indeed studying numbers of subjects representing a
few years. For obvious reasons there is no wholesale endeavor to encom-
pass the lifetime. The approximate graphs drawn from small and scattered
samples are apparently adequate to outline the relation between sleep
patterns and age.

The Volunteers of This Study

The young adults and elderly patients were studied at St. Blizabeths-a
huge Federal mental hospital located in the southeast portion of Washing-
ton, D.C. There, a special ward was established within a modern building,

525

where patients and noimal outsiders might live while their nightly sleep
was being studied. Nine young men and six young women were recruited
from among the hospital nurses and aides at the hospital and persuaded lo
sleep in the laboratory for five or six nights in succession, and to comply
with rules by not drinking alcohol or napping on those days. These people
weie between 1 9 and 36 yeais of age.

Normal Elderly

Nine men and six women who were between 65 and 95 years old-ati
average age of 77-were studied on a volunteer ward at the large modern
hospital on the NIH campus in Bethesda, the Clinical Center. These un-
usual people were members of a club called the Fossils, a wry veision of
the Golden Age clubs for retired and elderly persons. In contrast with the
young adults, these people were comfortable suburbanites, retired profes-
sionals, and generally cultured and highly educated people. Nobody has
been able to ascertain that educational level in any way alters sleep pat-
terns, but the good health of these older people may indeed be attribut-
able to their social class. This group was thoroughly screened for even
premonitory signs of impaiiment, and each individual was brought into
the research ward for 5 days of acclimatization before the sleep studies
began. These people also observed the no-napping and no-alcohol rules.

Chronic Brain Syndrome

Although this group, studied at St. Elizabeths, was matched for age
with the Betliesdans, it presented a sad contrast between a healthy old age
and old age attended by severe brain impairment. These were intellec-
tually damaged people, some of them professionals retired from exceed-
ingly responsible jobs, others were firemen and small businessmen. At the
time of the study they were incompetent to take care of themselves, sonic
wandered around in a disoriented manner, and all needed institutional
care. Most of them were not comparable with their healthy Bethesda peers
in either education or economic status, although the research team had
tried very hard to recruit volunteers of equivalent cultural status.

Children

Eight boys and girls, aged 4'/£ to 61/2 and 9 to 10 years, slept in the
laboratory atop the Downstate Medical Center's large new hospital, a
laboratory that is often referred to as the Dream Lab. The children were
brought by their parents to the laboratory before bedtime, coming on
several nights to accustom themselves to the laboratory, and for three
nights of recording.

526

All of the subjects, old and young, weie carefully and lullingly deco-
rated with electiodes around the scalp and face (in standardized place-
ment) before letiring. They were whed up to the electroencephalograph
amplifier system, through a cable system permitting them to sleep in a
private room, where they were undisturbed by the comings and goings of
researchers in the control room. In each hospital, the undeviating hospital
routine dictated the hour at which the sleep subjects had, perforce, to rise.
At St. Elizabeths the rising hour was 6:30, while at NIH and Downstate
the subjects slept until 7 or 7:30.

The High Water Marks

Throughout the night, changing brain waves and physiological functions
leave a sea of data, marked by only a few distinguishable tides, rhythms
that are recurrent and obvious. Within this sea of data, the investigators
selected certain intervals to act as the high water marks of the night. How
long did the volunteer sleep? How often did he awaken from sleep? How
[ong did it take him to fall asleep when the lights went out? How long did
it take to reach the first rapid-eye-movement period, and to show not only
the irregular low-voltage brain waves but also an eye movement? How
much of the night did he spend in REM sleep, and how much in slow-wave
sleep?

The investigators subdivided the usual EEG categories of sleep into
more refined intervals which they defined quantitatively. When they
looked for the amounts of deep delta sleep (stages 3 and 4), they would
count the number of slow waves of at least 50 microvolts in a 20-second
interval: They called it stage 3 if there were between 1 0 and 1 5 waves, and
stage 4 if the interval contained more than 15 such waves, They also
examined the EEG for bursts of activity that visually resemble wire
spindles, bursts that last about half a second or moie, and contain 12—14
cycles a second. This is a configuration that does not occur in the EEG of
sleep until after about age 3 months. It is reduced or absent in old age.

The investigators looked at intervals of rapid-eye-movement sleep in
several ways. Within the REM period are many moments when the eyes
are not moving at all, and the EEG shows a pattern of light, low-voltage
sleep. The researchers looked at density of eye movements. They meas-
ured the intervals of stage 1 sleep preceding and succeeding the eye
movements, and the percentage of the night's sleep spent in this variegated
REM period. They measured the night's total REM sleep. They looked at
the amount of time an individual slept, next to the amount of time he
spent in bed. They looked at the periodicity of nightly events-the timing
of recurrent stages of EEG. Their breakdown differed from that of various
colleagues, but the emerging picture from various laboratories has been
amazingly consonant.

527

The infant

In J955 Drs. Eugene Aserinsky and Nathaniel Kleitman observed that
infants in their first 7 months alternated predictably between active and
quiet periods in sleep, roughly every hour. Eight years later Dr. Howard
Roffwarg was able to report in considerable detail what happens during an
infant's sleep. Using tiny sensois they were able to track the heart rate,
the brain waves, the respiration, and muscle tone of the newborn baby as
he slept. A year later Dr. Arthur H. Parmelee and his colleagues at UCLA
confirmed these findings from their own detailed study of infants begin-
ning on the first day of life. During the quiet phase of sleep the infants
showed a tegular respiration and scarcely any body movements, but dur-
ing the active sleep that has resemblance to REM sleep, the infants smiled,
kicked, grimaced, made eye movements and sudden gestures with their
arms or legs.

For about his first 3 months, an infant naps. Only slowly does he begin
to sleep primarily at night. As infants have been recorded and observed
around the clock, during their first week of life, it was clear that there
were already some important individual differences in the amount of sleep
and the length of the baby's cycle. One infant would sleep for 40-mmute
intervals, 16 times a day, while another would sleep in 14 longer periods.
The UCLA laboratory has studied about 25 infants between birtli and 3
months, and found that in this period babies generally show a sleep cycle
of about 40-45 minutes, which lengthens as the infant matures.

Although many people imagine that infants sleep most of the time and
that they are awake more as they mature, this is not really the major
change in the first 2 years of life. A newborn infant who sleeps 1 4 hours a
day in his first weeks, may still be sleeping about 1 2 hours a day at age 2
years. The primary change is in the distribution of his sleep and waking:
He begins to sleep continuously through the night and be awake by clay,
in the almost 24-hour rhythmic pattern of his parents.

The EGG stages of sleep are shifting in their periodicity too. The brain
waves of a premature or full-term infant are not so coherent and defined
as those of the child or adult. EGG studies by Dr. Parmelee and his
associates at UCLA, and independent studies by Drs. Howard Roffwarg,
Joseph Muzio, and William Dement suggest that the infant spends about
one-third of his existence in a state resembling REM sleep. Premature
babies show even greater proportions of this sleep. In the first days of life,
this stage occupies about 50 percent of the baby's sleep, declining as the
infant's nervous system matures, The infant, like no normal adult, will fat!
directly from waking into this rapid-eye-movement sleep, so characterized
by dreaming in adults, and by subtle facial expressions and extraordinary
physical activity on the part of infants. Children over IH or 2 years will
sleep for almost 3 hours, and adults will sleep for about an hour before
falling into a rapid-eye-movement state. Although children and adults may
speak out, grimace, or exhibit certain twitches of muscles, they do not
wave their arms or legs, nor thrash about wildly during their REM dreams
as do infants. Signs of exaggerated activity, like those of the infant during

528

REM sleep, are highly abnormal in adults. Indeed, the only adults who
have exhibited as much REM sleep as a newborn infant have been drug
addicts, alcoholics during withdrawal, or people who were experimentally
deprived of this sleep stage on prior nights. Some of these people have
been recorded during a night that was half given over to REM sleep.

Children studied by Dr. Feinbeig and his associates, like children
studied by Dr. Roffwarg, et al., Dr. Kales et al., H. W. Agnew, Jr. et al.,
spent around 25 percent of their sleeping time in the REM state; this was
about the same proportion as the young adults. The amount of REM sleep
declined slightly in the older people, and noticeably in the patients with
chronic brain syndrome, but without greatly changing the relative propor-
tions of REM and non-REM sleep. The impaired old people sometimes
reacted to REM sleep in the manner of an alcoholic suffering withdrawal
symptoms. Three of the patients often awakened from a REM dream with
a start. They would try to pull off their electrodes and leave, saying they
had to make a train, be at work on time, 01 meet a business associate,
They were in effect delirious and had to be restrained from racing out of
the building to do the prosaic enands of their dreams.

It has been conjectured that the intensity of the dream experience may
be detected by the density of rapid eye movements. When a person has
been deprived of REM sleep, he compensates later by indulging in more
than usual REM sleep with more intense eye movements. The paucity of
eye movements in the REM sleep of retarded children is one bit of data
suggesting that the eye-movement activity of REM sleep may be related to
brain metabolism.

Curiously enough, the direction of the dreamer's eye movements
changes with age. Newborn babies and infants make many more vertical
eye movements than horizontal movements in their active sleep. Elderly
people and senile patients were found to make mainly horizontal eye
movements during their REM experiences. Children and young adults
showed both, although their eye movements have been reported to be
primarily horizontal by Dr. J. Antrobus.

Memory of the REM experience also seems to change with age, although
pure memory is hardly what the sleep researcher determines when he
awakens a sleepy child or an old person and asks what he was thinking.
Neither young children nor old people seem to recall— or to be able to
recite— dream experiences as well as young adults.

This has been an observation in a recent study by Drs. Edwin Kahn and
Charles Fisher at Mount Sinai Hospital in New York City. In contrast to
Dr. Feinberg's study, they found no correlation between age and the
amount of REM sleep in 11 old men. This discrepancy may be due to
differences in sampling or in procedure. The elderly gentlemen studied by
Drs. Kahn and Fisher slept with a strain gauge attached to the penis. They
still showed some degree of the erections that are seen in all males during
REM periods, but when they were awakened from REM sleep they had far
less dream recall than a young adult. The investigators conjectured that
vitality and psychological vigor may determine the amount of REM sleep
obtained at a later age in life.

529

Rapicl-eye-movenient sleep has been linked with activity in a primit
portion of the lower brain, the pons. It is not surprising that brain st<
activity might be dominant in premature babies, whose brains are in t
early stages of development. It has been speculated that this rudimenta
braJn activity, occurring in the womb, constitutes a state that would n
be dreaming in an adult's sense, but which stimulates brain experiences
preparation for life. Dr. Feinberg and other researchers have specula t
that the high amount of dreaming found during early childhood is relati
to the enormous amount that a child must learn- for in early life the bra
must absorb more than it ever will later. One emerging theory about RE
sleep suggests that tliis is a time when newly learned material is stored ar
filed in the brain.

The rhythm of infant sleep and waking has been traced back into tl
womb in a recording from a mother during her sleep. Until about
months' gestation, the activity of the fetus was often changing. After
months, there was a regular 44-minute activity period, with a ronght
half-hour rest in between. After delivery, the infant showed the sam
cycle of activity during his first days.

The newborn infant's sleep rhythm only faintly resembles that of cliil
dren and adults, although the EEC of REM sleep even in very youn;
infants bears a striking resemblance to REM and stage 2 in adults. Tin
sleep records have been compaied by three main criteria: The amount
and percentages of REM sleep and slow-wave sleep; the length of interval1
such as the transition from waking into active sleep; the periodicity of tlu
complete sleep cycle.

Striking changes occur in the first 3 months, as the proportion ol
rapid-eye-movement sleep declines in the infant and the baby begins to
sleep at night. The child of 6 months or a year spends about a third of his
sleep in REM dreaming. By age 2 this has dropped to about a quarter.
Somewhere between 3 and 5 years the proportion of REM sleep falls to
about 20—25 percent, which is close to the norm throughout much of
adulthood. The total amount of sleep obtained by a child decreases very
slowly.

Drs. J. Mendels and D. R. Hawkins, at the University of North Carolina,
have begun an extensive study of children from 2 to 16 years in which
they expect to study five children for each year. They find, in their first
recordings, that there is a slow diminution of actual sleep time, propor-
tionately diminishing the time spent in each stage of sleep, but without
changing the proportion to the total night of sleep. A very noticeable
difference between the sleep of young children and adults is the time
spent in a deep stage-4 oblivion.

A child not only obtains more of this slow-wave sleep, but he is much
slower to awaken. While the old person will awaken from sleep with
alacrity, almost with a nervous promptness, the child tends to be a
somnambulist for quite a while, fumbling around sleepily if awakened,
and often drifting back to sleep while on his feet or in the lap of a parent.
Delta sleep, which occupies much of the first 3 to 4 hours of a 2-year-old's
sleep, is the stage from which a child will sleepwalk, and this phenomenon,

530

which is not uncommon among 2- and 3-year-olds, becomes progressively
more rare with age— until a new and different kind of somnambulism
emerges in senility. The senile person perambulates out of REM sleep: He
has no stage-4 sleep.

The youngster between 2 and 10 is particularly hard to awaken early in
the night when he is drowned in the depths of delta sleep. His stage 4
sleep differs from that of his parents not only in the subjective experi-
ence: The delta waves of a child are of ex h emery high amplitude. A
5-year-old spends around 2 hours a night in this deepest forgetfulness,
whereas the adolescent spends only about 75-80 minutes. The fiist nightly
REM period may not appear for 3 to 4 houis in a 7-year-old, whereas it
will arrive within 50-70 minutes after an adult has fallen asleep. The speed
with which a person reaches his first REM dream after falling asleep
becomes more rapid as he matures. The child begins to show an adult
pattern of REM latency around mid-adolescence.

When H. W. Agnew, Jr., and Drs. Wilse B. Webb and R. L. Williams
compared 10-year-olds with teenage youths, they found only slight dif-
ferences in their sleep records. The 10-year-olds slept longer-averaging
around 9*/2 hours. The teenagers slept about 2 hours less. The 10-year-olds
showed a higher amplitude activity on their EGG's than the teenagers. If>
as some researches speculate, the power lepresented by the amplitude of
a biain wave is related to the intensity of sleep, then it may be possible to
watch the intensity of stage-4 sleep decline steadily over a lifetime.

Young Adults

Sleep has been well documented for the young man between 20 and 28.
There are, to be sure, notable differences among individuals, but the gen-
eral picture of a night's sleep in the twenties is by now familiar. The
so-called average person falls asleep, taking about 7—15 minutes. He drifts
down through lighter sleep, stages 1 , 2, 3, into stage 4. Then he drifts back
up toward light sleep and enters a REM dream in 70 minutes from the
time he went to sleep. The first REM dream is apt to be short, in the
neighborhood of 10 minutes. Once more, the person dives into the depths,
rising again so that after about 75 minutes he is ready to dream at greater
length. The dive into slow-wave sleep, which happens about five times,
goes less deep in the last cycle of night. The four last REM periods of the
night run about 20-25 minutes, but the intervals between REM sleeps
have diminished.

Delta sleep, and particularly stage 4, provides a visible demarcation
between childhood, youth, and that point at which adulthood slips over
the edge into old age. Even the young adult receives far less delta sleep
than he did before adolescence or during adolescence. His delta sleep, as
seen on the EEC, is much lower in amplitude. His complete sleep cycles
run between 70 and 90 minutes, yet within the basic pattern lurk many
variations.

531

In one preliminary study of good and poor sleepers there appeared I
be unmistakable physiological differences between them, differences i
body temperature, pulse, vasoconstriction, amount of REM sleep, and t]
distribution of stage 4, as well as marked personality differences. On
young adult in the comparative study of Dr. Fein berg and his associate
showed no delta sleep at all, yet seemed normal during his waking hour;

For the most part, however, a number of lesearchers observe that th
EEC amplitude of stage-4 sleep declines with age. Not only is there les
stage 4 until it disappears altogether, but it diminishes in what might bi
called intensity, and which would be defined on the EEC as prolongec
trains of high amplitude delta waves. The duration of stage 4 diminishe;
toward a vanishing point that may occur as early as the late thirties 01
early forties.

The decline of stage 4 has been observed by Dr. Wilse B. Webb and
H. W. Agnew, Jr., at the University of Florida, They have observed that as
delta sleep declines, sleep may become less continuous, and its fabric
becomes punctured with awakenings. At about age 45 people are likely to
awaken three times in a night. People who were allowed to go on sleeping
in the morning in the Florida laboratory until they were "slept out"
found themselves awakening five and six times toward the end, when their
sleep was largely the light stage 2 and REM. The absence of stage4 sleep
in the aged and increased awakenings were also observed at UCLA by Drs.
Anthony Kales and Allen Jacobson in their study of elderly people,

As Dr. Feinberg and his associates compared children, young adults,
and elderly people, they found that an adult, once beyond adolescence,
does not sleep longer at age 20-30 than his counterpart of 70-80. The
young adults recorded at St. Elizabeths, who were between 19 and 36,
slept about the same amount of time as their counterparts in Bethesda,
the aged members of the Fossils Club. However, they took less time falling
asleep than either the normal older counterpart or the patient with brain
damage. They did not awaken so much at night. The older person
obtained an equivalent amount of sleep only by staying in bed a longer
time. The person with chronic brain damage, who took a long time falling
asleep, did not really obtain as much sleep as his peer in age. The normal
older person in this study awakened so often that he spent 17 percent of
his time in bed wakeful. The young adult spent his supine hours asleep,
but the institutionalized person with symptoms of senility lay awake fora
third of his time in bed.

The studies conducted by Dr. Feinberg and his associates enumerated
many of the differences between the young and old that make anecdotes
in the conversations of large families. For instance, it is common for an
older person to rise and go to the bathroom once, or even more often,
during the night. Young adults do not, unless they are in exceptional
states: for instance, women in late pregnancy find their sleep interrupted
and do visit the bathroom. A UCLA study indicates that some enuretic
youngsters are never even awakened by their need. On the other hand,
some children with enuresis present problems, for even if they awaken

532

with a need, they would rather wet a bed and return to sleep than get up
out of bed.

The sleep iccords obtained from children, young adults, and both the
normal and abnormal older people show piecisely the expected pattern:
Neither the youngsters nor young adults interrupt their sleep to go to the
bathroom during the night, but the older persons rose once on the average
and the senile patients slightly more often.

Quite apart from rising to visit the toilet during the night, adults do
awaken from sleep intermittently. The normal young adult will awaken,
albeit briefly, around three times a night. The awakenings may be so brief
that he will not remember them. A normal older person will awaken five
or more times a night. However, the elderly person with chronic brain
syndrome will awaken eight or nine times a night.

Perhaps some of these differences in the number of sleep interruptions
stein from the different arousal speeds of the young and old. Youngsters
are typically very hard to awaken, especially during the first half of the
night when they are so much immersed in intense delta sleep. Older
people, by contrast, have less slow-wave sleep, no intense stage 4, and
often awaken like a shot. Children are virtually somnambulistic for a long
time after awakening, but the oldster is alert at once, his brain having
turned off sleep like a faucet.

The insomnia of age is a frequently discussed affliction. Most older
people would appeal to have no stagc-4 sleep and less deep delta sleep.
They are more easily awakened than the young. Indeed, they do take
longer to fall asleep, awaken more often during the night, and spend
considerable time lying in bed sleeplessly. This insomnia has been
observed by Dr. Fein berg and his associates, by Dr. Anthony Kales and his
associates at UCLA, and by Dr. Wilse B. Webb and his associates at the
University of Florida, From childhood on, as the Webb group in Florida
has shown, there is a decrease in delta sleep and an increase in awakenings.
The proportion of the night spent without sleep increases as a person
grows older, but as Dr. Fcinberg and his colleagues have shown, the most
egregious symptoms of insomnia are found in patients with chronic brain
syndrome, far worse than in normal persons of the same advanced age,
The children studied by this team were a marked contrast: They fell
asleep fast and slept soundly and long. The 5-year-olds averaged about 914
hours, while the 10-year-olds averaged about 8!4 hours. The patients with
chronic brain syndrome averaged approximately 5 hours of sleep, while
the young adults and normal older people slept around 7 hours a night.

Although a great many older people complain of the insomnia of their
waning years, they are not accurate judges of their own sleep. It has
become clear in the laboratory, where the EEC record, like the snore of
Hie sleeper, indicates that a person who may think himself awake is not
actually awake. Dr. Feinberg has said, "A lot of people who think they're
awake arc just aware of their mental activity, which to the EEG appears to
be sleep. Therefore they think they are awake."

People do not even estimate very accurately the amount of sleep they
obtain. Dr. Fcinberg asked his elderly volunteers from the Fossils Club

533

how much sleep they usually got at home. "There was zero correlaii
between what they thought and the amount of sleep they did get, Soi
of them would just take the time they spent in bed and say that was h<
much sleep they got. Others would say that they were awake most oft
night but that they'd simply lie there. They were both wrong."

In the lifespan picture of sleep, certain in variances are striking. 0
thing that seemed not to change between ages 5 and 95 was the number
REM periods during a night's sleep. Subjects studied by Dr. Feinberg,
al. consistently showed 4-5 REM periods. The children would sic*
between 2 and 3 hours before their first REM period; the young adul
took about an average of 70 minutes; the aged people about 58 minuti
average; and the chronic brain syndrome patients dropped into REM slet
relatively rapidly— within an average of 46 minutes. The amount of rap«
eye-movement sleep, on the other hand, distributed over the nigh
showed considerable changes with age,

Similarly there was a pronounced reduction in delta sleep-stages 3 an
4- from childhood to old age. Throughout life, whether the sleeper i
child or octogenarian, it appears that most of the delta or slow-wave slee
occurs mainly at the beginning of the night, and there is virtually non
during the last period of non-REM sleep. At that time, the intcus
oblivion seems to have run its course.

When Dr. Wilse B. Webb and H. W. Agnew, Jr., at the University o
Florida invited subjects to return to the sleep laboratory for a nap in tin
morning after a full night's sleep, they found that there was virtually nc
slow-wave sleep and no stage 4 in the naps of these young adults. REM
sleep, on the other hand, occurred more icadily in the morning than it did
at night. The distribution of stage 4 throughout the night might seem
related to some roughly 24-hour rhythm with a peak coming toward the
onset of sleep rather than at the end- But, the decrease of stage-4 sleep
over a lifetime is striking.

The children and young adults exhibited long stretches of rhythmic
delta sleep whereas these slow-wave movements were always interrupted
by fast rhythms in the EEGs of the older persons. Dr. Feinberg has sug-
gested that stage 3, which persists in lessened form into old age, may be a
less intense form of stage 4. In general, the older people showed a flat-
tened EEC tracing, meaning rhythms of lower amplitude. The stage 3 of
the very young, the adult, and the aged remained much the same in
quantity although qualitative differences were noted. Even in the senile,
stage 3 activity was distributed similarly across the night, as if controlled
by some unchanging and inherent periodicity.

age—and the way this life trend can be read in the EEC patterns of sleep.
Although waking EEGs for highly abnormal senile patients often show
little deviation from the normal, sleep EEGs display changes so gross they
cannot be missed. The senile patient, who seems to represent an extreme
of the normal process of aging, shows a sleep pattern that is also an
exaggeration of the changes observed in normal elderly people. The
investigators began to ask whether certain changes in the sleep record
might also correlate with intellectual functioning, memory, and per-
formance on certain tests.

All of the older subjects were given a Wechsler Adult Intelligence Scale,
a Wechsler Memory Scale, and another test. How did these test scores
correlate with particular changes in the prevailing stages of sleep? Among
the senile patients it was possible to see that a person who slept longer, lay
awake less than his fellow subjects, but had more REM sleep, also gave a
higher showing on the intelligence test. The normal older person, who
awakened often from sleep and showed a decline of REM sleep, also did
correspondingly less well on the psychometric tests. Since some senile
persons did so absurdly on the psychometric tests that it was impossible
to rate their performances, an observer was set the task of rating these
patients' abilities to carry out the simple chores of life. The more awaken-
ings and less REM sleep they showed, the less intact they seemed during
these daily observations. Awakenings and REM sleep appeared to be two
indicators that told approximately how much intellectual impairment to
expect in an individual. Among patients with chronic brain syndrome, the
sleep shows a proportionately greater reduction in REM sleep than in
slow-wave sleep.

In earlier work, Dr. Feinberg had noted that lifetime change in the rate
of cerebral metabolism was strikingly similar to the lifetime curves for
total sleep time, and amount of REM sleep. Not all of the changes in sleep
patterns parallel the changing cerebral metabolic rate. However, a decline
in cerebral metabolic rate commensurate with the sleep-changes exhibited
in senility also spells intellectual impairment But subtle changes in
intellect occur earlier in the nonnal elderly person. These changes, which
cannot be detected by measurement of overall cerebral metabolic rate or
in waking EEC, are nevertheless reflected in the EEC of sleep.

One suggestive parallel in the age graph of sleep is that the decline of
stage-4 sleep occurs as a person's mental agility also declines and as it
becomes difficult for him to learn psychomotor skills. The investigators
postulate that in some fashion the intensity of non-REM sleep, particular-
ly as represented by long trains of high-synchronous delta waves, repre-
sents the plasticity of the individual's learning capacity. Stage-4 sleep
declines as this plasticity also declines in age. The relation between the
known sleep stages and an individual's cognitive ability will be the focus

of further study, .

The relationship between cognitive power and sleep patterns in elderly
people may evoke a profoundly different attitude toward some of the
deteriorations we now accept as the usual penalties for growing old. As
one researcher has conjectured, it maybe that the hypenirousability of the

535

aged and their repeated awakenings in sleep cause a reduction in REIv
sleep and even in stage 4. Loss of cognitive power may in some way bi
related to an inability to maintain continuous sleep. As other researcher
point out, "insomnia," disturbed sleep patterns, and intellectual deteriora
tion are not found in all old people. Perhaps these sleep and menta
factors also have a very strong psychological component and relate to tin
sources of depression in so many old people. Perhaps such deterioratioi
should be considered pathological even in its most usual form, instead ol
accepted as normal for a given age. During the Elizabethan era, a person
had normally lost most of his teeth and might be considered old after age
25. We would hardly accept this as normal today.

A very practical outcome of these studies will be felt in the treatment
of the senile. By documenting the changed sleep patterns which occur in
old age and chronic brain syndrome, clinicians are acquiring a rational
basis for treating the insomnia of the elderly. It is necessary to reveise
these age changes in sleep or to alter the sleep schedule so that the effects
are minimized. Dramatic vagaries shown by some chronic brain syndrome
subjects as they awakened from REM sleep suggest that nocturnal
delirium and wandering may result from a confusion of dream and reality
in an impaired cerebrum, or perhaps from the lingering of dreams and
other REM experiences into the waking state. Drugs that reduce the
intensity of REM processes could be valuable in the treating of these
symptoms. Since nocturnal disturbances are often the main cause for
hospitalization, such treatment might allow elderly people to remain in
their homes and communities instead of spending their last days in an
institution. As Dr. Feinberg has commented, "The changes in the sleep
electroencephalogram which occur in normal old age as well as in chronic
brain syndrome, and the correlation of these changes with intellectual
function, suggest that the EEC of sleep may provide a far more powerful
diagnostic tool for geriatric psychiatry than any which has been hitherto
available."

Figure 1. -Total sleep time as a function of
age, In this and subsequent figures, the
crosses represent data points, the dots
represent the best-fitting curve chosen ac-
cording to standard statistical techniques.
The number of subjects contributing to
the mean for each data point is as
follows: 6 years, 4; 10 years, 4; 21 years,
6; 30 years, 9; 69 years, 7; 84 years, 6. A
cubic curve provided the best fit for the
changes in total sleep time with age. This
measure is high in childhood, declines to
a plateau which is maintained during
maturity, and then shows a further
decline in very old age. The slight dip and
subsequent rise shown by the theoretical
curve may be artifactual.

JOE IN YEiflS

536

Figure 2.-Number of awakenings during
sleep as a function of age. Awakening
here refers to changes in the EEC and was
not necessarily accompanied by gross
behavioral arousal- This measure is low in
childhood but shows a steady, linear
increase with age whidh is apparent
during maturity, where total sleep time is
constant.

Figure 3. -Percent of time in bed spent
awake as a function of age. This measure
which is an index of insomnia remains
low throughout life and then shows a
sharp, positively accelerated increase after
age 50 years. The data of Williams et al.,
on subjects in the sixth decade of life,
suggest that the increase starts a few years
later than shown by the present parabolic
curve.

ZO 40 60 BO 100

BGE

Figure 4. -Stage 4 EEG as a function of age.
This measure of the high-voltage, slow-
wave EEG of sleep shows a hyperbolic
decline with age. However, appreciable
change is manifested during maturity, as
may be seen by comparing the values at
age 20 and age 50 years.

Figure 5.— Stage 1 latency as a function of
age. This measure represents the total
sleep time recorded prior to the onset of
the first period of emergent stage 1 EEG
of REM sleep. It shows a hyperbolic
decline with age; part of this decline is
probably related to the declining need for
stage 4 EEG, half of which precedes the
first period of emergent stage 1, The
curve for the onset of another indicator
of REM sleep, the first rapid-eye-
movement, is essentially the same as that
shown here for emergent stage 1 EEG,

40 60

AdE {YEARS!

537

Figure 6. -Total time spent nightly in stage 1
EEC (REM sleep) as a function of age.
This measure shows a hyperbolic decline
with little appreciable change during
maturity. The curve for amount of
rap id-e ye -movement activity is essentially
the same as that for emergent stage 1
EEC.

40 60

AGE IN VEflRg

Figure 7 .—Cerebral oxygen uptake as a
function of age These data are taken
from various sources in the literature and
are based on the Kety-Schmidt method
for measuring oxygen uptake (CMRO2).
This variable shows a hyperbolic decline
with age with little change during the
mature years. Since the data did not fall
into well-defined age subgroups, individ-
ual values, rather than mean values for
the different ages, are shown here. This
largely accounts for the apparent
increased scatter shown,

Figure 8. -Cortical cell density and brain
weight as a function of age. The decline
in cell density during maturity is con-
siderable, whereas that for brain weight
is, in percentage termSj less marked.

Figure 9.-Wechsler Adult Intelligence Scale
(WAIS) verbal and performance scores as
a function of age. Both measures show a
parabolic decline. However, the verbal
scores show little change during maturity
whereas the performance scores show a
reduction of about 30 percent between
age 20 and age 50 years.

Research Grant: MH 10927
Date of Interview: Apr. 26, 1967

tt SO

co«ric«, ciu. tiNSilr

' -.7-

Pirlormonce
**

To"

"75

AGE {YEARS)

538

Irwestigutors:

Mark R. Rosenzweig, Ph.D.*

Edward L Bennett, Ph.D.

David Krech, Ph.D.

Marian C. Diamond, Ph.D.

University of California
Berkeley, Calif.

Propnrod by:

Gay Luco

In the last dozen years a Berkeley team of psychologists, biochemists,
ami anatomists has demonstrated that learning niters chemical activity and
increases cell growth in the cerebral cortex— changes that arc correlated
with an enhanced problem-solving ability. Thus, experiments with young
and with adult animals show that mental activity produces detectable
brain growth and that specific kinds of activity may develop particular
regions of the brain. Although there may be limits to an animal's respon-
siveness to its environment, a stimulating environment appears to
maximize the rate of learning.

Using rats as experimental subjects, the scientists maximized environ-
mental contrast by segregating littermates into two extreme situations.
Some were reared in isolation in small barren cages in a dim, quiet room.
Their brothers matured in groups of 10—12 in large complex ciigcs
supplied with toys, exercise wheels, and other diversions, as well as daily
training on laboratory apparatus. These enriched rats erred less on
problem-solving tests, and postmortem analyses showed thai their brains
differed from the brains of their impoverished siblings. The cerebral
cortex had grown larger in relation to the rest of the brain. H continued
more glial cells and showed greater activity in a particular biochemical
system.

Animals exposed to the differential environments after adulthood
exhibited much the same differences in brain growth.

A variety of genetic studies indicated that the proportion of the brain
occupied by cerebral cortex and also certain chemical attributes of the
brains of enriched rats were characteristic of "bright" strains, while "dull"
strains more nearly resembled impoverished animals.

*Now at the Department of Psychology, Washington University, St. Louis,
Missouri.

539

The experimenters are now testing for the role of formal train
brain development. Will a rat reared alone in an empty cage show
growth and an improved problem-solving ability after intensive daity
ing on an automated teaching apparatus? Can formal training be us
compensate for lack of environmental complexity and stimulation?

The grantees have begun to push this field forward during a time
it was believed impossible to detect physical effects of learning j
brain. Their data may help to explain many baffling statistical ob
tions about human beings, among them the role of environment in
ligence, the accomplishments of people with median childhood IQ
the curious incidence of eminence among first-born children. The la
tory has seen some evidence that specific kinds of environn
manipulation enhance growth in specific brain areas, suggesting thi
may learn to provide rehabilitation training for peisons with brain i
or sensory deficiencies—giving exercise to develop brain regions,

Since the brain is measurably modified by use, in a manner that c
correlated with adaptiveness and learning capacity, the laboral
findings have strong implications for the education of children. Ho\\
the extreme impoverishment of the experimental rat should IK
translated as if it were synonymous with cultural deprivation, A vat
of experience is rare among human infants, unless they are doaf-t
However, there is no escaping the conclusion that we literally mod if1
brains according to the use we make of them. Through research prog
like this one, we may gain the impetus to deliberately improve the ei
tion and mental equipment of our young, for there is some potency i
knowledge that we are measurably altering their brains.

Background

Scientists in the past have felt that use of the brain might affect its
size and composition. Charles Darwin, for instance, observed
domestic rabbits had smaller brains than wild rabbits. He surmised
the domestic rabbit had been so confined and protected for many gci
tions that it exercised its intellect and instincts very little. He implied
lack of mental activity and stimulation had caused brain atrophy, h
late 19th century, the autopsy of Laura Bridgman, a deaf-blind n
showed deficient development in visual, auditory, and speech regioi
the brain. Recently, animals reared in darkness have been given ca>
post mortem examinations. They showed shrinkage in the visual pai
the cortex.

The notion that thinking might cause the brain to grow was suggc
during the 18th and 19th centuries. A German phrenologist, J
Spurzheim, believed that the brain grew by exercise, but he had no m
of demonstrating the phenomenon. Even in his day the brain waskn
to be relatively stable, and there was no equipment for testing sti
changes. Ideally, it was proposed, one should examine the same pc
after a long period of intense brainwork and after long period!

540

desuetude; or perhaps compare twins, after keeping one isolated and idle,
while the other lived a rich and mentally active life. This experiment
was hardly suited for human beings. It had to await the modern animal
laboratory, with its refined instruments.

Today, people are aware that the brain is the physical machinery of the
mind and that it can be affected by physical means-by drugs, electrical
stimulation, etc. The rate of brain activity, memory, and emotions are
related to the actions of chemicals in the brain; and when we alter
behavior, memory, and emotions by drugs, we thus change many factors
in what we consider to be intelligence. In the last century intelligence was
presumptively equated with brain size. Many attempts were made to
compare the brain size of men with distinguished intellects and with those
of inferior intelligence. Gross brain sizes were found to be inconsistent,
however, and an idiot can have a larger brain than Einstein. Thus, this
unsuccessful approach was abandoned. The notion of anatomical change
and actual increase in brain weight as a result of cerebration became so
widely debunked that the grantees, themselves, ignored this possibility
until they were well along in their research program.

In J 953 they began search for some of the physiochemical correlates of
learning, an approach that required considerable courage in the skeptical
atmosphere of that time. They had started by observing the behavior of
hungry laboratory rats who were placed in a problem-solving test box
invented by Dr. Krech. The Krech Hypothesis Apparatus Box is a simple
piece of equipment that has been widely used for a number of purposes.
This is a long box, divided into four successive chambers. In each chamber
the hungry animal chooses between a left and a right alley. He must make
four correct choices to reach the food compartment at the end. In each
vestibule he may face a darkened alley on one side and a well-lighted alley
on the other. Which side has the open door? In one problem the light may
be a clue to the correct side. That is, the animal may find heisiewarded
when he chooses the lighted alley, whichever side this happens to be.
When he almost invariably chooses this avenue and quickly runs to his
food reward without digressions, he suddenly finds that light is no longer
the rules of the game. He must now learn to choose the dark avenue. At
first he has learned a simple discrimination, and then he must learn how to
reverse his discrimination. From the animal's point of view it may seem
that, as soon as he adopts the profitable habit, the rules are changed. His
adaptability— and presumably his survival in the outside world-depend in
part upon his ability in this reversal discrimination learning. Thus, in many
experiments, reversal discrimination provides a crucial performance test.
In the laboratory's original studies, the hungry rats were bucking an
insoluble problem in the Krech apparatus, The choices were prearranged
so that there was no correct solution. It was observed, however, that
individual rats had interesting choice patterns. They acted as if following
some hypothesis about the rules of the game. When, later, the situation
was made progressively more solvable-for instance when light became the
correct cue 70 percent of the time-an animal might still prefer to go
down the left side, much of the time, disregarding the light cue. Some

animals acted as If the visual cues of light and dark were irrelevant
would make a spatial choice, running down the right avenue each 1
Others seemed to fixate on visual cues, ignoring the possibility that I
did not indicate the correct exit. Because the hypothesis box rev*
these individual differences in the rats' ability to restrict their attentic
the pertinent cues, it seemed to provide an excellent test of prob
solving ability.

When lats were allowed to run through the alley as they wished, nc
being uniformly correct, some animals followed the light and otto
spatial pattern. The population seemed to divide into rats with a pr
ence for visual cues and rats with a proclivity for spatial cues. On rev
tests in the alley, the "spatial" rats proved better at adapting to the
rules. Thus, in a sense, they seemed brighter. The experimei
postulated that these rats might have brains that were richer ii
important enzyme -related in its quantity to a chemical that I
transmit excitement from one brain cell to the next. At the time it see
plausible that quick-learning animals might have an abundant suppl
this chemical so important in nervous activity. The experimei
expected to find that rat strains which demonstrated superior lear
ability would show more of a particular brain enzyme, but the dun
animals actually showed more. In testing more strains of rats, they bi
to wonder if the neuro transmitter and enzyme were genetically
dependent, or indeed whether they were seeking the right chemica
correlate with behavior.

Biochemical Indices: The Acetylcholine System

Brain activity is both electrical and chemical. When a person think
takes an action, neurons pulse out rhythmic coded messages. Each b;
cell influences neighboring cells biochemically. As it fires, it relcasi
chemical at the infinitesimal gaps or synapses between itself and surroi
ing cells. This chemical, known as a neuro transmitter, bridges the gar
can change the excitability of the neighboring cells so that they
readied to fire. The safeguard that prevents cells from conlinu
excitability and exhaustion is an inactivating chemical. Each lini
transmitter is released, it is quickly neutralized by an enzyme.

One of the first neurotransmitters to be studied was acetylcholme,
released at nerve endings throughout the body, in the regulation
muscular contraction and relaxation, throughout the periphery and in
heart and other organs. Acetylcholine is released at a synapse^ t
hydrolyzed by the enzymes acetylcholinesterase (AChFi) ,
cholinesterase (ChE)— whereupon the excitability of junction po
subsides and they return to their resting state. Acetylcholine and its
activating enzymes are found throughout the brain.

A high rate of brain activity, as in learning, might require a rapid :
constant output of the transmitter and the enzymes that destroy it.

542

In the mid-1950's the grantees began to explore the possibility that
bright animals had brains richer in acetylcholine and that an environment
enforcing a great deal of learning might change the output of this
transmitter in the brain. It was then technically difficult to obtain precise
evaluations of acetylcholine in brain tissue, and so they worked by
indirection. Presumably the activity of the destroying enzymes was related
to acetylcholine concentiation.

Because the enzymes AChE and ChE would remain relatively stable in
frozen brain tissue, it was possible to measure their speed of activity in a
relatively direct way. A sample of homogenized brain tissue was injected
into a fluid containing a great deal of acetythiolcholine, which the enzyme
breaks down at the same rate as acetylcholine. When light was passed
through the fluid, and an indicator was added, its color would change as
the enzymes acted, producing a yellow compound; the color change
could be measured, and told how fast the enzymes were destroying the
transmitter.

Originally, acetylcholinesterase was the better known enzyme. Its
activity was first used in an attempt to find a chemical index of brain
efficiency. However, as many strains of rat were tested on problem-solving
apparatus, the poorest performers often showed the highest AChE
activity. Yet some strains, with excellent learning records, showed high
ChE activity,

During the late 1 950's the experimenters were able to directly evaluate
the relation between the transmitter, acetylcholine, and the activity of the
enzyme AChE in two rat strains. Although one strain had significantly
greater amounts of acetylcholine, the AChE enzyme was not propor-
tionately more active. They found that learning experience or that
"brightness" in an animal was indicated in the ratio between ChE and
AChE activity in different portions of the brain. AChE activity bore moie
directly upon the output of the transmitter, and the role of the other less
specific enzyme ChE was not clear.

By selective breeding the researchers acquired strains of rats whose
brains differed in acetylcholine concentration, or in its enzyme, AChE. On
successive tests there has been considerable uncertainty about the relative
intelligence of different rat strains and their brain characteristics. Students
in the laboratory, and consultants, have begun to amass information about
the different rat strains. For example, a strain with high brain acetyl-
choline would react differently to electric shock than another strain. This
fact has some importance to other researchers, especially those who might
employ these animals in experiments using shock punishment. The genetic
studies have continued to be part of the laboratory program. The team has
attempted to control as much as possible for heredity, as it has begun to
track the way in which environment might change brain biochemistry.

Experience and Learning Capacity

During the 1950's a number of laboratory e
early experience would make an enormous

ability to cope with learning problems. In one laboratory, animals rea
in isolation with a minimum of handling turned out to have differ
adrenal responses and skittery reactions to tests or stresses, in comparii
with littermates raised in groups with considerable stimulation. Isola
animals performed poozly in comparison with handled littermates, I
were their nervous systems different?

The Enriched and the Impoverished

Over almost a decade the grantees have employed a method t!i
maximized the contrast between their animals, prior to testing and bra
assay. Littermates were segregated into extreme environments. In ea
case lats of a particular age, strain, and sex were used. During subseque
brain assays, the rats were identified only by codes so that the e
perimenters would not know from which group they came. This proc
dure, initially used with weanling rats, has since been used with adults at
othci species.

In general the infant rats were segregated when they were 25 days oli
At this point, some pups would land in barren individual cages in a din
quiet room unable to see or touch othei animals, being handled on!
during biweekly weighing. Their littermates, in groups of 10-12, foun
themselves surrounded by toys and were handled each day as they wei
set to play in a square box with wooden barriers or mazes. In man
experiments, the rats lived their extraordinarily different lives for 81
days-well into maturity. At this point they were tested and analyzed. Ii
some experiments the environmental exposure lasted only a month or 51
days.

By 1966, the laboratory had amassed data on well over 200 pairs 01
littermates. The evidence was consistent. Enriched rats were supeiior k
their impoverished brothers on reversal discrimination learning. Alonj
with their greater problem-solving abilities, they had a far different brain.
The cerebral cortex, often colloquially termed the thinking brain, had
grown larger relative to the rest of the brain, and they showed a different
balance in the activities of the enzymes cholinesterasc (ChE) and acelyl-
cholhtesterase (AChE).

Brain Growth

The research team had already found consistent biochemical changes
before they began to look at changes in brain weight. As Dr, Rosenzweig
has explained:

Brain anatomy was disregarded since we had inherited from our
predecessors the dogma of absolute stability of brain weight. Fortunately
we had to record the weights of our brain samples in order to measure
chemical activity per unit of tissue weight. After about 2 years of
contemplating the chemical effects, it finally dawned on us that the

544

weights of the brain samples were also being altered by the environmental
manipulations.

Had the researchers been comparing the weights of whole brains they
would have noticed no diffeiences. Indeed, from outside, one might have
judged that the impoverished animals showed the advantage. When
phrenologists practiced their art of measuring intelligence and personality
by comparing proportions of the skull, there had been some expectation
that a large skull encompassed a larger brain. The large forehead and
cranium which appear to encase a more sizable Intelligence turns out to be
a poor index of intelligence. The experimenters compared the inner
cranial capacity of some of their enriched and impoverished animals and
found that the two groups did not differ. By making meticulous rubber
casts of the skulls and measuring facial bones, after weighing the brains,
they saw that the larger facial bones were those of the impoverished
animals and were related to body size rather than to brain development.
The impoverished and inactive creatures were both heavier and larger than
their littermates.

Comparisons of whole-brain weights would not have defined the
anatomical differences between the enriched and impoverished rats.
However, in preparing brains for chemical analysis, the experimenters have
been dissecting the cerebral cortex into four sections and treating these
sepaiately from all the rest of the brain beneath. They had reasoned,
plausibly, that the cerebral cortex should be the part of the brain to show
(he most changes as a result of learning.

Indeed, as they soon saw, the enriched animals had a heavier cortex
than did their impoverished brothers. When 141 pairs of litteimates of a
particular genetic strain were compared, the enriched animal had a cortex
(hat was, on the average, 4 percent heavier than that of his impoverished
brother. The impoverished animals showed their growth -a very slight
one— in the more primitive subcortex. The delicacy of these procedures is
difficult to convey. Altogether, a rat brain is not much larger than a
healthy Brazil nut. Still, if the percentage differences do not sound
enormous, the brain is generally so stable that these signs of growth,
relative to experience, are impressive.

As the experiments were varied, and as the scientists looked closely at
the several regions of the cortex, they found that certain portions of the
cortex were changing more than others. The occipital region, the visual
area located at the back of the brain, showed the most growth. The
enriched animal had an occipital cortex that was 6 percent heavier than
that of his impoverished littermate, Other regions showed less difference.
A number of experiments were conducted in order to see whether specific
cortical areas could be caused to grow through the use of specific training
and environment.

Animals reared in darkness have been found to show a lack ol growth m
the visual cortex. When the researchers segregated blinded animals into
extreme environments, they found that the complex environment and
training of the enriched animals did compensate somewhat for lack ol
sight. Although the blind animals showed enzymatic differences that

545

distinguished them fiom the sighted animals, they too responded
impoverishment or stimulating surroundings in very much the s;i
manner. Signs of growth in the visual cortex of enriched blind i
suggested that it must participate in nonvisual functions. Thus, !>l
animals benefited from the envhonmental stimulation.

AH of the enriched rats have shown a thickening in the gray outer h;
of the brain and also in a subcortical region that has been associated \v
memory functions and refinement of emotion-the hippocampus. Sttnui
tion, enforcing active use of the brain, had caused palpable growth in t
cerebral cortex of the enriched animal. Just as specific sensory dcprivult
such as blindness or deafness would cause a deficiency in Lho relal
cortical region, particular experience seemed to produce growth in
relevant cortical area.

Chemical Changes

Originally, the investigators had expected to find that their
animals showed increases in the specific activity of the enzyme (AC '111:
indicating greater concentrations of the ncurotransmitter acelyleholin
Indeed, the enriched cieatures did have a cortex whose tissue showed mm
enzyme activity than that of their impoverished brethren. On the oUu
hand, the AChE activity was a smaller percent per unit weight In Eh
cortex of the enriched. The growth of the cortex outdistanced th
increase in this particular enzymatic activity. However, it showed ,
proportional increase in the subcortex. If AChE activity declined in tli>
cortex, it increased in the subcortex.

In order to pinpoint the biochemical change in the cortex, the re
searchers began looking at the proportions of the specific enzyme AClil
to the nonspecific enzyme ChE in the cortex and rest of the bmin: in Ifu
brains of the enriched animal, there was more cortical ChE relative to
AChE. The relative proportions of these two enzymes thus inclieiikul (hi
difference between the enriched and impoverished animals.

When the scientists looked at these enzyme ratios in specific portions in
the brain, they found especially striking differences between ihcir
enriched and impoverished animals in the visual cortex.

They were more than a little curious to find that the cholincslcNM;
activity was exceeding that of the more specific enzyme in their enncliol
animals. Cholinesterase is known to be concentrated in the gliul- or mm-
nerve cells-of the brain. As they searched to see precisely whfll livwtf
growth accounted for the heavier cortex of the enriched animals, (hey
suspected a proliferation of these important small cells.

A Multiplication of Brain Cells

The experimenters measured the diameter of capillaries in the
and found them to have grown larger in the enriched animals than in

546

impoverished animals. But this enlargement in the brain's blood supply
network did not explain the amount of cortical growth. They subsequent-
ly made a cell count in the cortex— the region that showed most growth in
the enriched animal. Frozen tissue, stained and cut paper thin, was made
into slides, and sizable photographic enlargements weie made of each
slide. Now began the count of individual cells.

Two anatomists would make sepaiate counts. A technician using
colored pencils would mark the location of each glial and neuronal cell on
a sheet of plastic placed on the photographic enlargement of a brain
section. The same process would be repeated by a second technician.
When the two sheets were superimposed, the discrepancies could be seen
and discussed, and finally all the brain cells would be classified and
counted.

A comparison was made of the visual cortex of 17 pairs of littermates.
The enriched animals had more glial cells than did the impoverished, a
higher ratio of glial cells to neurons. Glia are not well understood, but
tliey are believed to nourish the neuron and to modulate the brain
activity by altering the excitability of the neurons. Environmental
complexity and stimulation had caused a proliferation of these cells, thus
perhaps enhancing the efficiency of the neuronal activity.

The proliferation of the glia was confirmed at MIT by Drs. Joseph
Altman and Gopal D. Das. Using the same extreme environments, they
analyzed their animals' brains by a different method. Each animal was
injected with a radioactive substance that is used in formation of new
cells. This substance acts as a label. Wherever it was absorbed by a brain
cell, there would be slight emissions of nuclear particles which affect a
photographic emulsion in the manner that light darkens the silver grains of
any ordinary photographic emulsion. When the experiment was concluded
and brain sections were dipped in emulsion, photomicrographs showed a
significant increase in glial cells in the cortex and neocortex of the
enriched animals. The density of these labeled cells-showing up as dark
spots on the film-can be measured automatically.

Drs. Altman and Das have shown that in the rat neural cells also
multiply after birth. It has long been supposed that an animal possesses its
entire lifetime supply of neurons at birth. By injecting a radioactive
component of the genetic molecule, DNA, they have observed that the
radioactive substance was incorporated in the formation of many new
brain cells in adult animals. These cells-so small they are known as
microneurons -appear to migrate as they differentiate, moving into new
brain areas. It now seems possible that some of these small neural cells
have been mistaken for glia. The MIT team has opened a new question in
the role of experience in shaping the brain. Will older adults show an
increase in these microneurons? Do they multiply in response to the
exercise of the brain?

Even if glial cells are the only population to increase in the brain as a
result of environmental challenge, they may enhance brain efficiency in a
number of ways. Glia are important in the fatty white matter that
sheathes nerves. They surround the dendritic tentacles that interconnect

547

each neuron with multitudes of others. Perhaps their multiplict
permits the sprouting of new dendrites, new contacts from one ncurc
others. Perhaps their proliferation allows greater nourishment of ni
cells, or more refined modulation of neuron excitability. All that c.i
said at present is that environmental manipulation adds to the lissu
the ceiebral cortex in part by the multiplication of glia and that ani
exhibiting this brain growth excel on problem-solving tests.

How Do the Different Environmental Factors Cause Brain Changes?

Clearly, when an animal's entire environment is manipulated in
laboratory, it is difficult to pinpoint the particular factois responsible
this change. Isolation, for instance, has a potent effect upon man
beast. People who have undergone tests in silent chambers, or in tanfe
water at about blood temperature, have noted that the sensory ilcpi
tion began to evoke odd psychological experiences. Some people quit
leave such an experiment, while others have suffered little. Some stinin
rats, reared in isolation, have become increasingly aggressive and difft
to handle in the laboratory. They suffer enlarged adrenal glands, ;
irritations, and other symptoms known as isolation stress. None of
Berkeley animals showed any of these signs. Nonetheless, the h
inevitably wanted to ascertain what role isolation contributed to
contrasting brain and behavior effects of the impovereshed and ciiiic!
rats.

Social controls. -Some experiments were run with a third group
littermates, reared neither in isolation nor in a complex envJronnu
These rats grew up in the ordinary animal laboratory cages in gioups
three. The only sign that the social life was better than isolation i
detected in the visual cortex, where growth exceeded that of
impoverished littermates. On other counts, the brains more clos
resembled those of the impoverished animals than of the enriched.

Extreme isolation.— In recent studies a condition has been creal
approaching a sensory vacuum. Some littermates have lived where Itglu
dim and temperature constant. The rat cannot see outside the ciigc, n
even having contact with the experimenter as he changes food and wal
The only change is the alternation of light and darkness every 1 2 hours,

As might have been expected, rats who spent 80 days in this envirt
ment proved to be extremely different from their littcrmales iti I
enriched condition. The differences in cortical weight and enzyme flctbi
were markedly enhanced. Indeed, these extremely isolated unit ii
poverished animals were different from rats merely raised in the origin
loneliness of unadorned cages.

Isolated pairs.-W&$ social isolation, or lack of a stimulating cnvira
ment, the overriding factor in the lack of brain development in I!
extreme isolates? Pairs of rats were now placed in the extreme isoliitio
each one from a different litter. Post mortem analyses of these anuna

548

indicated that paired living could not compensate for the rarified sur-
roundings. The cerebral consequences were striking. These pairs did not
differ significantly in cortical weight or enzyme activity from those who
had lived in a solitary state in these cages, devoid of interest and stimula-
tion.

As they created graded situations, to test for the role of social inter-
action in learning and brain growth, the researchers saw that the extreme-
ly impoverished lone animal was not much worse off than pairs in the
same kind of cage. Moreover, triads, raised in barren cages, hardly differed
from their isolated littermates, although they did show a ratio of enzyme
activity which approached that of the enriched rats. A close comparison
of the visual and somesthetic regions of the cortex showed the jesearchers
that the effects of an impoverished environment were somewhat mitigated
by placing groups of lats together. Groups of 10-12 rats have been
studied after rearing in large cages, devoid of complexity or stimulation
Sheer numbers do not appear to compensate for lack of environmental
complexity. These rats fall somewhere between their impoverished and
enriched littermates.

The addition of toys, variation, and training appears to add some
further element -can sing the animal to use his brain more actively -with
concomitant biochemical and anatomical changes in the brain. .

Formal training. ~Vfh*t is the role of pure learning, of training. In
current experiments, the laboratory is rearing animals in isolation in cages
devoid of toys or social contact, but for an hour or two each day the
animal receives intensive training. Can formal training alone produce
cerebral growth commensurate with that observed from life in a rich ana
varied environment?

Differential Environment: Impact on Adults

Environment during the earliest, most formative years of life appears to
leave more of a mark than experience after maturity. Surely this has ueen
a tenet of human clinical psychology. It has been demonstrated in some
animal experiments, too.

The Berkeley team demonstrated that adult rats as well as weanlings
show cerebral effects of experience. Therefore, the cerebral consequences
of environment were not merely effects of normal growth processes
accelerated by the impact of environment during infancy.

Between the time of weaning (at about 25 days) and 105 days, ,a i rafs
brain grows appreciably. If he is kept in a standard colony cage here will
be a 20-percent increase in cortical weight and a 4<H»rGent mere,
rest of the brain by the time the animal is 105 days old. Alter
growth is observed. In the next 80 days there will be _ n
increment in brain weight. Since brain growth contnbnt
days, rats of this age were selected for exposure to
enriched environments. They had been sexually matim

Adult rats, left foi 80 days in an enriched or impoverished envi
ment, diverge even more than young rats in their ultimate cortical weif
The effects of enriched enviionment upon adult rats do, however, di
slightly from the effects on young nits. Adults gain more toliil b
weight than does the immature animal, and they show some weight g;ii
the subcortical legions.

This difference between young and old is notable in assays of on/}
activity. The older animals show a more pronounced drop in their eorl
AChE activity. Nonetheless, the adult rat exhibits an cncouiaf
plasticity, suggesting that the effects of mental activity and a ehallenj
environment can be induced long after maturity. The cerebral $m\
produced environmentally in this laboratory would not seem to be ;i ro-
of accelerated maturation, since enriched adult rats, compared willi IF
adult isolates, showed brain growth as did the young.

Nor are these effects of differential environment restricted U> rt
Comparable studies with mice have offeied very similar results.

Heredity and Environmental Manipulation

Within any family, variations among children vividly display how JMHI
an individual's response to his surroundings depends upon liis inherit
structure. Inbred rats show only slight variation among litlerinutos. Ho
ever, inbred strains differ exceedingly from each other.

Genetic studies and selective breeding have played an importnnl pint
the laboratory's program during the last dozen years. Using perfoiman
on certain learning tests as a criterion for brightness or dulliujs-i. (1
laboratory developed two stiains; One that was consistently hriphl, II
other consistently dull. In each strain, an impoverished and nn
littermate will show the behavioral, biochemical, and anatomical i
of environment. However, some rat strains show a greater bra In
and enzymatic change than others when exposed to environment
complexity and training.

When 240 rats-24 from each of 10 different strains-worts tested o
reversal discrimination, their errors were compared with bruin iinalysc
The individual rats who made the most errors had a smaller cortex rcliitu
to the rest of their brains. Their ratio of AChE/ChE activity was ah
closer to that of the rat from an impoverished cage. Even within ;my give
strain, there was a consistent correlation between an animal's pcrlbmuik
and the relative weight of his cortex and his enzyme ratio. The brigfi
animals more nearly resembled animals from an enriched environment. "

Both heredity and environment appear to influence an animal'!
adaptiveness and problem -solving ability through common bioclicmka1
and anatomical pathways in the brain. Whether from onvkomiK'nU
manipulation or endowment, the good problem solver has a brain with J
cortex that is larger in proportion to the rest of his brain and ii ratio of
enzyme activity that now seems to characterize the efficient learner

550

Other experiments have confirmed the correspondence between the
laboratory's brain measures and problem-solving capacities.

In 1962 the team left littermates in their extreme environments for
only 30 days. Then they regrouped them in colony cages, under the care
of an experimenter who had no way of knowing which rats had been
enriched and which impoverished. For 10 days they received preparatory
training on test apparatus and were acclimated to doing without food so
that they could be given a food reward on their performance tests. In
order to maintain control for possible weight loss due to food deprivation,
a third group of littermates was maintained with food and water always
available. These rats received no training or testing and, as it turned out,
body weight was not a factoi.

The enriched and impoverished animals were then tested m the
hypothesis apparatus. On their first trials they appeared similar. But these
were easy runs, merely asking that they run down the lighted side at each
choice point. When the cues were reversed and the problem became more
difficult, the impoverished group performed poorly, the animals many
times taking the wrong turn. The enriched group performed significantly
better,

Only 30 days of environmental difference had made a pronounced
difference This was especially striking, since the impoverished group had
received the benefits of 10 days' pretraining and handling. This short
period of relative stimulation probably attenuated the differences between
the two groups.

After the discrimination-reversal test had been scored for errors, brain
analyses showed that there was a correlation between the ^mber of
errors an animal made and the size of his cortex relative to the test of h.s
br™ Ratios of enzyme activity also distinguished the poor performer
from his more efficient littermates. On the brain measures _the re was no
loneer any difference between the two groups, md eating that the
Imp^eSd group had benefited from their 10 days of pretrammg and
the 20 days of testing.

Environment and Other Brain Components

!„ the course of a long and varied
been so consistent as to become predicta*. n n

solving scores, the experimenters can s ™f ~™ ^e rest of the
enzmes in the animal's brain and the ratio of ™™* a rebred

enzymes

This has bean borne out ^ gen^c^u^ -^ jmal exposed to

capacity is enhanced.

551

The laboratory has traced other biochemical systems as u
Hexokinase, an enzyme important in cell metabolism, did not seem tc
altered by differential experience. Nor did serotonin activity appear
change. In other laboratories, scientists have used differential envir
ments to measure for changes in other brain constituents.

Dr. Edward Geller and a team of scientists at UCLA have found tl
the transmitter substance, norepinephrine (also known as noradrenali
exists in greater concentration in the brains of enriched rats. An assay
specific brain regions has revealed, however, that the impoverished anim;
had about five times as much norepinephrine in a subcortical regi<
sometimes described as the brain's chemical storehouse-the cauda

nucleus.

The Berkley team had initially expected to find that protein met
bolism increased in the brain as a result of mental activity and tin
expected to find higher protein turnovei in the brains of enriched rat
Percent protein did not differentiate the enriched from the impoverishe.
groups. Recently, however, Dr. Joseph Altman and his MIT coworkei
have found that protein turnover is actually lower in the enriched animals
They injected radioactive leucine into their enriched and impoverishet
littermates. Leucine is an amino acid used throughout the brain in flu
construction of protein. Cells that utilized the radioactive leucine created
dark spots on a photographic emulsion, On examination, the slides were
a surprise, for the brains of the impoverished animals had absoibed the
most of the radioactive label and thus indicated a higher rate of protein
metabolism. Altman and his co workers have speculated that this high rate
of protein metabolism may be a sign of stress and that the isolated animals
may have experienced stress each time they were handled in the labora-
tory. Perhaps a lower rate of brain metabolism is a sign of greater ef-
ficiency.

Memory is, inevitably, a crucial factor in learning* and some ex-
perimenters have enhanced retention and speed of learning in animals by
chemicals. Under certain conditions oral doses of magnesium pemolinc
have been reported to cause rats to learn conditioned responses four to
five times faster than their untreated controls. Moreover, the treated rats
showed no signs of forgetting after 2 weeks. Recently this drug has been
tested on senile patients, and preliminary reports suggest that it may
improve memory.

Conclusion

From many quarters, diverse scientists of behavior are beginning to
illustrate the same proposition. An individual's behavior depends upon the
anatomical and chemical attributes of his brain. To a large extent this is
dictated by heredity, but many of the brain's characteristics can be altered
by experience as well as by drugs.

A rich and complex environment would seem to cause brain growth, In
infancy and also in adulthood. Some initial evidence suggests that it is

552

possible to selectively modify the brain by selective environmental
demands. Blinded animals, for instance, have shown some development in
the visual cortex after an interval in an enriched surrounding. Since adult
animals also show brain plasticity, it would seem that people with brain
damage might be rehabilitated by special training designed to cause
growth in specific portions of the brain. A recent study of elderly patients
with aphasia indicates that brain-damaged persons can be i cached and
possibly rehabilitated by carefully designed training. Aphasics, who are
unable to read or understand speech, have been given intensive training in
visual discrimination, a fundamental function in leading. Following
automated training, they have shown improved performance in discrimina-
ting the geometric shapes that make up the English alphabet and have
given signs of retention on subsequent followup tests. It appears that
programmed training might become an effective tool, helping disabled
patients to function once more, perhaps by causing growth in intact
portions of the brain. Surely, the Berkeley studies suggest that experience
can improve a creature's capacity for learning and adapting, by piomoting
change in certain anatomical and biochemical pathways of the brain.
These appear to be pathways by which cultivation can enlarge the in-
dividual's capacities, whatever his hereditary limits initially may seem to
be,

These studies have interesting implications for education. Perhaps we
can begin to understand factors in the development of the young, which
at present yield puzzling statistics. Many studies of birth order and life
achievement (for instance) suggest that the first-born child in a family is
more likely to be gifted and eminent. Twins, indeed, are found to have
lower IQ's than single-born children or pairs of children spaced farther
apart. On the other hand, studies conducted in orphanages suggest that
institutionally reared babies lag behind their counterparts raised in
families. Could these differences reflect the relative enrichment and
stimulation the children receive as they are growing? Does the first child
in a family receive far more attention and handling on the average than
the subsequent children? A scattering of studies suggests that this may be
the case. One cannot probe the development of the cortex in human
Infants; still a difference has been seen in the visual performance of
orphan infants who were given visual stimulation and toys in their cribs.
Cross-cultural studies have indicated that the early precocity of a very
young child in Uganda, for instance, may be related to the constant
company and attention of the mother.

A dramatic case in point has been reported recently. Thirty years ago
Dr. Harold M. Skeels of NIMH experimented with 13 toddlers who were
classed as mentally retarded. They were orphans in an Iowa institution,
where the usual procedure was to keep retarded children for a while and
then transfer them to a special institution. By chance two youngsters got
transferred early. Within a year they were mentally normal. Because they
were much younger they attracted a great deal of attention and care from
the older inmates, and this apparently accounted for their improved intel-
ligence. Dr. Skeels noted that the orphanage was so efficient that infants

553

received little individual attention. He placed 13 retarded orphans as
"houseguests" in a state institution at a very early age. Three of them
were categorized as imbeciles. Eleven of the children so profited from the
extra play and care that they became "normally" intelligent, were put up
for adoption, and later became self-supporting middle-class adults. A
control group of the same age had been left in the orphanage. These
children had average intelligence. After a number of years both groups
were tested. The normal children, within the orphanage, lost IQ points,
whereas the supposedly retarded children had gained in IQ, The two
groups had, indeed, switched positions, and one child who began with a
rating of good average intelligence had become an imbecile by age 1 9,
after a life in the orphanage. The impact of environment on intelligence
was heartbreakingly palpable.

In the rearing and education of our young, we can now see that the
environment we provide may enhance or retard brain growth— and perhaps
intelligence. However, one cannot liken the culturally deprived person to
the impoverished laboratory rat. Except for those people who are deaf
and blind, or seriously handicapped, a vacuum of experience is rare.
Rather, the experience of our young is random. We have done only little
to discover how a human education should be programed and placed to
maximize the potential of each individual. It is no news to say that the
resources of our population could be increased by an education that
induced mental growth. Now this same proposition has been put into
physiological language. We can deliberately cultivate more effective brains
and cause growth in the cerebral cortex.

Research Grants. MH 1292, MH 7903

Dates of Interviews: December 1965, May 1966

References:

Altmaii, J. Are new neurons formed in the brains of adult mammals? Science, 1962,

135,3509.
Altman. J. Autoradiographic investigation of cell proliferation in the brains of rats and

c9.ts.Anat. Rec.t 1963, 145,4.
Altman, J. Differences in the utilization on tritiated leucine by single neurones in

normal and exercised rats: on autoradiographic investigation with microdensi-

tometry. Nature, 1963, 199,4895,

Altman, J. The use of fine-resolution autoradiography in neurological and psycho-
biological research. In Haley & Snider (Eds.), Response of the nervous system to

ionizing radiation. Little, Brown, 1964.
Altman, J. Autoradiographic examination of behaviorally induced changes in the

protein and nucleic acid metabolism in the brain. In F. A. O. Schmitt (Ed.),

Macro molecules and behavior.
Altman, J., and Das, G. D. Autoradiographic examination of the effects of enriched

environment on the rate of glial multiplication in the adult rat brain. Nature, 1 964,

204,4964.
Altman, J., and Das, G, D. Postnatal origin of microneui'ones in the rat brain. Nature,

1965,207,5000.
Altman, J., and Das, G. D, Autoradiographic and histological evidence of postnatal

hippocampal neurogenesis in rats. J. Camp. Neural, 1965, 124, 3.
Altman , J., Das., G. D., and Chang, J. Behavioral manipulations and. protein metabo-
lism of the brain: effects of visual training on the utilization of Leucine-H3.

Physiology and Behavior, 1966, vol. l.Pergamon Press.

554

Bennett, E. L., Crossland, J., Krech, D., and Rosenzweig, M R. Strain differences in
acetylcholine concentrations in rat brain. Nature, 1960, 187,787.

Bennett, E. L.t Diamond, M. C., Krech, D., and Rosenzweig, M. R. Chemical and
anatomical plasticity of brain Science, 1964, 146,610-619.

Bennett, E. L., Diamond, M. C , Krech, D., and Rosenzweig, M. R. Heredity environ-
ment, learning, and the brain. AAAS Symposium, Behavior, Brain, and Bio-
chemistry. Berkeley, Calif., Dec. 28, 1965.

Bennett, B. L., Drori, J. B., Krech, D., Rosenzweig, M. R., and Abraham, S. Hex-

okinase activity in brain./. Biol Chem., 1962,237, 1758-1763.
Bennett, E. L., Drori, J. B., Krech, D., Rosenzweig, M, R., Diamond, M. C., and
Abraham, S. Determination of hexokmase activity in brain. Fed. Proc., 1961, 20,
343 (abstract).

Bennett, E. L., Herbert, M., Rosenzweig, M R., and Krech, D. Genetic selection of rats
for cerebral acetylcholme concentration. Proc. Sixth Int. Congr. Biochem , New
York, 1964.

Bennett, E. L., Krech, D., and Rosenzweig, M. R. Reliability and regional specificity
of cerebral effects of environmental complexity and training. J. Comp, Physiol
Psychol, 1964,57,440-441.

Bennett, E. L., Rosenzweig, M, R., Krech, D,, Karlsson, H., Dye, N., and Ohlander, A.
Individual, strain, and age differences in cholinesterase activity of the rat brain. /.
ffeurochem. , 1958,3,144-152.

Das, G. D.f and Altman, J. Behavioral manipulations and protein metabolism of the
brain: effects of motor exercise on the utilization of Leucine-H3. Physiology and
Betiavtoi, 1966, vol. 1, Permagon Press.

Das, G. D., and Altman, J, Behavioral manipulations and protein metabolism of the
brain: effects of restricted and enriched environments on the utilization of
Leucjne-H3, Physiology and Behavior. 1966, vol. 1 , Permagon Press.
Diamond, M. C., Diamond, R. M., Bennett, E. L., Krech, D., and Rosenzweig, M. R.
Distribution of acetylcholinesterase in cerebral cortical tissue slices from maze
bright and maze dull sliains of rats. Anat, Rec., 1961, 139,221 (abstract).
Diamond , M, C., Rosenzweig, M. R., and Krech, D. Relationships between body weight
and skull development in rats raised in enriched and impoverished conditions. J.
Exp, Zool., 1965, 160,29-35.
Krech, D. Heredity, environment, brain and problem solving, Internatl. Cong. Psych.,

Moscow, August 1966.

Krech, D., Rosenzweig, M. R., Bennett, E. L., and Krueckel, B. Enzyme concentra-
tions in the brain and adjustive behavior patterns. Science, 1954, 120, 994-996.
Krech, D., Rosenzweig, M. R., and Bennett, E. L. Dimensions of discrimination and
level of chalinest erase activity in the cerebral cortex of the rat J. Comp. Physiol.
Psychol, 1956,49,261-268.
Krech, D., Rosenzweig, M. R., and Bennett, E. L. Cholinesterase activity. Science,

1958, 12S, 1 176.

Krech, D., Rosenzweig, M. R., and Bennett, E. L Correlations between brain
cholinesterase and brain weight within two strains of rats.Amer.J. Physio!., 1959,
196, 31-32. , , .,

Krech, D., Rosenzweig, M. R., and Bennett, E. L. Effects of environmental complexity

and training on brain chemistry. J. Comp. Physiol. Psycho!,, I960, 53, 509-519.
Krech, D., Rosenzweig, M. R., and Bennett, E. L. Inter hemispheric effects of cortical

lesions on brain biochemistry. Science, 1960, 132, 352-353. . . • ,

Krech, D., Rosenzweig, M. R., and Bennett, E. L. Relations between brain chemistry
and problem solving, among rats raised in enriched and impoverished environments.
J. Comp. Physiol Psycho!., 1962,55,801-807. .

Krech, D." Rosenzweig, M. R., and Bennett, E, L. Effects of complex environment and
blindness on rat brain. Arch. Neurol, 1963,8,403-412. hphflvinr APA

Rosenzweig, M. R. Environmental complexity, cerebral change, and behavior, ArA
Chicago, Sept. 5, 1965.

555

Rosenzweig, M. R Evidence for anatomical and chemical changes m the brain dm

primary learning. Symposium on Biological Bases of Memory Traces, Intern

Cong. Psych., Moscow, August 1966
Rosenzweig, M R., Bennett. E. L., and Krech, D. Cerebral effects of envjronmer

complexity and training among adult rats / Comp. Physiol Psychol, J964,

438-439.
Rosenzweig, M. R., Krech, D., and Bennett, E. L. Acetylcholine metabolism a

behavior ot rats. Science, 1959, 129, 62-64.
Rosenzweig, M. R., Krech, D., and Bennett, E. L. A search for relations between brs

chemistry and behavior Psychol Bull, 1960, 57,476-492.
Rosenzweig, M. R., Krech, D., Bennett, E. L., and Diamond, M. C. Effects of enviro

mental complexity and training on brain chemistry and anatomy: a replication ai

extension,/. Comp. Physiol. Psychol., 1962, 55, 429-437.
Rosenzweig, M. R., Krech, D., and Bennett, E. L. Effects of environmental cotnplcxil

and training on brain anatomy and cholmeslerase m rat. Fed. Pwc,, 1962, 21,3j

(abstract)
Rosenzweig, M. R., Krech, D., and Bennett, E. L,, Effects of environment!

complexity and training on brain chemistry and anatomy among mature rats. Fee

Proc., 1963, 22, 515 (abstract).
Rosenzweig, M. R., Krech, D., and Bennett, E. L. Effects of differential experience o,

brain. AChE and ChE and brain anatomy in the rat, as a function of strain and age

A mer. Psychologist, 1963, 18,430.
Rosenzweig, M. R., Krech, D., and Bennett, E. L. Strain differences in cerebra

responses to environmental complexity and training. Fed. Proc., 1964, 23, 25;

(abstract).
Rosenzweig, M. R,, Krech, D., and Bennett, E. L. Modifying brain chemistry and

anatomy by enrichment or impoverishment of experience. In G. Newton and S.

Levine (Eds.), Readings in early experience, 1 965, in piess.
Woolley, D. E., Rosenzweig, M. R., Krech, D., Bennett, E. L., and Timiras, P. S. Strain

and sex differences in threshold and pattern of electroshock convulsions in rals.

The Physiologist, 1 960, 3, 1 82 (abstract).
Woolley, D. B., Timiras, P. S., Rosenzweig, M. R., Krech, D,, and Bennett, E. L Strain

and sex differences m electroshock convulsions of the lat. Nature, J961, 190,

515-516.
Woolley, D. E., Timiras, P. S., Rosenzweig, M. R., Krech, D., and Bennett, E. L. Strain

differences in seizme responses and brain cholinesterase activity in rats. Proc, we.

exp, biol Med., 1963, 112, 781-785.

556

Investigator-
James L. McGaugh, Ph. D.

University of California
Irvine, Calif.

Prepared by:

Gay Luce

Introduction

The suivival of mankind may now hang upon how well we can educate
each coming generation, since the capacity for sophisticated and coopera-
tive behavior has become the essential of modern civilization. Ironically,
GUI educational institutions lag behind us. Agriculture and medicine have
been transformed by modern scientific research while the schools remain
virtually untouched by the 20th century. One aspect of the scientific
Harvest that might indeed influence the training of the young is the
GP pl°Jatlon of biological bases for learning and memory. We can no longer
atlord the luxury of ignoring important problems of memory and their
implications for the manner in which we teach.

The social implications of some researches in this area are just beginning
to be known. Experiments with rats have divulged a genetic base for
learning. As University of California Professors Krech and Rosenzweig and
their associates have demonstrated, it is possible to influence the learning
ability of rodents by enriching or impoverishing the environment, thereby
also altering brain biochemistry. Without stimulation, there can be little
learning. Without memory, there can be no learning. There appear to be
many processes in memory, some related to short-term memory, some
related to long-term storage, and others related to retrieval. There are
sorne patients with brain damage, for example, who appear to be normal,
with unaffected IQ's, yet they cannot learn anything new, not even a new
iiome address. Dr. McGaugh and his associates have used electroconvulsive
sliock to produce a similar amnesia for new experience in animals. By this
device he has seen evidence that there may be two separate stages in the
Implantation of any memory.

The potentiality for improving education by using such research has
been demonstrated by Prof. Millard Madsen of UCLA. He has shown how
knowledgeable timing of information will permit children with low IQ
scores to learn quite as well and almost as fast as children with high IQ's.

Tn studying biochemical influences upon memory, Doctor McGaugh has
recently found that biological time of day may also influence learning
toeliavior.

557

History

In 1917, the psychologist William Lashley observed that rats would
learn to run a maze with more celerity than usual if they were fiist given
low doses of strychnine sulfate. The implication that drugs might in-
fluence learning or be useful in exploring memory processes was ignored
for several decades. Then, stimulant drugs, known as analeptics, were
again observed to facilitate learning.

In recent experiments strychnine, picrotoxin, pentylenetetrazole, and
diazadamantanol have been used on animals performing prescribed and
measurable tasks. Some have been discrimination problems, in which the
animals learned to choose among alternative paths in a maze in order to
reach a goal and to discriminate between black and white gates, while in
otheis they had to escape from a situation in which they would be
punished or learn restraint in order to avoid electric shock. The situations,
while limited, were precisely controlled and therefore quantifiable. In
most of the early studies, animals were first given stimulants a few
minutes before their first encounter with the training, and were trained
under drug influence. Did the drug influence the process we call memory?
Perhaps not. Perhaps it improved performance by making the animal more
attentive, by sharpening his perceptions, by enhancing his motivation, or
by improving muscular coordination. Research on learning resembles the
divergent reports from six blind men describing an elephant, each touch-
ing a different part of the beast. The impact of a drug upon "learning" is
partly dependent upon the measure of learning. One easily quantified part
of performance is the response latency-how long it takes the animal to
get around to making a response, When this measure is the criterion of
learning, a drug that increases response speed or alertness will also seem to
improve learning. Since no experimenter can afford to use many criteria
of performance, studies of learning and memory typically contain some
knotty problems of method and measurement. Doctor McGaugh and his
associates began to train unclrugged animals, but gave them "memory"
tests while they were drugged. Perhaps drugs altered dimensions of per-
formance, although they did not act upon memory processes during learn-
ing. It soon appeared that the memory process, itself, was exceedingly
subtle.

Retrograde Amnesia

One means of exploring memory is by training a creature on a narrowly
prescribed performance and administering drugs or convulsive shock at
intervals directly after the training. Will they interfere with the consolida-
tion of memory? Does it matter how soon after training the shock or drug
is given? In the course of many such experiments, the grantee and his
associates found that animals shocked immediately, or at short intervals
after training, seemed later not to recall. By strategically varying their
treatment of the animals, they saw that training or experience seemed to

558

initiate a kind of potential residue for memory. Yet this potentially per-
manent memory would remain labile for long periods, perhaps hours,
before being permanently etched into the neural code of memory. These
studies in which memory was impaired by drug or shock interference led
Dr. McGaugh to wonder about enhancing memory in a similar manner. If
there is a long period after training in which memory processes are active,
a memory-enhancing drug could be given in this interval after training and
should improve performance or make the animal resistant to calculated
amnesia-electroshock.

Drugs After Training

The training procedures were straightforward. Animals were condi-
tioned to push a lever or avoid a grid, etc. Then, after training, they were
injected ^ with drugs. After a suitable interval allowing the drugs to be
metabolized, the animals were tested on the original procedure. A number
of drugs appeared to enhance performance when injected after training-
and by implication seemed to be acting upon the mysterious processes of
memory. Strychnine, picrotoxin (at low doses), and amphetamine all
seemed to enhance the learning of animals who received doses directly
after training; on tests they outperformed the animals who had received
only a placebo.

^ During the last few years, several drugs have been given to animals on a
wide range of learning tasks, measured by various criteria. Clearly, the
outcome is some blend of the kind of learning (it is vastly different to
learn to discriminate between two colors than to learn to avoid a shock at
the toll of a bell), the experimental conditions, the drug, and the amount
of the drug used. Results from laboratories around the country are not all
in accordance, but most of the evidence suggests that certain drugs en-
hance learning. If so, presumably, they are acting upon memoiy storage in
Its labile period. Presumably, these same drugs would do nothing for
memory if they were injected at some maximum time after training. By
carefully graded experiments, therefore, one might expect to delineate
How long the labile period of memory lasts.

The procedure in this laboratory was straightforward. One group of
animals would be injected with a drug 5 seconds after they responded to a
learning procedure. Another group would be injected after 1 minute, still
another after 5 minutes, and so on. During 1962, Doctor McGaugh was
injecting strychnine at various intervals after training: he found that one
strain of mice "learned" better if injected immediately after training, yet
injections given a half hour after training were ineffective. On a discrimi-
nation task, one group of mice showed the greatest facilitation-by com-
parison with undrugged controls-if they were injected between 5 and 1<:
minutes before training.

The results of these studies of effects of time o
drugs indicate that the effects are time-dependei
of the facilitating effect decreases with the intervi

lllt-Sv I >Y\_) Hll^** \J L ^-rin^'i'^'v ^F-^*-v»** • - • j », *„ -^ --u j- i .- - - -

that memory storage processes are susceptible to both

and impairing influences for a relatively long period of time folio

ing training.

Environment

The elusiveness and delicacy of the process of memory and (lie sen-
tivity of rodents-has forced the investigators to take nil kinds ofprcviu
tions against slamming doors and disturbance in the laboratory. Tin." ML-L-
for constant temperature, quiet, etc., was underscored by an cxpuiimcn
in which mice, given strychnine, weie given a discrimination problem ;IIK
were disrupted by environmental stimulation. Like .sUuluiMs Irymji u
memorize a lesson in an "acid-rock" discotheque under strobe liphK tail]
these mice were rocked back and forth in their cuge, exposed to flash JIIL
lights and bursts of sound for 20 minutes. On tests, il WHS Hie oilier lull
of the animais-who remained in dark quiet cagos- -who showed unhnncfd
learning from strychnine while the stimulated animals did worse Hun
controls. It was an encouragement for control in the learning
and perhaps also a hint for humans.

Drug Attenuation of Retrograde Amnesia

Inevitably, the investigator wondered whether strychnine itml oilier
drugs enhanced learning by accelerating the rate a| which Jiiemory f,Viu'*
were consolidated into permanent memory. If so, the (IniKs sliouhl /in-
vent or attenuate the kind of retrograde amnesia (Jim is causwl l>y utiivtri.
sive shock. Animals were given saline solution or strychnine just he I ore ur
just after training: then, within a few minutes each uniimil WHS shockviJ A
day iater each mouse was tested on the training (ask. The i1miMii
•niniate die! a little better on tests than did the con.rols who r««. v S

Recently the experimenters have used a diffurent kind
^ar^tTS^ SUmdCm '^ " teSl Of IM^«-

Plairo"»

The mouse, seeking the comfort and security of a dark box, would step
in the hole and immediately get a shock on the foot. Twenty-four hours
later he would be placed on the platform again. Would he remember that
the hole leads to a "forbidden" place? Would he restrain his natuial urge
for security? The length of time that he would hesitate on the platform
before attempting to enter the box would represent a degree of learning.
This time, resisting temptation, as it were, was measured as the criterion
of learning.

Each mouse was placed on the platform, received a footshock when it
transgressed the boundary and then received electroconvulsive shock. For
some, the shock came within 18 seconds; for some, it came after 18
seconds; and other animals received it an hour or 3 hours later. Some of
these mice were on saline solution, whereas half the mice received
strychnine either 1 0 minutes before or a minute after the platform train-
ing.

Twenty-four hours later, each mouse was put to the test, placed on the
platform, and clocked. The amount of time an animal stayed on the
platform without moving through the hole was directly proportional to
the amount of time he had been allowed between his first experience and
electroconvulsive shock. The animals given shock 3 horns after training
showed to amnesia at all. Control animals given shock 18 seconds or a
minute after their training experience seemed to have forgotten that entry
was forbidden. Those pretested with strychnine were not rendered so
completely amnesic by shock at the same short inteivals. Oddly enough,
some memory also persisted in the animals who had injections of strych-
nine after the electroshock.

If strychnine affects learning by acclerating consolidation pro-
cesses, these injections should have no effect. However* as can be
seen, retention of these animals was superior to that of controls.
Why should strychnine counteract the amnesia effects when injected
after electroshock?

In the next series of experiments, strychnine was administered at vary-
ing intervals after the electroshock. Evidently the critical interval was the
timing of the shock. Strychnine did not improve the retention of an
animal given shock 8 seconds after training; yet if shock was delayed a
minute or 3 hours, strychnine made a difference. In subsequent experi-
ments, it became evident that the timing of the strychnine was also perti-
nent. The longer the interval between the drug and the shock, the less
effect the drug had in improving memory. If injected 9 hours aftershock,
it was totally ineffective. Further studies showed the same effects with
other stimulants such as picrotoxin.

It appeared that these drugs did not act-as previously believed-by
accelerating memory consolidation. Otherwise, why should they cancel
the amnesic effects of electroshock?

Doctor McGaugh and his associates conjectured that ther
distinct phases in memory consolidation. Pei
tion of a pattern in the brain which is then
ma terial in the permanent "files" of the mind

the first step if given instantly, within 8 seconds of an experiuntY, p
dedicating that initial pattern. However, if that first palteni wci
mitted time to form-taking more than 8 seconds— then shock Mov
second stage of permanent storage. This means that if one WL'IC lo
sufficiently long time after a training trial, the first phase ol' UK
would be impervious to erasure by shock. What shock docs ;it I
intervals, is to obliterate the second stage— and it was the sec'otul sl.i#<
was influenced by drugs.

Drugs might affect learning in many ways, and as we conic Id it
stand them it may be possible to evolve chemotherapies Cor if ok-
memory, such as those of the elderly or the retarded child, or tlisoi
like those of brain-damaged patients. In recent years ninny
have postulated that the synthesis of nucleic acids nuisl In1
memory. Antibiotics can inhibit the synthesis of RNA, yet
not impair learning in animals. Puromysin inhibits prole in synilu'iK.
when injected into the brain of a goldfish has produced effects rcscinh
those of electroshock. Some researchers have postulated Unit im-m
depends upon changing levels of ncurotiansmittcrs such iis acelyMjoJi
Using chemicals that effectually raise the levels of ncutylcltoJijio iit i
brain, one laboratory found that there were extraordinary cITtvls up
memory -an injection might eradicate memory of a task loimuMl I f Jj
earlier but enhance a performance learned 30 diiys bct'oic. rie.irl
memory processes arc time-dependent, and changes occur in MIIIIC rut
terious but presumably regular order.

A new dimension has been added to the study of
the finding that biological time of day influences memory-

Daily Rhythms of Memory

In recent studies with Dr. Gwen Stephens, the grantee has (ouml ih
mice retain inhibitory training best if trained at night, when (J,cv ..H
the peak of their body temperature and motor activity for the 24 ton*
indeed .just as training seems more vividly impressed at Iliis nmc. ili

? mSft^f? *CtS. Sh°Ck are als° more Pronounwd when they l^,,,,
3 minutes after training at this peak time.

Wilhin l

th ,f , int° the box> on|y to ^ sliockcil. -u

*heplatform withi» 5 seconds, somo seemed not to mnenilv

M ,

and would step agmn into the forbidden place. However, if

ndexperience ™ lhc P^tfonn after « del y el oy

hove , r « e y e oy

hover there for a long time before venturing in, A clay Inter II ev * .";

2±f "" l'rge t0 eiUCr the d«?k *** com^bJc ol * n
tor remembering was set at 30 seconcls-for If the anintiiJ hwmJ

•^ rccaiied uic ^*- ""ho r^;;;

^^^ ™ SOT Ioss 'n the strength of h^
between 30 seconds and zero seconds was taktn to

562

measure the loss of memory, since animals placed on the platfoim the fiist
time would often hesitate for a few seconds before exploring.

A recent seiies of experiments was run in several ways. Some animals
never icceived more than one experience on the platform. Others were
placed there seveial times in succession. Groups of animals destined for
the same schedule of treatment were trained and tested at trough and
crest of the 24-hour temperatuie-activity rhythm.

Ironically, much of the study of memory and the impact of drugs has
been conducted with rodents. They are nocturnal animals, whose peak
activity occurs in darkness. Yet these animals are treated in most labora-
tories as if they were diurnal. Usually they are tested at times when they
would normally be resting and sleeping.

The influence of experimental ordei within a training and treatment
session was noticed in 60 mice who were put on the platform, given
shock, and later tested. It seemed to matter whether the mice had been in
the early part, the middle, or the late portion of an experiment, since the
training and treatment of 60 animals stretched out across 2 or 3 horns,
Weie the animals responding to noise in the laboiatory, to lighting, to the
experimenters?

Under the suggestion that time of day might matter, experimental data
for one study were submitted to a double analysis. The usual analysis
scored groups of animals for their retention, grouping them according to
the elapsed time between training and shock. Those animals shocked
within 5 seconds of training showed almost no retention; those shocked
after 15 seconds or many minutes later showed signs of memory in pro-
portion to the interval between training and shock. However, the neat
correlation between retention and elapsed time before shock was thrown
into complication, if the data were grouped differently. If averages were
lakcn by cage of four animals in the order trained—regardless of the
interval between training and shock— the retest scores show a striking
fluctuation. No longer is there a clean straight line, suggesting the direct
correlation between elapsed time before shock and retention. Instead, it
would seem that physiologic status of the individual animal at the time of
training and treatment is a big factor in the retention of the animal-
icgardless of the length of time between training and shock.

The experimenters discovered, for instance, that the animals trained in
the first 20 minutes of a 2-hour experiment were very different from
those trained in the last third of the experiment. They showed no correla-
tion between retention and the interval between training and shock. Yet
those animals processed later did show a correlation. Was this due to the
experimenters, to noise in the laboratory, to time of day?

In these "time of day" experiments, the animals have been housed in
quiet cages, protected from disturbance, and maintained at a constant
temperature on a rigid lighting schedule. The mice live in darkness from 4
p.m. to 4 a.m. and in light between 4 a.m. and 4 p.m. This entrains the
animals' temperature rhythm so that at the 1 p.m. experimental hour they
are at trough values; at the 9 p.m. session they are at crest.

563

Time of Day and Amnesia

Recently, an experimental population of 72 mice was divided into six
groups. Half were processed at 1 p.m., the middle of the animals' res!
period; the other half were processed at 9 p.m. at about the peak of
activity, In each time period, for each of three conditions there was a
control group of animals who received only a sham electroshock, while
the others were put into convulsion. Each animal went through three trials
on the platform, one group receiving shock at 3 minutes and another an
hour after the training. The amnesia following shock was most pro-
nounced in the animals treated at 9 p.m. This surprised the experimenters
who commented:

"This could mean either that memory consolidation varies inverse-
ly with metabolic activities or susceptibility to electro convulsive
shock varies directly with metabolic activity. Whatever the basis,
such variations must be considered in experimental investigations of
memory storage processes."

The greatest response to shock occurred at the time when the animal's
temperature reached its daily peak, at a time when there would be a high
concentration of adrenal steroids in the blood. A correspondence between
biological time of day and the strength of conditoned fear and also extinc-
tion has been observed by Dr. Charles Stroebel of the Institute of Living
in Hartford. He postulated that certain emotional conditioning might
show a time-lock-iinking strength of memory and the biological time of
day at which the conditioning took place.

In a series of experiments conducted in Doctor McGaugh's laboratory,
it was found that animals treated by day differed from the animals treated
at night. Indeed, animals who were put into shock right after multiple
experiences on the platform were unusually resistant to the amnesic ef-
fects of shock if they had been trained between 9 and 1 1 o'clock in the
evening-their period of high temperature and peak activity. This sug-
gested again that retention of avoidance reactions must be strongly in-
fluenced by physiological cycles.

An animal's temperature curve over 24 hours is not a sine wave. Tem-
peratures were obtained from 72 animals on a schedule of 1 2 hours of
light and 1 2 hours of darkness. These animals were protected from dis-
turbance and maintained on a steady, even room temperature. In the
course of 3 days, it was clear that the rectal temperature would show a 2*
to 3° change each 24 hours. The peak would occur around 9 p.m., and the
nadir at about 1 p.m. These were now the hours selected for experimental
trials.

In repeated trials on the platform, with shock or sham shock delivered
at varying intervals, it was possible to examine the animals' rates of learn-
ing. How quickly did they learn and how resistant were they to forgetting
that they must hover on that platform and not enter the forbidden hole?
In several replications of the experiment at different hours of day and
night, the animals' rates of learning and persistence of memory were
scored, Both seemed to fluctuate, rising and falling rhythmically around

564

the clock. It suggested that the animals were fluctuating in some basic
level of arousal.

dearly, retrograde amnesia did not simply depend upon the time al-
lowed between the acquisition of some new learning and an electro convul-
sive shock. Indeed, it depended upon the phase of the animal's tempera-
ture rhythms and other factors. The most sizable fluctuation seemed to
correspond with the daily rise and fall in adrenal steroids and temperature.
Almost every system in the body is known to show a roughly 24-hour
fluctuation in function.

Recently, the investigators have been shifting the lighting schedule 3
hours in older to see how animals learn and remember during periods of
biological transition. If the lighting is shifted by 3 hours, there are some
days in which the animals are readjusting to the new day-night cycle, and
they no longer show their former crest of activity and temperature. Now,
when the experiments are run at 1 p.m. and 9 p.m., the memory gradients
aie no longer the same. Perhaps the disruption of sleep schedules, as this
experiment suggests, also disrupts some memory functions.

Implications

All of these facets in the exploration of the biology of memory are
bound to have their impact upon education. Clearly, the more we under-
stand about a student and his properties as a learner, the more effectively
he can be taught. Unfortunately, today's classrooms are not run according
to knowledge about the nature of memory and learning. Educators gener-
ally assume that such information is irrelevant despite the fact that
memory is one of the most important aspects of a child's learning. One of
the few contributions accepted by educators from psychology has been
the IQ test. This is not used as it should be, for diagnostic purposes. This
test attempts to say something about the child as a learner and to predict
something about the child's capacity to learn. But the current use of IQ
tests reminds one of a physician who examines his patient and says. "You
have a very bad liver and you are likely to die an early death. Very
interesting, you have 3 clays to live."

In a very crude sense, the IQ test appears to measure some adaptability
that is biologically based. Many deficiencies as well as talents are bio-
logically based. Some are metabolic, like phenylketonuria, which can be
counteracted by a diet starting at birth. If it is not corrected, the defect
leads to mental retardation. Biological variations of normal intelligence
and learning are not presently analyzed and used for the purposes of
education, Where there is strong evidence that family resemblances and
intelligence have a genetic basis, general notions of intelligence are useless.
There are diverse reasons why a person may be a good or a poor learner.
For example, a child's learning efficiency depends on many different
processes. He might be a poor learner because he has a deficiency in
several systems of storage and retrieval. He might have no short-term

565

memory or he may not be able to retrieve information. Some people have
good immediate memory but have lost the ability to memorize; that is, to
learn any new material and store it for any length of time.

Dr. Breiida Milner of the Montreal Neuiological Institute has studied
memory in patients with brain damage in both temporal lobes. Often
these people seem to be quite normal and their IQ scores are unaffected.
Some of them have completely lost the ability to acquire and retain new
information. For instance, one man had suffered brain damage 10 years
ago. While in many respects he seems normal, his family moved and he
never was able to learn the address of his new house. He does the same
jigsaw puzzles day after day without ever showing any effect from prac-
tice. He reads the same magazine over and over again, but doesn't find the
contents familiar. This type of amnesia has proved to be very helpful in
understanding various aspects of memory. Some of the problems of school
children may stem fiom kinds of memory-retrieval defects that normal
school techniques never overcome,

As Dr. Millard Madsen of UCLA has shown, high IQ and low IQ chil-
dren can be induced to learn at about the same rate under the appropriate
conditions. Undei the usual schoolroom conditions, the children with high
IQ's seem to learn much faster. However, if the low IQ children are
permitted a longer interval between the introduction of new information,
they learn at about the same rate as the high IQ children. Thus, rearrang-
ing the school environment to meet the individual's needs can eliminate
the differences predicted by IQ scores. The role of our present diagnostic
tests, such as the IQ test, should allow us to find out what information-
storing process is deficient and how to compensate for it.

The day of the memory pill is not quite here, but it is coming soon. In
the future it may be just as common to give Johnny his pill when he leaves
the house as it is to remind him to brush his teeth and to put on his
glasses.

The relation between long-term memory and the intervals between
training, the influence of shock, of drugs, may indeed revolutionize the
intellectual capacity of humankind. We may also learn how biological
rhythms enhance or detract from learning and may schedule learning for
the optimum time of day. We may learn how to encounteract memory
deficiencies with drugs- thus opening the doors for almost half of man-
kind more fully to realize their potential. Children may receive a more
compact and rich education when we begin to respond to the way they
learn. For the elderly, too, the advent of memory drugs may prolong the
philosophical years of intellectual activity now curtailed by poor memory.

Research Grant: MH 12526

Date of Interview: September 1968

566

Investigator:

Seymour Levine, Ph. D.

Stanford University Medical School

Stanford, Calif.

Prepared by:

Gay Luce

How Experience Shapes the Infant

A dominant child-rearing philosophy -often accredited to Freud-warns
that trauma and distress in infancy and childhood may be the source of
neurosis and adult instability. As Americans have become more affluent,
increasing numbers of financially secure and devoted parents have ab-
sorbed the popular distillations of child-rearing theoiies, and have been
striving to protect their offspring from some of the nastier realities and
contretemps that they themselves encountered in early life. A variety of
protections have been built into the environment of the middle-class child.
The amount of stimulation in infancy has been reduced by bottles and
pacifiers. In extreme designs for infant care with minimal handling there
are thermostatically controlled glass cribs with cloth rollers that obviate
even the necessity of changing diapers.

What arc the consequences of increasingly automated infant care fol-
lowed by a highly protected childhood? There is no way of telling from
case histories in retrospect. Instability and neurosis do not appear to have
decreased notably. Some observers have cited the Korean war as evidence
that Americans were growing soft, for jarring and exaggerated analyses of
American soldiers had implied that they capitulated to the enemy under
only moderate stress. Some reporters thought these POW's had be en i ren-
dered unduly vulnerable to stress by comfortable and protected back-
grounds. For a time, the middle-class home was blamed for defections and
surrenders in Korea although they were not statistically different trom
defections and surrenders in other wars and among other groups-Due rms

tor in

attempt to understand and explain behavior was not

retrospect there could be no measure of the Pf

tliat affected the POW's and might have been ^m*

war. It was an object lesson in the difficulty of dcternann« ca uw an

effect, and might have thrown into question the case study basis tor

prevailing theories of rearing the young.

567

Until the last decade there was remarkably little scientific study of the
developmental process. There was a small literature containing a few con-
trolled studies of human infants, observations by ethologists, and a few
neurophysiological studies. But extensive developmental research had to
await an instrumentation that was not yet developed, and also a change of
attitudes, A good many explorations could not be conducted upon human
infants; the physiological effects of infant experience had to be learned in
laboratory experiments on animals. Initially, these studies of infant experi-
ence made the presumption that the most important, most scarring effects
might be blamed on trauma-which was roughly assumed to be equivalent
to any intense and unpleasant experience. Scientists searched for the ef-
fects of trauma.

Many early attempts to find the effects of trauma in infancy were often
conducted with mice or rats, but the methodology left the question of
effects unanswered. During the 1950's Dr. Levine and his various col-
leagues began to compare groups of infant animals: one group would be
subjected to handling, shock, and other treatments while the control
group was strictly exempt from all treatment and handling. Comparisons
of these groups clearly suggested that even a seemingly mild experience
would leave perceptible traces in the animal's behavior and physiology.

In some studies the infant mice or rats were shaken in bottles, shocked,
or vibrated in their cages. The investigators expected their tests to show
that these manhandled infants were neurotic and emotional while the
controls behaved properly and matured faster. The effect seemed to occur
in reverse. Rats handled in infancy might be placed in a large open con-
tainer as adults and would show little timidity, They would rummage
about-white the controls cowered in corners. It seemed, indeed, that tlie
handled animal was less emotional.

One measure of distress and emotionality in rodents has been the open
field test. When taken from a cage and placed in a larger barren container,
the emotional and distressed rat will tend to crouch and cower in one spot
rather than wander around. He will also defecate and urinate a great deal,
The calmer animal will eliminate much less and will explore his new
setting. Involuntary elimination under fear and stress has provided a good
many battle jokes, for it has been a relatively common experience among
fighting men, It is one of the responses of the sympathetic nervous system
when activated by the neurochemical network that functions during
stress. The indices of nervous elimination and activity in open field tests
have been useful measures of behavioral-physiological differences among
rodents.

Another criterion used to judge a rat's adaptive behavior is avoidance
conditioning, a procedure in which the animal must press a bar or perform
some action in order to escape being shocked, A traumatized animal might
be expected to learn avoidance very slowly. With this expectation, an
early experiment divided the infant rats into three groups. One group was
shocked, one group was put through all the motions preparatory to shock
but not the shock itself, and the third group was left to rest quietly in
Isolated cages-never handled at all, The shock was evidently painful

568

enough that the infant rats squeaked and tried to escape, and the experi-
menters expected that these creatures might later show signs of trauma.
When they became mature they were placed in an avoidance cage. The
previously shocked animals learned to avoid shock most rapidly, with the
other handled animals pel forming nearly as well. However, when the
totally untieated animals were put to the test they did not react normally
at all. Instead of running and jumping when shocked they would tend to
freeze and defecate. Their lack of infantile experience had apparently left
them incapable of confronting a new situation, and they were so flustered
that they showed what might be called, in human terms, the signs of a
paralyzing fear.

Infant experience, even if quite noxious, seemed to gear the animal to
meet new situations and stress in adulthood, enhancing his capacity to
survive in a changing environment. As subsequent experiments indicated,
lack of infant experience was indeed a severe handicap in the survival
behavior of the rat. Even when quite thirsty the imhandled rats would
drink little in a new situation. When shocked they became so flustered
they drank less and less. Here again, the rats had been divided into three
groups. All were deprived of water for nearly a day. Then they were given
water and the amount they drank was measured. The untreated rats drank
least.

Latei , when the thirsty rats were shocked before receiving water, all of
the groups drank less, but as this procedure continued the untreated group
drank least of all. Throughout the study this naive group persisted in
drinking very little, whereas the shocked and handled groups apparently
got used to the situation and began drinking more. The creature with an
infancy empty of varying experience appeared so overcome when con-
fronted with stress that it would not even drink when very thirsty.

Further experiments began to suggest that the inexperienced animal
might react poorly to an aversive test, but if it were merely novel, the
animals would perform more adequately as they grew accustomed to it.
They simply took much longer to adjust than did the handled creatures.
A deeper exploration of the animals's reaction to shock indicated that
the handled animals would learn avoidance faster at low levels of shock.
Although these creatures were supposed to be less emotional, they gave
signs of being more responsive than the untreated creatures to acute
shock. Infantile stimulation did not merely make an animal less emo-
tional: apparently it enabled him to make discriminations about relevant
attributes of the environment and adjust his responses accordingly.

By what physiological process was this experience altering the emo-
tional reactivity and learning ability to the animals? The effects of infan-
tile stimulation were explored by Dr. Levine and others in a multitude of
experiments. They soon found themselves examining the endocrine sys-
tem which suffuses the animal with emergency chemicals during stress.
One measure of stress in an animal is the level of certain adrenal steroids

in the blood.

When crisis occurs and an animal is threatened, the central nervous
system transmits the alarm to two important and interconnected glandular

569

systems. Messages from the brain will set into action the pituitary, an
inconspicuous egg-shaped gland which in man resides just above the nasal
passages, at the base of the brain. The pituitary controls quite a few
specialized sex hormones, exerts influence upon the thyroid gland, con-
tiols growth hormones. On occasions of stress it releases ACTH, adreno-
corticotropic hormone, which mobilizes the adrenal glands far distant
The adienals, two small yellowish capsules above the kidneys, respond to
ACTH by releasing several hormones into the system. One of these is r?ie
familiar stimulant adrenalin, whose action increases the available blood
sugar. When a stress outpouring continues for some time, the adrenal
glands will enlarge. Thus a good deal can be learned about an animal's
intensity of stress reaction by measuring adrenal steroids in the blood or
by weighing the adrenal glands themselves, and analyzing their content.
These measures indeed may be used to predict how an animal will behave,
although not invariably, for many genetic factors must be taken into
account.

As Dr, Levine and his associates probed the consequences of infant
handling, they found that a detectable and sometimes exceedingly largo
physiological change occurs and persists into adulthood. In lecent studies
they have been examining the differences between handled and untreated
rais after shock-now by a combination of behavioral and physical tests.

Manipulated infant rats and untouched controls were subjected to
shock: during the first 15 minutes after shock it was clear that the
handled animals produced more adrenal steroids. At first it looked as if
handling had made the animal overreactive, but in fact it simply caused
him to respond fast. When shocked, he would summon his adrenal forces

rapidly, receiving a rich suffusion of emergency steroids for the period of
stress, bvit the steroid level would subside quickly buck to normal. The
unhandled rats, by contrast, were reacting slowly to stress. They con-
tinued pouring emergency fuel into their systems long after the crisis had
passed, They might be like the peison who awakens slowly to disaster and
remains perturbed and anxious long after it has passed. They are not
geared to act promptly but endure a prolonged aftermath of stress physi-
ology which might indeed incur damages, perhaps causing ulcers or suscep-
tibility to infection,

Handling in infancy, as a long program of studies began to show,
improved the stress physiology in the maturing animal and adult. The
investigators began to find that the handled animals matured far more
rapidly than the unhandled controls. There were many indices for the
maturation rate. Ordinarily the pituitary gland of the laboratory rat docs
not release ACTH until the animal is about 16 clays old. Handled rats,
however, were producing and releasing ACTH by about 1 2 days.

The brains differed. As the infant brain matures, all the nerve fibers in
the brain and body that link neural cells become ensheathed in a fatty
substance, myelin. This white matter, as it is often called, is largely com-
posed of cholesterol. When the brains of handled infant rats were com-
pared with those of the unmanipulated controls, the handled rats showed

570

considerably more cholesterol, perhaps indicating that myelinization was
occurring faster.

Experiments with deliberately induced brain damage again pointed up
the difference between the handled and unhandled animals, Lesions
within the septal region of the hippocampus cause hyperexcitability and
even viciousness. After lesions were made, however, the handled animals
did not become as excitable and certainly not as vicious as did the un-
handled controls. These brain-damaged innocents, the experimenters
noted, were the most vicious little i orients they had ever seen and would
pursue people around the laboiatory, attacking their ankles and legs.

The overt maturation rate of the handled infants was well in advance of
the unhandled controls. They simply grew faster, opened their eyes
sooner, moved in a coordinated fashion earlier, developing a good coat of
fur and gaining weight rapidly. Since these handled rats did not appear to
be eating more than their naive controls, it was speculated that they might
have developed a more efficient protein metabolism. The handled group
survived the longest. The infant who had been moved around, joggled,
shocked, and subjected to various stresses was clearly better equipped to
survive than the creature who had been raised in a completely sterile
fashion.

There is, of course, a huge difference between the laboratory environ-
ment of the piotected animal and the environment he would enjoy living
in a cage with his mother and littermates or, surely, in the natural environ-
ment. His is a world composed of a temperature-controlled cage, aseptic
and barren, with light and sound artificially regulated. Thus, the labor-
atory study of infant experience compares animals subjected to extreme
and purified conditions. The experimenters have consequently been very
cautious about generalizing their findings to human beings. What they
Jiave indicated, in their purification of environments, is that some range of
stimulation even if noxious must be necessary to the normal maturation
of a mammal and will leave its imprint upon the physiology and behavior
of the adult animal. Ordinarily neither animals nor infant humans pass
their earliest months in an environmental vacuum. Perhaps the closest
approximation to a vacuum in the life of the human baby is that of the
orphanage. Orphanage babies are rarely picked upon, spoken to, played
with, and moved around.

Clinical studies of orphans have indicated that they were often retarded
in development, timid, unable to adapt to changing situations, and indeed
more susceptible to disease than the home child of the same age. Some
investigators attributed the debilities of the orphan to the (ack of a
mother. Today, however, animal studies make it possible to offer another
conjecture. It is the mother who lifts and handles a child, speaking to it,
Teeding, playing, and expending upon it a prodigious amount of "atten-
tion." Perhaps the missing element in the orphan's development is the
motion,, the stimulation ordinarily provided by a devoted parent. Very
recently, in the first controlled study of its kind, White and Held have
sliown that aspects of visual acuity and adaptation in very young infants
in orphanages can be "speeded up" by enriching the visual environments

571

of the crib. Only longitudinal studies will tell whether the infant in his
usual orphanage environment wiU later catch up with the experimental
infants and whether, indeed, effects of early visual experience can be
detected in future behavior.

In the animal studies of Dr. Levine and his associates, employing a
purified situation, it has become increasingly clear that manipulation
during infancy has produced more responsive adult animals. Rats that
were left strictly alone in infancy were shocked as adults: they took
considerable amounts of shock before they would learn to avoid it. Their
counterparts, handled during infancy, required little shock to teach them
avoidance. These handled rats appeared to be making more sensitive discri-
minations about events impinging on them.

The experimenters predicted that early handling would help the cre-
ature respond to new situations, and one measure of response would be
the amount of stress registered physically-the output of steroids. Rats at
the weaning stage weie set down in a new cage for a few minutes and then
their steroids weie measured, The handled rats showed n much lower
output of stress steroids than did the nonhandled controls. Laboratory
tests of this kind may seem remote from their human analogy-in part
because we have precious little empathy for rats. It may be possible to
imagine the experience of the infant rat by recalling the reactions of a
small infant. He is accustomed to sleeping in his own crib and he is
suddenly tiansported to a new house where he is left upon a large open
bed. Some infants will cry and fuss showing signs of fear and distress. The
first time an infant is lifted up and held by slrangeis he may produce a
fearful wail. Surely, the infant human shows signs of intense distress when
fust exposed to a night in a hospital, and many parents comment that
their babies are generally resistant to change. One may imagine that the
laboratory rat endures drastic experiences when he is first picked up by a
giant, removed from his familiar cage, and set down in a new place with
unfamiliar smells. The first time he con ft on ts the open-field test he must
experience some of the emotional upheavals that might occur in an infant
who is lifted from his crib and set into a gigantic baby-pen the size of a
huge loom.

Both handled and unhandled rats reacted similarly to their first open-
field test, if judged by their movements around the container and amount
of elimination, But when the experience was repeated, the effects of
handling were apparent. The handled rats now acted more "at home" and
explored more and eliminated less. The "protected" rats, on the contrary,
acted more emotional than before. They reduced their activity and defe-
cated more. They did not seem to have benefited from their first experi-
ence but acted as if this were a new and even more threatening situation.

The effects of infant handling have been studied in other species than
the rat. The monkey is, of course, much closer to man phylogentically,
and quite a number of investigators have been looking with interest at the
effects of early experience on monkeys. Recently, in collaboration with
Dr. William Mason of the Delta Primate Center, Dr. Levine and his col-
leagues have been able to follow the behavioral and physiological effects

572

of handling in infant monkeys. They have recently compared the stress
and learning responses of three pairs of monkeys. One pail was raised with
only a stationary dummy for a mother. Another pair had a robot mother
that was motorized and capable of certain motions. The third pair had
been laised natuially by their mothers in the wild. When the six animals
anived at the Stanford laboratory they bore no identification tags saying
which was which. Their behavioial differences were so great, however,
that it was quickly apparent. One of the fhst examinations required plac-
ing each monkey in a restraining chair while blood samples were drawn
and steroid levels analyzed. The experimenteis wanted to determine some-
thing about the baseline responses of the animals to ordinary laboratory
handling and to get some reactions to very minoi and standard stimula-
tion.

When seated in the restraining chair, a small blood sample was taken
from the leg, A half hour later, electrodes were applied so that galvanic
skin responses could be obtained as tones were played. Another blood
sample was drawn at this time. The motherless monkeys gave a low-
galvanic skin response and at first their steroid levels were also very low.
As the procedure continued, the steroid response rose high above the
others. When this procedure was repeated, the reactions of the feral and
motherless monkeys were more nearly the same. The experimenters began
to look at behavior.

Now the animals were placed in a choice situation. They were presented
with two different visual patterns on two panels. When the monkey
pushed the right panel he automatically received a peanut. The feral
monkeys, once they mastered the notion of pressing panels, learned the
correct discrimination in few trials. The monkeys raised with robot
mothers took more practice but learned the problem in far less time than
did the monkeys raised with stationary dummies.

When the monkeys were transfeired to a new discrimination apparatus,
the differences again became evident. The wild monkeys learned with their
former speed. The monkeys mothered by robots learned, also. But the
monkeys raised on stationary dummies never learned. Now the problem
was reversed so that the formerly wrong choice was correct and rewarded.
This taxed the adaptiveness and resilience of the monkeys. On the first
day of reversal there was no reinforcement. The feral and robot-raised
monkeys continued responding rapidly as before. On the next day, when
reward was offered for the formerly wrong choice, the feral monkeys
adapted with no show of difficulty. The robot-raised monkeys took twice
as long, behaving as if this were a totally new situation. The monkeys
raised on stationary dummies never did learn. The experimenters observed
that the maternally deprived monkeys reacted to each change, each varia-
tion in a situation as if it were now totally novel and inimical. Character-
istically, steroid samples indicated that these deprived monkeys were both
sluggish in responding and then overreactive.

An initial look at these three pairs of monkeys suggested that infant
experience might have far-reaching and subtle effects upon the animal's
later ability to learn, to generalize from one situation to its variant. From

573

the very fiist it was clear that novelty of any kind presented the mother-
less monkeys with a painful problem of adjustment. They would crouch
and cower and scream when placed in the learning cage-as if learning to
push a panel were a wretched torture. When they finally mastered a discri-
mination between two panels and were asked to make the reverse choice
among the alternatives, they acted as if they had been dropped into an
entirely new situation.

These maternally deprived animals have, of course, lacked more than
just the handling, training, and affection that their mothers would have
bestowed during critical peiiods. They lacked, as well, whatevei bio-
chemical modulation they might have received through their mother's
milk. A program of controlled studies with monkeys, beginning in 1967,
may begin to illuminate the weight of these many factors in mature ad-
justment. These studies may also begin to explain why the inexperienced
infant fails to become adept at making discriminations later. Is h& handi-
capped by a physiological system that is slow to react to a stimulus and
then overreactive? Is he thus placed in a state of anxiety, feeling so con-
tinually overwrought that he cannot take note of the subtle distinctions in
his surroundings and must react to everything?

In a decade of animal studies, the investigator has consistently observed
that infant handling influences later behavior by adjusting the animal's
level of physiological response to stress, What is the physiological mecha-
nism through which cxpei fence regulates the stress sensitivity of the mature
animal? Why does early handling instill responses to "crisis" that are
prompt and brief? Why does lack of infant experience create an adrenal
response that is sluggish and disproportionate?

Adrenal Hormones and Experience

Early studies of internal responses were often, of necessity, acute, and
the animal had to be killed to find the physical effects. Determinations of
adrenal responses were made by examining the glands themselves and by
measuring steroid levels in sizable quantities of blood. More recently the
experimenters have devised a method of determining steroid levels in very
small amounts of tissue, in a drop of blood. Very precise determinations
can be made on one drop of blood by fluorescence measurement, follow-
ing this new method— a distinct advantage when several measures are re-
quired and the experimental animal is a small rat or mouse. Using this
technique, it has been possible to follow the steroid responses in handled
and nonhandled animals over repeated tests without harming the animal,
Now it is possible to obtain frequent steroid determinations during stimu-
lation and behavior.

Since adrenal steroids seem to correspond to a level "emotionality" or
reactivity, the experimenters have inevitably asked what role these ste-
roids play in behavior. Early studies suggested that steroids appeared to
alert the animal for emergency action. When the steroid levels were exces-
sively high, however, the animal appeared to be too keyed up to make

574

cnsitivc discriminations. OIIL- ol UK* IMM whulw In, olwrving the effect
if steroid Icvols on bchnvior was llu; HVOU.IIIU. situation. Hats plnccd I in ,1
mall cage or maze with «u <'kvtid.nl Inor |intl would burn quile a
aricty of behaviors to avoid ^'H^', Nli»i'UI. In a recent Viinanl on this
voidance conditioning, rats h:ivc h;uJ lo IIMIII a U-mpomi rhythm. Jlns
as allowed the experimenters to IIUMSHIO di(inr.i's in holmvior wilhprcai
iccision.

When shocks wore piicx-d to anivr i-\viy ,M) wimls, the animal luitl lo
rcss a bar before the shock in ordn tt> pn-voni il, IK> would quickly learn
int he did not need to press tin- h.ir < uniinuothly. lie would develop a
iyUim> pi ess ing about every I H MI P> swimls. l;adi day. when he was
rat placed in (he nvoidnnci' vluimlu-i, Mir iiniiuul would receive n few
ocks as he rcailjustod to tin- apjMr.itus. 1ml his pi-rlurinmict' would soon
icoinc steady, his rhyllnn aci'tii-id', (Jivvii .in injct'liiui ol' ACTII, then
ion his performance was alivjitly M.iMi'. hi* \vniild hecoine even more
ficicnt. When placed in tin- I'hmnhrr. hr wunl<l receive fewer .shocks
ring the warimip puruxl. Ik* »|MI niailc II'WIT u^pniiiics, more nminiloly
nctl. The priming sloroid ;ip|H%tu'il l<t fiih.uuv his precision. When the
. was injected with an uvi-n IIM»JC |n»ln»( Nlerniif, he would hccoinc
rceptibly more ufficionl. "I hi' sli'mul u-riiii-d lo he enlminrinB liienlcrt-
^s of the animal.

In the noimnl course ol'evcnls. llu- .uhoii.tl sUMoiiKsiifntse Ihchodyby
:hain of reactions. A stiinultis WOM'V Ihr IU.IHI to sentl .slgnuls lo Ilio
uilary, which relenses A<vni, ihtiMJiiwnn Die iulronids to release their
milnting chemicuis into the MomMiiMMi ,iml nc«r the hotly for action.
.1 immcduiio supply of hlo.t.1 -.U}.',ii h tiu reused hy tt hrcakdown^ol
red glycogen, giving I In- huily iHMr.t eiternv lo use. Hut I lie alerting
jcls, the potent inlluoiKV iipini In'h.iviur0 must come from (he effects
ilcroids upon the ccnlnil ncrAHi^ ivsinn itwll OneMepin nscorlniinng
v these steroid hormones inlpht %h.ipr nc«r,il rtsptmscs, 1ms been lo
eh hehavior when sterofils ;m< ;ul*uirii^m-tl tliratly lo the brain- Poos
hrain rcspontl to those sleiouNh|

)r. Luvino and his nssociates IIAU* twyun ;i H-rks of experiments ^ in
i'h they are imphmlin^ Mcrnhh thuvily iniu ilw brnln llirou.gh.t!|i;
ow cannulae. In otTecl, tliis V.IIMC"I (lit* hniin tu "believe" t«« r"F
>tcl level of the hotly is a grcar ik^il liiplicr than il bHiiactwaliiy- U : so.
t'xporhncnlcrs have postviluUxl lh«l ihc animal should be exceedingly

nml learn to avoid punishnu'ni my *wil'rly. When hydrocortison^
ol the adrenal steroids, lins hern nurclcd Into the hypotlinlmnus 01 »
ulhcr no lon«er onuses a strc«\ rc^ptinw Mwcowr.wJionihB*™"1;"
.'il iii a sluittlchox. il will (turn to jmtd shock very quickly n"« *J
.'uver to escape being shocked •!* Hmc* out of 60. A control ™ j
w hrain has received instead «»r «lcroid»n implant of cholesterol, w
I this .same shock much les* ^irkkntly and Hwrofore rece w nw
^ in tlie shutdchox. U would .ipp<r«r Hint (Jie adrenal steroj cs^
inlluonco upon the brain, lfu>, enhandng «ho abfHiy 10 dlscnmin
oulance situations situations in which the stakes must be n«M

575

the animal's point of view, for the failure to respond means a very unplea-
sant consequence.

These studies of adrenal steroids, as they directly influence the cental
nervous system, are filling in the puzzle of infant handling and its conse-
quences in adulthood. Handling, a wide variety of stimulations, and in-
deed stresses, appeals to alter a basic hormonal system during infancy,
Without handling and stress, the infant and the adult animal learns avoid-
ance slowly, does not discriminate readily, and exhibits disproportionate
emotionality when placed in a novel situation for the first time. He exhi-
bits a slow but inordinate level of adienal steroids whenever confronted
with a challenge or shock. His excessive output after repeated stiess is
reflected in adrenal glands that are unusually large and heavy, by stunted
growth, by susceptibility to disease. How does experience create the dif-
ferent and healthier pattern of the handled infant?

The experimenters have postulated that handling during critical periods
of infancy must allow the system to adjust its "stress" system by a series
of approximations. Perhaps the system operates in a manner analogous to
a thermostat. The animal reacts to an event and pours out adrenal steroids
by which the brain responds according to their level, thus regulating
further behavior. The shaping that occurs is in the organization of emo-
tional responses via the pattern of released corticosleroids, a pattern thai
is thought to be controlled by the hypothalamus. Release of these fear-
inspired steroids may have a permanent effect upon the hypothalamus
during critical periods of infancy, thus calibrating the animal's level of
response to stimuli. Without experience, there can be no comparison,
none of the raw material by which the central nervous system could
establish a ratio of response and event-setting up an internal sense of
proportion. If the nervous system of the experienced infant is calibrated
so that responses are rapid, brief, and create little aftermath-cues to this
development must lie within the steroid levels of the newborn and infant
animal.

A number of factors-among them size and cost -have made rodents the
most convenient animals for many of these studies. Different strains of
inbred rats and mice show distinctive characteristics in their stress re-
sponses, their normal excitability, and their responses to handling. When
laboratory animals are purchased, moreover, it is hard to determine
exactly how the infants have been handled by the breeder and, therefore,
how much of their response pattern is genetic, how much due to experi-
ence. Genetic differences exert a profound influence upon the develop-
ment of the individual animal as he interacts with his environment. For
this reason, the laboratory has engaged in a number of genetic studies and
has compared quite a few strains of rodents on a number of criteria. There
have been studies of differences in adult emotionality, studies of the
effects of infant handling, and of the adrenal output of various strains of
mice under stress. These genetic studies add some important dimensions
to the study of infant development.

576

Genetic Differences in Hormonal Predisposition

No creature is completely plastic at birth. The shape of the body, the
color of its fur, the arrangement of vital organs, and many other internal
patterns are clearly laid out by the genes. Many behaviors are determined
by the structure of the nervous system, and neural circuitry is preset. The
nervous system contains a network of communication fibers determining
which cells shall receive and transmit impulses to other particular cells.
These nerve circuits determine how cells in the eye shall transmit to the
visual brain thus structuring the animal's vision. Even relatively specific
predilections such as preferred kinds of food may be determined before
birth. Early studies have indicated that experience in infancy may perform
the next modification, adjusting factors that control the readiness of an
emergency response, the extent of a reaction, and the length of time that
an animal is alerted, his intensity. Here, too, the animal has some genetic
predilections. Steroid measurements are being used to distinguish some of
the basic differences which appear in the magnitude of the animal's re-
actions to handling. Some strains of mouse, for example, exhibit a hyper-
reactiveness and a strong steroid response to stress. These animals show
great differenes in the production and release of steroids following shocks.
The reactive mice, placed in an open-field test, exhibit a notable tendency
to cower in corners, to defecate. The others show less emotionality.

The genetic foundations of a response pattern are important if the
effects of experimental manipulation are to be measured precisely. In
working with mice it has been advisable to explore this baseline of respon-
siveness and a number of studies have been undertaken recently. Wild
mice, as compared with domestic mice, show differences that may tell a
good deal about the way in which the creature becomes tailored to the
particular demands of his environment. Wild mice, for instance, show a
much larger steroid response to laboratory tests, and have, indeed, larger
adrenals than domestic mice.

In a current study, genetic differences showed up when the creatures
were presented with a conflict situation. The mice were given no water to
drink in their cages. They were placed in a special cage where they had to
drink from a water bottle. After 5 days they received a shock while
drinking from the bottle. The strain with a high-steroid response quit
drinking after 5 shocks, but the low-steroid strain took 15 shocks before it
would give up, After an injection of a substance that blocks steroid out-
put, the high-steroid strain would behave like the low reactors and would
also require 1 5 shocks to prevent them from drinking on.

A very interesting adrenal adaptation was discovered in a variety of
desert mouse that showed none of the usual elevation in stress steroids
after shock and other tests. Here, as it turned out, the assays that were
suitable for most of the laboratory mice were totally inappropriate. It was
not that this mouse lacked a stress response, but that his glandular func-
tions were differently organized. Steroids usually indicative of stress were,
in this mouse, functioning to maintain water in tissue.

577

Some recent experiments have indicated that the startle response of a
mouse may be a good indicator of his steroid level. Two strains, one high
steroid and another low, have been tested in a delicate little box that
registers the height and impact of a jump. At 1 -minute intervals a cap
pistol is fired. The reactive mice leap high at first and take some time
befoie they get used to the noise so that they no longer jump. The other
strain consistently shows a much smaller startle jump and takes tittle time
to habituate.

A recent study of four strains has indicated that the reactive mice
(showing the most marked steroid response to shock and novelty) are the
ones in whom infant handling makes the greatest change. Handled mice
from these two strains have indicated the greatest reduction in stress upon
testing. The two strains initially showing least response to shock and
novelty in steroid output have also least change after handling.

Current studies of genetic factors in the developing responses of animals
do suggest a variety of differences in initial responsiveness to environment.
Experience makes littie dent upon the adrenal response of low-steroid
strains. However, the high-steroid strains appear to become "less emo-
tional" with experience and show less reaction to stress. It seems likely
that analogous genetic differences operate within human beings. Parents,
certainly, speak as if the "high strung1' infant makes his presence known
in early infancy. The intimations of a temperament seem to be conveyed
by the baby in the first weeks of life. Tests, such as the startle test used
with laboratory animals, might indeed give a very early estimate of the
infant's adrenal responsiveness. Even so, one would not know whether the
infant's intensity of reaction were truly genetic or whether it were in-
fluenced by its mother's own steroid levels prior to delivery and through-
out the nursing period. In addition, maternal behavior presumably plays
its role in shaping of the infant.

Concurrent with a number of genetic studies, the laboratory has also
been investigating the mother's own prenatal and postpartum adrenal re-
activity and the effects on the young.

The Development of Endocrine Responses in Critical Periods

Whatever the genes that formed him, a child does spend almost a year
in the chemical bath of his mother's womb. In the first months afterbirth
he will be nourished by her milk and suffused with her chemicals. Most of
his experience will derive from her ministrations. Will he feel the conse-
quences if his mother endures an ordeal while she is bearing him? Will he
suffer from the debilities of a high-steroid start in life if she is high strung
while nursing him? Does her stress chemistry give the infant's own physi-
ology a push, helping him to set the level of his endocrine responses to
life?

There is a period in the life of an infant rat which seems to be strongly
influenced by the mother's chemistry perhaps through her milk, perhaps
through her behavior. It was discovered when rat pups were being tested

578

for their steroid responses to shock and other stresses. The experimenters
noticed that the neonate, on the day of birth and for 3 days of life, gave a
marked response to stress. But after the third day, there was no sign of
adrenal stress hormones. The usual response had vanished. Ether, electric
shock, exposure to cold-all of the standard laboratory stresses failed to
elicit a steroid response in the infant. They injected ACTH-the hormone
that instructs the adrenals to release stress steroids-but nothing seemed to
happen. There was no detectable release of steioids from the adienals. The
animal was in a dormant period, and it lasted somewhere between 6 and
12 days.

Handling the infant seemed to shorten this dormant period. Creatures
handled in the laboratory would again show a steroid response to shock
after about 6 days. Protected creatures would not resume their response
for about 1 2 days.

Why did the animal give a shock response right aftei birth and then
exhibit none for almost 2 weeks? The probable answer was that the fiist
responses depended upon an endowment of steroids from the mother, but
thereafter the creature was in a state of transition. His endocrine system
might be setting up its steroid response levels dictated, in part, by genetic
endowment mediated by the mother and by his expeiiences during the
critical period. The experimenters postulated that the infant animal must
adjust his adrenal hormone release to the stresses of the outside world,
after a fashion-setting a glandular thermostat that would dictate the in-
tensity of his responses to events.

The research team began to take a close look at the mother's influence
upon this ciitical adjustment-a glandular calibration that would seriously
affect the animal's rate of growth, his learning ability, his resistance to
disease, and his long-term ability to adapt and survive.

They saw, with mice* that the mother's steroid levels appeared to exert
far more influence than the father's. Genetically identical mothers bred to
antipodal fathers bore infants whose steroid response patterns inevitably
followed those of the mothers-not the fathers.

If the mother's own steroid levels were mediating the endocrine re-
sponses of her newborn, the experimenters expected that a shift in her
own steroid levels might show up in the young. After delivery, in one
study, they injected the mother with ACTH. The newborn showed a
higher steroid response than did the neonates whose mothers received
only saline injections. The next test was to drastically reduce the mother's
steroid level with a chemical that blocks the release of ACTH. Her young,
as might be expected, exhibited decidedly lower steroid levels. She was
undoubtedly still capable of transmitting steroids, for an ACTH blocker
would prevent further release of the hormones but would not reduce the
high level already circulating in her system. The steroid level of the
mother distinctly influenced that of her young.

Ordinarily a mother's stress responses are at their lowest JUF*
delivery. The steroid levels rise before labor and diminish after t
and during nursing. After the offspring are weaned, there is n
increase in steroid levels. A similar pattern of steroid decrease aft

found in human beings. Perhaps this is a built-in safety device. It works to
insure low-stress responses in the mother directly after birth, Since high
levels in the infant will stunt growth, it is essential that the newborn not
experience high-steroid concentrations too early. Afterbirth and through-
out the nursing period natvue has provided a mechanism for generating a
particular calm in the mother. This appeais to be the period in which she
mediates the initial steroid levels of the young, perhaps through her be-
havior and her milk.

Within the laboratory, in a study of mice, the calmest, low-steroid
mothers had offspring who showed a sluggish response to shock. It has not
yet been determined whether these low-steroid infants will learn discrimi-
nations faster than the more excitable creatures. One thing was ctear-the
infants reflected, in their steroid respones, the steroid levels of (heir
mothers. A good many factors might be icsponsible for the base steroid
fevel of the mother.

In a study of rats, the experimenters started at the beginning-in the
infancy of the mothers. They handled one gioup of females in infancy,
and protected another group. Now the effects of handling might be traced
through to the next generation, for all the mothers were of the same
genetic strain. Would the infant experience of the mother show up in her
own offspring? The study seemed to answer in the affirmative. The
handled mothers had lower resting levels of steroids, and their young
showed low-stress responses. The protected, or nonhandled mothers had
higher resting levels of steroids, and their young showed higher stress
responses. The handling of the infant female appeared to help set an
endocrine level that was then transmitted to the next generation.

Although a round of different studies has not yet indicated whether the
mother's steroid influence comes mainly through her milk or from her
behavior-it seems clear that her steroid levels dictate the early levels of
the infant. A cross-fostering study of rats indicated very dramatically the
importance of the mother's role in mediating the early steroid levels of the
young. The offspring of high-steroid mothers were given to low-steroid
mothers, and vice versa. Initially, it seems that the offspring of high-
steroid mothers show a very low-steroid response to stress when they are
raised by low-steroid mothers.

The importance of the mother's steroid level has been evident through-
out these studies and raises many questions about infant care and the
postnatal experience of the mother. Is the nursing period a time in which
shock to mothers will raise their steroid levels and transmit to the young a
tendency for emotionality that would reverberate throughout infancy, re-
tard growth, and even appear in the adult responses to alarm? Although
many questions must be answered about the mother's own experience and
her steroid state-the other half of the question must be answered by
determining what the infant body is doing during this critical time after
birth.

His mother's steroid levels give him an initial presetting, yet during
these first weeks experience will shorten the transition period and may
alter his glandular calibration of intensity and stimulus. The process might

580

be likened to the setting of an automatic thermostat, so that it will turn
on only when the temperature drops below a certain point.

If a kind of hormonostat is developed during the time when the infant
begins producing his own steroids, its effect upon his behavior may come
about through a feedback to the central nervous system. When his steroid
levels rise, his brain will move him in the diiection of alertness, of excita-
bility. This early development of the hormonal system, according to the
data now amassed, would seem to exert a pi o found and endless effect on
the growth, the adaptability, and health of the animal. Basic metaboJic
processes may be altered through this system and so the steroid response
to shock or other stress may be the key to many of the adult animal's
adaptive capacities.

The hormones issued by the adrenals, in response to pituitary command
during situations of potential danger, aie evidently affected by the sex
hormones. Normal male rodents when shocked show a lower stress re-
sponse than do the noimal females. The typical pattern of reaction diffeis
according to sex. The male reacts by pouring steroids into his system
much more lapidly and more rapidly declines when the stimulus is over.
The female takes longer to respond and characteristically shows high-
steroid levels for some time after the shock. This description of stress
reaction may sound familiar for it describes a male-female difference in
reaction that is encountered in the human species. The typical emotional
female remains emotional long after her baby has returned home safely or
the near accident has been averted. One might almost expect that the
difference between the male and female stress pattern is a pivot of conflict
between the sexes. Surely it is clear that the hormones of stress are in
some way related to the apportionment of sex hormones,

Just as the stress responses of an organism appear to be shaped during
critical periods of infancy and remain irreversible thereafter, the shaping
of sexual traits by hormonal settings is irreversible after a critical period in
infancy.

Sex Hormones and Behavior

The sex of a child can be determined very early-while still in the
womb. The overt sexual features are pronounced by birth, and a doctor
can after one glance say, "It's a boy." Nevertheless, the maleness of the
child may be deceptive. His internal hormonal states will determine
whether he acts like a boy or a girl. Quite a number of experiments have
demonstrated that sex can be altered aftei birth.

Sexual differentiation determines the behavior and reproductive ability
of the adult, also the nature of physical responses to stress, «nH thk

female animals grew up without any sign of the usual cstrus cycle, hut
rather seemed to be in a continuous estrus. They did not, however, show
cyclical iclease of important hormones and ncvei became prugmml al-
though given ample opportunity, and so it was thought that these jinimata
refused to mate. As latei studies were to demonstrate, the transplanting uf
testes was unnecessary. A single injection of testosteione in a fcwiikj
during her first 5 days of life sufficed to create a lack of cstrus cycle niu)
an inability to be impregnated. Hormone ticatmcnts to the mother in
other studies indicated that prenatal estiogcn could produce fcniini/uljoii
in the male offspring, and testosteione resulted in mate behaviors in IV-
males. By the mid-1 950's, it seemed clear that in the life of the guinwi pig
there was a critical period during its first week when sex hormones could
enter the brain and influence the sexual differentiation of the nniniitl.

Dr. Levine's laboratory has been using several techniques to find out
how, indeed, the hormones cause sexual differentiation «nd (hereby in-
fluence behavior. The investigator had found that when mule rats were
given the ovarian hormone, estrogen, their gonads atrophied, und (hey
were forever feminized. Indeed, if the male rat was castrated during his
first 24 hours, it was possible to implant an ovary in him in adulthood nnd
the ovary would ovulate in the cyclical fashion typical of female cstrus.

Further transplantation studies suggested that hormones must inl'Iut-neo
the brain. The cyclic pattern of ovulation is stimulated by the piluilnry
gland in the base of the brain. This gland produces the foWdc-sliimilntinu
hormone that prepares the ovary to produce egg cells. When (his process is
finished, the pituitary issues a luteinizing hormone which ripens the folli-
cles so that they rupture and release the eggs. When a female ovary is
transplanted mto a male who has been castrated nconatally, that oviiry
will ovulate This suggests that neither the ovary nor the pituitar alone Is

gess a neer te ovary nor the pituitary alone Is
°r *'1* °l

acuo on .

react to hormonal stimulation and that during critical periods (how brain

"

wtiMn . final «'no,i

rats v^ffi^S^^f"0*';1'8110™01108' Whon rclllillu
sterone, in infancy hev^n,! ?' mal° Eomidal hormone, leslo-
normal sex b^SrsUtevL °° "^"^ ""<' ™^^ »b~

they would growTnto id" K ItT" *?" W? d°SCS ol' l«Mlostoro»c,
Any observefwatSg on of tt! e ,'10 tnco,?ta™* ^nlo boj.uvtors
interacted with another feLte , th horraomlllr tr™^ romalos ». she
*e was . male. CmoS h ,* v C<1ge *?'1llld llilV° to COIldl1^- """
pearance of ejaculations o of femTh^° " ^ C°"cludo '" tllc 1^
would not reverse the effw? ™?frm J ?orn)on°-Mt«>8cn-[n odulthoocl
ever after would behave as a mak ' 8'W" lestostcro'^ '" infancy,

582

These studies suggested thai hormonal balance in infancy was crucial,
and that its effects upon the central nervous system would subsequently
influence behavior and reproduction. The castrated neonate— effectively
losing his testosterone—appeared to be suffering from an alteration in the
brain tissues which would Hilcr become receptive to male hormones. This
male, it should be said, was being feminized in many ways. He would
subsequently show feminine responses in nonsexual behaviois. His
hormonal balance, causing differentiation within bruin tissue would also
affect his adrenal system.

Male and female rats differ in their open-field behavior, but the dif-
ference can be reversed with a single hormone shot. A castiatcd neonate
will act like a female in the open field. A female injected with testosterone
will act like a male, The activity cycles of the male and female rat ordi-
narily show characteristic differences, for the female becomes quite active
before cstrus and inactive afterwards. A castrated neonate with trans-
planted ovary will show female activity cycles.

A variety of studies has indicated thai these changes in sexual orienta-
tion occur during a very short time after birth in the rat. Moieovcr, the
critical period is especially brief for the male. Male differentiation seems
to occur during the first 48 hours, while female differentiation takes as
long as 120 hours in the same species. The investigators have postulated
that the female state is in some sense more primitive and that without the
addition of a particular male hormone every rat remains a female.

The evidence comes from a number of studies. In essence, it seems to be
lack of testosterone that creates female differentiation, A newborn male
rat, if castrated, will not produce testosterone. If he is given no supple-
menting injections of testosterone he will behave as a female, exhibit
female sex behavior with the lordosis position of receptivity; he will ac-
cept implanted ovaries, show ovulation, and behave like a female on open-
field tests, in response to shock, and with other females. Injections of the
female hormone, estrogen, do noi have the overwhelming impact on a
male that is caused by Inck of testosterone. Hslrogcn causes the testes to
wither, however, and creates some female behavior.

Hormonal influences can be profound after birth, for it appears that
sexual functions and behaviors arc exceedingly malleable during critical
periods. Rats in prenatal development, arc structurally developed in part
by the influence of androgcn. Written into the very chromosomes are the
directions that determine whether the animal will have testes or ovaries.
Once the testes and ovaries develop, however, their hormones exert the
final influence upon the sexual nature of the animal. During infancy, the
hormones perform an organizational function, perhaps by creating a regu-
latory system within the brain that finally governs sex behavior and biolo-
gical function. Gonadal hormones appear to act upon the brain, thus
dictating whether the creature shall be male or female.

By this differentiation, the sex hormones dictate many behaviors that
are not strictly sexual. Recent studies within this laboratory have indi-
cated how estrogen and testosterone injections will influence the emo-
tional behavior of the animal. Inbred strains of highly reactive rats and

583

unreactive rats, as defined by open-field behavior, were injected with hor-
mones and observed in their subsequent open-field behavior. Before injec-
tion, the reactive males defecated more than the females. In all the strains
tested, the females weie more active. Neonatal hormone injection had a
profound effect on the nonsexual behavior of the animals. Estrogen
caused considerably more defecation and activity, producing what might
be called a more emotional rat, and affected both sexes by heightening
this reactivity. Testosterone, on the other hand, produced a more male
response by one criteria; it increased defecation in the reactive females to
roughly the level of the males, In another study, male neonates had been
injected with estrogen, and when stressed, their output of corticosterone
was like that of the female.

In current studies, the investigatois have been exploring for reliable
behavior differences among male and female rats. It will then be possible
to perform more sensitive tests of the effects of hormone changes, In
mice, aggressive behavior is exhibited mainly by males. Males fight when
placed together but females do not. Females given testosterone, however,
will engage in some fighting, Further studies may indicate whether neo-
natal castration will remove fighting from the behavior repertoire of the
male and whether neonatal injection of testosterone produces an adult
female who does battle like a male. Female rodents show more rapid
avoidance conditioning than males and this, too, may turn out to be
amenable to hormonal change.

Stress responses have been alteicd by hormonal treatment. In measuring
the adrenal steroids in the blood, samples are drawn from the jugular vein
under ether. This procedure, in itself, constitutes a stress for the animal.
Ordinarily, the normal male will respond to the ether and surgical cut by a
rise in corticosterone lasting about 20-45 minutes. The male treated with
neonatal estrogen shows almost double the normal concentration of adre-
nal stress hormone, but his response is still considerably less than that
generated by the normal female. If, however, the female has been given
testosterone during the initial days of life, her output of adrenal stress
steroid will fall between that of the normal male and the estrogen-treated
male. It might seem that testosterone, acting upon the central nervous
system, mediates the response of the emergency system, the emotional,
or stress output of the animal. There are many questions and one very
striking conclusion to be drawn from this program of research. The en-
vironment of the newborn animal and its changes have a very significant
and life-long effect upon his adult biology and behavior. Sex hormones,
the interplay of environment factors, and adrenal steroids, all appear to
have an organizing influence upon the central nervous system of the new-
born. Hormones cause the infant to become a male or female in function,
And infantile stimulation appears to organize the emotional responses of
the creature.

Both neonatal manipulation and hormonal treatment may have an
effect upon the neuroendocrine system as a whole, for they are inter-
linked, An event occurs, and a brain activity signals the pituitary to release
ACTH, which in turn stimulates the adrenal cortex to produce steroids.

584

The pituitary produces growth hormone, and hormone enabling the
ovary to produce eggs, and a male hoi mone that stimulates the production
of testosterone in the cells of the lesles. I he hormones of the pituitary
stimulate the thyroid gland. Thus the systems ol endocrine regulation and
hormonal regulation of mating behavior and specific sex mil functions me
to some extent joined. The n comic, at critical periods, win be shifted in his
responses to stress by manipulation, treatment wilh lulrcnal steroids, 01 by
gonadal hormones. These iKHirophysiolo^ical systems, wilh (heir enor-
mous impact upon adult behavior, lie ul UK- huh of the formative periods.

The mysterious and vast influence ol (he honnones has been probed at
the cellular level by many biologists. Today I here is 11 growing body of
data that suggests how Ihc hormones exeil their influence. Hormones
within the tissues appear Lo direct the activity of tin- t-cnes, the controllers
Df individual cells. It is by their influence upon Hie gene that the
lormoncs appear to steer cell differentiation, Thus (he hormone, while
jxternal to the cell, can niter (he |>n Item of uene activity am! (hereby
nfluence protein synthesis. This influence is nowheie more clearly seen
:han in some of the recent t mnspliintalioii experiments in which a eas-
:rated male wears a healthy, cyclically nourished ovary in an eye-socket,
vhile a testosterone-treated female remains amwulalory and shows no
cycles.

These studies demonstrate llu- power of m-omilnl hormones in shaping
he sexual differentiation of the animal. Insofar as Hie mother's own
jalancc may influence the infant, there aie some very strong implications
or medicine. It would seem thai an indiscriminate use of prenatal
iormon.cs holds some possible dangers for Hie differentiation of male
nfants, since the male appears particularly vulnerable to lack of Icsloslcr-
>nc or to an overbalance of female hormones. Uecauso neonatal honnones
ppear to play such a cleeisive role in sexual dilTci'eiitiitlion, hormone
Indies of infancy should instigate new approaches and reseaieh into sex-
lal deviation and its possible (real men Is. If it is possible to develop sensi-
ive hormone assays for the infant aiul child, it seems possible that prcvcn-
ive treatment might be attempted. Surely, these studies raise important
tueslions about the extent to which deviant behaviois are the result of
lormonal imbalances that have gone undetected in the clinic but have
:d to a life of social cnstigulion niul punishment. To what extent might
fiesc hormonal imbalances be caused by the sleroiils ingested by pregnant
mothers'? To what extent might they he altered after the critical and
orirmtivc periods have passed. Sexual differentiation is crucial to human
djustmcnl. It may seem ama/ing that the Mory of neonatal hormones and
licir bearing on differentiation and behavior have just begun to be ex-
•lorcd in the animal laboratory, but the liummi being for whom those
icchanisms have long social and personal reverberations is fur loss under-
Loocl und studied.

The psychophysiology of the luiiunn formative period will undoubtedly
:ikc its guidelines from the animal laboratory. Already studies performed
y this laboratory suggest techniques thai should enable safe and Illumina-
ng studies of human infants. It should be possible to devise an equivalent

585

of the startle apparatus permitting measurements of the amplitude of an
infant's response and the detection of early signs of overreactivity. By
techniques allowing a determination of steroid concentrations within a
diop of blood, it should be possible to measure the steroid elevations an
infant shows in response to mild stimulation. Thus we may hope to dis-
cover something about the critical periods in which the neuroendocrine
system sets up its hierarchy. Controlled studies of infants and mothers
should give us information about the prenatal and postnatal influences
and how they are molded by outside events. It should be possible to detect
the eavly warning signals of hypeneactivity in a child pretending sus-
ceptibility to sickness and perhaps to psychosomatic illness. We can finally
hope to establish child-watching and child-reaving practices that will en-
hance the adaptability and health of the individual.

Some of the more mysteiious questions of large populations may
finally be answered as more data become available from studies of the
ways in which hormone mechanisms influence differentiation. Indeed, it
may be possible to answei some of the recurrent questions of the demo-
grapher. During war and its aftermath, for instance, it has been noticed
that theie is an increase in male births. Is this a response of the parents'
hormonal systems to stress?

These studies of the neuroendocrine system in its developmental stages
have far-i caching implications, but perhaps the ones that fall closest to
home are really hopeful questions-what do high doses of steroids do to
the piegnant mother and her young? How could our clinics begin to
obtain predictive information about the endocrine adjustments of the
newborn? Can we begin to look for hormone balance and stress-steroid
concentrations as usefully as we look for blood types?

Research Grants. Mil 1051, MH 1630, K3-MII 19,936, MH 7435
Dates of Interviews: Fall, 1965;spring, 1966

References:

Davidson, E. H. Hormones and genes. Scient. American, 1965, 212, 36-45.

Hanis, G. W.> & Levine, S. Sexual differentiation of the biain and its experimental

control,/ PhysioL 1962, 163, 42-43.
Click, D,, Von Redlich, Dorothy, & Levine, S. Fluorometric determination of corti-

costerone and cortisol In Q.Q2-O.Q5 milHHters of plasma 01 submilligram samples of

adrenal tissue. Endocrinology, 1964, 74, 653-655.
Gray, J. A,, & Levine, S. Effect of induced oestrus on emotional behavior in selected

strains of rats. Nature, 1964, 201, 1198-1200.
Gray, J. A., Levine, S., & Broadhurst, P. L. Gonadal hormone injections in infancy and

adult emotional behavior. A mm, Behav,, 1965, 13, 1, 33-45.
Levine, S. Stimulation in infancy. Scient. American, 1 960, 202, 80-86.
Levine, S. The effects of infantile experience on adult behavior. In A. J. Bachrach

(Ed,), Experimental foundations of clinical psychology. New York: Basic Books,

1962.
Levine, S. The psyche-physiological effects of infantile stimulation. In E. Bliss (Ed.),

Roots of behavio). New York: Hocbei, 1962.
Levine, S. Some effects of stimulation in infancy. In S, H, Bartlett (Ed,), Lessons from

Animal Behavior foi the Clinician. London: Nat'l Spastics Soc, (Little Club Clinics

in Developmental Medicine, No. 7), 1962, 18-24.

586

I.evinc, S., & HnmillnnM, I1. I . (it-ru'lk uml oiiloneiuMic determinants of adult be-
havior In llic ml ,f <'<>int>. riivxt'il. /'.NT/HI/., |»)d.l, Sf>, <123-4iR.

I.cvim1, S., A Miillius, K. l-sirtt|ton iitlniinisti'rcil nvnniilally aflccts ;uliill sexual be-
havior in niiik iiiul li'inali' nils, >Vnr;iM', I''M, I'l-l, IKS-IK7.

Ijjvinw, S.. A t'lt'lnuui. I), M OiltViontlul ptiiMiiu vorlicnsU'ronc response to stress in
lour inliR-il iti.nnMil niici1. /'.'m/iit'/imi/fw, I''M,7,S, U2-M4.

.ovine, S. A Wot/rl. A. Itituntili- i-xpt'iii'iHVi, stniin tlitlt'ivncus, and avoklnncc learn-
/'/M'u.i/ /'.u'c/H'/,, I'Kil, Sti. «7»>HK1.

587

ACKNOWLEDGEMENTS

Major contributions in the editing and preparation of this volume weie
made by Sherry Prestwich and Muriel Reich. They were assisted by Emily
Barren, Lillian Becker, Mary Carmody, Helen Fussell, and Sandy Snider,

588

ft U.S. GOVERNMENT CHINTING OFFICE . l»7f O—Hll-tltl