Brain (1993), 116, 1303-1335

Impaired visual motion perception in the
contralateral hemifield following unilateral
posterior cerebral lesions in humans
Gordon T. Plant,1-4 Kenneth D. Laxer,2 Nicholas M. Barbaro,2
Jade S. Schiffman3 and Ken Nakayama1*
The lSmith Kettlewell Eye Research Institute, the 2Northern California Comprehensive Epilepsy Centre
and the Department of Neurological Surgery, University of California, the 3Letterman Army Medical
Centre, San Francisco, USA and the ''National Hospital for Neurology and Neurosurgery, London, UK

Contrast thresholds for a number of tasks were measured in the contralateral and ipsilateral
upper quadrants of the visual field (eccentricity = 10°) before and after an occipito-parietal
surgical resection, in one patient, carried out for intractable epilepsy. Postoperatively
the contrast thresholds for discriminating the speed of movement of drifting sine-wave
gratings were elevated by greater than a log unit in the contralateral field with little or
no change in the detection thresholds for the same stimuli. Contrast thresholds for opposite
direction-of-motion (DOM) discrimination of a contrast modulated (CMod) grating (a
'non-Fourier' motion stimulus) were also elevated by about a log unit in the contralateral
hemifield but thresholds for DOM discrimination of a sine-wave (luminance modulated,
LMod) grating were unaffected. Contrast thresholds for orientation discrimination of
stationary gratings (a non-motion task) were unaffected. This general pattern of results
was found in two other patients following lateral occipital surgical resections. Eight other
patients with occipito-temporal (two cases), parietal (three cases) and medial occipital
lobe lesions (three cases) showed no difference between the two hemifields on any of
the tasks. Comparison of die location of the lesions leads to the conclusion that damage
to the lateral occipital gyri is responsible for the pattern of visual deficit observed. Damage
to an extra-striate visual area concerned with motion perception (the human homologue
of primate V5-MT) may have occurred. There has been no previous description of impairment of motion perception localized to a hemifield in humans. The characteristics of the
residual motion perception in these cases is described further in the accompanying article
[Plant and Nakayama (1993), Brain, 116, 1337-1353].

INTRODUCTION

Evidence from primate anatomy and physiology suggests that projections from VI (striate
cortex) and from extra-striate visual area V2 to a cortical area designated die middle
•Present address: Department of Psychology, Harvard University, Cambridge, MA 02138, USA.
Correspondence to: Dr Gordon Plant, The National Hospital for Neurology and Neurosurgery, Queen Square, London,
WCIN 3BG, UK.

© Oxford University Press 1993

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

SUMMARY

1304

G. T. PLANT AND OTHERS

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

temporal area (MT) (Allman and Kaas, 1971) or V5 (Zeki, 1974) (referred to as V5-MT
in this article) subserve a special role in the processing of visual motion. Also involved
are the adjacent areas known as the medial superior temporal area (MST) (Allman and
Kaas, 1974a; Desimone and Ungerleider, 1986) and the visual areas of the parietal lobe
to which V5-MT and MST project. The physiological properties of neurons encountered
in these areas are compatible with V5-MT having an important role in visual motion
processing: neurons in V5-MT encode direction of motion (Dubner and Zeki, 1971; Zeki,
1974), and probably speed and velocity (e.g. Maunsell and Van Essen, 1983; Rodman
and Albright, 1987). Maunsell and Newsome (1987) and Zeki (1990a) have provided
recent reviews of this evidence. The location of an area in human cortex which may be
the analogue of primate V5-MT was first identified on the basis of myeloarchitectonics
(Clarke and Miklossy, 1990) and evidence of a similar functional specialization in human
extra-striate cortex has been obtained in human PET studies (Corbetta et al., 1990; Zeki
etal., 1991).
Examples of damage to the posterior cerebral hemispheres in humans giving rise to
selective impairment of motion perception are rare (see Zeki, 1991, for a recent review).
The most extensively studied case was first reported by Zihl et al. (1983) and has been
the subject of further investigation (Hess et al., 1989; McLeod et al., 1989; Baker et al.,
1991; Zihl et al., 1991). Vaina et al. (1990) have reported a case of bilateral parietooccipital damage which resulted in impaired spatial localization and stereopsis who also
showed impaired performance on a speed discrimination task. This case did not report
the marked symptomatic perceptual difficulties experienced by the case described by Zihl
et al. (1983). Both of these cases had sustained bilateral lesions and the anomaly was
present throughout the visual field.
The purpose of our study was to better specify disorders of motion perception in humans.
Specifically we have exploited the observation that, in primates, V5-MT is a relatively
small visual area which contains a complete representation of the contralateral hemifield
(Allman and Kaas, 1971; Gattass and Gross, 1981; Van Essen et al., 1981; Desimone
and Ungerleider, 1986; Fiorani et al., 1989). Other extra-striate areas differ in this respect:
in the case of the dorsal and ventral portions of V2 and V3, representations are found
of the two contralateral quadrants of the visual field which are largely separate in cortex
(Allman and Kaas, 1974ft; Gattass et al., 1981) or in the case of V4 the contralateral
visual field is represented in a much larger strip of pre-striate visual cortex (Gattass et al.,
1988; see Zeki, 19906, for a recent review). It therefore seemed plausible that impaired
motion perception in a hemifield may occur in humans following discrete damage to a
homologue of V5-MT or its afferent or efferent projections, provided that the pathway
is interrupted at or before the extensive transfer of information to the opposite hemisphere
through callosal connections. Such a lesion could occur more commonly than bilateral
lesions responsible for impairment of motion perception and the ability to use the ipsilateral
field as a control is a considerable experimental advantage.
We describe three patients with impaired speed discrimination restricted to a hemifield.
The lesions responsible were surgical resections for (i) a medically refractory seizure
disorder; (ii) a tumour; (iii) an occipital haematoma. In two we therefore had the unusual
opportunity (in human studies) to examine them before and after the resections. Apart
from the fact that the anomaly is localized to a hemifield there are other important
differences from the perceptual deficit experienced by the case described by Zihl et al.

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1305

PATIENTS AND METHODS
Eleven cases of unilateral posterior hemisphere damage were studied. The clinical details of the patients
and the results of perimetry are summarized in Fig. 1. Case 1 was the first patient to be studied and the
abnormalities found determined the experimental protocol used subsequently. Informed consent was obtained
from all the patients in accordance with the Declaration of Helsinki.
Cases 1 and 2 were tested before and at various time intervals following surgery. The other patients, who
had established lesions when first seen, were tested at least 3 months following the acute event and in two
sessions of 5—6 h each, separated by an interval of 2 - 3 days. The stimulus was masked to provide a circular
aperture subtending an angle of 12° at a viewing distance of 114 cm. The aperture was centred at an eccentricity
of either 10° or 30° of visual angle on the 45° oblique meridians in the quadrant of the visual field to be
tested. Viewing was binocular. A chin and headrest were used to minimize head movements and the display
itself was positioned directly in front of the subject at eye level. A fixation mark was positioned on a square
white screen surrounding the display which measured 100 x 100 cm and the subject was instructed to fixate
it by an eye movement and not by a head movement. This surround was illuminated to give approximately
the same mean luminance and hue as the oscilloscope screen; no light fell directly on the oscilloscope display.
During data collection the experimenter was positioned behind the screen so that he could observe the subject
and reject responses to stimuli which may have coincided with an eye movement.
Visual stimuli
The visual stimuli employed in this study were sine-wave gratings generated digitally and displayed on an
oscilloscope with a white phosphor (P4; Joyce screen) having a mean luminance of 200 cd/m 2 and a frame
rate of 100 Hz (1024 lines displayed on each frame). The contrast response of the oscilloscope was linear
up to 98 % contrast. In all the experiments employing a sine-wave stimulus the spatial frequency of the grating
was 0.5 cycles/degree and the orientation vertical (except where this was varied to set thresholds for orientation
discrimination). One digital-to-analogue convenor (DAQ generated the sine wave, a second DAC controlled
the contrast across the screen by a vertical one-dimensional raised cosine trapezoidal spatial window. In the
case of the CMod gratings there was no spatial window and the first DAC was used to generate the carrier
signal and the second to generate the modulating signal. The overall contrast of the pattern was controlled
by a third DAC and the pattern was ramped on and off by controlling the contrast of the pattern by a raised

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

(1983) {see the accompanying article, Plant and Nakayama, 1993). In our investigation
we have measured contrast thresholds for five basic visual parameters. These tasks have
been selected to study the effects of the cerebral damage on basic and higher visual functions
subserving motion perception, (i) Contrast detection in order to determine the integrity
of mechanisms subserving the detection of the stimuli used, (ii) Direction-of-motion (DOM)
discrimination of a luminance modulated (LMod) sine-wave grating in order to determine
the integrity of mechanisms subserving the basic motion task of direction discrimination
of a 'first order' stimulus, i.e. one in which information about motion can be derived
from the contrast of the signal, (iii) Direction-of-motion discrimination of a contrast
modulated (CMod) grating. This stimulus is 'second order', i.e. information about the
motion of the stimulus cannot be obtained without a non-linear transformation. Unlike
the bright and dark bars of the LMod grating, in this stimulus the motion is of regions
of high and low contrast which have the same average luminance. Detection of motion
in this stimulus may require additional processing in the brain in comparison with stimulus
(ii). (iv) Speed discrimination (which is also likely to require additional processing in
the brain), (v) Orientation discrimination (a non-motion task which employs the same
stimulus as the other tasks).
Some of the findings have been published in abstract form (Plant et al., 1990; Plant
and Nakayama, 1991).

1306

G. T. P L A N T AND OTHERS

LATERAL OCCIPITAL LOBE LESIONS

CASE 1
EXTENSIVE LEFT OCCIPITO-PARIETAL
SURGICAL RESECTION FOR INTRACTABLE
FOCAL ERLEPSY

CASE 2
LEFT LATERAL OCCIPITAL RESECTION
FORTUMOUR

LEFT PARTIAL OCCIPITAL LOBECTOMY
FOR HAEMORRHAGE

MEDIAL OCCIPITAL LOBE LESIONS
L

R

CASE 4
LEFT MEDIAL OCCIPITAL
INFARCTION

CASE 5

RIGHT MEDIAL OCCIPITAL
INFARCTION

CASE 6

LEFT VENTRO MEDIAL
SURGICAL RESECTION
OFTUBERCULOMA

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

CASE 3

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1307

PARIETAL LOBE LESIONS

CASE 7
RIGHT PARIETAL VASCULAR
MALFORMATION WITH
ASSOCIATED HAEMORRHAGE

CASE 8

CASE 9
RIGHT PARIETAL LOBE
RESECTION FOR TUMOUR

OCCIPITO-TEMPORAL LESIONS

CASE 10
RIGHT OCCIPITO-TEMPORAL
SURGICAL RESECTION
FOR TUMOUR

CASE 11
LEFT OCCIPITO-TEMPORAL
SURGICAL RESECTION
FORTUMOUR

Fio. 1. The visual field defects and diagnoses are shown for the 11 patients who participated in this study. The field defect
indicates an absolute field defect either to Me on the GoWmann perimeter OT to a target with an incremental contrast of
99% on the Humphrey automated perimeter. For each case the location of the test stimulus in the contralateral visual
field is shown, the results obtained at that location have been compared with those obtained at the mirror symmetric location
in the ipsikteral hemifield.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

LEFT OCCIPITO-PARIETAL
RESECTION FOR VASCULAR
MALFORMATION

1308

G. T. PLANT AND OTHERS

cosine trapezoidal function of time. These waveforms are shown schematically in Fig. 2. Stationary gratings
were in spatial cosine phase; where temporal modulation was used, the spatial phase of the grating was changed
at the start of each frame. In the case of the CMod grating the contrast of a 0.5 cycles/degree grating (the
carrier, six cycles per 12° stimulus field in cosine phase) was modulated in space across the screen by a
0.083 cycles/degree cosine (die modulating signal, one cycle per stimulus field). This is trigonometrically
equivalent to summing two cosine gratings of 0.417 cycles/degree and 0.583 cycles/degree each of half the
contrast of the CMod grating and gives rise to an apparent 'beat' frequency of 0.17 cycles/degree (two cycles
per stimulus field; see Fig. 2).
In preliminary experiments it was found that DOM discrimination was possible in control subjects at the
eccentricity used when the modulating grating was drifted at a temporal frequency of 0.3 Hz. A very low
spatial frequency of the modulating signal was necessary, otherwise DOM discrimination was not possible
at this eccentricity. In this stimulus the carrier grating is temporally modulated in counterphase when the
modulating grating is drifted and hence gives an equal rightward and leftward DOM signal. Therefore the
subject must judge DOM from the direction of drift of the modulating waveform or 'beat' frequency. The
contrast of the complex waveform was controlled in the same way as the sine-wave gratings by means of
the diird DAC and thus the contrast of the carrier and of the modulating grating were always equal.

Spatial waveform - 0.5 cycles/degree

Distance
Spatial waveform - contrast modulated grating

Distance
Temporal waveform - 8 Hz modulation

Time
FIG. 2. The visual stimuli employed in the study are illustrated. The sine-wave luminance modulated (LMod) grating was
a 0.5 cycles/degree vertical grating modulated in space by a vertical raised cosine trapezoidal window (A). The stimulus
was presented in a circular aperture centred eccentrically in the visual field, usually at 10" from fixation. The contrast
modulated (CMod) grating or 'beat' stimulus was a 0.5 cycles/degree cosine grating modulated in space by a 0.08 cycles/degree
cosine-wave giving rise to an apparent 'beat' frequency of 0.17 cycles/degree (2 cycles per stimulus field) (B). When
the modulating grating is drifted horizontally the 0.5 cycles/degree grating is modulated in counterphase and so information
as to the direction of drift is available only from the motion of the envelope of the waveform. The stimulus presentation
time was usually 800 ms. The stimulus was ramped on and off slowly by a raised cosine temporal modulation (c) this
is similar to the spatial modulating function shown in (A).

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Specific test procedures
Contrast detection thresholds (LMod and CMod gratings). A temporal two-alternative forced-choice staircase
procedure was employed to measure contrast thresholds for stimulus detection. The stimulus was randomly
assigned by the computer to one of two temporal intervals, the other interval being blank, and it was the
subject's task to identify the interval in which the grating appeared. The subject's responses were in the
form of button presses which were monitored by the computer. In these experiments a correct response resulted

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1309

Contrast thresholds for speed discrimination. Contrast thresholds for speed discrimination were estimated
in an identical staircase to the detection thresholds, except that a pair of gratings differing in temporal rate
of drift were presented in the two intervals and the subject reported the interval in which the grating drifting
at the fester speed had appeared. The spatial frequency was 0.5 cycles/degree and the other stimulus parameters
were identical to those employed in the threshold detection estimations. Direction of drift was randomly
assigned by the computer and the stimulus duration was also chosen randomly from four preset values in
the range 360—420 ms to minimize the possibility that the subject might utilize distance and duration cues
in the speed discrimination task. Contrast thresholds for two temporal frequency pairs were determined:
8 and 10 Hz (i.e. velocities of 16°/s and 2 0 7 s , representing a Weber fraction of 25%) and 4 and 8 Hz
(velocities of 87s and 167s representing a Weber fraction of 100%). The two staircases were interleaved.
The contrast detection thresholds had been previously determined for the three temporal frequencies employed
in the two staircases and a pair of stimuli were always presented at the same contrast ratio as their respective
detection thresholds in order to minimize any contrast cues in the discrimination task. Furthermore, as the
contrast of each of the paired stimuli changed with every trial it is unlikely that the subjects were able to
utilize contrast cues. The contrast threshold measured and used in the subsequent data analysis was in both
staircases the contrast of the 8 Hz stimulus. The staircase procedure used to measure the contrast thresholds
for speed discrimination was in all other respects identical to that employed in the threshold detection
experiments described above.
Direction-of-motion discrimination (LMod and CMod gratings) and orientation discrimination. A single stimulus
forced-choice procedure was used for DOM and for orientation discrimination. The drifting grating was
shown in a single interval: the direction of drift was randomly assigned and the subject reported the direction
of motion. The stimulus was either a 0.5 cycles/degree LMod sine-wave grating drifting at 8 Hz or the CMod
grating described above drifting at 0.3 Hz. The two staircases were interleaved. The contrast thresholds
for direction discrimination were determined in an otherwise identical staircase procedure to that used to
determine detection thresholds. In the case of orientation discrimination in the single interval a grating was
presented which was oriented either plus or minus 3° from vertical and the subject reported the direction
of tilt. The contrast threshold was determined in the same manner as for DOM discrimination.
Cortical mapping
One patient (Case 1), as a standard procedure in the assessment of his epilepsy, underwent subdural implantation
of an 8 x 8 array of stimulating and recording electrodes (stainless steel discs; 0.5 cm diameter, centre to
centre separation 1 cm). A standard stimulation protocol was followed: 60 Hz, 0.2 ms width square pulse
trains of 1 s duration. Stimulus intensity was increased until an effect occurred.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

in the next stimulus or stimulus pair being shown at a 20% lower contrast level and an incorrect response
(a reversal) resulted in the contrast being increased by 100% on the next presentation of the stimulus. A
brief preliminary test run was carried out to establish the approximate range of the thresholds to be measured
and to give the subject practice at the task. Only in this preliminary run was feedback given to the subject
(audible tones). The staircase proper was then commenced at a contrast level a factor of 10 higher than the
predetermined approximate threshold. More than 40 observations were made in each staircase which was
continued until at least seven reversals had occurred: the arithmetic mean of the contrast values at all the
reversals except the first two (which were disregarded) was then computed and taken as the threshold for
each stimulus. Each staircase was repeated at least once and the contrast sensitivity values presented below
are the reciprocals of the means of two or three such values. Each observation interval was preceded by
an audible tone and lasted 1 s: within that interval the stimulus duration was 800 ms with the maximum
contrast being attained for 400 ms (the plateau of the temporal window). Contrast thresholds were measured
for stationary and drifting sine-wave gratings of 0.5 cycles/degree. The drifting gratings were modulated
at one of three temporal frequencies, 4, 8 and 10 Hz; the direction of drift being randomly assigned by
the computer. In the case of the CMod grating the procedure was identical, except that the stimulus was
the complex waveform described above drifting at 0.3 Hz in a randomly assigned direction. Following the
two presentation intervals (total duration 2 s, giving an interstimulus interval of 400 ms) a third tone signalled
the subject to respond. Unlimited response time was given during which the subject was encouraged to rest
periodically. The next trial was initiated by the subject's response. These different stimuli were randomly
interleaved in the staircase and the paired visual field locations were tested alternately.

1310

G. T. PLANT AND OTHERS

Neuro-imaging
All of the patients included in the study had had neuro-imaging studies as a part of their standard clinical
work-up and these were MRI studies in all patients except Case 8 in whom X-ray CT images were available.
The MRI studies described were carried out on a 1.5 tesla imager. Both T,-weighted images with and without
gadolinium-DTPA enhancement and T2-weighted images were obtained in coronal, axial and parasagittal
planes.
The authors consider that detailed case histories of the three cases showing abnormal motion perception
are desirable to establish the anatomical features underlying the visual abnormalities in these cases, so that
comparisons can be made with other cases. These case histories are given in the Results.
RESULTS

The first subject to be examined in this study is Case 1 in whom contrast thresholds were
measured before and after surgery. The results obtained in this patient established the
experimental protocol used in the remainder of the study.
Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Case 1: history
Case 1 was a 25-year-old male. Seizures were observed at age 7 years when he was
described as having 'staring spells' lasting 1 —2 min. At age 13-14 years there was an
alteration in the character of his seizures which became brief tonic/clonic seizures preceded
by an aura of 'blurred vision' and a lightheaded feeling. This remained the most common
seizure type but he reported that on some occasions he experienced aright-sidedor bilateral
sensory aura. On other occasions he had observed a colourless 'fog' moving from the
periphery of the right visual field and stopping abruptly at the midline. From his simulation
in the form of a hand gesture this fog appeared to be travelling at ~40°/s. Objects in
his visual field were never seen in motion and he was uncertain as to the extent that objects
remained visible as the 'fog' passed across the right hemifield.
Clinical neurological examination was normal except for an uncertain right plantar
response. Visual acuity was 20/20 bilaterally and performance on the Farns worth—Munsell
100 Hue test and Goldmann perimetry were within normal limits. There was no evidence
of parietal or pyramidal dysfunction and no hemiatrophy. A preoperative MRI scan (Fig. 3)
revealed focal atrophy of the left posterior parietal convexity with an enlarged superficial
vessel and appearances suggesting a left parietal heterotopy (Fig. 3, arrows). The remainder
of the brain appeared normal with the exception of mild cerebral atrophy. Left temporal
spike and sharp wave discharges were recorded on scalp EEG. Scalp telemetry recordings
demonstrated the seizure onsets to occur from the left posterior quadrant. Concern over
the risk to normal function located in this region resulted in a decision to improve the
localization of the seizure onsets, as well as perform cortical mapping by means of a
subdurally placed multi-contact grid.
A left fronto-parieto-occipital craniotomy was carried out (Fig. 4) with the inferoposterior limit of the craniotomy 1 cm from the occipital pole (Fig. 4, upper panel,
asterisk). The enlarged superficial vein noted on MRI is indicated in Fig. 4 (arrow). The
gyral anatomy was distorted. Electrocorticography revealed continuous epileptic activity
in the occipito-parietal region with phase reversal at points labelled 'A' and 'B' in Fig. 4.
Motor responses were observed following stimulation of points 1, 2, 3, 5 and 8 and this
area represents the motor strip. It therefore seems that the large anterior vein is situated
in an equivalent location to the posterior ramus of the lateral sulcus in a normal brain.

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1311

The subdural electrode grid was positioned (Fig. 4, lower panel) and the dura and skull
flap closed. The grid is shown diagramatically in Fig. 5 with the asterisk again marking
the position of the occipital pole and the position of the large vein indicated.
The patient was allowed to recover from his craniotomy for 2 days before electrical
stimulation was begun. During the period that the grid was in place both EEG and timesynchronized video monitoring were carried out. On electrical stimulation of electrodes
marked 'M', movements of right wrist, fingers, forearm or right corner of his mouth
were observed. On stimulation of electrodes marked 'S', a variety of somatosensory
sensations were reported, some involving contralateral limbs, some bilateral. 'A' indicates
a single auditory phenomenon. Unmarked electrodes produced no visible or subjective
change when stimulated.
Visual phenomena occurred on stimulation of electrodes marked 'VB' when he reported

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

FIG. 3. T2-weignted axial MRIs of Case 1. In the preoperative images (upper row) the left parietal developmental anomaly
is seen (arrows) and in the lower cuts asymmetry of the occipital horns of the lateral ventricles with cortex of normal
appearance. Postoperatively the extensive occipho-parietal resection is seen (arrows) together with tbe increased dilatation
of the left lateral ventricle {lower row). See also Fig. 12.

1312

G. T. PLANT AND OTHERS
Dorsal

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Ventral
FIG. 4. The upper photograph shows me left occipito-parietal region of the brain of Case 1 exposed at craniotomy. Anterior
is on the left arid ventral is at the bottom of this and all subsequent operative photographs. The numbers 1-8 (inverted
because the surgeon stood at the patient's vertex) mark regions which on stimulation generated motor responses in the
contralateral face or arm. Other symbols mark locations at which visual sensations were generated. 'A' and 'B' are points
where there is phase reversal. The large vein (arrow) continues anteriorly along the Sylvian fissure. The dashed line shows
the limits of the resection which was subsequently carried out. The occipital pole was measured to be 1 cm from the
posterior limit of the craniotomy (asterisk) .The lower photograph shows the 8 X 8 array of electrodes implanted subdurally
for stimulation and recording before the resection was carried out. The centre to centre separation of the electrodes is
1 cm. A schematic diagram showing the results of stimulation is shown in Fig. 5.

I M P A I R E D MOTION PERCEPTION AFTER O C C I P I T A L DAMAGE

s
m

KEY:

•

a

•

vm

•

vc
vh

••
••

vb

•

A

somatosentory
motor
auditory
visual - motion
visual - colour
visual - homianopic blurring
visual - blurring in both hemifields

FIG. 5. Diagram of the subdural electrode grid. The position of the occipital pole is shown by the asterisk and the position
of the large cortical vein is shown diagramatically at top left. Somatosensory responses, often mimicking symptoms occurring
at the onset of seizures, were abnormally widely distributed throughout the occipito-parietal cortex. Electrodes at which
stimulation gave rise to visual and motor phenomena are also indicated. The arrows denote the direction of motion of
the evoked percept in the contralateral visual field. The cortical region which, on stimulation, evoked moving phosphenes
in the contralateral field is located from 30 to 50 mm from the occipital pole.

non-specific visual blurring, ' V C where he reported the appearance of a coloured spot
in the contralateral field and 'VM' where he reported motion of a colourless fog. The
latter experience was the same as the subject's aura mentioned above. In the case of the
moving fog, and as was the case with his seizures, the patient reported that there was
no motion of objects seen within the visual hemifield affected. Specific descriptions of
the results of stimulation are given in the Appendix. In view of the difficulty in locating
a precise epileptogenic focus, and having established that stimulation of none of the locations
stimulated interfered with language functions it was decided to proceed with resection
of all the cortex which, when stimulated, had reproduced symptoms associated with

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

B

H

1313

1314

G. T. PLANT AND OTHERS

seizures. The extent of this resection is shown by the dashed lines in Fig. 4 and it included
lateral occipital and parietal cortex. In the postero-superior aspect of the resection there
was found to be an abnormal area of cortical material beneath the most superficial grey
matter. Pathological examination of the specimen was consistent with a migration defect.
Postoperative MRIs are shown in Figs 3 and 12.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Case 1: experimental results
Post-operatively visual acuity remained 20/20 bilaterally. Goldmann perimetry was normal
before surgery including isopters to coloured targets. The field defect 28 days
postoperatively (to the I4e target) is shown in Fig. 1. It had been anticipated that a resection
in this location was likely to damage the upper fibres of the underlying optic radiations
and the field defect shown is typical of a radiation lesion: it is wedge shaped and shows
minor incongruity. Comparison with larger targets revealed a 'sloping' lateral border
to the field defect. There was slight depression of function in the upper quadrant also
immediately postoperatively which subsequently recovered. Isopters to the identification
of coloured targets were also preserved in the upper quadrants. Although he has continued
to have seizures following the resection he has never again experienced the visual aura
of motion in the right hemifield.
In view of the anticipated field defect, ah* testing was carried out in the upper quadrants
with the stimulus centred at an eccentricity of 10°. Before surgery contrast sensitivity
was determined for three tasks: detection and DOM discrimination of a 0.5 cycles/degree
grating drifting at 8 Hz and discrimination of an 8 and 10 Hz grating pair. There was
no difference between the results obtained at an eccentricity of 10° in either the left or
right (ipsilateral or contralateral) upper quadrants (Fig. 6). In the immediate postoperative
period these tests were repeated and it was found that there was a twentyfold rise in the
contrast threshold for the speed discrimination task (open triangles in Fig. 6) and there
was a slight increase in the threshold for DOM discrimination (by a factor of two) closed
squares in Fig. 6). In the ipsilateral quadrant there was no change in the thresholds after
surgery. In view of the marked impairment of speed discrimination for the 20% speed
difference found postoperatively it was decided also to test, using 4 and 8 Hz as the panto be discriminated (closed triangles in Fig. 6). The contrast threshold for this task was
also elevated, being a factor of 20 higher in the contralateral than in the ipsilateral hemifield.
These measurements were repeated at intervals over the following 7 months. The detection
and DOM discrimination thresholds returned to normal. There was a substantial improvement in the threshold for discriminating 8 and 4 Hz but little change in the threshold for
the more difficult speed discrimination task.
At 28 days postoperative, tests were carried out in addition to those plotted in big. 6.
The purpose of these tests was to include a second DOM task using the CMod grating
stimulus described in Patients and methods. It was also decided to include a task which
did not depend upon intact motion sensitive mechanisms and the orientation discrimination
task was chosen for this. These results and the 28 day results in Fig. 6 are replotted in
Fig. 7 {upper panel), not as absolute contrast sensitivity values but as ratios of the
discrimination to the detection contrast thresholds. The paired bars give the results for
the ipsilateral field (solid bars) and the contralateral field (shaded bars). The small difference
in the ratios for sine-wave DOM discrimination (Fig. 7, task 1) and the lack of any
difference in the ratios for orientation discrimination (Fig. 7, task 5) should be compared

Ipsilateral field
Contralateral field

-10 f
Operation

1000i

Days

190

FIG. 6. Contrast sensitivity is shown for four tasks tested at an eccentricity of 10° (SF = 0.5 cycles/degree; drift frequency = 8 Hz) in the left (ipsilateral) and right (contralateral)
upper quadrants of Case 1. In the ipsilateral quadrant there was no change in his contrast sensitivity for detection (closed circles), direction discrimination (closed squares)
or speed discrimination (triangles: closed = 4 and 8 Hz; open = 8 and 10 Hz) following surgery. In the contralateral quadrant there was a small fall in sensitivity in the detection
and direction discrimination tasks which subsequently recovered but there was a tenfold fall in sensitivity in his ability to discriminate two stimuli drifting at 8 and 10 Hz,
respectively. Post-operatively we also examined his ability to discriminate 4 and 8 Hz and this was initially impaired but recovered (see also Fig. 7). Data in this and all subsequent
figures are the means of at least three threshold estimates each of which was obtained from a 40 trial staircase.

-10 4
Operation

1000

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016
tn

a

r
o
2

n
n

O

m

H

z

O

to

o

30

m

•0

2
o
H
o
z

tn
O

53

S
>

1316

G. T. PLANT AND OTHERS

-Case 1

1000T

•

Ipsilateral field

VX Contralateral field

Fie. 7. These results are detection and discrimination threshold ratios for five tasks: direction discrimination of a sine
wave and of a contrast-modulated grating (beat); speed discrimination of two gratings drifting at 4 and 8 Hz and 8 and
10 Hz; and orientation discrimination of a stationary grating. The results shown are obtained at an eccentricity of 10°
in the ipsilateral and contralateral upper quadrants of Cases 1 and 2 and at an eccentricity of 20° in Case 3. In all cases
the threshold ratios do not differ for the orientation discrimination and direction discrimination of sine-wave tasks. The
ratios are greater (indicating a relative impairment in the ability to discriminate the quantity as opposed to detect the presence
of the stimulus) in the contralatera] field for the direction discrimination of the beat and the speed discrimination tasks.
The results shown for Case 1 were collected 28 days postoperatively, subsequently there was an improvement in the ability
to discriminate 4 and 8 Hz (see Fig. 6) by which time the deficit more closely resembled that shown by Case 2.

with the elevated discrimination to detection ratios for the other three conditions. Note
from Fig. 5 that there was subsequent improvement in the thresholds for 4 and 8 Hz
discrimination. The data in Fig. 7 are shown together because they were all obtained
at the same test session (28 days).
It is of interest to note that Case 1 has not noticed any gross symptomatic consequences
of the perceptual deficit that we have quantified here. When asked to view the same moving

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

ISine wavel I Beat I |4and8Hz||8and 10Hz| IQiientationi
Direction discrimination Speed discrimination
Task

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1317

stimulus in right and left hemifields he has reported only that the object moving in the
right (affected) hemifield appears to be moving more slowly than when viewed in the
unaffected hemifield. From a functional point of view he has not noticed any problem
except when playing racketball: he has noticed that, if a ball passes from left to right
across his visual field in a more or less tangential trajectory, it appears to slow down
when it enters his right visual field and his subsequent forehand racket stroke (he plays
racketball right-handed) is always inaccurate. For other shots, when he is able to track
the ball or he is playing backhand shots he has noticed no difference in his performance.

Case 2: experimental results
Postoperatively visual acuity remained 20/20 in each eye, there was no change in the
Goldmann fields (Fig. 1) including isopters to coloured targets. When this patient was
seen the results obtained in Case 1 were known and therefore the entire range of tests
was carried out both before and after surgery in both upper quadrants. Figure 10 {upper
panel) shows the pre- and postoperative results obtained in Case 2 for detection and DOM
discrimination of a 0.5 cycles/degree grating drifting at 8 Hz presented at 10° eccentricity
in the left and right upper quadrants of the visual field. Testing was carried out on the
fourth and 28th days following surgery. The resection did not affect either threshold in
either quadrant (closed circles and closed squares). Also plotted are the thresholds for
speed discrimination of the 4 and 8 Hz pair (closed triangles) and the 8 and 10 Hz pair
(open triangles), there was a tenfold elevation in the contrast threshold for discrimination
of the 8 and 10 Hz pair in the contralateral field only.
The lower panel of Fig. 10 shows the results obtained in the orientation experiment

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Case 2: history
Case 2 was a 20-year-old woman who developed minor seizures at the age of 7 years.
At the onset of her seizures her vision is blurred and observors have told her that her
eyes deviate to the right. She may occasionally clench her teeth but is able to comprehend,
talk and follow commands. The episodes last for 30—45 s and had occurred once or
twice a month. On examination there were no abnormal neurological findings and
Goldmann visual fields were normal. Scalp EEG revealed non-specific abnormalities only.
Magnetic resonance images demonstrated a ventral occipito-temporal lesion that did not
enhance with gadolinium-DTPA (Fig. 8). Electrocorticography following a left temporooccipital craniotomy under general anaesthesia revealed occasional spikes, sharp waves
and spike and wave discharges from the inferior lateral occipito-temporal surface. No
electrical stimulation was carried out. On inspection of the brain a pale lesion could be
seen at the base of the temporo-occipitalregionjust behind a large temporo-parietal bridging
vein (Fig. 9). As the bulk of the lesion arose from the ventral temporo-occipital region
it was necessary to remove some cortex of normal appearance in the lateral temporooccipital region. The extent of the removal is also shown in Fig. 9 and postoperative
MRIs are shown in Fig. 8, the location of the normal-appearing lateral occipital cortex
which was damaged is shown in Fig. 12 {upperpanels, arrow). Note that the vein draining
into the transverse sinus was preserved. The superior and medial aspects of the lesion
blended imperceptibly with the surrounding tissue and a complete removal was not
attempted in order to minimize damage to the optic radiations. Histological examination
revealed a dysembryoplastic neuroepithelial tumour.

1318

G. T. PLANT AND OTHERS

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

FIG. 8. Axial Tj-weighted MRIs of Case 2 are shown before (upper pair) and after (lower pair) surgery. The ventral
occipito-temporal mass is shown together with the normal appearing lateral occipital cortex which was damaged at operation
(arrow). See also Figs 9 and 12.

1319

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE
Dorsal

a
to

i.
Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Ventral
FKJ. 9. The upper photograph shows the appearance of the brain at the left lateral temporo-occipital craniotomy in Case
2. Anterior is on the left and ventral is at the bottom of the photograph. The inverted letter 'A' marks the site at which
spike discharges were recorded. The asterisk marks the position of the occipital pole. Because the tumour was largely
ventro-medial it was necessary to remove some normal appearing cortex in one of the lateral occipital gyri as indicated
by the surgeon in die lower photograph. The damaged region was 3 —4 cm lateral to the occipital pole and 2 cm above
the tentorium.

Operation

t

5
Days

Ipsilateral field

15

25

1
-15

10

100

1000

1-15

10

100

J
•

-5

-5

Operation

Days

ft

Contralateral field

15

15

25

25

Fie. 10. Upper panel. Contrast sensitivity measurements (eccentricity = 10°; SF = 0.5 cycles/degree; drift rate = 8 Hz) for four tasks are plotted against time for Case 2.
Postoperatively the only change was an elevation in the threshold for speed discrimination of 8 and 10 Hz drift rates in the hemifield contralateral to the surgical resection.
Closed circles = detection; closed squares = direction discrimination; triangles = speed discrimination (closed = 4 and 8 Hz; open = 8 and 10 Hz). Lower panel. Contrast
sensitivity measurements (eccentricity = 10°) for four tasks are plotted against time for Case 2. Postoperatively the only change was an elevation in the threshold for direction
discrimination of the contrast modulated grating (beat) in the contralateral hemifield. There was no change in the threshold for direction discrimination of a sine-wave grating
(upperpanels). Closed circles = detection (beat); open circles = direction discrimination (beat); closed triangles = dectection (0 Hz drift); open triangles •= orientation discrimination
(3°, 0 Hz).

-15

1000

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016
on

a
m
7)

H

o

•z
H

r*
>

o

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1321

and the DOM discrimination of the CMod grating. Results were obtained preoperatively
and on the 10th postoperative day. As with Case 1, there was no change in the thresholds
for detection or for orientation discrimination in the case of the stationary grating, and
no change in the detection threshold for the CMod grating. There was, however, a tenfold
elevation in the contrast threshold for discrimination of the direction of drift of the 'beat'
or CMod grating.
The results obtained on postoperative days 10 and 28 have been replotted as discrimination : detection contrast threshold ratios in the centre panel of Fig. 7. Unlike Case 1 there
was no threshold elevation for the 4 and 8 Hz discrimination task. In this respect the
pattern of the deficit resembles that obtained in Case 1 several months after the resection
(see Fig. 6).

Case 3: experimental results
In view of the field defect which involved the central portion of the upper field, visual
function was tested at an eccentricity of 30° in the upper quadrant and the diameter of
the target was 24°. Unlike Cases 1 and 2, and in accord with the visual field changes
and likely involvement of the optic radiations, there was a modest elevation of the detection
threshold for the 8 Hz drifting grating. The contrast sensitivity to this stimulus was 220
in the ipsilateral quadrant and 128 in the contralateral quadrant. However, as can be seen
from the discrimination : detection ratios plotted in the lower panel of Fig. 7, there was
a disproportionate elevation in the contrast thresholds for DOM discrimination of the
CMod grating (beat) and for the two speed discrimination tasks. There was therefore
a similar pattern of loss to that shown in Cases 1 and 2.
Cases 4, 5 and 6: lesions principally affecting striate cortex
Magnetic resonance images of these cases are shown in Fig. 13. All had homonymous
quadrantanopias due to medial occipital lobe lesions involving striate cortex and the terminal
optic radiations (see Fig. 1). Any involvement of extra-striate cortex was limited to the

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Case 3: history
A 55-year-old man was unable to read upon awakening. Clinical evaluation shortly after
this episode revealed a right, mainly lower quadrantic, homonymous field defect. Corrected
visual acuity was 20/25 o.d. and 20/30 o.s. Isoptres to coloured targets were normal in
the right upper quadrant on Goldmann perimetry. Magnetic resonance images of Case 3
were obtained shortly following the ictus. A left occipital lobe lesion was demonstrated
which gave high signal on both T r and T2-weighted images and which did not enhance
following the administration of intravenous gadolinium-DTPA (Fig. 11). A diagnosis of
occipital haemorrhage was made and an associated primary or secondary neoplasm was
considered likely. A left occipital craniotomy was carried out and at operation a limited
occipital lobectomy was performed extending superiorly ~ 3 cm off die tentorium and
laterally 4 cm from the most lateral portion of the visible mass. Pathological examination
of the resected material confirmed a haematoma without a neoplasm but there was evidence
of amyloid angiopathy. Postoperative MRIs are shown in Figs 11 and 12. On MRI,
involvement of medial occipital cortex was found to be limited to the upper calcarine
area posteriorly. This is compatible with the most dense portion of the field defect which
involves the lower quadrant and spared the peripheral field.

G. T. PLANT AND OTHERS

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

FIG. 11. Axial MRIs of Case 3 are shown. In the upper pair an acute occipital hawnatnma is shown (arrow). The lower
pair were obtained following the lateral occipital resection described in the text in which lateral occipital cortex was removed
[see also Fig. 12).

medial occipital lobes. Discrimination : detection threshold ratios were not elevated at
locations in the residual contralateral hemifields when compared with the performance
at mirror symmetric locations in the ipsilateral hemifields (see Fig. 14).

I M P A I R E D MOTION P E R C E P T I O N AFTER O C C I P I T A L DAMAGE

1323

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

FIG. 12. Magnetic resonance images of Cases 1, 2 and 3, all of whom demonstrated impaired motion perception in the
contralateral visual field, are shown. The upper pair are lateral parasagittal and coronal images in Case 2. These are prcoperative images and the region of normal-appearing lateral occipital cortex damaged at surgery is indicated by the arrow
(see also Figs 8 and 9). The lower two images are lateral parasagittal images from Case 1 (left) and Case 3 (righ:). Although
the lesion in Case 1 is very extensive, the major overlap with the other two lesions is in its ventral extent, involving the
lateral occipital gyri.

Cases 7, 8 and 9: parietal and parieto-ocdpital lesions
Magnetic resonance images of Cases 7, 8 and 9 are shown in Fig. 15. These were all
examples of parietal lobe damage without a visual field defect (Fig. 1). None showed
evidence of impaired motion perception in the contralateral visual field (see Fig. 16).

1324

G. T. PLANT AND OTHERS

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

FIG. 13. Axial MRIs of Cases 4 (upper left), 5 (upper right) and 6 (lower panels) arc shown. All had homonymous
quadrantanopia due to medial occipital lobe lesions (arrows) involving striate cortex and the terminal optic radiations. If
extra-striate cortex was also damaged then it was limited to the medial occipital lobes.

I M P A I R E D M O T I O N P E R C E P T I O N AFTER O C C I P I T A L DAMAGE

100
•
a

1325

-Case 4
Ipsilateral field
Contralateral field

14 and 8 Hz | |B and 10Hz| I Orientation!

FIG. 14. Set legend to Fig. 7. Cases 4, 5 and 6 all had field defects due to lesions of striate cortex and adjacent optic
radiations, but the lateral occipital gyri were unaffected by the pathology, the discrimination/detection threshold ratios
for all tasks were not elevated at locations in the residual contralateral hemifields when performance was compared at
mirror symmetric locations in the ipsilateral hemifields.

Cases 10 and 11: occipito-temporal lesions
These two cases are of importance because the lesions are very close to those in Cases 1 - 3 .
Both were tumour cases and both had partial upper quadrantanopias due to involvement
of the lower fibres of the optic radiations (Fig. 1). Magnetic resonance images of both
cases and operative photographs of Case 11, indicating the extent of the resection, are
shown in Fig. 17. Neither showed any abnormality in the contralateral hemifield in the
tests performed (Fig. 18).

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

ISine wave! I Beat I

Direction discrimination Speed discrimination
Task

1326

G. T. PLANT AND OTHERS

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

FIG. 15. Magnetic resonance images of Case 7 (upper left) showing a parietal vascular malformation with associated
haemorrhage; Case 9 (upper right) showing the parietal lobe restion and CT images of Case 8 before (lower left panel)
and following (lower right panel) surgical excision of a vascular malformation with associated damage to the occipitoparietal region (not well shown because of clip artefact). None of these lesions caused a visual field defect. None of these
cases showed evidence of impaired motion perception in the contralateral visual field. The lesions coincide with much
of the parietal damage which occurred in Case 1, reinforcing the view that it is the more ventral and posterior component
of that lesion which was important in giving rise to the abnormality.

I M P A I R E D MOTION P E R C E P T I O N AFTER O C C I P I T A L D A M A G E

1327

-Case 7
•
Ipsilateral field
ezi Contralateral field

100

10-

10

1

Not done
ISine wave] [Beat]

Not done

Not done

|4and8Hzlfeand 10H3 IQrientationl

Direction discrimination Speed discrimination
Task
FIG. 16. See legend to Fig. 7 for explanation. Cases 7, 8 and 9 were cases of parietal lobe lesions which did not cause
a field defect. There was no abnormality in the contralateral visual field on any of the parameters measured. A limited
study was carried out in Case 9 as indicated.

DISCUSSION

In this study we have demonstrated that a motion deficit localized to a hemifield can
occur following posterior hemisphere damage in humans. The deficit is observed using
'higher order' motion tasks (speed discrimination and DOM discrimination of a CMod
grating) but not DOM discrimination of a LMod sine-wave grating or orientation discrimination. To our knowledge this has not been previously demonstrated, although
impaired colour vision within a hemifield has been reported several times since the earliest
case reported by Verrey (1888) [see also Zeki (1990*) and Plant (1991) for recent review
articles]. The bidirectional ocular motor pursuit anomaly within a hemifield described
by Thurston et al. (1988) may represent a similar phenomenon, although perceptual testing
was not carried out in that case and we have not studied eye movements. In Cases 1

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

o 100

1328

G. T. PLANT AND OTHERS

and 2 we were able to show that there was no impairment of motion perception in the
ipsilateral hemifield after the resection. Case 3 did have a higher threshold for speed
discrimination in the ipsilateral field {see Fig. 7) but we do not know what his performance
would have been before the resection. He also was an older patient with generalized vascular
disease and we cannot exclude contralateral cerebral damage not revealed on MRI. Case 3

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1329

Dorsal

is also a less 'pure' example than the other two cases in that threshold contrast sensitivity
was slightly impaired at the tested location in the contralateral hemifield.
As far as the anatomical basis of the observed effects in our three patients is concerned,
in Case 1 we are not necessarily dealing with normal anatomy. However, we do think
it is significant to note that the region of extra-striate visual cortex which gave rise to
a perception of motion in the contralateral hemifield on electrical stimulation (Figs 4, 5),
reproducing his epileptic aura, was located on the lateral occipital cortex within 4 cm
of the occipital pole. This cortical region is normal in appearance on the MRI (Fig. 3)
and is not necessarily involved in the developmental anomaly. In the cortical region
corresponding to the anomaly, electrical stimulation gave rise predominantly to non-visual
effects (see the Appendix).
The lateral occipital region is relatively accessible compared with the ventral occipitotemporal or striate cortex: for this reason it was the visual area most commonly encountered
by Penfield in his studies of the effects of electrical cortical stimulation of patients
undergoing surgery for intractable epilepsy (Penfield and Rasmussen, 1950; Penfield and
Jasper, 1954). Penfield's experience was that seizures originating in the lateral convexity
of the occipital lobe gave rise most commonly to the perception of moving, twinkling
or pulsating lights either in the contralateral visual field or directly in front of the patient

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Ventral
Fie. 17. Postoperative axial and lateral parasagittal MRIs of Cases 10 (upper row, opposite) and 11 (middle row, opposite).
Both were cases of tumour and both had upper partial quadrantic hemianopia due to involvement of the lower fibres of
the optic radiations (see Fig. 1). The operative photograph of Case 11 (above) indicates the extent of the occipito-temporal
resection that was carried out (dashed lined). Anterior is on the left and ventral is at the bottom of the photograph. The
labels numbered 1 - 5 indicate the location of die motor strip. Continuous spike and sharp wave activity was recorded
from the location labelled 'A'. The occipital pole was measured to be 4 cm further posterior than the posterior edge of
the craniotomy. These lesions are very close to those of Cases 2 and 3 being more anterior but neither showed evidence
of impaired motion perception in the contralateral hemifield reinforcing the view that it was the damage to the lateral
occipital gyri that was important in producing the abnormality found in Cases 1, 2 and 3.

G. T. PLANT AND OTHERS

1330
100

-Case 10
Ipsilateral field
ContralateraJ field

|4and8Hz|[8andiOHz| IQrientationl

FIG. 18. Case 10 had an extensive temporo-parietal lesion (see Fig. 17) which caused an upper partial quadrantic field
defect due to involvement of the optic radiations. The tests described previously were carried out in the lower quadrant
of the hemifield contralateral to the lesion at an eccentricity of 10° and the results are compared above with those obtained
at the mirror symmetric location in the ipsilateral hemifield. There was no evidence of a relative reduction in contrast
sensitivity on the discrimination tasks in the contralateral field. Case 11 had a similarly situated occipito-temporal
oligodendroglioma and following excision had an upper quadrantic field defect. Once again, however, there was no evidence
of impaired motion perception in the contralateral lower quadrant.

(Penfleld and Jasper, 1954). In a summary of the effects of electrical stimulation of the
lateral occipital lobe (Penfield and Rasmussen, 1950, pp. 139 —141) it was a general
(although not absolute) observation that stimulation near the occipital pole and the calcarine
fissure was more likely to produce coloured phosphenes, whereas stimulation more
anteriorly was more likely to give rise to phosphenes which were colourless and moving,
flickering or dancing.
We regard the lesion produced in Case 2 as being particularly important because the
normally appearing cortex excised was 4 cm from the occipital pole and as such overlaps
only the most inferior and posterior portion of the cortical resection in Case 1 (see Fig. 12);
also corresponding to the region where moving phosphenes were induced on stimulation.
The resected cortex in Case 3 included lateral occipital cortex extending anteriorly several
centimeters from the occipital pole (4 cm anterior to the limit of the visible haemorrhagic
change) and dorsally 3 cm above the tentorium. We suggest that it is also significant
that our two examples of extensive occipito-temporal pathology (Cases 10, 11) are very
close to the resections in Cases 2 and 3. These lesions are only 3 —4 cm more anterior
at the occipito-temporal junction and yet have produced no anomaly of motion perception
identifiable by our test procedures. This, and the lack of any effect in our three cases

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

Bine wave| | Beat |

Direction discrimination
Speed discrimination
Task

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1331

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

of parietal lobe pathology (Cases 7, 8, 9) reinforces our view that it is the damage to
the lateral occipital gyri in Cases 1, 2 and 3 which was critical in producing the deficit
in visual motion processing documented in this article.
From work carried out in many laboratories and conveniently summarized by Ungerleider
and Desimone (1986), it is known that in the macaque V5-MT is the earliest visual area
which is predominantly concerned with the projections to the parietal lobes mediating
oculomotor and visuospatial performance (extra-striate areas V2 and V3 also process
information destined for V4 mediating colour and form vision). Furthermore, V5-MT
contains a retinotopic organization. Although there are dense callosal connections between
left and right V5-MT and some representation of the ipsilateral field is found (Newsome
and Allman, 1980; Van Essen et al., 1981), an extensive representation of the ipsilateral
visual field is not found until higher projection zones (e.g. in the floor of the superior
temporal sulcus). Therefore to produce a motion deficit limited to a hemifield it may
be necessary for a lesion to interrupt the inputs to V5-MT or damage V5-MT itself. We
cannot distinguish these possibilities in our cases. It is possible that the closely located
occipito-temporal lesions in Cases 10 and 11 may be further anterior than V5-MT itself,
thus giving rise to damage after substantial callosal transfer of information has taken place.
Lesions affecting higher visual areas than V5-MT in occipito-parietal cortex might be
expected to produce a unidirectional pursuit anomaly throughout the visual field by analogy
with the ocular motor impairment of ipsilateral pursuit (Thurston et al., 1988). Such a
deficit would not have been detected by the methods used here because the direction of
motion of the stimuli was randomly varied in the speed discrimination task.
Striate cortex in the macaque and other non-human primates occupies a considerable
portion of the lateral occipital lobe, whereas in the human it is located largely on the
medial surface of the occipital lobe and in the calcarine fissure. Dorsal and ventral V2
are also likely to be located more medially in the occipital lobe if, in humans as in primates,
these areas are contiguous with dorsal and ventral VI along the representation of the
vertical meridian (Horton and Hedley-Whyte, 1984). Dorsal and ventral V3 are, in turn,
contiguous with V2 along the representation of the vertical meridian. If this is the case
in humans, then the lateral occipital damage in our three cases is unlikely to have involved
dorsal or ventral V2. The preserved chromatic fields make it very unlikely that V2 was
involved in these lesions as these areas continue to process chromatic information (see,
for example, Livingstone and Hubel, 1984; Zeki, 19906). Indeed evidence has been
presented suggesting that lesions of dorsal or ventral V2 in humans may give rise to dense
lower or upper homonymous quadrantanopic field defects (Horton and Hoyt, 1991).
Lesions which, in humans, give rise to loss of colour vision in the contralateral hemifield
are located in ventral occipito-temporal cortex (Verrey, 1888; Kolmel, 1988), similar
in location to the region activiated in PET studies (Corbetta et al., 1990; Zeki et al.,
1991). Thus, in the human, the region damaged in cerebral achromatopsia may be located
more ventrally than macaque V4.
These differences suggest that the human homologues of V2, V3 or V4 are unlikely
to be located in lateral occipital cortex, although we do not know what more fundamental
rearrangements may have taken place in the organization of extra-striate visual cortex
in humans. Clarke and Miklossy (1990) have demonstrated a strikingly heavily myelinated
region on the convexity of the lateral occipital cortex in human pathological material,
the inferior margin of which received very dense callosal afferents. The region was centred

1332

G. T. PLANT AND OTHERS

ACKNOWLEDGEMENTS
We wish to thank W. F. Hoyt, M. Aminoff and J. C. Horton of UCSF and S. Bowne of Smith-Kettlewell
for assistance and helpful discussion. This work was supported of the Medical Research Council of the United
Kingdom and the Brain Research Trust.
REFERENCES
ALLMAN JM, KAAS JH (1971) A representation of the visual field in the caudal third of the middle temporal
gyrus of the owl monkey (Aotus trivirgatus). Brain Research, Amsterdam, 31, 85 — 105.
ALLMAN JM, KAAS JH (1974a) A crescent-shaped cortical visual area surrounding the middle temporal area
(MT) in the owl monkey (/{otus trivirgatus). Brain Research, Amsterdam, 81, 199—213.
ALLMAN JM, KAAS JH (19746) The organization of the second visual area (VH) in the owl monkey: a second
order transformation of the visual hemifield. Brain Research, Amsterdam, 76, 247—265.
BAKER CL, HESS RF, ZIHL J (1991) Residual motion perception in a 'motion-blind' patient, assessed with
limited-lifetime random dot stimuli. Journal of Neuroscience, 11, 454—461.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

on the upper bank of the inferior lateral occipital sulcus in the two hemibrains studied.
The posterior and anterior margins were, in one brain, 28 and 36 mm, respectively, from
the occipital pole and its inferior border was 10—20 mm from the tentorium {see Figs 6
and 17 of Clarke and Miklossy, 1990). As such this region would certainly be included
in the region of lateral occipital cortex included in the resections carried out in our cases.
It also corresponds well with the region of lateral occipital cortex stimulation in which
evoked moving phosphenes in Case 1 (Fig. 5). Beckers and Homberg (1992) have shown
that discrimination of direction of motion of random dot patterns presented in a background
of uncorrelated random dots was impaired in the contralateral hemifield following
transcranial magnetic stimulation. The critical location to produce this effect was
4—6 cm lateral to the medio-sagittal Nasion-Inion line; this result is compatible with our
findings and those of Clarke and Miklossy (1990). However, stimulation of this type
does not give rise to illusions or hallucinations of movement.
It seems likely that unilateral lesions of the type we have described may be considerably more common than bilateral lesions giving rise to selective impairment of motion
perception. We would predict that cases of more anterior parietal lobe lesions showing
impaired ipsilateral pursuit eye movements, an anomaly which is found throughout the
visual field and which is likely to occur at a level where transcallosal transfer of information
is extensive (Thurston et al., 1988; Morrow and Sharpe, 1990), will not show anomalous
motion perception in the contralateral hemifield. We further suggest that in all studies
of anomalous motion perception the possibility of hemifield asymmetry should be explored.
Without this information studies of unilateral posterior lesions under conditions of free
fixation are difficult to interpret (e.g. Vaina, 1989). The paradigm employed in the present
study (i.e. seeking examples of anomalous motion perception limited to a hemifield in
unilateral lesions) is likely to provide the opportunity to investigate a larger number
of human cases of impaired motion perception than has been possible hitherto. With
contemporary imaging techniques a clearer idea of the locations of damage giving rise
to visual motion perceptual deficits of varying severity should emerge. The characteristics
of the motion anomaly in these three cases are further explored in the accompanying
article (Plant and Nakayama, 1993) and are placed in the context of other human studies
and studies involving cerebral lesions in animals.

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1333

BECKERS G, HOMBERG V (1992) Cerebral visual motion blindness: transitory akinetopsia induced by transcranial
magnetic stimulation of human area V5. Proceedings of the Royal Society of London, B, 249, 173-178.
CLARKE S, MIKLOSSY J (1990) Occipital cortex in man: organization of callosal connections, related myeloand cytoarchhecture, and putative boundaries of functional visual areas. Journal of Comparative Neurology,
298, 188-214.
CORBETTA M, MIEZIN FM, DOBMEYER S, SHULMAN GL, PETERSEN SE (1990) Attentional modulation

PLANT GT, LAXER KD, BARBARO NM, NAKAYAMA K (1990) Impaired speed discrimination restricted to

the hemifield contralateral to an occipito-parietal surgical restion. Society for Neuroscience Abstracts,
16, 5.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

of neural processing of shape, color, and velocity in humans. Science, 248, 1556-1559.
DESIMONE R, UNGERLEIDER LG (1986) Multiple visual areas in the caudal superior temporal sulcus of the
macaque. Journal of Comparative Neurology, 248, 164 — 189.
DUBNER R, ZEKI SM (1971) Response properties and receptive fields of cells in an anatomically defined
region of the superior temporal sulcus in the monkey. Brain Research, Amsterdam, 35, 528-532.
FIORANI M, GATTASS R, ROSA MGP, SOUSA APB (1989) Visual area MT in the Cebus monkey: location,
visuotopic organization, and variability. Journal of Comparative Neurology, 287, 9 8 - 1 1 8 .
GATTASS R, GROSS CG (1981) Visual topography of striate projection zone (MT) in posterior superior temporal
sulcus of the macaque. Journal of Neurophysiology, 46, 621—638.
GATTASS R, GROSS CG, S A N D E L L J H ( 1 9 8 1 ) Visual topography of V2in the macaque. Journal of Comparative
Neurology, 201, 519-539.
GATTASS R, SOUSA APB, GROSS CG (1988) Visuotopic organization and extent of V3 and V4 of the macaque.
Journal of Neuroscicnce, 8, 1831-1845.
HESS RH, BAKER CL, ZIHL J (1989) The 'motion-blind' patient: low-level spatial and temporal filters. Journal
of Neuroscience, 9, 1628-1640.
HORTON JC, HEDLEY-WHYTE ET (1984) Mapping of cytochrome oxidase patches and ocular dominance
columns in human visual cortex. Philosophical Transactions of the Royal Society of London, B, 304,
255-272.
HORTON JC, HOYT WF (1991) Quadrantic visual field defects: a hallmark of lesions in extrastriate (V2/V3)
cortex. Brain, 114, 1703-1718.
KOLMEL HW (1988) Pure homonymous hemiachromatopsia. Findings with neuro-ophthalmologic examination
and imaging procedures. European Archives of Psychiatry and Neurological Sciences, 237, 237—243.
LIVINGSTONE MS, HUBEL DH (1984) Anatomy and physiology of a color system in the primate visual cortex.
Journal of Neuroscience, 4, 309—356.
MCLEOD P, HEYWOOD C, DRIVER J, ZIHL J (1989) Selective deficit of visual search in moving displays
after extrastriate damage. Nature, London, 339, 466—467.
MAUNSELL JHR, NEWSOME WT (1987) Visual processing in monkey extrastriate cortex. Annual Review
of Neuroscience, 10, 363-401.
MAUNSELL JHR, VAN ESSEN DC (1983) Functional properties of neurons in middle temporal visual area
of the macaque monkey. 1. Selectivity for stimulus direction, speed, and orientation. Journal of
Neurophysiology, 49, 1127-1147.
MORROW MJ, SHARPE JA (1990) Cerebral hemispheric localization of smooth pursuit asymmetry. Neurology,
Cleveland, 40, 284-292.
NEWSOME WT, ALLMAN JM (1980) Interhemispheric connections of visual cortex in the owl monkey, Aotus
trivirgatus, and the bushbaby, Galago senegalensis. Journal of Comparative Neurology, 194, 209—233.
PENFIELD W, JASPER H (1954) Epilepsy and the Functional Anatomy of the Human Brain. Boston: Little,
Brown, pp. 401-406.
PENFIELD W, RASMUSSEN T (1950) The Cerebral Cortex of Man: A Clinical Study of Localization of Function.
New York: MacMillan, pp. 136-147.
PLANT GT (1991) Disorders of colour vision in diseases of the nervous system. In: Inherited and Acquired
Colour Vision Deficiencies: Fundamental Aspects and Clinical Studies. Edited by D. H. Foster.
Basingstoke, England: Macmillan Press, pp. 173 — 198.
PLANT GT, NAKAYAMA K (1991) Impaired motion perception following unilateral occipital damage.
Investigative Ophthalmology and Visual Science, ARVO Supplement, 32, 824.
PLANT GT, NAKAYAMA K (1993) The characteristics of residual motion perception in the hemifield contralateral
to unilateral lateral occipital lesions in humans. Brain, 116, 1337 — 1353.

1334

G. T. PLANT AND OTHERS

RODMAN HR, ALBRIGHT TD (1987) Coding of visual stimulus velocity in area MT of the macaque. Vision
Research, 27, 2035-2048.
THURSTON SE, LEIGH RJ, CRAWFORD T, THOMPSON A, KENNARD C (1988) Two distinct deficits of visual

tracking caused by unilateral lesions of cerebral cortex in humans. Annals of Neurology, 23, 266—273.
UNGERLEIDER LG, DESIMONE R (1986) Cortical connections of visual area MT in the macaque. Journal
of Comparative Neurology, 248, 190-222.
VAINA LM (1989) Selective impairment of visual motion interpretation following lesions of the right occipitoparietal area in humans. Biological Cybernetics, 61, 347—359.
VAINA LM, LEMAY M, BIENFANG DC, CHOI AY, NAKAYAMA K (1990) Intact 'biological motion' and

ZEKI S, WATSON JDG, LUECK CJ, FRISTON KJ, KENNARD C, FRACKOWIAK RSJ (1991) A direct

demonstration of functional specialization in human visual cortex. Journal of Neuroscience, 11, 641—649.
Z I H L J , CRAMOND VON, MAI N (1983) Selective disturbance of movement vision after bilateral brain damage.
Brain, 106, 313-340.
ZIHL J, CRAMON D VON, MAI N, SCHMID CH (1991) Disturbance of movement vision after bilateral posterior
brain damage. Further evidence and follow up observations. Brain, 114, 2235—2252.
Received September 21, 1992. Revised August 16, 1993. Accepted September 14, 1993

APPENDIX
Results of cortical stimulation in Case 1 (please refer to Fig. 5)
Electrode
Al
At 4V he immediately responded 'colour' and later described a red spot (the size of a coin at 10
feet) that had appeared in the right visual field.
A2
At 8V he saw a change in his vision which he described as a wave travelling downwards in the right
hemifield.
A3
At 8V he noticed a slight change in his vision in the right visual field. He described this as blurring
and said that it was similar to the onset of a seizure.
A4
Slight hemianopic visual disturbance and tingling in his right hand.
B3
Slight hemianopic visual disturbance.
C2
At 4V he reported a change in his vision like a wave moving from right to left. At 6V he reported,
in addition to the visual disturbance, facial tingling.
C3
At 8V he reported a similar change in his vision like a wave moving from right to left and stopping
abruptly at the midline.
C8
At 8V he was aware of tingling in his chest and right leg.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016

'structure from motion' perception in a patient with impaired motion mechanisms: a case study. Visual
Neuroscience, 5, 353-369.
VAN ESSEN DC, MAUNSELL JHR, BIXBY JL (1981) The middle temporal visual area in the macaque:
myeloarchitecture, connections, functional properties and topographic organization. Journal of Comparative Neurology, 199, 293-326.
VERREY D (1888) He'rniachromatopsie droite absolue. Conservation partielle de la perception lumineuse et
des formes. Ancien kyste hernorrhagique de la partie infe'rieur due lobe occipital gauche. Archives
d'Ophtalmologie, Paris, 8, 289-300.
ZEKI SM (1974) Functional organization of a visual area in the posterior bank of the superior temporal sulcus
of the rhesus monkey. Journal of Physiology, London, 236, 549-573.
ZEKI S (1990a) The motion pathway of the visual cortex. In: Vision: Coding and Efficiency. Edited by C.
Blakemore. Cambridge: Cambridge University Press, pp. 321-345.
ZEKI S (1990*) A century of cerebral achromatopsia. Brain, 113, 1721-1777.
ZEKI S (1991) Cerebral akinetopsia (visual motion blindness). A review. Brain, 114, 811-824.

IMPAIRED MOTION PERCEPTION AFTER OCCIPITAL DAMAGE

1335

Dl
D2

At 3V he reported a sensation of warmth throughout his body.
At 4V there was a change in his vision, moving as a wave from right to left. At 6V there was the
same change in his vision, but stronger, associated with deviation of head and eyes to the right. These
sensations were similar to those occurring at the start of seizure.
D3
At 8V there was a similar change in his vision travelling across to the midline.
E6, E7, F6, G l , G6. At these electrodes he reported visual blurring affecting both hemifields, sometimes
associated with somatosensory phenomena.
E8, F8, G7, G8, H6, H7, H8. At these electrodes motor phenomena involving the contralateral arm and face.
G5
At 12 V the patient reported the sound of a lady screaming in the distance with an associated echo.

Downloaded from http://brain.oxfordjournals.org/ by guest on June 7, 2016