Journal of Neurology, Neurosurgery, and Psychiatry, 1980, 43, 719-724

Colour anomia restricted to the left visual hemifield
after splenial disconnexion
J ZIHL AND D VON CRAMON
From the Max-Planck-lnstitute for Psychiatry, Munich
S U M MARY In a patient with damage to the right occipital lobe and to the splenium of the
corpus callosum, an incomplete colour anopia in the left upper quadrants and a colour anomia
was found for the complete left visual hemifield beyond 20 eccentricity. The patient had no

difficulty in recognising coloured targets when presented in the periphery of the left visual halffield and in the foveal region, but could not name them correctly. The results suggest that the
lesion of the splenium of the corpus callosum disconnects the right visual cortex from the
language areas of the left hemisphere, and the specific disturbance of colour naming is the
consequence.

since her childhood. Menorrhagia and metrorrhagia as well as many "bruises" were observed
during the last four years. In June 1979 she suffered again from acute sinusitis. A few days later
a left-sided hemiplegia with ptosis on the right side
developed within hours, suggesting a right-sided
Weber syndrome. The patient was admitted to
another hospital and became drowsy and eventually comatose. She was said to have remained
unconscious for about two weeks, but no detailed
information about this period is available. The
CT-scan showed haemorrhage in the medial
temporo-occipital region of the right hemisphere
penetrating into the inferior horn of the lateral
ventricle and into the superior, ambient and interpenduncular cisterns. The right peduncle had been
slightly compressed by a space-occupying blood
clot.
Upon admission examination showed a pale
patient with petechiae and generalised bruising.
The spleen was slightly enlarged. The main laboratory findings were a hypochromatic anaemia, a
thrombocytopenia (<3000/mm3) with giant platelets, a prolonged bleeding time, a reduced
maximal thrombus-elasticity, a pathological
prothombin-consumption-test and an increased
CASE HISTORY
number of megakaryocytes in the sternal marrow.
A 41 year old Caucasian suffered from frequent
On the basis of these findings idiopathic thromacute sinusitis with secondary bronchial infections bocytopenic purpura (chronic-continuous type)
Address for reprint requests: Dr J Zihl, Max-Planck-Institut fur was diagnosed. The patient was given steroids in
Psychiatrie Kraepelinstrasse 10 D-8000 Munchen 40, W. Germany.
high dosage, and the number of thrombocytes inAccepted: 20 March 1980
creased to normal values.
719
Colour anomia may be a symptom of amnestic
aphasia (anomia), a primary aphasic disorder
arising from damage to the cortical structures
directly subserving language function.' A second
variety of colour anomia is very often accompanied
by a right sided hemianopia and a dyslexia. The
latter disturbance of colour naming has been
attributed to an anatomical disconnexion between
the visual cortex and the cerebral structures subserving language function. This kind of colour
anomia is secondary to this disconnexion while
language function remains intact.2 3 In these cited
cases colour anomia was present in the remaining
intact visual field. The loss of one or both upper
quadrants is often accompanied by an inability
to recognise colours.4 No disturbance of naming
colours, however, was observed for a particular
part of the visual field, for example, for one
hemifield. Here we report a case with a lesion
including the right occipito-temporal region and
damage to the splenium of the corpus callosum.
As a consequence colour perception was partially
lost in the left upper quadrants and a disturbance
of colour naming for the left hemifield beyond 20
eccentricity was found.

J Zihl and D von Cramon

720

The patient was transferred to our department
in November 1979. By that time the neurological
signs had receded except for a latent paresis of
the left arm. The patient was now complaining of
intermittent blurred vision.
The CT-scan was repeated in November 1979
(fig 1). There was an enlargement of the collateral
trigone and the inferior horn of the right lateral
ventricle (1A, B, C). The superior cistern was
widened at least on its right side. A well marked
low density zone was present mainly in the white
matter of the medial temporo-occipital gyri of the
right hemisphere. The lesion extended cranially
into the precuneus (ID) and rostrally into the
right half of the splenium of the corpus callosum
(IC). Parts of the right hippocampal gyrus and
the cingulate gyrus at its isthmus were probably
also affected (IB,C).
The patient was right-handed, alert and orientated in person, place and time. She did not show
any difficulty in verbal expression and comprehension or in naming objects. She had no alexia

Fig 1

or agraphia. Visual acuity was normal in both
eyes.

Methods
Visual perimetry Visual fields were mapped with
the Tubinger perimeter,56 using a target of 116
min arc diameter and maximum intensity (320
cd/M2) for 900 dynamic perimetry. In addition,
static perimetry was performed to determine the
profile of the gradient of light sensitivity along the
horizontal axis and along the 450 axis (right upper
quadrant) and the 1350 axis (left upper quadrant)
respectively. Visual fields were also mapped with
coloured targets of 116 min arc diameter and
maximum intensity (320 cd/M2). The spectra values
of the coloured targets were 480 nm for "blue,"
525 nm for "green" and 656 nm for "red":
Colour fields were also determined using the
Tubinger perimeter. The patient's task was to
press a button whenever she could identify the
coloured target.

Results of CT scans obtained at four different levels. For details see text.

721

Colour anomia restricted to the left visual hemifield after splenial disconnexion
In addition critical flicker fusion (CFF) was
determined by measuring CFF-fields for 15, 20
and 25 Hz (target: 116 min arc diameter; 32 cd/
m2). The flickering target was moved with a velocity of about 10/sec from the periphery towards
the centre of the visual field. The patient was
asked to push a button whenever she could detect
the target flickering.
Colour perception and naming The tests on
colour perception and naming within the visual
field were performed with the Tubinger perimeter.
Targets of different wavelength (blue: 480 nm;
green: 525 nm; yellow: 578 nm; red: 656 nm)
were presented at different eccentricities along the
horizontal axis and along the 450 meridian in
the right half-field, and the 1350 and 2250 meridians in the left half-field. Target size was always
116 min arc diameter, intensity was 320 cd/M2;
stimulus duration was either 100 ms or unlimited.
Fixation was controlled with an infra-red camera.
The patient's task was to compare the colour of
a given target presented beyond 20 eccentricity
with the colour of several foveally presented targets, and to indicate, whether the colour was the
same or different. For colour naming, targets
were presented at a given position in the left or
right visual hemifield, or foveally. The patient was
instructed to name the colour of the target.
In addition to the colour tests with the perimeter, the Farnsworth-Munsell 100-hue test was
used.7 Furthermore, a slide with 60 different
coloured written colour names (for example,
"blue" written in red letters) was presented and
the patient was asked to name the colour of the
words and-in a second session-to read the
words.
Object naming in the visual field In order to test
object recognition and the naming of objects,
symlbols (circle, triangle etc) and figures of objects
(a house, an apple etc) were presented tachistoscopically (exposure duration: 100 ms) in the
left and right visual hemifield. The patient's task
was to identify the presented figures by naming
them.
Reading in the left and right hemifield Letters
and words of appropriate size (for visual acuity
beyond 20 eccentricity) were presented tachistoscopically (exposure duration: 100 ms) in the left
and right hemifield. The patient's task was to name
the letters and to read the words.

Results
Perimetric data Fig 1 A shows the binocular
visual field for white light and the colour visual

fields for blue, green and red test targets. There
was an incomplete colour anopia in the left upper
quadrants indicating a partial loss of colour per-

ception in this region of the visual field. Contrast
sensitivity, however, was not diminished in the
left hemifield, neither for the horizontal axis nor
for the diagonal meridian (1350 meridian as compared to the 450 meridian in the right hemifield;
cf fig iB). Furthermore, no restriction was found
for the CFF-fields in the left as compared with the
right hemifield.
Colour perception and colour naming The results
for colour comparison in the left hemifield are
shown in table 1. No deficit in colour comparison
was found along the horizontal meridian and for
the upper and lower quadrants except for a small
decrease in correct responses for green at 200
eccentricity along the 2250-meridian (lower
quadrant) and along the 1350-meridian (upper
quadrant).
Table 1 Colour comparison (fovea-left hemifield)
Correct responses in percent (n=30) 100 ms stimulus
duration
Colour

Red
Blue
Green
Yellow

Lower quadrant
10°
15°

200

Upper quadrant
5°
JO

15J

100
100
100
100

100
100
86-7
100

100
100
100
100

100
100
93-3
100

100
100
100
100

100
100
100
100

Colour naming was found to be correct for the
fovea and for the right hemifield.
In contrast to these observations, colour naming
was clearly affected in the left hemifield. In the
upper quadrant, the test was performed only up to
10° eccentricity, because the border of the colour
field for "red" was at 120 eccentricity. When
coloured targets were presented along the horizontal meridian, a gradual decrease of correct
responses was obtained. "Green" was first affected
(at 30 eccentricity), followed by "red" and
"yellow." At 200 eccentricity, the percentage of
correct responses dropped markedly also for "red"
and "yellow." "Blue" was in all cases of incorrect
response the substituted colour name. Blue targets were always called "blue" (cf table 2).
Similar results were obtained for different positions
along the 1350-meridian and 2250-meridian (upper
and lower quadrants) in the left hemifield. A
comparable decrease in the percentage of correct
responses was found; the order of the decrease
was quite the same (green, red and yellow; tables
3 and 4). There was no clear difference between
the two quadrants. Also in these tests, "blue" was
always named correctly; the other coloured targets

./Zil.il

-7 ''

A

(C tnom

antid 1)

B

01
03
r 1-0
c 32
E

10

L,, 32

a

100

320

600

200

00

200

400

400

60'

RVF

LVF

Fig 2 Binocular visual field (up to 60 eccentricitv) for twlite l/lit and coloured targets (A ) and profile of
lig/lt sensitil'it3' along tile hloriolontal axi.s and thle diagonial meridians in tile uipper quadrants (B). The visual
field for colours (A ) is restricted in tile left upper quadrant (R: red target, G: green target: B: blue target).
No difference was folund for liglht sensitivityv between tlhe left and rigllt Ileinifield along the hlorizontal axis
(filled circles) and thle 4S°-meridian (R F) and 13i.5-mzeridian (LVF), respectivelY.
Table 2 Colour n1amingtg alotlg the left iorizontal
meridian Correct responmes in percen t (=-30) A:.
100 in s stimnu!u.s duration B: unlimnited duratiotn

Table 4 Colour nam7?ing in tile left upper quaclral-t
Correct re.spon.se.s in percent (nt30) A: /00 inis
stimnulu.s duration B: unzlitmzited durationi

(/lour

Colour

Red

Blue
Green

Yellou

Eccentricit v
2
3

A
B
A
B
A
B
A
B

100

100
100
100

100
100
100
66-7
73 3
100

100
100
100
100
100

100

IS

5

I()

100
100
100
100

73356' 7
73 3
100
100
6 7
10 0
73 3
70 0

23-3
33 .3
100
10()

100
13 3
100
100
0 00
0 00
16 7

60 0
100
100
6-7
6 7
60 0
63 3

20()

Table 3 Colour nam?inlg in thle left lower quadratit
Correct responses in percentit (n=30) A: 100 inu
Stim ulus dulration B. unilimited durationi
Colour

Red
Blue
Green

Yello,A

Eccentricit v
2
3

A
B
A
B
A
B
A
B

100
100
100
100
100
100
100
100

90.0
93 3
100
100
53 3
60 0
100
100

25)

10

5
40 0

50s0
100
100
10 0
10 0

100
100

23 3
26 7
100
100
6.7
13 3
80 0
76 7

Eccentoriitv

20(

0 00

000
100
100
0 00
0 00
60 0
73.3

(green. red and yellow) were mostly called as
"blue," when they were named incorrectly. The
patient was not aware of the consistent

misnaming.
The performance on the Farnsworth-Munsell
100-hue test was diminished in all parts of the
spectrum except for blue (fig 2). The patient had

2

Red
Blue
Cireen

Yellou

A
B
A
B
A
B
A
B

100
100
100
100
100
100
100
100

3

1i

5

90.0

40 0

23'3

867

43*3

26 7

100
100
533
60 0
100
100

100
100
100
10 0

10(
1t(1
6-7
13 3

100

80()

100

76-7

no problems to assign the correct name to the
colours in which words are written (for example.
"blue" when the term "red" was written in blue
letters), or to read the word for a colour written
in an another colour (for example. "green" when
written in yellow or blue letters).
Object naming All presented figures (circle. triangle. square, house. apple. dog, faces) were
identified and named correctly by the patient
irrespective to whether they were presented in the
left or right hemifield.
Letter and word recognition The patient had no
difficulty in recognising letters and words presented
in either half-field.

Discussion
The main findings described in this report are (I)
a partial loss of colour vision in the left upper

Colour anomia restricted to the left visuial hemifield after splenial disconnexion

723

Fig 3 Farnsworth-Munsell 100-hue test (binocular condition). The extent of radial
displacement indicates the degree of disturbed hue discrimnination in that part of the
spectrum. Note the good performance in the blue part of the spectrum.

reduced performance on the
Farnsworth-Munsell 100-hue test except for the
blue part of the spectrum, and (2) a colour anomia
in the left hemifield beyond 20 eccentricity (where
colour vision was preserved). This colour anomia
was not accompanied by an alexia or agnosia for
visual objects.
The partial loss of colour vision in the
Farnsworth-Munsell test is in agreement with the
findings in patients with unilateral brain damage
reported by other authors, see fcr example references8 9. The good performance for the blue part
of the spectrum in the Farnsworth-Munsell test
may be related to the observation that the colour
visual field for blue in the upper left quadrant is
reduced by a lesser extent than the colour fields
for green and red targets. The extent of shrinkage

quadrant and

a

in the colour visual fields in the upper left
quadrants may indicate a lesion involving the connexions from the striate cortex to prestriate cortical areas, and to the inferior occipitotemporal
regions. Zeki'0 11 has shown that cells in Area V4
of the prestriate cortex of the rhesus monkey are
specifically colour-coded. Even though it is not
possible to correlate the damaged region revealed
by computed tomography in our patient, with the
colour-coded areas found by Zeki, one might
speculate about a functional connection of the
lesioned area in our patient with Area V4 as
defined by Zeki. It seems, however, that total
loss of colour vision only occurs after bilateral
lesions in the infero-temporal region in both man4
and monkey (Fries and Zeki, in preparation). In
man, total loss of colour vision was found to be

724

accompanied by visual field defects, and there is a
strikingly high incidence of bilateral defects
affecting the upper quadrants (for a review, see
reference 4). The observation that a partial visual
field shrinkage in the upper left quadrants was
only found for coloured targets suggests a dissociation of colour vision from other visual functions
(for example, light sensitivity, critical flicker
fusion) and supports the idea of a localised representation of colour vision in the visual cortex of
man.
The colour anomia found in the left hemifield
can be considered as a disturbance due to splenial
damage.'2 13 One may assume that the lesion of
the splenium of the corpus callosum disconnects
the areas involved in processing colours of the
right visual cortex from the language areas of the
left hemisphere. Colour anomia has not to be
accompanied by an alexia, or an agnosia for
visual objects, as our case indicates. This may be
taken as evidence that a splenial disconnexion
syndrome can be restricted to only one visual
function (see also reference 14).
It has been shown that the peristriate regions
important for colour coding have their own callosal connections.' 15 16 If we suppose that fibres
carrying colour information from the right prestriate cortex to the left hemisphere via the splenium
of the corpus callosum are damaged, the isolated
disturbance of naming colours in only the left
hemifield would be the consequence of this splenial
disconnexion. Colour anomia, therefore, should
be considered as a single unitary disturbance. It is
not clear, however, why this anomia did not occur
in the complete left hemifield, but only beyond 20
eccentricity. Probably either the foveal region is
represented bilaterally'7 18 or there is an overlap
in the cortical representation of the visual field
across the vertical axis.'9 20 It remains also unclear
why performance in colour naming is better along
the left horizontal meridian as compared with the
upper and lower quadrants, as density of colourcoded cells or colour carrying fibres in relation to
visual field regions (for example, the horizontal
meridian) has not been clarified.
In conclusion the results presented here support
the notion of splenial disconnexion of colour
naming from colour perception. The observations
were obtained by testing the two visual half-fields
for different visual and visual-verbal tasks. Under
normal (that is, foveal) test conditions the presented disconnexion phenomenon would not have
been observed.
This work was supported by Deutsche Forschungsgemeinschaft.

J Zihl and D von Cramon
References

Kinsbourne M, Warrington EK. Observations on
colour agnosia. J Neurol Neurosurg Psychiatry
1964; 27:296-9.
2 Geschwind N, Fusillo M. Color naming defects
in association with alexia. Arch Neurol 1966; 15:
137-46.
3 Oxbury JM, Oxbury SM, Humphrey NK. Varieties
of colour anomia. Brain 1969; 92:847-60.
4 Meadows JC. Disturbed perception of colours
associated with localized cerebral lesions. Brain
1974; 97:615-32.
5 Sloan LL. The Tubinger perimeter of Aulhorn
and Harms. Arch Ophthalmol 1971; 86:612-22.
6 Aulhorn E, Harms H. Visual perimetry. In:
Jameson D, Hurvich LM, eds Handbook of Sensory Physiology VII/4. Berlin: Springer Verlag
1972; 102-45.
7 Farnsworth D. The Farnsworth-Munsell 100-hue
and dichotomous tests for colour vision. J Opt
Soc Am 1943; 33:568-78.
8 Scotti G, Spinnler H. Colour imperception in
unilateral hemisphere-damaged patients. J Neurol
Neurosurg Psychiatry 1970; 33:22-8.
9 Capitani E, Scotti G, Spinnler H. Colour imperception in patients with focal excisions of the
cerebral hemispheres. Neuropsychologia 1978; 16:
491-6.
10 Zeki SM. Colour coding in rhesus monkey
prestriate cortex. Brain Res 1973; 53:422-7.
11 Zeki SM. Colour coding in the superior temporal
sulcus of rhesus monkey visual cortex. Proc R
Soc Lond, B 1977; 197:195-223.
12 Geschwind N. Disconnexion syndromes in animals and man. Part I Brain 1965a; 88:237-94.
13 Geschwind N. Disconnexion syndromes in animals and man. Part II Brain 19651b; 88:585-644.
14 Vincent FM, Sadowsky CH, Saunders RL,
Greeves AG. Alexia without agraphia, hemianopia, or color-naming defect: A disconnexion
syndrome. Neurology 1977; 27:689-91.
15 Zeki SM. Interhemispheric connections of prestriate cortex in monkey. Brain Res 1970; 19:63-75.
16 Rockland KS, Pandya DN. Laminar origins and
terminations of cortical connections of the occipital lobe in the rhesus monkey. Brain Res 1979;
179:3-20.
17 Bunt AH, Minckler DS, Johanson GW. Demonstration of bilateral projection of the central
retina of the monkey with horseradish peroxidase
neuronography. J Comp Neurol 1977; 171:619-30.
18 Huber A. Homonymous hemianopia after occipital lobectomy. A J Ophthalmol 1962; 54:623-9.
19 Cragg BG. The topography of the afferent projections in circumstriate visual cortex of the
monkey studied by the Nauta method. Vision
Res 1969; 9:733-47.
20 Zeki SM. Representation of central visual fields
in prestriate cortex of monkey. Brain Res 1969;
14:271-91.
1