;csj

;0
•Is-

Lashley, Karl Spencer

Studies of cerebral
function in learn:*

QL

TS5

111

v-3

[November, 1921.]

o I 20 i I

^ .7 . s-s-

BRAIN

PART 3, VOL. 44.

STUDIES OF CEKEBEAL FUNCTION IN LEARNING.
No. III.— THE MOTOR AREAS.

BY K. S. LASHLEY,
Department of Psychology of the University of Minnesota.

In their first study of the electro-stimulable cortex of the dog,
Fritsch and Hitzig [5] ascribed to it the function of control of
voluntary movement, although they recognized that its destruction
resulted in a difficulty rather than a complete loss of ability to make
voluntary movements. The latter fact was largely disregarded by the
investigators following them and the view that the fibres descending
from the stimulable area form the chief efferent path from the cortex
was adopted by Ferrier, Carville and Duret, Luciani, Bechterew, and
other early workers, so that this theory has come to be accepted as one
of the most firmly established facts of cerebral localization. From the
first, however, certain facts, derived chiefly from studies of recovery
from paralysis following lesion to the motor area, have seemed incom-
patible with this view. Fritsch and Hitzig emphasized the importance
of other descending tracts, and Carville and Duret [4] showed that
recovery from hemiplegia is not due to the vicarious functioning of the
uninjured stimulable area. The belief that recovery from paralysis is
only a revival of depressed reflex functions and that the power of
voluntary movement is permanently abolished by complete destruc-
tion of the motor area was disproved by Rothmann [16] who observed
the formation of motor habits in monkeys after destruction of the
stimulable area of one hemisphere and the pyramidal and rubrospinal
tracts of the opposite side.

Such experiments show that the extra-pyramidal fibres are capable
of mediating complex activities of the habitual or voluntary type but

BRAIN. — VOL. XLIV. 17

256 ORIGINAL ARTICLES AND CLINICAL CASES

leave undecided the question whether this is their normal function or
merely a vicarious function due to the destruction of the stimulable
cortex. In a previous study of the role of various parts of the cerebrum
of the white rat in the formation and retention of visual habits the
writer found that the habit of discrimination of light intensities survived
the total destruction of the stimulable area. Injury to an area on each
occipital lobe corresponding to Brodmann's area striata [2] resulted
in the complete loss of the visual habit, but any other third of the
cortex might be destroyed without in the least disturbing the animal's
ability to ma.ke the discrimination accurately. The data reported in
that study seem to prove that the conditioned-reflex arcs involved in
visual habits descend from approximately the same area of the cerebral
cortex to which they are projected and do not traverse the cortex to
descend from the stimulable areas. The visual habits have cerebral
representation, but the supposed motor areas have no direct function
in their performance [10].

In higher animals, especially the primates, complete paralysis of
voluntary movement may follow destruction of the stimulable cortex,
but this does not necessarily imply an interruption of the direct
conditioned-reflex arcs. Such an explanation is to be preferred on
grounds of simplicity, if contradictory facts do not appear, bat dis-
turbances of some of the mechanisms of tonic reinforcement, not in the
direct conditioned-reflex path (section of the afferent nerves of a limb,
cerebellar injury, &c), may produce somewhat similar disturbances of
voluntary movement and indicate an alternative explanation of the
function of the stimulable areas. The demonstration that, in the rat,
the stimulable area is not traversed by the conditioned-reflex arcs of
visual habits seems sufficient evidence to raise the question of whether
the so-called motor area is really a part of the mechanism for voluntary
movement or is merely a part of the subsidiary tonic and postural
mechanism.

Generalization from the rat to higher forms is complicated, how-
ever, by the probable transfer of function from the corpus striatum to
the stimulable cortex, with ascent in the evolutionary scale. In the
rat, destruction of the stimulable areas produces no detectable motor
disturbances, but simultaneous destruction of the motor area and the
corpus striatum produces a paralysis which is relatively permanent and
which resembles the effect of destruction of the stimulable area in
higher forms [10]. This indicates that the function of the stimulable
area of primates is represented in the rat by the combined function of

STUDIES OF CEREBRAL FUNCTION IN LEARNING 257

the stimulable area and the corpus striatum.1 In order that data
obtained from the rat may be applicable to higher forms it is neces-
sary, therefore, that both the stimulable cortex and the corpus striatum
be considered. My earlier study eliminated only the stimulable cortex
from the habit mechanisms. In the experiments reported below I have
tested the effects of simultaneous destruction of the stimulable areas
and the caudate portions of the corpora striata upon the rat's ability to
form and retain visuo-motor habits. The data outlined above make it
fairly certain that in the combined lesion we are dealing with structures
analogous in function to the stimulable areas of the primates and
that conclusions valid for higher forms may be drawn from the
experiments.

The chief question which the following experiments were designed
to answer is, then, whether or not the structures of the rat's cerebrum,
injury to which produces motor disorganization, are directly concerned
in the performance of habitual acts. Visuo-motor habits are best
suited for such tests, since they are easily acquired and recognized. A
visual area in the occipital region is clearly defined and it has been
shown that cortical representation of the habit is retained after 1,400
trials of overtraining [11], so that the possibility that the habits are
carried out at subcortical levels is ruled out.

The Topography of the Stimulable and Visual Areas of the
Cerebrum and of the Corpus Striatum in the Eat.

The stimulable area in the rat includes the antero-dorsal third of
the cerebral cortex. It extends from above the anterior margin of the
hippocampus forward to the frontal pole, covering the median half of
the dorsal convexity in this region. On the frontal pole it extends
laterally and then caudally over the inferior orbital surface. Fig. 1
shows a composite map of the area in about twenty-five animals, with the
movements elicited by bipolar stimulation. In the majority of animals
the area does not extend so far caudally as in the diagram and only
movements a, d, i, j, and p can be elicited.

The visual function seems limited to a small area on the dorsal

1 Since the pyramidal fibres are scattered throughout the caudate nucleus it is impossible
to destroy the latter alone. In earlier work [10] I found one case with extensive degenera-
tion of the right caudate nucleus involving few of the pyramidal fibres and leaving the
greater part of the stimulable area intact. This animal showed an unusually rapid and
complete recovery from the motor disturbances following operation and led to the tentative
conclusion that the striate nucleus and stimulable cortex have interchangeable function.

258

ORIGINAL ARTICLES AND CLINICAL CASES

convexity of the occipital pole. Fig. 2 is a composite diagram of the
lesions in nine animals in which habits of visual discrimination persisted
after bilateral injury. Detailed descriptions of these animals have been
given in previous papers [10, 11]. In each animal the area destroyed
included about one-fourth the blackened area of the diagram. Fig. 3
is a composite diagram of the lesions in seven animals which lost the
habit of visual discrimination after operation. These two series of
animals show clearly the restriction of cortical function in vision to the

Fig. 1. — Composite diagram of the electro-stimulable points on the cerebral cortex of
the rat. a = Head turned to opposite side, b = Nose retracted, c — Vibrissa? moved.
d = Chewing movements, c = Tongue protruded. / = Eye closed, g = Ear adducted.
h = Ear erected, i = Shoulder drawn forward, j = Fore-arm retracted, k = Elbow
flexed. I =Elbow extended, m = Wrist flexed, n = Fore-arm rotated, o = Back flexed
to opposite side, p = Hind leg drawn forward, q = Homolateral leg flexed, ipselateral
extended, r = Ankle extended, s = Tail drawn to opposite side. All movements are
contralateral to the hemisphere stimulated, except where indicated.

occipital pole and the survival of the habit after destruction of the
motor area. The area included in fig. 3 corresponds roughly to
Brodmann's area striata in other rodents [2].

The corpus striatum in the rat is very large in proportion to the
volume of the cerebral cortex. Fig. 4 shows it in frontal and hori-
zontal sections through the levels used in later diagrams. The caudate
nucleus consists of masses of cells scattered among the fibres of the
internal capsule and is nowhere distinct from the descending fibres.
The lenticular portion is relatively free from pyramidal fibres. The

STUDIES OF CEREBRAL FUNCTION IN LEARNING

259

experimental lesions described below are restricted to the caudate
nucleus and this is also the part whose injury produces paralytic
symptoms [10, 12].

Experimental Methods.

Training. — For the study of visual habits a discrimination box of
the type designed by Yerkes was used. It consists of a central com-
partment opening into two parallel alleys which lead by lateral passages

Fig. 2. — Composite diagram of the lesions in animals which retained the habit of visual
discrimination after operation. The habit survived the destruction of any third of the
blackened area.

Fig. 3. — Composite diagram of the lesions in seven animals which lost the habit of visual
discrimination after operation. The area which was destroyed in all is shown in solid black.

to compartments in which food is placed. At the end of each alley is
a small electric bulb, 5-watt frosted " Mazda," visible from the dis-
crimination compartment. One bulb is lighted and the other darkened
in irregular alternation and the animal is trained to choose the illu-
minated alley, receiving food there and punishment in the darkened
one. Training was continued with ten trials per day by the usual
method of irregular alternation until the animals made thirty consecu-
tive trials without entering the darkened alley. Controls were intro-

260

ORIGINAL ARTICLES AND CLINICAL CASES

duced to assure that the reaction was to the light and not to accidental

cues.

Operative methods. — Under ether anaesthesia the skull was pierced
on each side of the median line at the fronto-parietal suture and the
bone clipped away to make an opening, about 2 by 6 millimetres, above
the caudate nucleus. Through this opening a small electric cautery,
heated to redness, was plunged to a measured depth, then drawn slowly
back and forth at a nearly constant depth throughout the length of the

Fig. 4. — Frontal and horizontal sections through^the corpus striatum at the levels of its
maximum area. c.a. = Commissura anterior, c.c. = Corpus callosum. p.f. = Columna
fornicis. c.o. = chiasma opticum. /. = fornix, g. = Genu corporis callosi. h. = Lobus
hippocampus, l.v. = Veutriculus lateralis, n.c. = Nucleus caudatus. n.h. = Nucleus
habenulse. n.l. = Nucleus lentiformis. sp. = Septum pellucidum.

caudate nucleus. A fine scalpel was next passed from the posterior
margin of the skull opening to the base of the olfactory bulb and drawn
back and forth with its point in contact with the floor of the cranial
cavity, so as to sever the frontal pole from the remaining cortex. In
the majority of cases this method resulted in the complete destruction
of the stimulable areas and in extensive lesions to both caudate nuclei.
When haemorrhage was stopped, the scalp was closed with interrupted
sutures and covered with a cotton-collodion dressing.

STUDIES OP CEREBRAL FUNCTION IN LEARNING '261

Retention tests. — Animals trained before operation were tested in
the problem box as soon as their general condition permitted ; usually
twenty-four hours after operation. In these tests for retention of the
habit the same method was employed as in training, except that no
punishment was given for errors during the first fifty trials ; that is,
the retention test is essentially a retraining in the habit and evidence
for retention consists in the reappearance of accurate discrimination
in significantly less time than was required for the original learning.

When a habit is lost after operation the question of post-operative
shock must be considered. In earlier papers I have discussed methods
of distinguishing between loss due to shock and that due to destruction
of a functional area. In the present experiments positive evidence for
retention following operation was obtained in every case, so that the
question of shock does not enter, save as an explanation of the few
errors made in the trials immediately following operation.

Reconstruction of lesions. — After completion of the retention tests
the brains were removed and serial frontal sections stained with iron
hematoxylin were prepared. Camera outlines of sections at intervals
of one half millimetre were made and the extent of the lesions indicated
on them. The positions of the sections were determined from internal
structures and indicated on diagrams of these structures projected to
the surface of the brain. The dimensions of the lesions were then
transferred with proportional dividers to the diagrams, the points so
determined connected by lines and the areas inked in. The reconstruc-
tions were finally verified by reference to the sections.

Only easily recognizable lesions, absorption or complete degenera-
tion of the cortex and subcortical nuclei were recorded. The diagrams,
therefore, represent the minimal destruction produced by the operations.
In all cases, probably, larger areas than those shown were destroyed,
but, since the experiments deal with the non-function of given areas
and do not seek to localize functions accurately, this fact does not in-
validate the results.

Graphic and tabular presentation of results. — In the preparation of
the figures a uniform system has been followed. Diagrams of the
superficial lesions made from camera outlines of an average brain are
given at the top of the figure. Below at the left is a diagram of a
horizontal section through the cerebrum at the level of the maximum
area of the caudate nucleus. Practically all of the ganglion cells of
that nucleus are included within the dotted line, the posterior horn
being made up almost altogether of the fibres of the internal capsule.

■J!". -2 ORIGINAL ARTICLES AND CLINICAL CASES

At the right, below, is a camera outline of a frontal section through the
cerebrum at the level of the anterior commissure (except in fig. 12
where the section lies at the level of the anterior horns of the lateral
ventricles). In these outlines complete replacement of nervous structure
by scar tissue or amorphous material is indicated in solid black ; marked
degeneration with some nerve elements intact is shown by stippling.

In the records of training and retention the "number of trials
required for learning" represents the number of times that the animal
traversed the problem box from the starting compartment to the food,
before thirty consecutive trials were made without entrance into the
dark alley. Ten trials were given each day in training and retention
tests except where indicated. The records of these trials are given as
horizontal rows of figures, each number representing the number of
trials, in the ten of that day's practice, in which errors were made.
Thus in the record of No. 17, 4:2:1:2, &c, means that on the
first day four of ten trials contained errors, on the second two of
ten, &c. Five errors in ten trials shows no discrimination. When
no error is made for thirty trials, it is rare that errors occur in
later trials unless the animal is disturbed by being roughly handled or
by some new elements in the situation, such as those introduced by
scrubbing the problem box.

The Formation op Habits after Destruction of the Stimulable

Areas and the Caudate Nuclei.

The formation of maze, latch-box, and visual habits after destruction
of the stimulable area in the rat has been reported in earlier papers
[10, 12]. In the present study I have attempted to train animals after
destruction of both the stimulable areas and the caudate nuclei. Only
three animals have survived the operation long enough to give evidence
of learning and they have formed only simple habits. The individual
records of these animals follow.

No. 17. l Adult male. — The cautery was plunged into each caudate
nucleus and the frontal pole of each hemisphere was incised. Training for
visual discrimination was begun eight days after operation. For several days
be spent most of the time climbing to the top of his cage and falling back to
the floor. When placed in the problem box he promptly climbed out and
repeated this more than two hundred times in spite of severe punishment. He

1 For convenience in reference these animals are numbered consecutively with others of
the visual series wbich have been reported earlier [10, 11].

STUDIES OF CEREBRAL FUNCTION IN LEARNING

263

was then placed in the left-hand feeding compartment in contact with the food,
and began to eat immediately. By the process of removing him a few inches
farther from the food at each trial, he was trained to find his way through the
training box from the starting compartment to the food. He was then given
regular training in visual discrimination for seven days. In this time he
showed no evidence of visual discrimination, but formed a position habit
toward the left alley, which was fixed in a normal learning curve, with errors
(turns to the right) as follow, in successive tens of trials : —
4:2:1:2:1:0:0.

Fig. 5. — Extent of lesions in No. 17. In this and the following diagrams a uniform
arrangement has been adopted. Above, the lesions to the cortex are shown in lateral and
dorsal aspects. Below, at the left, is a diagram of the horizontal extent of the lesions to the
caudate nuclei, in this case embracing the entire extent of both nuclei. At the right is a
camera outline of a frontal section at the level of the anterior commissure, showing the
depth of the lesion. In these outlines degenerated areas are stippled and scar tissue, which
does not appear in this one, is shown in solid black. The outlines of the frontal sections are
reversed, the right side appearing on the left. The diagrams are correctly orientated.

Ten days after operation he developed a marked spasticity and died on the
eleventh day. Autopsy showed the wound infected, with extensive abscesses in
both hemispheres, extending into the lateral ventricles.

Extent of lesions (fig. 5). — Eight hemisphere : The entire cortex of the
dorsal surface of the hemisphere, from the frontal pole to the level of the

'264

ORIGINAL ARTICLES AND CLINICAL CASES

pillars of the fornix, is replaced by a cyst. The dorsal half of the striate
nucleus, including all of the caudate nucleus, is absorbed. .The lenticular
nucleus is uninjured. Left hemisphere : The lesion is identical with that on
the right.

No. 18. Small adult male. — The frontal lobes were transected and the
caudate nuclei cauterized. For eleven days the animal showed marked dis-
turbances of movement, rotating to the right, so that he was unable to eat
without assistance. By the twelfth day this trouble had partially cleared up,
so that he could walk in a straight line — although he still tended to rotate

Fig. 6.— Extent of the lesions in No. 18. Arranged as in fig. 5.

when eating. Training in visual discrimination was begun on the fourteenth
day after operation. At first he had great difficulty in making the turns in the
training box and in finding food, since the operation rendered him anosmic.
In five days (fifty trials) he formed the somaesthetic-motor habits of the training
box, so that he could go directly from the starting compartment to the food,
correcting the error and orientating promptly when he failed to find food in the
darkened alley. He still had some difficulty in recognizing food, and would
eat shavings and faeces in the neighbourhood of the food dish, but not else-
where. He died twenty-three days after operation.

Extent of lesions (fig. 6).— Eight hemisphere. The lesion to the cortex

STUDIES OF CEREBEAL FUNCTION IN LEARNING

265

extends as a diagonal incision from the level of the thalamico-mamillary tract
to the frontal pole, destroying all of the cortex of the dorsal convexity and
extending over the superior part of the orbital surface. The caudate nucleus
is completely destroyed from the level of the anterior commissure to its
posterior margin. The thalamus is invaded at the level of the anterior and
lateral nuclei. Left hemisphere : The lesion to the cortex is similar to that
on the right, but does not extend so far laterad. The posterior part of the
caudate nucleus is destroyed, as on the right. The anterior and lateral
thalamic nuclei are destroyed.

Fig. 7. - Extent of the lesions in No. 19 Arranged as in fig. 5.

No, 19. Large adult male. — The frontal lobes were transected and the
caudate nuclei cauterized. Training in brightness discrimination was begun
ten days after operation. The habit was formed in 130 trials with errors
in successive tens of trials distributed as follows: —

6:6:2:5:5:2:1:3:1:1:0:1:1:0:0:0.

Extent of lesions (fig. 7). — The dorsal convexity of both hemispheres was
destroyed from the level of the thalamus to the frontal pole. The orbital
surfaces and olfactory tracts were not injured. The lateral portion of the
right caudate nucleus was destroyed from the level of the knee of the corpus
callosum to its posterior limit. The left nucleus was uninjured.

266 OEIGINAL ARTICLES AND CLINICAL CASES

The data from these animals, in conjunction with that presented in
earlier papers, establishes the fact that the rat may form somsesthetic
motor habits in the absence of the stimulable cortex and of the caudate
Duclei. I have shown that destruction of the stimulable area alone does
not lead to an appreciable reduction in the rate of formation of visual
habits, and that a difficult latch box, the "double platform box," may
be learned at normal rate after complete destruction of the stimulable
areas and severe injury to one or both caudate nuclei [10]. Of the
three animals reported in the present study, No. 19 alone survived
long enough to allow of acquirement of the visual habit and in this
animal the left caudate nucleus and part of the right were uninjured.
In the other two, however, the stimulable cortex was destroyed and the
caudate nuclei were either destroyed, as in No. 17, or severed from
their thalamic connections, as in No. 18. These animals survived
only a short time, but both gave unmistakable evidence of the formation
of simple somaesthetic-motor habits. From the data on the retention of
visual habits after the same operation, presented in the following section,
it is probable that the animals would have formed the visual habit if
training could have been continued.

The problem of the relation of the extent of cerebral destruction to
the complexity of the habits which may be formed must be reserved for
later discussion when more adequate experimental data are available.
The interest of the present study is chiefly in the question of whether
or not habitual " voluntary " movements are mediated by the stimulable
areas. These experiments show clearly that neither the stimulable area
nor the caudate nuclei are necessary for the acquirement of such move-
ments. This is in agreement with the results of Kothmann [16] with
monkeys and justifies the conclusion that the so-called motor areas are
not essential to the acquirement of voluntary activities.

This at once suggests the more important question of whether under
normal conditions the motor areas are directly functional in the
performance of habitual reactions. This problem has been attacked by
destroying the structures after a complex habit had been acquired and
testing for retention of the habit after the operation.

Tin Retention of Visuo-motor Habits after Destruction of

the Stimulable Areas and the Caudate Nuclei.

For tests of retention after operation the animals were trained in the

crimination box until they made thirty consecutive trials without an

When this record was attained the problem was considered

STUDIES OF CEREBRAL FUNCTION IN LEARNING 267

learned, but in some cases training was continued for from 30 to 60
additional trials, as it was not always convenient to operate earlier.
Since I have found that over-training up to 1,400 trials does not
influence cortical localization, this additional training is not objection-
able. When the problem was learned the animals were subjected to
operation and their retention of the habit was tested as soon as their
physical condition permitted.

In training ten trials were given daily. When the animal entered
the darkened alley he was punished and the trial was recorded as an
error, although he was allowed to turn back and pass through the
illuminated alley to the food. In the retention tests the same
method was used except that the animals were not punished. In
the records given below the number of errors made in each successive
ten trials of training and of the retention tests is given. Five errors are
to be expected from chance if the animal is not discriminating. Fewer
than four errors in ten trials suggest discrimination, and no error in thirty
consecutive trials usually means that the animal will discriminate
accurately thereafter. The individual records of the animals follow.

No. 20. Large male. — Trained in visual discrimination. Sixty trials were
required for learning, with errors in successive tens of trials distributed as
follows : —

7:6:2:5:3:2:0:0:0:0:0:0.

The frontal lobes were transected and the caudate nuclei cauterized. The
animal was first tested three days after the operation. He was slow and
spastic, but made no error in twenty consecutive trials. For the next three
days he was in an excited state and showed great fear, either refusing to run
or dashing through the training box at random and paying no attention to the
food. On the eighth day he was again normal in behaviour and gave clear
evidence of discrimination.

Post-operative retention tests ; errors in successive tens of trials on con-
secutive days after operation : —

0:0: (frightened, 6 : 4 : 3) 0 : 1 : 0 : 1 : 0 : 0.

Extent of lesions (fig. 8). — Eight hemisphere : the lesion includes all of
the cortex of the dorsal convexity from the level of the thalamus to the base
of the olfactory bulb. It extends over the orbital surface at the level of the
caudate nucleus where the cautery passed through the external capsule. The
caudate nucleus is destroyed from the level of the anterior commissure to its
posterior limit. There is some injury to the anterior thalamic nuclei. Left
hemisphere : the lesion to the cortex is similar to that on the right, but does
not extend on the orbital surface. The injury to the caudate and thalamic
nuclei is similar to that on the right.

This animal showed unmistakable evidence of retention after

2I-.K

ORIGINAL ARTICLES AND CLINICAL CASES

destruction of almost the entire stimulable cortex and of the posterior
two-thirds of both caudate nuclei.

No. 21. Small female about 90 days old. — Trained in visual discrimina-
tion. Ninety trials were required for learning, with errors in successive tens
of trials distributed as follows : —

8:5:5:3:2:3:0:4:1:0:0:0:1:1:0:0:0.

The frontal lobes were transected and the caudate nuclei cauterized. For
the first week after operation the animal was unable to walk except in 6-inch
circles and could not make her way through the discrimination box. By the

Fig. 8.— Extent of the lesions in No. 20. Arranged as in fig. 5.

ninth day after operation the motor disturbance had cleared up so that she
could walk without rotation, although she still rotated when attempting to
change direction and could not turn her body to the left. Eetention tests
were begun at this time and the animal at once gave certain evidence of
discrimination. The tests were continued for twelve days. Throughout this
tune siie was unable to turn to the left but compensated by rotating 270
degrees to the right wherever the problem box demanded a turn of 90

STUDIES OF CEREBRAL FUNCTION IN LEARNING

269

degrees to the left. Fig. 14 shows some of her simpler paths in traversing the
problem box.

Post-operative retention tests ; errors in successive tens of trials on con-
secutive days. : —

1:1:4:3:0:1:0:1:1:0:0.

Extent of lesions (fig. 9.) — All of the cortex of both hemispheres from
the level of the thalamus to the base of the olfactory bulb is destroyed. The
right caudate nucleus is completely destroyed except for a small antero-medial
region laterad to the forceps of the corpus callosum. The left caudate nucleus

Fig. 9. — Extent of the lesions in No. 21. Arranged as in fig. 5.

has a superficial transverse lesion at the level of the anterior commissure.
The right anterior thalamic nucleus is invaded.

After destruction of the stimulable cortex and severe injury to
both caudate nuclei, resulting in marked motor disorganization, this
animal gave certain evidence of retention of the habit of visual dis-
crimination.

No. 22. Large male. — Trained in visual discrimination. Eighty trials

270

ORIGINAL ARTICLES AND CLINICAL CASES

were required for learning, with errors in successive tens of trials distributed
as follows : — ■

7:7:4:2:1:1:0:1:0:0:0:0:0.

The frontal lobes were transected and the caudate nuclei cauterized. The
animal was stuporous for the first day after operation, but on the second day
was in fair condition and reacted promptly to the problem box, although
tending to rotate to the left. On this day he made four errors in twenty
trials, but the errors were obviously due to failure to compensate for the
motor difficulty and his behaviour in the discrimination compartment clearly
indicated discrimination. On the following and later days he made no errors.

Post-operative retention tests ; errors in successive tens of trials : —

2:2:0:0:0:0.

Fig. 10. — Extent of the lesions in No. 22. Arranged as in fig. 5.

Extent of lesions (fig. 10). — Eight hemisphere. The lesion extends from
the level of the thalamus to the base of the olfactory bulb, separating all of the
cortex above this plane from the underlying structures. The cortex of the
orbital surface was destroyed by the cautery. The lateral portion of the
caudate nucleus is destroyed by a lesion which embraces the full width of
the nucleus at the level of the anterior commissure and grows narrower
cephahul and caudad to this level. Left hemisphere : The incision extends

STUDIES OF CEREBRAL FUNCTION IN LEARNING

271

from the level of the anterior commissure to the base of the olfactory bulb,
destroying all of the cortex above this plane. All of the caudate nucleus in
front of the knee of the corpus callosum is degenerated ; the posterior part
is uninjured.

This rat retained the visuo-motor habit after extensive but incom-
plete lesions to the stimulable cortex and caudate nuclei.

No. 23. Small adult male. — Trained in visual discrimination. Forty
trials were required for learning, with errors in successive tens of trials dis-
tributed as follows : —

7:5 : 2 : 4 : 0 : 0 : 0 . 0:0:0:0:2:0:0:0.

Fig. 11. — Extent of the lesions in No. 23. Arranged as in fig. 5.

The frontal lobes were transected and the caudate nuclei cauterized. For
three days after operation the animal's eyes were closed and he could not be
tested. On the fourth day he was in good condition and gave clear evidence
of discrimination.

Post-operative retention tests ; errors in successive tens of trials ; —

2:1:0:0:0.

Extent of lesions (fig. 11). — -Right hemisphere: The lesion to the

liKAIN- VOL. XI.'V.

18

272 ORIGINAL ARTICLES AND CLINICAL CASES

cortex begins at the level of the thalamus as a narrow band and broadens
cephalad to include all of the cortex of the frontal pole. The lesion to the
corpus striatum extends from the caudal limit of the nucleus throughout its
length. All of the lateral half of the nucleus is destroyed, but the lesion
probably did not involve the median part in front of the anterior commissure.
The depth of the lesion is the same throughout its length, involving a little
more than the dorsal third of the corpus striatum. Left hemisphere : The
lesion to the cortex is practically identical with that on the right. The
greater part of the caudate nucleus is destroyed, only the extreme poles
remaining uninjured.

This animal showed perfect retention to the visuo-motor habit after
nearly complete destruction of the motor areas.

No. 24. Small adult male. — Trained in visual discrimination. One
hundred trials were required for learning with errors in successive tens of
trials distributed as follows : —

4:5:3: 5:1:3:4:0:2:2: 0:0:0:0.

The frontal pole was transected and the caudate nuclei cauterized. The
animal was stuporous for forty-eight hours after operation, then was tested for
discrimination. He reacted slowly, but found his way through the problem
box and gave some indication of discrimination, although he made frequent
errors. On the fourth day of the tests he began to discriminate accurately
and made no further errors.

Post-operative retention tests ; errors in successive tens of trials :- —

3:5:3:0:0:0:0.

The fact that the habit was reacquired in thirty trials, in contrast to the
hundred of original training, shows retention rather than relearning.

Extent of Unions (fig. 12). — Eight hemisphere. The lesion to the
cortex begins at the level of the anterior commissure and includes all of the
dorsal convexity of the frontal pole except a narrow median band. The hind
leg and part of the fore leg region remain intact. All of the caudate nucleus
in front of the anterior commissure is destroyed ; its posterior part is
uninjured. Left hemisphere: The lesion is almost identical with that on the
right.

After extensive injuries to the stimulable area and destruction of
the anterior halves of both caudate nuclei this animal gave evidence of
retention.

No. 25. Small adult female. — Trained in visual discrimination. One
hundred and ten trials were required for learning, with errors in successive
tens of trials distributed as follows : —

4-5-4:5:2:3 -.5:3:1:1:1:0:0:0:0:0:0.

The frontal pole was transected and the caudate nuclei cauterized. For
three days after operation the animal was unable to walk except in a very
narrow circle to the right and could not get through the problem box. On the

STUDIES OF CEREBRAL FUNCTION IN LEARNING

•273

fourth day she still rotated but was able to find her way through the box by
keeping near the partitions, and made only one error in ten trials. On the
following day she developed pronounced fear reactions and climbed out of the
training box more than one hundred times. She was finally induced to
make ten trials, but made six errors. Training was discontinued for five days.
When she was next tested the fear had disappeared and she showed perfect
discrimination.

Post-operative retention tests ; errors in successive tens of trials : —

1:6:0:1:0:0:0.

Fig. 12. — Extent of the lesions in No. 24. Arranged as in fig. 5.

Extent of lesions (fig. 13). — Right hemisphere: The lesion to the cortex
begins at the level of the commissura habenulae and extends to the base of the
olfactory bulb, including all of the motor cortex except a part of the hind leg
area. The cautery passed throughout the length of the caudate nucleus and
the entire nucleus is degenerated. Left hemisphere : The lesion to the cortex
is similar to that on the right, but misses the antero-lateral surface of the
frontal pole, and does not extend so far over the orbital surface. The caudate
nucleus is entirely degenerated.

Practically the entire stimulable cortex and both caudate nuclei

274

ORIGINAL ARTICLES AND CLINICAL CASES

were destroyed in this animal. There was clear evidence of retention
following the operation.

This group of animals is not selected, but comprises all that
survived the operation long enough to allow of retention tests. Every
one of them gave evidence of retention early in the tests. Table I
gives the number of trials and the number of errors, in learning and
and in the post-operative retention tests. The average number of
trials required for learning is eighty. Only 33*3 trials on the average

Pig. 13. — Extent of the lesions in No. 25. Arranged as in fig. 5.

were required after operation to reach the same degree of accuracy in
discrimination. Errors were made in an average of 26'6 trials in
learning and in only 6'3 trials in the retention tests. These figures
include errors obviously due to motor disturbance and fright following
operation. Thus, No. 21 was never able to turn directly to the left
and all her errors seemed due to failure to compensate for the motor
difficulty ; No. 24 was slow and stuporous during the time when
errors were made ; six of the eight errors made by No. 25 were made

STUDIES OP CEREBRAL FUNCTION IN LEARNING

275

on a day when the animal was too disturbed to eat. If allowance is
made for such errors, retention in every case was practically perfect.

TABLE I.— Numbers op Trials and op Errors preceding the First Thirty Consecu-
tive Trials without Error in Learning and Post-operative Retention Tests for
Animals operated after Training.

Training Retention tests

No. 20
No. 21
No. 22
No. 23
No. 24
No. 25

Average

Trials

Errors

Trials

Errors

60

25

0

0

90

31

90

12

80

23

20

4

40

18

20

3

100

29

30

11

110

34

40

8

80

26-6

33-3

6-3

The small operative field presented by the rat's brain and the
necessity for avoiding haemorrhage on the floor of the cranial cavity or
injury to the optic nuclei of the thalamus make the complete destruc-
tion of the caudate nuclei difficult. It was accomplished only in
No. 25. In this animal also the entire motor area, except the pos-
terior part of the hind leg region, was destroyed. The animal showed
motor and emotional disturbance, but gave certain evidence of reten-
tion of the habit. In No. 20 the posterior halves of both nuclei were
destroyed ; in No. 24, the anterior half : in the others the lesions
were irregular, but even more extensive. The lesions to the cortex
in all cases included the neck, fore-leg, and the greater part of the
hind-leg regions.

Earlier experiments [10] had shown that no part of the stimulable
cortex is necessary for the performance of the visual habit. The
alternative explanations of this fact were: (1) that the striate nuclei
function either vicariously for the stimulable cortex when the latter is
destroyed, or normally as the source of efferent impulses from the
cerebrum ; or (2) that the supposedly motor regions of the cerebrum do
not lie in the direct path of conditioned-reflex arcs and are of only
secondary importance in the performance of voluntary movements.

The foregoing results seem to prove that the second hypothesis is
correct. In the absence of the stimulable cortex, injury to the caudate
nuclei does not affect retention of the visual habit and perfect discrimi-
nations may appear, even after complete destruction of both nuclei. It
seems clear, therefore, that neither the stimulable cortex nor the
caudate nuclei are directly functional in the performance of the visuo-
motor habit.

276 ORIGINAL ARTICLES AND CLINICAL CASES

The experiments have not dealt with the lenticular nucleus and the
possibility that this forms a link in the conditioned-reflex arc remains.
The data previously presented on the effects of injuries to the striate
nuclei [10] indicate that the caudate nucleus and the stimulable cortex
together in the rat are equivalent in function to the stimulable cortex
of higher forms, since their combined destruction results in relatively
irrecoverable paretic symptoms. On embryological grounds there is
less evidence for assuming homologous function between the lenticular
nucleus and the stimulable cortex than between the latter and the
caudate nucleus. In some of the earlier experiments of this series,
Nos. 7 and 8 [10], the cerebral cortex and underlying association
fibres were almost completely transected at the level of the anterior
thalamic nuclei ; in the present experiments the greater part of the
internal capsule has been destroyed without abolishing visual discrimi-
nation. The cortical relations of the lenticular nucleus are not well
established, but any occipito-lenticular fibres which may exist were
almost certainly destroyed by these operations. It seems very improb-
able, therefore, that the lenticular nucleus has any more important
place in the conditioned-reflex arc than has the caudate nucleus.

The evidence presented in this series of papers all points to the
conclusion that in the rat the area striata on the dorsal convexity of the
occipital pole is the only cerebral structure which functions in the habit
of discrimination of light intensities. The habit may be formed or
retained after the destruction of any other part of the cerebrum (with
the possible exception of the ectorhinal and inferior temporal regions,
which have not been explored but are almost certainly not functional in
this habit). This means that the conditioned-reflex arcs which mediate
the habit must pass to and from the cortex by way of the occipito-
thalamic fibres, that the reintegrations occur within the limits of the
visual area, and that long transcortical association fibres do not function
in the formation or retention of the habit.

Are the conditions similar for other types of habit ? There are no
significant data for more complex habits. The poor vision of the rat
makes the formation of complex visual habits so slow that I have not
been able to carry out similar experiments with them. I have not
yet located an auditory area. Certain types of latch-box habits are
abolished by injury to the stimulable area [12]. The effective lesions
here do not correspond in extent to the stimulable area, but only to the
excitable region for the neck and fore limbs, which are used no more
than the trunk and hind limbs in the manipulation of the latch. It is

STUDIES OP CEREBRAL FUNCTION IN LEARNING 277

probable that the loss in this case is due to the fact that the frontal
region is an important somaesthetic projection area as well as electro-
stimulable. The one case for which there is clear evidence is the
group of somaesthetic habits of the discrimination box which determine
general orientation, finding of food, responses to doors, prompt correc-
tion of errors, &:c. These survive the destruction of any given third of
the cortex [10] and of the subcortical cerebral nuclei. It is possible
that they are formed at levels below the cerebrum ; it is certain that
they are not dependent upon the so-called motor structures.

There is some evidence, to be reported later, that the habit of the
" double platform box " [10] is disturbed but not completely abolished
by destruction of either the frontal or the occipital regions. This
suggests that more complex habits involving diverse sense organs may
demand co-ordination of distant portions of the cerebrum and that the
visuo-motor habit dealt with in this study is too simple to give a typical
picture of cerebral function. It does not, however, indicate any greater
importance of the pyramidal tracts in complex habits, and we may
conclude that the existing evidence for the rat all opposes the direct
participation of the " motor areas " in the performance of any habit.

The Function of the Stimulable Area and the Caudate

Nucleus in the Rat.

The evidence that the stimulable cortex in the rat has no direct
function in the performance of habitual motor adjustments seems
conclusive. What then is the function of this portion of the cortex ?
Data for a final answer to the question are not available, but some
suggestions are given by the motor disturbances following unilateral
lesion to the stimulable area and the caudate nucleus.

Immediately following combined unilateral destruction of the
caudate nucleus and the stimulable cortex the animal shows a very
pronounced rotation toward the injured side. Left undisturbed, he
assumes a normal position, but immediately upon stimulation he bends
sharply toward the side of the injury and walks in a circle, sometimes
so narrow that the head is brought across the hind quarters or trips up
the hind feet. The legs of the side opposite the injury are hyper-
extended and spastic (fig. 15c). This condition may persist for several
weeks, but usually clears up within a few days to such an extent that,
when the animal is placed in a large open space and allowed to become
orientated, he is able to walk in a straight line. But the tendency to

278

ORIGINAL ARTICLES AND CLINICAL CASES

rotate still appears in many situations. When the animal is placed on
the floor he turns around several times before taking a direct course.
The necessity for avoiding an obstacle frequently results in several
complete rotations. The tendency to rotate becomes more pronounced
in narrow quarters and in the problem box the animals have a good deal

Fig. 14. — Tracings of paths followed by No. 21 in the discrimination box. The animal
was unable to turn to the left but compensated for this motor difficulty by rotation to the
right, as shown. The discrimination was in each case correct. The path varies with the
alley illuminated and with the initial orientation.

of difficulty in making the necessary turns. Fig. 14 shows tracings of
the path of animal No. '21 at a time when she was making no errors of
discrimination. .She was able to walk in a fairly straight line, but any
need for change of direction resulted in rotation toward the right.
(Only the right caudate nucleus was injured in this animal.) Her path

STUDIES OF CEREBRAL FUNCTION IN LEARNING

279

was usually more complicated than those shown, the turns involving
not one but several complete rotations.

In feeding and scratching the animals show similar difficulty in

B

Fig. 15. — The behaviour in feeding of an animal with total destruction of the right
stimulable area and caudate nucleus. For explanation sec I

adjustment. Fig. 15 shows various positions assumed by an animal
with complete right unilateral destruction of the stimulable cortex and

280 ORIGINAL ARTICLES AND CLINICAL CASES

caudate nucleus, three weeks after operation. She approached the
dish, walking in a rather wide circle, but as soon as her mouth came
in contact with the food, which stimulates the normal rat to squat and
assume a definite feeding position, she turned sharply to the right, the
position shown at A. She then rotated several times to the right until
by chance her nose was braced against the side of the dish (B). A
similar position with hyperextension of the left legs is shown at C.
Once such a feeding position has been attained it may be held for some
time and a gradual relaxation of the strained posture of the body then
appears. Thus, at D, the animal was held in contact with the food for
about thirty seconds, then gradually released. She then maintained the
normal feeding position for nearly a minute, turning only gradually to
the right, as successive bites were taken from the edge of the cube of
bread, and finally becoming overbalanced and falling.

In all the activities of such animals, the greatest disturbance of
co-ordination appears when some new adjustment to the environment is
necessary. Initiation of locomotion, change in direction of locomotion,
assumption of a sitting posture, stretching forward for food, scratching,
&c, are attended by sudden rotations as though the difficulty were
chiefly in gaining a new posture or in initiating a new activity. Yet
the activities are initiated and the disturbances are compensated for in
ways that indicate that the difficulty is not at the level of the most
complex integrations. Thus, in 1917, Franz and the writer reported
the case of an animal with complete destruction of the motor cortex and
degeneration of one striate nucleus, which learned the " simple maze."
The maze used had essentially the same ground plan as the problem
box shown in fig. 14. The animal was trained to go constantly to the
left. He rotated and fell constantly to the right and at first could not
make the turn to the left. He finally learned to grasp the end of the
partition with his left hind foot and, with this as a pivot, to swing his
body to the left and so enter the food compartment. Fig. 14 shows a
more common type of adjustment where the drive toward the goal is
maintained in spite of the motor difficulty.

Such adaptations seem to prove that the interference is not with the
more complex integrations of voluntary reactions. The compensations
rather suggest those described by Luciani [13] for dogs after cerebellar
injury, where the tendency to rotate is compensated for by walking
with the body curved, and where the lesion admittedly interferes only
with reflex tonic control.

The recent discussions of postural reflexes by Sherrington [17] and

STUDIES OF CEREBRAL FUNCTION IN LEARNING 281

Langelaan [9] and of their relation to decerebrate rigidity [Wilson,
19], suggest a possible explanation for the results of destruction of the
cerebral motor structures in the rat. The initiation of almost any new
activity demands a modification of postural tone, the assumption of a
different pattern of tonic contractions, which may then persist almost
unaltered for some time during the performance of finer manipulative
movements. The behaviour of the operated rats suggests strongly that
their chief difficulty is in taking a new posture. Once an activity has
been initiated and maintained for a moment even artificially, as when
the animal is held in contact with the food, it may persist for a rela-
tively long time until some stimulus to a new posture occurs. The
disturbances seem most pronounced in the attempts to perform those
activities which require the most complex and unusual postural co-
ordinations : scratching and feeding. The compensations for the motor
disturbances also suggest that the latter involve less complex, more
nearly reflex, integrations than habitual acts. The primary compensa-
tions are not by the direct restitution of motor control (recovery from
the paresis) but by the acquisition of new patterns of co-ordination
which tend to counteract the motor difficulty, much as a normal animal
would adapt to a heavy weight attached to one leg. After the destruc-
tion of the stimulable cortex the complex adaptive mechanisms are
still intact.

These facts argue strongly for a relative independence of the cerebral
mechanisms for control of postural activity from those for voluntary
movements. The hypothesis which seems best to fit the facts for the
rat is that the stimulable area and caudate nucleus are concerned chiefly
with the regulation of postural reflexes and that upon these are super-
imposed the more complex integrations constituting voluntary acts,
transmitted to the final common paths by extrapyramidal fibres. This
would make the cerebral motor structures an additional stage in the
hierarchy of spinal, vestibular, and cerebellar mechanisms for the con-
trol of postural reflexes. Such a function has already been ascribed to
the corpus striatum by Wilson [18], on the basis of work with monkeys,
and the present experiments add to his conclusions only the fact that
the stimulable area of the rat is to be classed in function with the
striate nucleus rather than with the cerebral habit-mechanisms.

The Function of the Stimulable Cortex in Higher INI vmmai.s.
Is such a view of the function of the electro-stimulable cortex con-
tradicted by evidence from the study of higher animals? It is certainly

282 ORIGINAL ARTICLES AND CLINICAL CASES

contrary to the accepted view of its function. In current text-books of
physiology and psychology it is frequently stated that the motor area is
the (sole) efferent path of voluntary movements from the cortex. Even
investigators like Rothmann, who have studied recovery from hemiplegia
in animals, look upon the stimulable area as the chief source of the
efferent fibres concerned with habitual activity. In speaking of con-
duction through extrapyramidal paths Rothmann says, " Diese extra-
pyramidal Leitung diirfte fiir die Erlernung neuer Bewegungen von
grosster Bedeutung sein, wahrend die in festen Besitz des Individuums,
iibergegangenen gut eingelernten Bewegungen vorwiegend die direkte
Verbindung von Grrosshirnrinde una Riickenmark, also die cortico-
spinale Bahn, benutzen werden " [16], Von Bechterew advances a
similar view in describing the course of conditioned-reflex arcs across
the cortex from the sensory projection areas to the motor area [1].

If this conception is correct for higher forms, it means that a pro-
nounced change in the function of the stimulable cortex has occurred
somewhere above the rodents in the evolutionary scale. But such
changes in function are rare and in this case the evidence does not
seem to show more than that the stimulable cortex of higher forms
has taken over a part of the function which the caudate nucleus
exercises in rodents.

King [8] , using the Marchi method, concluded that the pyramidal
tract in the rat is small and unimportant, but Ranson [15] has shown
that it is quite large, though made up chiefly of thinly medullated
fibres. The area of the tract shown on Ranson's figures is propor-
tionally greater than in primates, so that it cannot be argued that there
is any great increase in the anatomical importance of the tract with
ascent in the evolutionary scale.

The physiological evidence does not seem sufficient to prove that
the pyamidal tracts, even of primates, are the efferent path for
impulses to voluntary movement. Fritsch and Hitzig [5] pointed out
that the paralysis following destruction of the stimulable area is only
partial. Writing of the motor centre they stated that " . . . es ist
sicher, dass eine Verletzung dieses Centrum die willkiirliche Bewegung
des von ihm sicher in einem gewissen Abhangigkeit stehenden Gliedes
nur alterirt, nicht aufhebt, dass, also irgend einem motorischen Impulse
noch andere Statten und Bahnen offen stehen um geboren zu werden
und um zu den Muskeln jenes Beines zu eilen . . ." Nothnagel ob-
served recovery from the effects of unilateral injury to the motor area
and Carville and Uuret [4] first showed that the recovery is not due to

STUDIES OF CEREBRAL FUNCTION IN LEARNING 283

the vicarious functioning of the motor cortex of the opposite side.
Kothinann [16] showed that in monkeys the extra-pyramidal paths
are able to mediate learned activities. Thus it is certain that what-
ever the normal function of the stimulable cortex in higher mammals
it is not absolutely necessary for the performance of voluntary acts.

The chief evidence for the voluntary function of the motor areas
comes from the cerebral paralyses in monkeys and man. In them finer
manipulative movements are greatly affected and the paralysis is much
more severe than in lower mammals, yet even so it is rarely complete.
The condition has been characterized as an enormous difficulty rather
than an inability to move the paralysed limbs, and there is some
evidence that the degree of paralysis varies with the general tonic
condition of the organism.

Thus after cauterization of the stimulable cortex on one side the
rhesus monkey may show a well-marked hemiplegia. But if he is
badly frightened and chased about the room, he uses the limbs of both
sides in an apparently normal manner. Left undisturbed, he again
lapses into the paralytic condition. Gierlich [6] summarizes other
examples of this sort. Similar incidents are occasionally reported for
man. A girl with hemiplegia of fourteen years' standing, following
diphtheria, with paralysis of the right limbs and marked contractures,
tells me that when her brother pushed her from the wharf into a lake
recently she used her paralysed limbs actively in climbing out. This
patient is mentally retarded and perhaps not reliable. I know of no
well-authenticated case of the sort, but evidence of less pronounced
effects of emotional disturbance upon paralyses is frequently encoun-
tered. Depressing stimuli, such as discussion of the above patient's
very distressing home conditions, increase the extent of contracture
and limit the power of voluntary movement. Exciting stimuli, on the
other hand, will sometimes increase the extent of voluntary movement,
and have been used to advantage in re-education (personal communica-
tion from Dr. S. I. Franz).

All these facts point to the conclusion that, even in primates, it is
not so much the mechanism of voluntary movement which is affected in
cerebral paralysis, as some facilitating mechanism whose action is usually
essential to the movement, yet is not a part of the direct conditioned-
reflex arc. Graham Brown's study of facilitation after section of the
connections between pre- and post-rolandic areas [3] shows that the
descending fibres from both areas act upon the same final common paths,
and that transcortical connections are not necessary for the mutual

284 ORIGINAL ARTICLES AND CLINICAL CASES

facilitation of the areas. Voluntary movements are possible in the
absence of the motor areas ; there is not conclusive evidence that they
function directly in habit ; the phenomena of cerebral paralysis can
be explained on the assumption that the motor area is postural and
facilitating in function ; there is clear evidence that the analogous
structures in lower forms are not directly concerned with the per-
formance of learned activities. Although not conclusive, these con-
siderations seem sufficient to cast doubt upon the accepted explanation
of the function of the motor cortex, and to demand further experimental
investigation before we can conclude that the pyramidal tracts form
the efferent path for impulses to voluntary movements.

Summary.

It has been shown that the albino rat is able to acquire soma?sthetic-
motor habits after destruction of the electro-stimulable cortex and the
caudate nucleus. Visuo-motor and simple somsesthetic-motor habits
which are acquired before the operation are retained after the destruc-
tion of these structures and probably after the section of occipito-
lenticular fibres. It is clear, therefore, that neither the cerebral-
motor areas nor the subcortical nuclei are directly concerned with the
performance of learned activities.

Combined destruction of the motor area and the caudate nucleus
results in relatively permanent motor disturbances resembling hemi-
plegia in monkeys. Evidence is given that the difficulty is primarily
in assuming new attitudes, and it is suggested that the primary func
tion of the cerebral motor structures in the rat is the regulation of
postural reflexes. Existing evidence does not seem to exclude such an
explanation of the function of the stimulable cortex, even in man.

REFERENCES.

[1] Bechtebew, W. v. " Die Funktionen der Nervencentra," Jena, 1908.

[2] Beodmann, K. " Vergleichende Lokalisationslehre der Grosshirnrinde," Leipzig, 1909.

[3] Brown, T. G. " The Motor Activation of Parts of the Cerebral Cortex other than those

included in the so-called ■ Motor ' Areas in Monkeys, with a Note on the Theory of

Cortical Localization of Function," Quart. Jotirn. Exp. Physiol., 1916, 10, 103-144.
[4] Carville, C, et Duret, H. " Sur les fonctions des hemispheres cerebraux," Archives

de Physiol, 1875, 7,352-490.
[5] Fritsch, G. , and Hitzig, E. " Ueber die elektrische Erregbarkeit des Grosshirns,

Arch. f. Anat. u. Physiol., Jahrg. 1870, 300-332.
[6] Gierlich, N. " Ueber Symptomatologie, Wesen und Therapie der hemiplegischen

Lahmung," Wiesbaden, 1913.
[7] Gor/rz, F. "Ueber die Verrichtungen des Grosshirns," Arch. f. d. ges. Physiol, 1881,

26, 1-49.

STUDIES OF CEREBRAL FUNCTION IX LEARNING 285

[8] King, J. L. " The Cortico-spinal Tract of the Rat," Anat. Record, 1910, 4, 245-25-2.
[9] Langelaan, J. W. " On Muscle Tonus," Brain. 1915, 38, 235-380.
[10] Lashlby, K. S. " Studies of Cerebral Function in Learning," Psychobiology, 1920, 2,

55-135.
[11] Lashley, K. S. Studies, &c, II: "The Effects of Long-continued Practice upon

Cerebral Localization," Journ. Compt. Psychol., 1921.
[12] Lashley, K. S., and Franz, S. I. "The Effects of Cerebral Destruction upon Habit

Formation and Retention in the Albino Rat," Psycliobiology, 1917, 1, 71-140.
[13] Luciani, L. " Human Physiology," London, 1915, vol. 3.
[14] Monakow, C. v. " Die Lokalisation im Grosshirn," Wiesbaden, 1914.
[15] Ranson, S. W. " The Fasciculus Cerebro-spinalis in the Albino Rat," Amer. Journ. of

Anat., 1912-13,14.411-424.
[16] Rothmann, M. " Ueber die physiologische Wertung der cortico-spinalen (Pyramiden-)

Bahn," Arch. f. Anat. u. Physiol. (Physiol. Abth.), 1907, 217-275.
[17] Shebrington. C. S. " Postural Activity of Muscle and Nerve, "Brain, 1915, 38,191-234.
[18] Wilson, S. A. K. "An Experimental Research into the Anatomy and Physiology of the

Corpus Striatum," Brain, 1913-14, 36, 427-492.
[19] Wilson, S. A. K. " On Decerebrate Rigidity in Man and the Occurrence of Tonic Fits,"

Brain, 1920, 43, 220 268.

A PSYCHOLOGICAL INQUIKY INTO THE NATUEE OF
THE CONDITION KNOWN AS CONGENITAL WOBD-
BLINDNESS.

BY LUCY G. FILDES.

The aim of the investigation about to be described was to discover
something of the psychological characteristics of the condition commonly
called by the misleading term of congenital word-blindness — a condi-
tion which shows itself most clearly in the subjects' extreme difficulty,
or even total failure, in learning to read, and appears to be closely
related to the various forms of acquired alexia met with commonly as
the result of brain injury in later life.

Three explanatory theories, it will be remembered, have been put
forward as indicating the nature and cause of this condition, viz. : —

(1) A theory which assumes the existence of definitely localized and
circumscribed visual and auditory word-centres in the brain, the
destruction or isolation of which will destroy language in either its
visual or its auditory aspect ;

(2) A theory which interprets word-blindness as only one symptom
in a general lowering of mental ability ; and

(3) One which attributes the condition to a more specialized lowering
of power in the primary visual centres, rendering true visual percep-
tion of words and of other complex sense-data difficult.

The main problems raised by these different interpretations are two
in number. Is inability to learn to read or the loss of the power of
reading due to specific or to general defect ? If the former, does the
defect show itself only in reading, or does there appear to be any
general lowering of visual power? A psychological investigation should
throw light on these points.

The subjects mainly used for the investigation were twenty-six in
number — all children between the ages of 9 and 16 years, who were in
attendance either at ordinary elementary schools (4) or at special schools
for mentally defective children (22). They were selected on the report
of their teachers as finding reading a very great difficulty. Before the
work peculiar to the investigation was begun, all the children were
tested (a) with the Stanford revision of the Binet scale, in order to get

L37
v.3

Lashley, Karl Spencer

Studies of cerebral
function in learni~£

Biol

,i'"dl

& Medical

PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCI

UNIVERSITY OF TORONTO LIBRAR