Journal o f the Neurological Sciences, 1976, 28: 317-330
© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

317

A STUDY OF VISUAL PROCESSES IN A CASE OF INTERHEMISPHERIC
DISCONNEXION

F. LHERMITTE, F. CHAIN, F. C H E D R U and C. PENET
The Neuropsychology Research Unit - - 1 N S E R M U.84 H6pital de la SalpdtriOre, 47, Boulevard de
l'Hdpital, 75634 Paris Cedex 13 (France)
(Received 6 November, 1975)

SUMMARY

A 75-year-old right-handed woman, after a probable cerebral infarct, developed
an irregular constriction of the visual fields, a left-sided agraphia, and an anomia for
objects in the left hand. Subsequent testing demonstrated an inability to name,
though ability to recognize, letters and objects flashed in the homonymous left visual
field. An inter-hemispheric disconnexion syndrome was inferred from these findings.
The present publication concerns mainly the visual aspects of this disconnexion
syndrome.
Tasks were devised to test the abilities of the major and minor hemisphere: (a)
the left hemisphere demonstrated a complete dominance for language expression and
an incomplete dominance for written language comprehension; (b) the right hemisphere appeared to be dominant for some visuo-spatial tasks including number
comprehension ; (c) when the hemispheres were given contradictory visualinformations
on a non-verbal task (chimeric stimuli) there was a predominance of the right hemisphere. The i ight hemisphere appeared able to process complex information. Specialization of functional activities in each hemisphere is briefly discussed.

INTRODUCTION

After complete section of the corpus callosum, sensory events occurring in one
hemisphere are not available for processing in the other hemisphere (Sperry 1961).
Therefore, human cases of interhemispheric disconnexion permit an independent
study of the capacities of the major and minor hemispheres, notably of their verbal
and visual capacities (Sperry, Gazzaniga and Bogen 1969). Our paper presents a
study of visual processes in a case of interhemispheric disconnexion due to a probable cerebral infarct. We have employed methods developed in the testing of split-

318
b r a i n p a t i e n t s as well as o r i g i n a l tests. O u r case c o n f i r m s findings in cases o r .~urgical
commisurotomy

and, m o r e o v e r , i n d i c a t e s t h a t t h e m i n o r h e m i s p h e r e is c ~ p a b l e o f

processing complex information without language mediation.
CASE HISTORY
The clinical history has been reported in a previous publication (Barbizet, Duizabo, Degos
and Chartier 1974).
The patient was a 75-year-old right-handed woman, without previous illness except arterial
hypertension (170/1 t0). She suddenly complained of headache and of an impression of "hazy vision"
on 26 February 1971. Over the following few days she developed difficulty with her gait, impaired
memory and spatial and temporal disorientation. In May 1971, she became suddenly somnolent and
incontinent with a marked sensorimotor deficit of the right side. After several days she improved,
at first rapidly then slowly, over the following months. In October 1971, her condition was stabilized
and since then, particularly during the period of our observation (March 1972 to January !974). no
marked clinical change has occurred.
Neurological examination
In 1972 this disclosed : (1) a mild right-sided pyramidal syndrome; (2) a mild sensory deiicit of the
right side (hypoaesthesia, impairment of two-point discrimination); (3) a constriction of the visual
fields (see below). There was no sensory or motor deficit on the left side. However, objects placed
in the left hand were named correctly in only 28 ~ of cases (105 trials) while, on a tactile multiple
choice, a correct response occurred in a proportion of 83 ~ (84 trials). This "tactile anomia" led to the
investigation and demonstration of an agraphia of the left hand (see Fig. 1). Gestures to command
performed by the left hand were perturbed in the first few weeks of evolution. Later oil, while symbolic movements were correctly carried out with either hand, sequences of gestures (such as "raise
your arm, make a fist and extend your forefinger") were failed wiih the left hand and not with the
light.

Fig. 1. Samples of the patient's handwriting A : with the left hand: B: with the right hand. l. Patient's
name; 2. Le boulanger fait du pain (the baker is making bread). (With kind permission from Prof.
J. Barbizet and the publisher Masson, Paris, reproduced from J. Barbizet et al. (1974) Syndrome de
d~onnexion interh&nisphrrique, Rev. neurol., 130:127-141, Fig. 2).

319

.

o

Fig. 2. Visual fields of the patient. Stimuli were presented tachistoscopically, on either side of the
fixation point, in the square areas outlined (15' ;-; 15~).

320
WAIS full scale I.Q. was 96 (information: 10; comprehension: l I ; arithmetic: 7; similarities:
10; digit span: 9; vocabulary: 12; digit symbol: 8; picture completion: 9; picture arrangement: 8;
block design: 6; object assembly: 8). Spontaneous speech, repetition and naming were normal as
was oral comprehension, as tested by simple and complex commands. Reading was slow but correct.
Visual recognition of images and objects was normal. There was no disturbance of behaviour.
The disorder of memory and disorientation noted at the onset had been completely resolved. The
patient cooperated fully during the examination and subsequent testing.
An EEG obtained in 1971 showed bilateral slow polymorphic delta activity with a left hemisphere predominance. In June 1971 a brain scan (Ptc 99) showed two foci of uptakc, one in the
left temporo-occipital region and one in the right parieto-occipital area. A follow-up brain scan done
in December 1973 was normal.

Ophthalmological examhlation (March 1972)
Uncorrected visual acuity was 6/10 in each eye. Fundal examination showed signs o1 hypertensive retinopathy. Opto-kinetic nystagmus was present and symmetrical. Goldmann perimetry
showed irregularly contracted visual fields, the defect being greater in the right homonymous
field than in the left (Fig. 2). Fusion threshold for flickering light (CFF) was tes*ed for each visual
field, first in the right, then in the left eye. For each eye, the left visual field was tested first. A flickering lamp was placed 10° from the point of fixation. The flicker frequency was increased, then decreased
in a step-wise fashion. For each frequency the patient was requested to say whether or not she saw
a flicker. In these conditions, the C F F was 31 for the left visual field and 26 for the right visual
field. These figures were identical for the right and left eyes.
In summary, a 75-year-old woman, 8 months after the sudden onset of a transient neurological
syndrome characterized by disorientation and disturbance of vision and gait, presented with a stable
sensorimotor deficit of the right side, a constriction of the visual fields, and features of an interhemispheric disconnexion syndrome (tactile anomia, agraphia and praxic difficulties) on the left
side. The sudden onset, the progression of the clinical signs, and the clearing of the brain-scan foci
suggested an ischaemic vascular accident. There were probably lesions of the pathways between the
two hemispheres, in the corpus callosum and/or in the adjacent white matter.
TESTING METHODS
The patient was seated facing a screen with her head immobilized by an ophthalmological
support. She was asked to fix her gaze on a red spot marking the centre of the screen and to try to

rt6tvt
A
I;aft

B
Fig. 3. Sample of EOG recording of eye movements. The upper tracing (A) represettts the gaze
displacement and the lower (B) the stimulus presentation. An arrow indicates the onset of the stimulus
presentation on tracing (B). Note, on tracing (A), the small eye movement following Stimulus presentation (300 ms).

321
recognize the items which were flashed on the screen. The pictures were presented tachistoscopically,
laterally to the central red spot, either in the right (RVF) or in the left (LVF) homonymous half
visual fields. Each tachistoscopic presentation lasted 250 msec. The immobility of gaze was
checked by electro-oculographic recording of eye movements (Fig. 3). Although the gaze did not
move during stimulus presentation, a gaze displacement was frequently observed immediately after
the end of the presentation (300-350 msec after the onset of projection).
Based on previous studies of commissurotomized subjects, 3 series of tests having different
objectives were used: (1) tests designed to demonstrate a visual hemispheric disconnexion; (2) tests
designed to study the perceptual capacities of each hemisphere tested separately; (3) simultaneous
stimulation of the two hemispheres by chimeric figures.

1 - Findings of visual disconnexion
Namhlg and designation Of pictures and letters. Twenty-four photographs of objects and 24
letters were serially presented in the LVF, then in the RVF. The patient was requested to "name
what she saw". Then the test was repeated after she was requested to keep silent and to indicate the
picture which she saw among a group of 6 pictures. For each test, the designation was effected serially by the two hands (first the left and then the right hand). The results are expressed as percentages
of conect responses. In the tests of designation, the scores were compared to the chance level score
(17 %) by a chi-squared test.
Visuo-tactih, matchings. Twenty-four photographs of objects and 24 letters were presented
successively in the LVF and in the RVF. The patient had to choose tactually among 3 objects or 3
wooden letters the one corresponding to the visual stimulus. She was asked to palpate successively

•

OO0
O0
i
i

A

B

OO0
OO0
O0

0
O0

O0

•
•

0
0

C

D

•
•

•

•
•

•

•

O

OOO

t::

•

oo

•

• OO
o o o

Fig. 4. Visuo-spatial tests:A: recognition of nonsense shapes; B: position of a dot in a square; C:
number of dots from an identical spatial arrangement; D: variable disposition of 5 dots; E: number
of dots from a different spatial arrangement (for this test, a multiple choice set is shown to the right
of the stimulus). In test C, the patient had to match a certain number of dots (from 2 to 8) to the
same number of dots disposed in an identical fashion. In test E, the patient had to match a certain
number of dots (from 2 to 8) to the same number disposed differently.

322
the 3 objects or letters while blindfolded. For each presentation, the response was tested serially for
either hand, first the hand contralateral to the presentation in the half-field, then the ipsilateral one.
Finally the patient was asked to name the stimulus.

I1 - - Perceptual capac#ies o f each hemisphere
In each of the following tests the stimuli were presented to the LVF and then to the RVF.
Comprehension 0 / written language. The stimulus word was printed. The response was a cho ice
from 6 pictures of objects or from 6 words in cursive script (72 words were presented for the choice
of pictures and 96 for the choice of words in cursive script). In another test, the response was by
multiple choice among 3 palpated objects (12 different words were presented).
Visuo-spatial matehhlg tests (Fig. 4). Different stimuli were used to test the ~isuo-spatial
capacities of each hemisphere: (a) nonsense shapes; (b) position of a dot in a square; (c) the number
of dots from an identical spatial disposition; (d) 5 dots variably disposed; (e) the number of dots
from a different spatial arrangement (we will refer to this task as the "number comprehension'"
test). Foreach of these tests, 24 different stimuli were projected serially. As soon as a stimulus had
been flashed, the patient was shown an array of 6 items belonging to the same category (e.g. : nonsense shapes) and she was requested to point to the one which matched the stimulus which had
just been flashed.
111

Chimeric figto'c.s
In this last series of tests, composite right-left chimeric stimuli were presented cc2"~lrally in the
patient's visual fields (Levy, Trevarthen and Sperry 1972): (a) i6 pictures of asymmetrical objects:
(b) 24 pictures of symmetrical objects: (c) 24 letters; (d) 24 pictures of objects belonging to the same

A

B

D

C

E

Fig. 5. Examples of chimeric stimuli used in the experiment. A: asymmetrical objects; B: symmetrical
objects; C: letters; D: cars; E: faces.

323
category; cars were chosen to correlate with the findings of a previous study (Lhermitte, Chain,
Escourolle, D u c a r n e a n d Pillon 1972); a n d (e) 40 faces (Fig. 5). The right-left pattern chimeras were
joined at the midline a n d projected on each side of the fixation point so that a different picture was
presented in each hemifield. For each variety of stimulus, the patient was requested to respond
initially by designation (left h a n d then right hand), a n d afterwards verbally (naming, or, for faces,
description). For each test, the percentage of stimuli correctly recognized in the L V F (L) and the
R V F (R) was recorded as well as the percentage of erroneous responses not corresponding to the
two stimuli presented (E).
RESULTS
I - Fimlings¢~fvisuahliscommxion
Pictures o/'objects and letters (Table 1). Stimuli presented in the R V F were easily named.
T h e only two errors concerned pictures which were slightly larger than the limits of the "seeing"
visual field.
In 83 % of cases, pictures of objects presented in the L V F could not be n a m e d correctly. T h e
c o m m e n t s of the patient are interesting. She sometimes said: "1 could not s e e . . , that goes to f a s t . . .
1 d o n ' t k n o w . . . " or " l k n o w I have recognized it but l can't r e m e m b e r the n a m e o f it" (17% of
responses). Otherwise, she described the general form of the picture and eventually gave a n a m e to it
(49%). T h u s , when presented with a c o m b , she said " t h a t was a long object with t e e t h . . , it could
be a rake" and, when presented with a car-key: "it is egg-shaped, with a tail, a kind of grater for cooki n g ' . On other occasions, her response had no relationship to the picture presented: a '~food-mill"
for a thimble (17~o). Purely semantic associations between the response and the stimulus, as noted
in some cases of visual agnosia (Lhermitte, Chedru a n d Chain 1973) and of optic aphasia (Lhermitte and Beauvois 1973), were not observed. In these tasks the patient was not told how m a n y
stimuli were to be presented but she was told whether the stimuli were going to be letters or pictures.
W h e n she was u n a w a r e of the nature of the stimulus, verbal responses were totally inaccurate; in
particular, she was often unable to say whether she had been flashed a letter or a picture. W h e n asked
to point with either hand to the letter or to the picture which matched the stimulus, the patient
p r o m p t l y a n d correctly responded regardless o f which h o m o n y m o u s field was stimulated.
Visuo-tactile matching (Table 2). On the " L V F / l e f t - h a n d m a t c h i n g s " , the scores were significantly better than chance level. D u r i n g the task, the patient tended to verbalize the stimulus presented in her LVF, a l t h o u g h she was asked not to speak. Quite often, the erroneous verbal identification
did not prevent a correct matching. T h u s , when presented the picture o f an electric torch she said

TABLE I
NAMING AND MULTIPLE CHOICE RECOGNITION
LETTERS
Naming
LVF

Pictures of
objects (24)
Letters (24)

17%
0%

OF PICTURES OF OBJECTS AND

Multiple choice
RVF

92%
I O0 ~,/

LVF

Chance
level

RVF

LH

RH

RH

LH

83%"
75 %"

7500~`'
87 %"

92%
100 %

9~'~......
,,,
96 %

17%
17 oi;

The n u m b e r of stimuli presented in each experimental condition is indicated in brackets. Scores are
expressed as a percentage of correct responses. Each multiple choice score was compared to the chance
level score by a Z2-test. Significance of the differences are schematized as follows: ap ,: 0.05; ~ P
0.02; "P :~ 0.01.
For this table a n d the others, the following abbreviations were used: LVF: presentation in the
left visual field; R V F : presentation in the right visual field; R H: pointing with the right h a n d ; L H :
pointing with the left h a n d ; ( R H ) palpation with the right h a n d ; (LH): palpation with the left h a n d ;
NS: not significant.

324
TABLE 2
VISUO-TACTILE M A T C H I N G

Pictures of
objects (24)
Letters (24)

LVF
............
(RH)
(LH)

(RH)

(LH)

29)o
35%

100%
100%

100%
100%

62%'
79%'

RVF

Chance
level

33.3%o
33.3%

See Table 1 for abbreviations and comments on statistical analysis.
TABLE 3
C O M P R E H E N S I O N O F WRITTEN L A N G U A G E
Multiple choice
sets

LVF

RVF

Chance
level

Pictures (72)
Words in cursive
script (96)
Palpation of
objects (12)

41 O/oe

100%

7%

33 %,,

100 %

17 %

92%"

100%

33.3%

See Table 1 for abbreviations and comments on statistical analysis.
"it is a kind of double comb standing up"; a moment later, after palpating the 3 objects, she picked
up without hesitation the electric torch and added "that's i t . . . it's a comb". Scores for the " L V F /
right-hand matching" tests were not better than chance. When the stimuli were presented in the RVF,
the patient could identify, by palpation with either hand, the corresponding object or w ~ e n letter.
II - - Perceptual capacities o f each hemisphere
Comprehension o f written language (Table 3). As expected, words projected in the RVF were

correctly read aloud and matched with their corresponding images (or words in cursive script).
When projected in the LVF, no stimulus words could be read aloud. By contrast, the scores obtained on the matching tasks were much better than chance. The rate of correct identification was
higher for short words (3 letters) than for longer words. When asked prior to matching to say what
she saw, the patient answered 'I don't k n o w . . , that went too f a s t . . . " . On the other hand, after
matching, she was sure she had previously been shown a picture and not a word.
TABLE 4
VISUO-SPATIAL TESTS

Test A (24)
Test B (24)
Test C (24)
Test D (24)
Test E (24)

LVF

RVF

Chance
level

LVF/RVF
comparisons (Z2)

67%"
46 %'~
67 %~'
50 ~
57 %c

30%
37 % t'
17 %
50 %c
27 %

17~
17 %
17 %
17 %
17 %

NS
~'
NS
~

See Table 1 for abbreviations and comments on statistical analysis.

325
Matching a written word with a palpated object was possible provided the 2 stimuli were
presented to the same hemisphere.
Visuo-~patial m a t c h i n g tests (Table 4). In tests B and D both hemispheres performed similarly.
In tests A, C and E, the right hemisphere (i.e. the LVF) performed significantly better.
I11 - - Chhneric stimufi

The patient was not surprised or uneasy when confronted with chimeric images. On the other
hand, she did not realize that she was presented two different stimuli simultaneously, in these respects, she differed from normal subjects and behaved like the 4 split-brain patients studied by Levy
et al. 0972).
The data pertaining to this part of the study are summarized in Fig. 6. When the response
was verbal (naming or description), the RVF was clearly privileged whatever the type of the stimulus.
When the response was non-verbal (designation): (a) the RVF remained markedly privileged in case
of objects or letters; (b) the LVF was privileged in case of faces; (c) in case of cars, both visual fields
were equivalent.
Of note was the large number of "E" responses, i.e. of responses leading to the designation of
an item not belonging to the chimeric stimulus. In this respect, our patient appeared to differ from the
split-brain subjects. Such a difference might be accounted for by the choice of stimuli: (a) the faces
used could not be recognized by the presence or absence of a beard, a moustache or glasses; (b)
half-cars and half-letters are difficult to identify on a tachistoscopic presentation even for normal
subjects.

ASYMMETRICAL
OBJECTS

SYMMETRICAL
OBJECTS

LETTERS

CARS

FACES

I00

too% r
B7.5

e7

s4

Is

L

R

E

L

R

E

L

R

E

70

<

L

R

E

L

R

R

E

L

R

E

L

"
R

R

E

'L

I ....

oLj o L

L

E

L
L

R

n
E

E

too%'
a7.5

a7

i
i

I
,

13

L

R

E

NOT TESTED

Fig. 6. Results of the tests with chimeric figures.

L

R

A
E

NOT TESTED

U
L

R

326
DISCUSSION

The existence of an agraphia and of a tactile anomia on the left side makes the
present case similar to surgical and non-surgical cases of interhemispheric disconnexion (Geschwind and Kaplan 1962: Sperry et al. 1969; Schott, Michel and
Dumas 1969; Brion and Jedynak 1972). The visual disorder also corresponds to cases
of callosal sectioning, either limited to the splenium (Trescher and Ford 1937;
Maspes 1948; Gazzaniga and Freedman 1973; lwata, Sugishita, Toyokura, Yamada
and Yoshioka 1974), or total division (Gazzaniga, Bogen and Sperry 1965). One
could argue that the left visual field failed to recognize stimuli because of left field
constriction and amblyopia. In fact, stimuli were flashed within the limits of the
"seeing" visual fields as determined by perimetry. On the other hand, thresholds for
flickering light were higher in the left than in the right visual field. Finally, recognition
of highly contrasted letters (black block letters on a white background) was worse
than that of lowly contrasted photographs of objects, a finding hardly consistent
with an amblyopia but consistent with the hypothesis of disconnexion ~vide h~'a).
As for the failure of the right hand to recognize stimuli presented to the left visual
field, it cannot be accounted for by sensory loss, since objects palpated by this hand
could be normally named.
lnterhemispheric disconnexion thus appears to be the most reasonable explanation for the patient's inability to name stimuli presented in her left visual field
in spite of their recognition as determined by correct matching to corresponding
pictures or objects. In this respect, a comment by the patient is of interest : "l know
that I have recognized it but I can't remember the name of it".
Considering the clinical data and the initial brain-scan, it is probable that there
were bilateral posterior ischaemic lesions involving commisural fibres at the level of
the splenium and/or adjacent white matter. The lesions certainly extended rostral
to the splenium since there were signs of sensory and motor disconnexion. The lesions
did not lead, however, to a complete disconnexion since, as opposed to split-brain
patients, some pictures presented in the left visual field could bc described
and occasionally named. Of note was the absence of a single correct oral
response when written verbal material was projected. Such a dissociation between
words and objects recalls cases of splenial section where, in spite of alexia in the left
homonymous visual field, pictures were named correctly (Maspes 1948; lwata et al.
1974). We may suppose that callosal transmission of information connected to written
symbols requires a "coding" which is more accurate, less redundant and therefore
more fragile, than that related to objects. Another feature indicating that the disconnexion was not complete was the patient's capacity to retrieve with her right
hemisphere stimuli presented to her left hemisphere. Thus, by feeling a set of objects
with her left hand, she was able to indicate the one which corresponded either to the
picture previously projected in her right visual field, or to the object palpated by her
right hand (Barbizet et al. 1974). On the other hand when the stimulus was presented
to the right hemisphere and the multiple choice set was presented to the teft hemisphere, matching was not possible. One can assume that, when presented a stimulus,

327
the left hemisphere informed the right hemisphere through (silent) verbalization; in
the reverse condition (stimulus presented to the right hemisphere, multiple choice
made by the left hemisphere) disturbing effects of incoherent (silent) verbalization
would have prevented the same mechanism from occurring.
Control of the upper limbs. In matching tests, the patient performed correctly
with either hand whichever visual field had been stimulated. Gazzaniga et al. (1965)
have suggested that the proximal musculature of the upper limbs, which is involved
in pointing movements, is controlled by the ipsilateral hemisphere almost as much
as by the contralateral hemisphere. Another, and more likely explanation (Trevarthen 1974) is that the patient seeks the stimulus among a set of pictures with her
maculae, i.e. with both hemispheres at the same time. The hemisphere "stimulated"
by the tachistoscopic flash would then localize the picture and fix the gaze when a
choice is made. Thereafter, either hand would be able to perform the pointing since
both hemispheres would be aware of the choice.
Visuo-tactile matching. As anticipated from split-brain studies, the patient was
able to match tactile to visual stimuli solely with the right hemisphere. These crossmodal matchings suggest that the right hemisphere, without verbal mediation, is
able to elaborate "concepts", at least concepts of familiar objects. We will refer again
to the significance of this fact and its relevance to the problem of the relationship
between language and thought.
Comprehension of written language by the right hemisphere. Written words were
"understood" by the right hemisphere: when words were flashed in the LVF, the
patient retrieved, on visual or tactile multiple choice, the corresponding item at a
frequency greater than chance. These performances cannot be explained by a partial
disconnexion for, if linguistic information had reached the left hemisphere, it could
have been at least partially expressed, something which did not occur in the patient.
Similar observations have been made in fplit-brain patients (Gazzaniga and Sperry
1967) but their significance has been controversial: in view of the epileptogenic lesions
in their cases, a reorganization of brain function might have occurred. In this respect
our case provides evidence that, in an adult without previous neurological damage,
the right hemisphere, although unable to formulate words, can understand some
written language. This fact could account for the "paradoxical" performances of
certain pure alexic patients, e.g., the patient of Albert, Yamadori, Gardner and
Howes (1973) who, though unable to "read" could nevertheless group words according to their meanings and indicate a neologism in a series of English words. In such
cases, the remnants of written comprehension may be performed by a right hemisphere
which - - because of the left lesion - - receives and processes all the visual information.
Visuo-spatial tests. The right hemisphere appears equal or superior to the left
one in the appreciation of the relative position of various visual patterns and in the
recognition of meaningless shapes. These findings correspond to those of splitbrain studies (Sperry et al. 1969: Nebes 1972), and fit with the now classical notion
of minor hemisphere dominance for visuo-spatial performance (Milner 1971).
The results of the test of number comprehension (test E) deserve further comment. it has already been stated that in this test the left hemisphere was unresponsive

328
while the right hemisphere performed at a level very significantly superior to chance.
While it appears surprising that the right hemisphere, though unable to verbalize,
could perceive the equivalence of 2 series of dots having in common only their number, studies in experimental psychology have demonstrated that a normal individual
can grasp immediately a small number of elements (less than 10) without vcrbal
enumeration (Woodworth 1949). It is probably through such a direct "comprehension"
that the task is performed by the right hemisphere. As for the inferiority of the left
hemisphere in this regard, it is in agreement with that of normal subjects (Kimura
1966). It is suggested that, in normal subjects the immediate comprehension of numbers is performed by the right hemisphere, while other modes of number perception,
requiring the intervention of language, are performed by the left hemisphere acting
in cooperation with the right one. Whatever their significance in relation to normal
performance, our data indicate that the right hemisphere is able to carry out complex
intellectual operations without language mediation. Some animals comprehend
numbers in a manner similar to that of the right hemisphere of our patient : "'we are
here so close to man that this ability (number comprehension) is shown by animals
and man provided the stimuli be not verbally enumerated. In such situations, human
performances are not better than performances of b i r d s . . . It is putting thought
into words that makes the difference (between man and animals) and not thought by
i t s e l f . . . " (Chauvin 1969).
Chimeric figures. When chimeric figures are flashed to split-brain patients, the
verbal response always corresponds to what appears in the right visual field while the
non-verbal response (designation) generally corresponds to what appears in the left
visual field (Levy et al. 1972). This does not depend on the nature of the chimeric
figure. In the present case, the verbal response also corresponded to what appeared
in the right visual field. However, the non-verbal response was not as expected; when
the stimulus consisted of letters or pictures the right visual field was privileged, while
when it consisted of faces, there was a predominance of the left visual field.
This dissociation conforms to what is known of the role played by each hemisphere
in visual recognition (Tzavaras, H6caen and Le Bras 1970: Rizzolati, Ulmita and
Berlucchi 1971). This difference in performance between our patient and the splitbrain cases may, however, be due to the fact that our patient tended to verbalize
everything presented to the right visual field. In fact, the right visual field was privileged with easily verbalized stimuli (letters and simple pictures of objects), while
it was not with stimuli difficult to name or describe (faces or cars). This
interpretation is supported by the following observation. When chimeric symmetrical objects were flashed to the patient for the initial 6 times, the patient
pointed without saying anything to those pictures corresponding to the left
field; it was only on the seventh trial that, while naming correctly the right-field
picture, she hesitatingly pointed to both "right" and "left" pictures. Thereafter, she
systematically pointed to the right-field pictures. Conceivably at a certain stage, the
choice was facilitated by verbalizing the picture flashed, biasing performance to the
right-half field.
In the present case, as in those of split-brain patients, the right hemisphere

329
was shown to be capable of processing complex nonverbal intellectual tasks, such as
retrieving, by palpation, an object similar to another object visually perceived, matching a written word to a picture, pairing patterns of dots having in common only
their number. It might seem astonishing that a "language-deprived" hemisphere
could perform such tasks, since we are accustomed to think of language as a prerequisite for intellectual processes. In fact, studies of infants and children, of deafmute children, of aphasics and of animals have clearly demonstrated that high level
intellectual activity (i.e., "thinking", "reasoning") may be present without, or with
limited, language intervention. The data obtained in disconnexion patients are further
evidence of this ability. The intellectual abilities of the tight hemisphere appear to be
different from those of the left. Experimental and clinical evidence suggests that the
right hemisphere plays a role in tasks requiring holistic or immediate comprehension
while the left hemisphere is concerned with serial analysis and language-bound
activity. From an ontogenetic view-point one may speculate that, at birth, the potentialities of both hemispheres are equivalent. In the course of childhood, under the influences of genetically-induced cerebral dominance and of environment, language
mechanisms of right-handed subjects develop in the left hemisphere. This leads one
to ask whether the powerful urge for the acquisition of language does not result, in
the left hemisphere, in a trend to process every piece of information in terms of
language. On the other hand, the right hemisphere, being hardly involved in language
activities, becomes able to develop and, later, to use other functional activities, more
particularly those which are responsible, to a large extent, for visual processes giving
rise to cognitive activities at a very elaborate level.
Whatever conclusions can be drawn from such hypotheses, interhemispheric
disconnexion yields evidence that high level intellectual operations can take place
without the support of language.
ACKNOWLEDGEMENTS
We wish to thank Pr. Barbizet for allowing us to study his patient, Dr. J.
Brown for help in the translation and Miss Bescheron for help in the statistical analysis.

REFERENCES
Albert, M. L., A. Yamadori, H. Gardner and D. Howes (1973) Comprehension in alexia, Brain,
96:317 328.
Barbizet, J., P. Duizabo, J. D. Degos and B. Chartier (1974) Syndrome de d6connection interhOmisph6rique d'origine ischSmique, Rev. neuroL, 130: 127-142.
Brion, S. and C. P. Jedynak (1972) Troubles du transfert interh6misphOrique - - Apropos de trois
observations de tumeurs du corps calleux, Rev. neuroL, 126: 257-266.
Chauvin, R. (1969) P,wchophys[o[ogie, VoL 2 (Le Comportement Animal), Masson, Paris.
Gazzaniga, M. S. and H. Freedman (1973) Observations on visual processes after posterior callosal
section, Neurology (Minneap.), 23 : 1126-1130.
Gazzaniga, M. S. and R. Sperry (1967) Language after section of the cerebral commissures, Brain,
90: 131-148.

330
Gazzaniga, M. S., J. B. Bogen and R. W. Sperry (1965) Observations on visual perception after
disconnexion of the cerebral hemisphere in man, Brain, 88: 221-236.
Geschwind, N. and E. Kaplan (1962) A human cerebral deconnexion syndrome, Neurologi ! Mbmecq~,
12: 675-685.
lwata, M., M. Sugishita, Y. Toyokura, R. Yamada et M. Yoshioka (1974) Etude sur le syndrome
de disconnexion visuo-linguale apr6s la transection du spl6nium du corps calleux
Troubles
de la verbalisation des informations visuelles dans I'h6misph+re mineur, J. neurol. Sci., 23:
421-432.
Kimura, D. (1966) Dual functional asymmetry of the brain in visual perception, Neurop.~vchologic;,
4 : 275-285.
Levy, J., C. Trevarthen and R. W. Sperry (1972) Perception of bilateral chimeric figures following
hemispheric deconnexion, Brain, 95 : 61 78.
Lhermitte, F. and M. F. Beauvois (1973) A visual speech disconnexion syndrome - Report of a case
with optic aphasia, agnostic alexia and colour agnosia, Brain, 96:695 714.
Lhermitte, F., F. Chedru and F. Chain (1973) A p r o p o s d'un cas d'agnosie visuelle, Rev. neurol.~
128: 301-322.
Lhermitte, F., F. Chain, R. Escourolle, B. Ducarne and B. Pillon (1972) Etude anatomo-clinique
d'un cas de prosopagnosie, Rev. neurol., 126: 329-346.
Maspes, P. E. (1948) Le syndrome exp6rimental chez l'homme de la section du spl6nium du corps
calleux - - Alexie visuelle pure h6mianopsique, Rev. nenrol., 80: 101Yl13.
Milner, B. (1971) lnterhemispheric differences in the localization of psychological prccesses in man,
Brit. reed. Bull., 27:272 277.
Nebes, R. D. (1972) Dominance of the minor hemisphere in commissurotomized man on a test of
figural unification, Brain, 95: 633-638.
Rizzolatti, G., C. A. Ulmita and G. Berlucchi (1971) Opposite superiorities of the right and left
cerebral hemisphere in discriminative reaction time to physiognomical and alphabetical material,
Brain, 94: 431-442.
Schott, V., F. Michel and R. Dumas (1969) Apraxie ideomotrice unilat6rale gauche avec main gauche
anomique, Rev. neurol., 120:359 365.
Sperry, R. W. (1961) Cerebral organization and behavior, Science, 133 : 1749 1757.
Sperry, R. W., M. S. Gazzaniga and J. E. Bogen (1969) Interhemispheric relationships: the neocortical commissures - - Syndromes of hemisphere disconnection. In: P. J. Vinken and G. W.
Bruyn (Eds.), Handbook of Clinical Neurology, Vol. 4 (Disorders of Speech, Perception, and
Symbolic Behaviour), North Holland Publ. Comp., Amsterdam, pp. 273-290.
Trescher J. H. and R. R. Ford (I 937) Colloid cyst of the third ventricle, Arch. Neurol. PsLvchiat., 37:
959 973.
Trevarthen, C. (1974) Personal communication.
Tzavaras, A., H. H6caen and H. Le Bras (1970) Le probl6me de la sp6cificit6 du deficit de la reconnaissance du visage humain Iors des 16sions h6misphdriques unilat6rales, Neuropo,chologia,
8 : 403-416.
Woodworth, S. (1949)Psychologie Expdrimentale, Vol. 3 (French translation), Presses L!niversitaires
de France, Paris.