WHAT THE EYES PERCEIVE, THE BRAIN IGNORES: A
CASE OF PURE UNILATERAL REPRESENTATIONAL
NEGLECT*
Nicoletta Beschinl, Gianna CocchinF, Sergio Della Sala3 and Robert H. Logie3
('Unit of Rehabilitation Somma Lombardo, Gallarate Hospital, Varese, Italy; 2Salvatore
Maugeri Foundation "Clinic a del Lavoro e della Riabilitazione" IRCCS, Medical Centre
of Veruno, Novara, Italy; 3Department of Psychology, University of Aberdeen, U.K.)

ABSTRACT

Bisiach and Luzzatti (1978) provided evidence that unilateral spatial neglect is not only
a disorder of visual perception, but also can affect mental representations such that patients
fail to report the left side of scenes or objects in mental imagery. However in case reports
of representational neglect generally it is accompanied by perceptual neglect.
We report a rare occurrence of a patient (NL) who presents a persistent unilateral neglect
which appears to be limited to visual imagery. The deficit appeared in tasks which require
the formation and manipulation of new visuo-spatial representations as well as those which
require access to information about familiar scenes. The patient, who had a lesion in the
right parietal lobe, showed no evidence of perceptual or personal neglect, although there
was some evidence of visual extinction.
We argue that the concept of visuo-spatial working memory can provide a framework
within which to interpret aspects of the representational form of neglect, whether or not it
is accompanied by perceptual neglect.

INTRODUCTION

Unilateral spatial neglect (USN) generally is defined as an impaired ability
to detect or respond to stimuli in spatial locations contralateral to the damaged
cerebral hemisphere in the absence of peripheral deficits. It is more frequent,
severe, and long-lasting after a right parietal lobe lesion (Vallar, 1993). This
definition of USN has been based primarily on observed deficits in visualperceptual tasks. However it has become clear that USN is not an impairment
limited to the processes of visual perception (Zingerle, 1913) but also can
compromise the "left side" of visual mental representations, sometimes referred
to as visual mental images (De Renzi, Faglioni and Scotti, 1970).
Two of the best known cases of representational neglect were described by
Bisiach and Luzzatti (1978). They tested two patients showing left USN, asking
them to describe from memory a very familiar piazza, the Cathedral Square
(Piazza del Duomo) in Milan. The patients had first to give their description
from a given imagined viewpoint at one end of the piazza, and then to describe
the same piazza from the opposite end. From both viewpoints they accurately
* Names of authors are listed in alphabetical order, and this does not imply differential contribution to the paper.
Cortex, (1997) 33, 3-26

4

Nicoletta Beschin and Others

described only the buildings on the right side of their mental image, failing to
mention those on the left side. This study gave rise to the suggestion that USN
also may involve mental representations in the absence of physical stimulation.
This impairment was thought to arise from damage to the brain hemisphere
opposite to the side of the neglected information in the mental representation.
In this sense, the representation was functioning like the visual perceptual
environment. Subsequent studies have supported the idea that USN involves
mental representations (Bisiach, Luzzatti and Perani, 1979; Bisiach, Capitani,
Luzzatti et aI., 1981; Baxter and Warrington, 1983; Ogden, 1985; Barbut and
Gazzaniga, 1987; Caramazza and Hillis, 1990a, 1990b; Halligan and Marshall,
1992) and it is now agreed that disorders in the perceptual and the
representational domain can occur together (Grossi, Modafferi, Pelosi et aI., 1989;
Rizzolatti and Berti, 1993; Bartolomeo, D'Erme and Gainotti, 1994).
The interpretation of these findings is still very much a matter for debate.
USN has been attributed to an impairment within a system that usually modulates
the ability to select and orient to stimuli in the visual world (Popperlreuter,
1923). It has also been attributed to an impairment in orienting attention to the
contralesional side (Riddoch and Humphreys, 1983); or to a strong tendency to
orient attention to the ipsilesional side (Kinsboume, 1970); or to the impaired
disengagement of attention from stimuli located in the ipsilesional side (Posner,
Walker, Friedrich et aI., 1987). However these attentional theories do not
incorporate the nature of the cognitive system that might be responsible for the
mental representation. Nor do these theories adequately address how such a
cognitive system might support covert attention to different parts of the
representation.
What kind of theoretical model might account for the representational form
of neglect? Clearly one approach to resolving this issue is to explore theories of
mental representation in the healthy brain and investigate whether any such theory
can offer an account of representational deficits. One candidate theory arises from
the mooted close relationship between processes involved in mental representation
and those involved in working memory. Working memory refers to a set of
cognitive mechanisms thought to be responsible for online processing and
temporary storage of information. In one widely adopted model, working memory
comprises an executive coordinator and two systems specialised for dealing
respectively with visuo-spatial information and with verbal information (Baddeley
and Hitch, 1974; Baddeley, 1986, 1996). This model has been singularly successful
in accounting for a wide range of cognitive deficits, although its use in accounting
for specifically visuo-spatial deficits has thus far been rather limited (for recent
reviews see Della Sala and Logie, 1993; Logie, 1995).
Baddeley and Lieberman (1980) suggested that patients such as those
described by Bisiach and Luzzatti " ... have a defect which is associated with
spatial working memory" (p. 23) and, using the analogy of the specialised vi suospatial system as a screen on which spatial information may be represented,
they also stated that " ... one might conclude either that part of the screen was
inoperative or that the process of scanning the screen was defective ... " (p. 23).
The view has been supported by other authors (e.g. Sunderland, 1984; Shallice,
1988), and Bisiach (1993, p. 439) has endorsed the suggestion of a partially

Representational neglect

5

inoperative visuo-spatial component of working memory. However as yet the
hypothesis has not been investigated in any detail. Moreover, in the vast majority
of cases of representational neglect that have been reported, there has been
concomitant visual perceptual neglect. Therefore it is not clear whether the
impairment arises primarily from impoverished perceptual input on the neglected
side or from an impaired representational system. One way to explore this further
would be if patients were found with a representational impairment but with
intact visual perception.
Guariglia, Padovani, Pantano et al. (1993) described one such patient who
showed a persistent disorder in visual imagery for familiar piazzas in the absence
of perceptual USN. During the description of familiar places from different
vantage points his performance was very similar to Bisiach and Luzzatti's
patients. However on perceptually based tests he appeared to perform normally.
Subsequently, Bartolomeo et al. (1994) analysed the relationship between vi suospatial and representational USN in 30 right- and 30 left-brain-damaged patients
and they found no cases of representational neglect in the absence of perceptual
deficits. In a follow-up eight months after the stroke one of their patients, with
a right hemisphere lesions, showed neglect limited to mental representations,
and the signs of visual perceptual neglect were no longer present. These authors
suggested that different compensatory mechanisms rather than lesions in different
components of the cognitive architecture were responsible for this result.
Bartolomeo and colleagues noted that the same could be true for the Guariglia
et al. patient who was tested 16 months after the stroke. Notwithstanding this
observation, the patient reported by Guariglia et al. suggests that the dissociation
is possible. In this paper we pursue this possibility by reporting the case of a
patient (NL) who, following an ischaemic lesion in the right anterior parietal
lobe, presented a persistent unilateral neglect limited to visuo-spatial imagery.
We examined the patient for a period of several weeks after the stroke, and
subsequently. Following the formal description of the patient, we shall discuss
the pattern of data obtained in the context of the visuo-spatial working memory
hypothesis referred to above.
CASE REPORT

NL, a 67 year old right-handed man with eight years of education, was admitted on
December 28, 1993 to a Neurological Unit after an ischaemic stroke which caused a left
hemiparesis and left hemianaesthesia. A Computerised Tomography scan (13/01/1994) showed
a large area of hypodensity in the right parietal lobe. Two months after the vascular accident
he was co-operative and was oriented in time and in space. His speech was fluent and he
did not show difficulties referable to aphasic deficits. He could easily recognise familiar
faces, showing no evidence of prosopagnosia.
Although no formal psychometric testing for perceptual neglect took place immediately
after the stroke there are a number of sources of evidence to suggest that perceptual neglect
was not present even at that stage. Doctors and nurses did not notice any typical neglect
behaviours: he was reported to perceive stimuli coming from each side of the visual field and
successfully to reach out to targets in both the right and left hemifields. Also his wife was
asked to complete a questionnaire concerning several behavioural aspects of neglect,
and her responses all were negative. He had a paresis in his left limbs, for which he was
given a programme of motor rehabilitation which commenced one week before formal testing.

Nicoletta Beschin and Others

6

Fig. 1 - CT Scan of patient NL, which shows an ischemic lesion in the right parietal lobe.

He was aware of his motor problems, indicating the absence of any anosognosia. Relatives
reported that NL was depressed and often cried.
Two and a half months after the stroke, he was sent to the first two authors for a routine
neuropsychological assessment. A new CT scan (22/03/1994) confirmed the previous findings
(see Figure I).

General Neuropsychological Examination
NL's scores on a variety of neuropsychological tests are shown in Table I.
His performance on the Mini Mental State Examination (Folstein, Folstein and
McHugh, 1975) was adequate for his socio-educational background (Measso,
TABLE I

NL's General Neuropsychological Examination Scores adjusted for age, education and, whenever
appropriate, sex (Score range is indicated in parentheses. For details of tests see text).
Test (score range)
- Global assessment
Mini Mental State Examination (0-30)
- Short term memory
Bisyllabic word span (0-10)
Corsi block-tapping span (0-10)
- Long Term Memory
Learning three words lists (0-30)
Corsi block-tapping spatial supraspan (0-29.16)
Maze learning (l5-l)bc
- Intellectual functioning
Ravens Progressive Matrices (0-36)
Verbal abstract reasoning (0-60)
- Visual Perception
Unfamiliar face recognition (0-54)
- Spatial cognition
Constructional apraxia (0-14)
Rey complex' figure A (0-36)
Rey complex figure B (O-lI) (elements plus details)
- Acalculia
Acalculia test (0-21)C

NL's scores adjusted
for age and education

Controls' inferential
5th centile or cut-off'

28.53

23.80

4
4.25

2.75
3.50

18.5
18.86
12*

7.50
5.50
6.2"

22.50
60

14.75
32

38

34

10.75
25.63*
11

7.75
27.99

21

, Cut-off is the score two standard deviations below the mean score of the control group.
b Error score as trials to criterion.
C Not adjusted for age and education.
* Score below the cut-off.

Representational neglect

7

Cavarzeran, Zappala et al., 1993). He had no verbal or spatial short or long
term memory deficits (bisyllabic words span - Baddeley, Thomson and
Buchanan, 1975b; learning three words lists test - Novelli, Papagno, Capitani
et ai., 1986; Corsi's block-tapping test - Milner, 1971; and Corsi's block-tapping
supraspan test following the procedure described by Spinnler and Tognoni, 1987).
NL was also given a maze learning test (Milner, 1971) which consists of a long
pathway through a 10 X 10 grid of buttons. As can be seen in Table I, the
number of trials he required to reach criterion performance was higher than that
of 5 normal control subjects, but his pattern of errors showed no suggestion of
asymmetry. Visuo-spatial and verbal intellectual functions, assessed with Raven's
Coloured Progressive Matrices (Basso, Capitani and Laiacona, 1987) and verbal
abstract reasoning (Spinnler and Tognoni, 1987) were unimpaired. NL had a
normal score on a test for recognition of unfamiliar faces without involving
memory (test of facial recognition - Benton and Van Allen, 1968). On two tests
for figure copying (constructional apraxia test - Arrigoni and De Renzi, 1964,
and the Rey complex figure A - Rey, 1942, 1959) he did not show constructional
apraxia nor was there evidence of neglect, but he copied two pictures (one was
the Rey figure) rotated 900 anticlockwise (Turnbull, Beschin and Della Sala,
1996). Although scores were independent of the orientation of the pictures, on
Rey's figure copy test NL's score fell below the cut-off for normal performance
(see Table I). The rotation phenomena could have been the main reason for his
poor performance on this copying task but the important point is that there was
no evidence of neglect on these tasks. He performed flawlessly on a test assessing
mathematical abilities.
In summary NL showed no mental global deterioration, he had no verbal
memory deficits, no abstract reasoning difficulties, or signs of acalculia and no
deficits in recognizing unfamiliar faces. Nor were there signs of neglect in
copying geometrical and abstract patterns. He performed normally on all spatial
tasks when the stimuli were present. His only observed spatial impairment was
in learning a long pathway through a matrix. This pathway test has been found
to be particularly sensitive to spatial disorientation in daily life (Hecaen, Tzortzis
and Rondot, 1980).
Assessment of Unilateral Spatial Neglect
Personal Neglect

Personal neglect was assessed with four tests that required movements
towards both sides of the body. NL's scores are shown in Table II. Normal
scores in the Table are those originally reported by the authors of the tests.
Using the Bisiach, Vallar, Perani et al. (1986) one item test of personal
neglect the patient was blindfolded and was required to touch his left hand with
his right hand. This test has a score range of 0 to 3 depending on the speed
and accuracy of performance. NL promptly reached for the target (score 0 = no
personal neglect).
The Fluff test (Beschin and Robertson, in preparation) required the subject
to reach for targets in the form of 2 cm circles attached to the front of clothes
that he was wearing. On the torso there were three stickers on the left and three

Nicoletta Beschin and Others

8

on the right of the central midline area, there were six stickers along the patient's
left arm, six along the right leg and six along the left leg. No targets were
placed on the right arm because the task was performed with this arm. In total
there were 15 targets on the left side of the body and nine targets on the right.
The patient was blindfold when the targets were attached and was not informed
as to how many targets were present. In the first condition he had to remove
all the targets while blindfold. In the second condition the blindfold was
removed. Results are shown in Table II from which it is clear that NL
successfully reached all the targets in the "no blindfold" condition, but that in
the "blindfold" condition he missed 12 of the 15 targets on his left side and
performed flawlessly with the targets on his right. Because of his poor
performance in the blindfold condition, this condition was repeated the following
day and very similar results were obtained (see Table II).
On the Comb test and Razor test (Beschin and Robertson, in preparation) NL
was asked to simulate respectively combing his hair and shaving himself for 30
seconds. The tasks were performed twice with an intervening delay of 15 days.
Results are shown in Table II presented as the difference in the number of strokes
on the right and left side of the body. NL's scores were compared with norms
collected from 64 brain damaged and normal subjects (Beschin and
Robertson, in preparation. His performance on the test was above the cut-off
score on the first test, but below the cut-off on the second test.
In summary NL performed normally on tasks that permitted visual input and
feedback. He showed personal neglect only on two tasks. In one of these he
was blindfold and therefore would have to rely on a mental representation of
his body. In the Comb and Razor test he scored just below the cut-off on retest.
In the case of this test, there is only partial visual input as to the position of
his arm, and he would be unable visually to monitor the parts of his body
involved (hair and chin). The fact that he could watch his arm movements but
did not have complete visual information may have led to his borderline
performance on this task which relies at least in part on a mental representation
of his face and head.
TABLE II

NL's Peiformance on Tasks Assessing Personal Neglect (Score range is indicated in parentheses.
For details of tests see text)

Tests (score range)
Personal neglect test (3-0)
Fluff test - eyes open
(left 0-15; right 0-9)
Fluff test - eyes closed
(left 0-15; right 0-9)
Comb and Razor test (O-inf.)

NL's scores
Left
Right

Cut-off
Left

Right

o (no defect)
test
retest
test
retest

15

9

15

9

3*
0*

9
9

13.5

9

36.6%

34.6%a

29.9%*

Cut-off is the score two standard deviations below the mean score of the control group.
+ Score below cut-off.
a The score is expressed as a percentage of strokes on the left relative to the total number of strokes on the left,
centre, and right.

Representational neglect

9

Perceptual Neglect

NL was given tests to assess the presence of visual perceptual neglect. The
patient's scores are shown in Table III. NL performed at ceiling on the Stars
Cancellation Test (Wilson, Cockburn and Halligan, 1987), Lines Cancellation
Test (Albert, 1973), Wundt-Jastrow Area Illusion Test (Massironi, Antonucci,
Pizzamiglio et aI., 1988) and Perceptive Discrimination Between Objects Test
(Bisiach and Rusconi, 1990). On the Bells Test (Gauthier, Dehaut and Joanette,
1989) he missed only one target on the left side the first time, and one target
on the right side the second time, and with this test presence of neglect is
assumed only with three or more omissions on one side (Vallar, Sterzi, Bottini
et aI., 1990). On the Letter Cancellation Test (Diller and Weinberg, 1977) his
performance was very good. He missed one target on the left side and the
presence of neglect in this task is assumed only if there are four or more
omissions on one side only (Zoccolotti, Antonucci, Judica et aI., 1989). NL's
scores on two horizontal lines bisection tests and a vertical lines bisection test
were compared with those of five controls and no difference emerged. Moreover,
NL's performance on these tests showed no difference between the right and
left side. NL also was tested on four separate tests using verbal stimuli (see
Table III). The tests comprised the Reading Horizontal Sentences Test (De Renzi
and Spinnler, 1966; Zoccolotti et aI., 1989), the Reading Vertical Sentences
Test (Zoccolotti et aI., 1989), and two single word reading tests devised for this
study, one involving normal words and the other involving mirror reversed
words. In these last two tests a list of thirty words of varying frequency,
concreteness and length (4-8 letters) were printed in large capital letters in normal
TABLE III

NL's Performance on Tasks Assessing Perceptual Neglect (Score range is indicated in parentheses.
For details of tests see text)
Test (score range)

NL's scores

Star cancellation (0-54)
Lines cancellation (0-40) test
retest
Wundt-Jastrow area illusion (0-40)
Perceptive discrimination between objects (0-48)
Bells test (0-35) test
retest
Letter cancellation (0-104)
Horizontal lines bisectiona
(9 single lines) (-720/ + 720 mm)
Horizontal lines bisectiona
(5 multiple lines sets) (- 1060/ + 1060 mm)
Vertical lines bisectiona
(5 single lines) (- 460/ + 460 mm)
Word reading (0-30)
Mirror word reading (0-30)
Reading horizontal sentences (0-6)
Reading vertical sentences (0-2)

54
40
40
40
48
34
34
103

Cut-off range

(L-R=O)
(L-R=O)
(L-R=O)

L-R,,",,±2
L-R,,",,±1

(L-R=-I)
(L-R=+I)
(L-R=-I)

L-R,,",,±3
L-R,,",,±4

+0.17

+6/-5.6

+0.11

+2.8/-3.9

+0.28
30
30
6
2

+5.1/- 8.7

Cut-off is the score two standard deviations below the mean score of the control group. L = left; R = right.
a The distance from midpoint is expressed in millimetres. A minus sign indicates an error towards the left or towards
the bottom. A plus indicates errors towards the right or the top.

Nicoletta Beschin and Others

10

orientation or mirror reversed and presented one at a time for the subject to
read aloud. On all four tests, NL performed at ceiling (see Table III)
Extinction

Visual extinction was assessed with a newly devised automated test (visual
extinction test). NL was asked to fixate a l.5 cm X l.5 cm cross displayed in
the centre of the computer screen. His task was to detect single or double targets
comprising 5 mm white dots displayed for 100 ms. In the case of single targets
these appeared either about five visual degrees to the left or to the right. In the
case of double targets, these involved either one single target on each side, or
two dots presented five degrees apart either on the left or on the right. When
the two dots were presented on the same side, they could be presented one
above the other or side by side, and in this last case the target nearer to the
centre of the screen was about 2.5 visual degrees from the point of fixation.
The order of presentation was randomised. The patient had to decide if one or
two targets had been displayed and then to indicate on which side of the screen
they had appeared. He indicated his decision by pressing one of two different
response-keys with his right hand.
Results are shown in Table IV, from which it is clear that he had difficulty
only with left sided targets when they were presented simultaneously with one
target on the right.
To assess the presence of auditory neglect and of auditory extinction, NL
was presented with sounds via headphones. Ten sounds were presented to the
left ear, ten sounds to the right, and ten sounds to both ears simultaneously.
The order of presentation was random and the sound was always the same
recorded repetitively. Table IV shows that NL had very little difficulty with this
task, although on five occasions he reported that he had heard the stimulus in
both ears when it had only been presented on the right, and in the double
stimulation condition he missed 20% of left stimuli. With few exceptions (e.g.
De Renzi, Gentilini and Pattacini, 1984), this level of performance usually is
TABLE IV

NL's Performance on Tasks Assessing Extinction (Score range is given in parentheses)
NL' s scores

Test (score range)
Visual extinction
Single stimulus presentation
Double unilateral presentation
- Vertical
- Horizontal
Double bilateral presentation
Auditory extinction
Single stimulus presentation
Double bilateral presentation
Tactile extinction
Single stimulus presentation
Double unilateral presentation

Left

Right

(0-10 each side)

10 (100%)

9 (90%)

(0-12 dots each side)
(0-12 dots each side)
(0-8 each side)

II (92%)
12 (100%)
0(0%)

II (92%)
12 (100%)
8 (100%)

(0-10 each side)
(0-10 each side)

10 (100%)
8 (80%)

10 (100%)
10 (100%)

(0-10 each side)
(0-10 each side)

10 (100%)
10 (100%)

10 (100%)
10 (100%)

Representational neglect

11

not considered to be evidence of extinction (see e.g. Bisiach et aI., 1986; Grote,
Pierre-Louis, Smith et aI., 1995; Arboix, Junque, Vendrell et aI., 1990).
We also assessed tactile extinction, but only with his right hand because of
the sensory problems in his left hand (hemianaesthesia). NL was given ten stimuli
on the left side of the back of the hand, ten on the right side and ten on both
sides following a procedure used by Moscovitch and Behrmann (1994). NL's
performance was flawless in all conditions.
In summary, NL showed neither a disorder of perceptual neglect nor of
directional hypometria. In fact he performed normally on all of the tests requiring
movements of the arm on the left side. He showed no asymmetry in perceiving
targets in upper or lower space. It appears that he did show some evidence of
visual extinction but did not show auditory of tactile extinction, although tactile
extinction cannot be completely ruled out as we were unable to test his left
hand. It is worth emphasising that extinction was clearly dissociated from visuospatial neglect. This finding of visual extinction in the absence of visuo-spatial
neglect adds to other similar cases which can be gleaned from the literature
(Vallar and Perani, 1986; Barbieri and De Renzi, 1989; Vallar, Rusconi and
Bisiach, 1994a; Di Pellegrino and De Renzi, 1995). This distinction coupled
with reports of the converse, namely visuo-spatial neglect without extinction
(Kartsounis and Findley, 1994), supports the notion that the two phenomena
have to be considered symptoms of two different processes.
Representational Neglect
Representations from Long-term Memory. Data for the tests of
representational neglect are shown in Table V. NL was given a spelling test in
which he was presented aurally with 80 words and 10 non-words, and had to
respond by spelling each word or non-word aloud. Thirty of the words were
TABLE v

NL's Performance on Tasks Assessing Representational Neglect (Score range is indicated in
parentheses)

Test (score range)
Mirror reversed reading (0-30)
Spelling test (0-90 word error score)
Letters in 24 words spelled incorrectly
Left (0-75 letter error score)
Right (0-75 letter error score)
Centre (0-11 letter error score)
Verbal description of complex scene
(Scores Left 0-22, Right 0-18)
Immediate recall
Recall delay of 15 minutes
Recall delay of 2 days
Tactile maze
Eyes closed (0-48) test
retest
Eyes open (0-49)

NL' s scores

Cut-off

30
24
(L= 11; R=9 ; mid=4)
23
13
6
L=5
L= 1*
L=2*

R=5
R=4
R=4
32*
29*
48

Cut-off is the score two standard deviations below the mean score of the control group.
L = left; R = right; mid = middle of the word.
* Score below cut-off.

L=5.0
L=4.0
L=4.5

R=3.l8
R=2.28
R=0.28

45.31

12

Nicoletta Beschin and Others

chosen to have transparent sound to spelling correspondence in Italian (e.g.
c-a-s-a). The remaining words and non words had no such simple correspondence
(e.g. scienza - the initial sc is pronounced as "sh" in English). From Table V
it is clear that the patient made errors on 24 of the items. Within those 24 items,
Table V also shows the number of letters incorrect according to whether they
would have been on the left, on the right or in the middle of the words, had
they been printed or represented in a visual mental image. Although there are
no control data for this test, more errors were derived from the left of the words.
NL showed clear differences between the left side and the right side of mental
images of very familiar piazzas (Bisiach and Luzzatti, 1978; Guariglia et aI.,
1993) and geographical areas (Halligan and Marshall, 1992; Bartolomeo et aI.,
1994). In one form of the task, NL was asked to mentally visualise a piazza
(see Figure 2) in the town where he lives and with which he is very familiar.
In a second version of the task he was asked to imagine himself in Sardegna
looking towards Roma or vice versa (see Figure 3). When he had to describe
them from a given vantage point (Figures 2a and 3a), he consistently reported
several elements on the right side (8 details for the image of the piazza and 4
details for the geographical locations) and very few on the left side (2 details
for the piazza and 9 details for the geographical image). Ten days after this
first test session the patient was asked to describe the same mental images from
the opposite perspective (Figures 2b and 3b), he again reported more elements
on the right (9 details for the piazza and 3 details for geographical locations)
than on the left (2 details for the piazza and I detail for the geographical
locations). The ten day delay between changes of viewpoint was introduced to
avoid the possibility that he used verbal labels for some of the details when
changing between vantage points (Logie, 1995). On these tasks the patient
showed some characteristics of neglect behaviour. Specifically he always started
to describe the images from the right side, he was much more accurate and he
spent more time reporting the details from the right of the imagined scene. On
two occasions he reported details that were on the right as being on the left or

Fig. 2 - Maps of a familiar piazza described from mental imagery. The viewpoint is indicated by
the arrow pointing in the imaged direction. Numbers indicate the order in which the details were
reported.

Representational neglect

V.ugo point a

13

VM1tagc point b

Fig. 3 - Maps of Italy described from mental imagery. The viewpoint, from the Sardegna coast
(a) and from Roma (b), is indicated by the arrow pointing in the imaged direction. Numbers indicate
the order in which the details have been reported. Numbers surrounded by a square refer to the details
transferred from the right to the left side.

in the centre. For example when describing his image from Sardegna he said
" ... on the left of Roma, towards the North, there is Napoli...". The city of
Napoli is actually south of Roma (to the right from Sardegna), and this would
be familiar to most Italians.
Another interesting aspect of NL's visual imagery ability comes from
drawings from memory. He was asked to use a pencil to draw each of the
following: A tractor, a house, a daisy, a watch, a telephone, a person, a horse,
a bear, a table, a car, a cat, and a map of Italy. Using coloured pencils he also
spontaneously drew a landscape with which he was familiar. In all cases there
were striking differences in the level of detail depicted on the left and right.
The right side of his drawings were very detailed, but on the left he missed
salient elements and those details which were present tended to be drawn in
the wrong place (see Figure 4a). After 20 days NL was shown some of his
drawings and was asked to describe them. He recognised that they had been
drawn by him and he identified omissions and erroneous spatial relationships
among elements of the drawings. One notable example was his description of
the drawing of the landscape that he used to see when he was driving toward
his holiday house (Figure 4b): he noticed that the picture was unbalanced on
the right side of the sheet but when he was asked the reason for this, he
answered: " ... the left of the mountain is all grey because it is all rock ... ". One
possible alternative interpretation is that the difficulty shown in these tasks arose
from a sensory motor processing difficulty.
However were this the case, we would have expected the difficulty also to
show up in the copying tasks described earlier. Therefore the impairment seems
to arise when NL has to rely on information in memory.

Nicoletta Beschin and Others

14

3

3

a

b

Fig. 4 - Drawings from memory of a persoll (a) and of a familiar landscape (b). In picture a.
the patient drew two ears on right side. Picture b was originally in colour. and in this reproduction
numbers indicate the colours that the patient used: 1 = green; 2 = brown; 3 = blank.

Representations Based on Novel Material. The pattern described thus far
suggests that NL showed evidence of neglect when performing representational
tasks based on mental images processed and stored in long term memory before
the lesion. The next obvious question is how he coped with representational
tasks that required the generation of new mental representations of stimuli to
which he had been exposed only in the testing sessions and he had not seen
before. NL was asked to remember objects, people and animals shown in a
complex picture which depicted a farm scene. The picture had 22 discrete items
on the left and 18 on the right. He examined the picture in silence for 2 minutes
after which the picture was covered and he was asked for immediate spoken
recall of the items in the picture. Items were scored correctly only if he recalled
a complete item (e.g. a tree, a tractor, a hen etc.). After recalling as many items
as possible, he was asked to produce a drawing of the picture from memory,
using a blank sheet identical in size to the original scene (25 X 24 cm). Results
are reported in Table V. The number of items reported orally was the same
from both sides of the picture but in his drawing from memory the spatial
relations were incorrect among the details from the left of the picture. Subsequent
spoken recalls were required 15 minutes and two days later. In contrast to
immediate recall, on both occasions he recalled fewer details from the left of
the picture than from the right (see Table V). It is clear from the cut off scores
shown in Table V that, if anything, the controls recalled slightly more items
from the left than from the right of the picture.
NL's ability to create and manipulate mental representations of new

Representational neglect

15

information also was assessed by a tactile task, the Tactile Exploration Maze
Test (Beschin, Cazzani, Cubelli et al., 1996). This test has been thought to rely
on spatial representation and not visual perception (De Renzi, Faglioni and Scotti,
1970). The test involves a 33 cm square maze cut into a 40 cm by 50 cm
wooden board (see Figure 5c). Four alleys (2 cm wide and 3 mm deep) spread
from the centre of the maze branching into eight lateral arms, each of which
ends in a 10.5 cm segment. At the extreme ends of each segment there is a
hollow designed to hold a 2.4 cm marble which is used as a target. Each target
position is therefore at the same distance from the centre. NL had his forefinger
in the centre of the maze and he was asked to move it along the alleys of the
maze. His task was to find as quickly as possible a marble placed in one of
the extremities of the maze. The whole test consisted of 48 randomised trials
(three for each of the 16 possible positions). The time limit for each trial was
120 sec and any failure to locate the target within the time limit was scored as
an omission. A frequency of omissions on the left side greater than 3 times that
on the right was considered evidence of neglect (Beschin et al., 1996). The test
was given three times. On the first two occasions (which were one month apart),
visual input was prevented (the patient was blindfold). During the first test NL
missed the marble 16/48 times, 3 times when it was on the right side and 13
times when it was on the left. On the second occasion NL missed the marble
19/48 times, l7 times when it was on the left side and two on the right. Normal
subjects are virtually at ceiling on this task (Beschin et al., 1996). At the end
of this session NL was required to draw what he thought the unseen maze looked
like. NL's drawing of the maze is shown in Figure Sa, along with a normal
subject's typical reproduction (5b) and the real maze configuration (5c). NL's
representation is completely devoid of details from the left side of the maze.
On the third test session, with eyes open and uncovered, NL found all the
marbles (48/48) without difficulty. In this condition he took longer to search
on the right (mean = 4.2 s) than on the left (mean = 3.3 s). However the bottom
right corner of the maze was occluded by his right elbow during much of the
task, and when the data based on this part of the maze were excluded, there
was no difference in the mean search times for left and right (F= 3.85; d.f. = 1,
38; ns). In either case, there is no hint of a left neglect in his searching time
with eyes open.
In summary, NL showed no evidence of perceptual neglect but he did show
a unilateral neglect restricted to visual imagery similar to the pattern reported
by Guariglia et al. (1993). However unlike the patient described by Guariglia
and colleagues, NL also showed a representational neglect in tasks that require
the creation and manipulation of new mental representations.
Brooks Matrix

As mentioned in the introduction, one focus of this paper is to explore the
feasibility of the visuo-spatial working memory model hypothesis in the context
of representational neglect. One of the tasks most widely used in the vi suospatial working memory literature (Baddeley, Grant, Wight et al., 197 Sa;
Baddeley and Lieberman, 1980; Farah, Hammond, Levine et al., 1988; Hanley

16

Nicoletta Beschin and Others

0~
. --Y

\

Q

c -\
D

_\

~
a

b

Fig. 5 - Graphical reproduction of tactile maze. (a) NL's reproduction. (b) performance of a
typical normal subject. (c) configuration of the real maze in which the x indicates the starting point.

and Pearson, 1990; Hanley, Young and Pearson, 1991) was originally devised
by Brooks (1967). A spatial version of this task has been shown to rely heavily
on visuo-spatial working memory, while a more verbally oriented version relies
on the verbal working memory system. In particular, performance of the spatial
version of the task has been shown to be disrupted by concurrent overt movement
such as tracking (Baddeley et al., 1975a), or by iapping out a pattern on a table
(e.g. Quinn and Ralston, 1986). This pattern of data has been taken to support
the view that the production of overt movement and the encoding and recall of
imagined spatial patterns rely on visuo-spatial worki :1g memory (for a review
see Logie, 1995).
In the Brooks' task, the subject is asked to im,,~;'"~ placing consecutive
numbers in consecutive squares of a four by four matrix, st:,;ting from the second
column on the second row. In one (spatial) version of the task, NL listened to
the instructions as to the sequence of directions around the matrix, moving from
square to square (e.g. up-2, right-3, down-4). Once the prese, . ·~ d sequence was
completed the patient responded with verbatim vocal recall of the series of
directions and numbers. Next, a complementary (verbal) form was used where
the patient again had to retain a sequence of sentences, except that the words
"up, down, left and right" were replaced by the words "good, bad, slow and
quick". This resulted in a set of nonsense items (e.g. good-2, slow-3, bad-4),
which were to be retained without the use of any form of visual image, perhaps
by subvocal rehearsal. In both conditions, NL was presented with sequences of
increasing length (range 2-9) with 3 trials for each length of sequence and the
score (span) was the maximum length for which NL correctly recalled at least
two out of three sequences.
NL's scores are shown in Table VI together with those of the control group
matched for age. His performance was in the normal range in the verbal task,
but when performing the spatial task he obtained a score which was two standard
deviations below the mean for the controls.
To investigate if these results arose because of heavy reliance on visuo-spatial
memory in the spatial version of the task, NL was again given the spatial task
but with the matrix present. This then provided perceptual information (stimulus

Representational neglect

17

TABLE VI

NL's Pelformance on the Brooks Matrix Test (Score range is indicated in parentheses)

Conditions (score range)

Verbal span (0-9)
Spatial span (0-9)
Spatial span with stimulus support (0-9)
"No-memory" perceptual task (0-9)
"No-memory" imagery task (0-9)

NL's scores

Cut-off

Controls'
mean

Controls'
standard
deviation

3
2*
8
8
2*

2.40
2.46
2.48
7.62
5.00

3.2
4.8
6.4
8.6
6.6

0.4
1.2
1.9
0.5
0.8

Cut-off is the score two standard deviations below the mean score of a control group.
* Score below cut-off.

support) about the form of the matrix, thereby reducing the load on the visuospatial memory system. Under these conditions, NL's performance improved
dramatically (see Table VI). Next NL was tested with two further tasks in which
he was not required to retain the sequence of directions (no memory load). NL
was given nine sequences and each sequence continued until the patient reported
that the path described by the sequence of directions led off the edge of the
matrix. In one case he could see the matrix during the task ("no memory
perceptual task"), and his performance did not differ from the mean of the control
subjects (see Table VI). In the other case there was no stimulus support provided
and the patient had to form and retain a mental image of the matrix ("no-memory
imagery task"). His scores in this condition fell two standard deviations below
the mean of the control group.
In conclusion, NL demonstrated that he was able to perform spatial "scan"
operations when he was provided with external support via visual perception
and did not have to rely heavily on visual imagery.

DISCUSSION

The main goal of the present study was to document the existence of
unilateral representational neglect in the absence of visual perceptual neglect,
thereby indicating that representation and perception can be dissociated in this
form of deficit (Brain, 1941; Anderson, 1993; Guariglia et aI., 1993). The patient
we have described appears to offer a body of evidence that is entirely consistent
with this main goal. He had no apparent perceptual neglect from two months
after the stroke, and as far as we were able to discern from the reports of relatives
and medical and nursing staff, there were no signs of perceptual neglect from
the onset of his stroke. Yet he had great difficulty in reporting details from the
left side of an imaged representation based either on tactile or on visual
information. This was true when 3e was asked to generate mental images drawn
from information in his long-term memory. It was also true when he attempted
to generate images from novel visual material which had been learned (longterm memory) or which had been seen only seconds before (short-term memory).
Conversely, perceptual neglect has been reported in a number of patients in the
absence of representational neglect (e.g. Bartolomeo et aI., 1994). Therefore it

18

Nicoletta Beschin and Others

is highly unlikely that patient NL would have suffered from perceptual neglect
at the onset of his stroke but then have recovered from this relatively quickly
while continuing to be affected by representational neglect.
The site of the patient's lesion in the right parietal lobe is in agreement with
the proposal by Farah et al. (1989) and Kosslyn (1987, 1991) that the parietotemporal region is a possible locus for the neural structures involved in mental
imagery. However the patient reported by Guariglia et al. had a lesion in the
frontal lobe, and therefore the neuroanatomical mappings for imagery are by
no means clear (Marshall and Halligan, 1993). Also, the dissociation found
between perception and imagery in both patients appears to present a challenge
to theories of imagery and of visuo-spatial working memory which assume that
perception and imagery rely on overlapping cognitive systems. This challenge
is however more apparent than real. In neither theory do the cognitive systems
completely overlap, thereby allowing scope for aspects of imagery to be impaired
while perception is intact and vice versa. So too, the neural structures and
pathways that have been identified as being responsible for visual perception
do appear to project on to the neuroanatomical areas linked respectively with
object identity and object location (Ungerleider and Mishkin, 1982) and then
to a region of the posterior parietal cortex (e.g. Stein, 1992; Zeki and Shipp,
1988), which as we have mentioned above, has been linked to imagery. Therefore
a lesion in the "imagery areas" might affect mental imagery but leave much of
perception intact, while a more extensive lesion in the same region might affect
both perception and imagery.
A second goal of this study was to explore whether the concept of visuospatial working memory could offer a vehicle for a theoretical account of
representational neglect (with or without perceptual neglect). It is clear that the
representational deficits in patient NL could not have arisen as a result of
impoverished visual perceptual input, because for several of the tests the
necessary information was drawn from information encoded in long-term
memory before the stroke. Nor could the deficit be the result of a loss of
information stored in long-term memory because he could access visual details
of familiar piazzas simply by changing his imagined viewpoint. This finding
suggests that all of the information was available but not all of it was directly
accessible as it is in the healthy brain. Consistent with the speculation by
Baddeley and Lieberman (1980) mentioned in the introduction, these observations
point to a deficit in the activation of a part of the "mental screen" where the
image would be represented or to a deficit of the process of scanning this
"screen" .
Information relevant to this argument is provided by NL's performance on
the Brooks (1967) matrix task which is thought to tap the resources of visuospatial working memory (however see Salway and Logie, 1995, for a caveat).
NL performed very poorly on the spatial version of this task, with scores close
to floor. Notably he performed extremely well when he had stimulus support
available in the form of the blank matrix, and did not have to rely heavily on
visuo-spatial working memory. This effect of stimulus support also can account
for NL's normal performance on the Corsi Blocks task in which the blocks
themselves are visible throughout. Stimulus support is thought to reduce rather

Representational neglect

19

than to remove the temporary memory load, because NL would still have to
retain the sequence of blocks. This account parallels the positive effects of
stimulus support (e.g. a drawing or other physical stimulus) on working memory
which have been shown in a range of imagery tasks with normal subjects
(Chambers and Reisberg, 1985, 1992; Pearson, Logie and Green, 1996; for a
review see Comoldi, Logie, Brandimonte et aI., 1996). Therefore visuo-spatial
working memory can be supported by perception but perception is not as closely
linked to working memory or to mental imagery as has widely been assumed.
Hence, the enhanced performance with stimulus support found in normal subjects
is consistent with the higher scores on Corsi block performance found in NL
when the blocks were available to provide stimulus support. In sum, NL's deficit
can be interpreted as suggesting that he has damage to important aspects of
visuo-spatial working memory but that impairments are not apparent when he
has support from his intact perceptual system.
The hypothesis that representational neglect reflects a deficit in vi suo-spatial
working memory applied not just to the impairment of our own patient NL, but
also to patients who suffer from both perceptual and representational neglect.
For example, the patients reported by Bisiach and Luzzatti (1978) showed
symptoms of perceptual neglect on some tests. However, when they failed to
describe items on the left side of the "Piazza del Duomo" as imagined from a
given vantage point, their representations were derived from knowledge stored
in long-term memory prior to their brain damage. Therefore their difficulty with
the imaged left side of Piazza del Duomo could not have arisen from an
impairment of their visual perceptual abilities at the time of the test. We are
left with the inevitable conclusion that there was damage to the system
responsible for generating and maintaining the temporary visual mental
representation or for extracting information from that representation. That is,
the patients had damage to their visuo-spatial working memory system which
was independent of any damage to their perceptual system.
How might we further specify the nature of the deficit in such a system? It
is worth stressing that the visuo-spatial working memory hypothesis of neglect
has never been fully developed. One possible reason for this is that until recently
the characterization of visuo-spatial working memory was poorly specified,
although this situation is now changing (e.g. Smyth and Pendleton, 1989; Hanley
et aI., 1990, 1991; Logie, 1995; Salway and Logie, 1995; Quinn and McConnell,
1996). Logie (1995, Reisberg and Logie, 1993) has argued that visuo-spatial
working memory may be further fractionated into a partnership of two specialised
components, one referred to as the "visual cache" which stores information about
visual form and colour. The second component deals primarily with spatial and
movement information and is referred to as the "inner scribe". This latter partner
in the system is thought to rehearse the contents of the visual cache, to transfer
information from the visual cache to the central executive, and to be closely
involved in the planning and execution of body and limb movements. Damage
to the visual cache would permit the inner scribe to rehearse or transfer only
an impoverished representation. On the other hand, damage to the inner scribe
would cause great difficulty in retaining visual information, in transferring visual
information to other parts of working memory such as the central executive,

20

Nicoletta Beschin and Others

and in carrying out tasks that require retention of sequences of movements in
the absence of stimulus support. Within this framework, NL's performance with
the various versions of the Brooks' (1967) task could be interpreted as pointing
to impairment of the visual cache. For example, he could retain the spatial
sequence, possibly in an intact inner scribe, when he had support in the form
of the blank matrix for a damaged visual cache. His difficulties with the "piazza"
tests and other tests of mental representation are also consistent with the idea
that information was available in the visual cache but not to the rest of working
memory. This interpretation is speculative, but it does offer a coherent framework
within which to understand the nature of representational neglect whether or
not it is accompanied by perceptual neglect.
This interpretation requires a revision of the visuo-spatial working memory
hypothesis to account for how information becomes available to working
memory. The most widely held assumption is that working memory serves as
a form of gateway between perceptual input and long-term memory. This
assumption was explicit in the early models of short-term memory (Atkinson
and Shiffrin, 1968), is maintained in most introductory textbooks on memory
(e.g. Gross, 1992; Atkinson, Atkinson, Smith et aI., 1993; Searlman and
Hermann, 1994), is a feature of Kosslyn's theory of visual imagery (e.g. Kosslyn,
1991), and is implicit in the working memory model (e.g. Baddeley, 1986).
However the "gateway" model of working memory would have difficulty
interpreting representational neglect because it does not depend upon perceptual
input. Moreover perceiving and interacting with the environment demands the
generation of mental representations of that environment. Such representations
would incorporate the location of various objects in relation to one another and
the location of objects in relation to the observer. Thus when retrieving information
about objects or scenes held in long-term memory, the mental representation
contains information about the scenes such as when they were previously
experienced, the names of objects depicted and so on (e.g. Chambers and Reisberg,
1985, 1992; Reisberg and Logie, 1993; Comoldi et aI., 1996). In this sense the
contents of working memory necessarily comprise some of the products of
perception and subsequent mental processing, including the semantic properties.
Working memory can then be considered as a multi-component workspace which
stores and manipUlates information which has been activated from the long-term
store, rather than as a primary route into long term memory from the environment.
Therefore working memory is better viewed as a system accessed by information
stored in long term memory and by sensory input only after that input has been
processed by long term memory (for more detailed discussion of this argument
see Logie, 1995, 1996; Ellis, Della Sala and Logie, 1996).
Moreover, if we see the visual cache and inner scribe as temporary visual
and spatial stores which are used either after perceptual input has been removed
or when vi suo-spatial information is extracted from long-term memory, then
this offers an account of the distinction between perception and mental
representation.
Experimental support for this view comes from a wide variety of sources
(e.g. Ellis, Plude and Young, 1987; Marshall and Halligan, 1988; Farah, Monheit

Representational neglect

21

and Wallace, 1991; Berti and Rizzolatti, 1992; McGlinchey-Berroth, Milberg,
Verfaellie et at, 1993; Vallar, Rusconi and Bisiach, 1994b) which have
demonstrated that on-line processing of information is accomplished by means
of access to stored knowledge in long-term memory, without the information
necessarily being available within working memory. For example, McGlincheyBerroth et at (1993) gave two neglect patients a lexical decision task, with
word or non-word letter strings presented centrally. Prior to the presentation of
the letter strings' pictures appeared in either the right or the left visual field.
Lexical decision times for real words were faster when the pictures (presented
in either hernifield) matched the meaning of the word. Therefore the pictures,
even on the neglected side, appeared to activate relevant semantic information
in long term memory and the level of activation was sufficient to affect on-line
cognition for lexical decisions.
These data, together with those cited above, are consistent with the view that
perceptual input is not processed in working memory prior to long-term memory
access. These data are inconsistent with the idea of working memory as a form
of gateway, as are data from the substantial literature on normal adults
demonstrating priming effects (see e.g. Roediger and McDermott, 1993) and preconscious priming effects (e.g. Marcel, 1983a, 1983b). It is worth noting that the
view that the contents of working memory comprise activated information from
long-term memory already appears in alternative models of working memory (e.g.
Cowan, 1988; Hasher and Zacks, 1988; Just and Carpenter, 1992; for a detailed
discussion see Richardson, Engle, Hasher et at, 1996).
How successfully can this modified form of vi suo-spatial working memory
be applied to unilateral neglect? As stated above, it is by no means uncommon
to find patients who have both perceptual neglect and representational neglect.
One possible account of their deficit is that, as a consequence of their perceptual
neglect, perceptual input reaches the long-term store but input from the left side
is absent or activated less efficiently. The asymmetry in the levels of activation
would then be passed on to the on-line representation system in working memory
which (we would argue) comprises the visual cache, the inner scribe and the
central executive of working memory. This could explain why representational
deficits are so often associated with perceptual ones. Indeed on this view it
would be possible in principle for a patient to show evidence of a representational
deficit arising entirely from perceptually based neglect, that is with unilaterally
impoverished perceptual input being dealt with by an intact representational
system. Whether this form of dissociation is detected depends on the range of
tasks given to the patient, since a representational impairment can only be
revealed by tasks which present information not known to the patient before
onset of the lesion.
Representational neglect cannot only be the result of a perceptual deficit
because, in some patients, the representational problems arise in tasks that require
the retrieval of information acquired prior to the onset of the lesion (e.g. Piazza
del Duomo test). Moreover our patient NL shows representational neglect in
the absence of perceptual input difficulties.
The same appears to be true of the Guariglia et at (1993) patient.

22

Nicoletta Beschin and Others

Our hypothesis would run into difficulty were a patient to be found showing
perceptual neglect in the absence of representational neglect for material
presented after onset of the lesion. Some studies (Anderson, 1993; Bartolomeo
et al., 1994) described patients who showed left neglect in perceptual tasks but
not in imagery tasks. Anderson reported an instance of a clockface drawing in
which the patient's performance was better with eyes shut (imagery task) than
with eyes open (perceptual task) (see also Chedru, 1976, and Mesulam, 1985).
Bartolomeo et al. reported a group of 14 out of 60 patients who performed
poorly on traditional perceptual tasks (such as line cancellation) but who did
not show representational neglect during descriptions from memory of familiar
piazzas, countries and towns. Although these results suggest that perceptual
neglect can be found in the absence of representational impairment, Anderson's
(1993) and Bartolomeo et al.'s (1994) clinical cases did not show the dissociation
between perceptual and representational processing that is the focus of our
discussion. In fact, information used in the imagery tasks had been encoded
prior to the onset of the brain lesion, and information from the left visual field
would have been stored normally in long-term memory. The crucial test would
be to ask the patients to describe from memory an unfamiliar pattern or array
of objects which they had just seen. As far as we are aware no patient has yet
been described showing perceptual neglect in the absence of representational
neglect on this kind of task (Marshall and Halligan, 1993).
Marshall and Halligan (1992) argued that the understanding of neglect
suffered from a lack of explanatory constructs due largely to the range of
dissociations based on test performance. The working memory approach does
not provide a complete account of the many behaviours that have been classified
as different forms of neglect. However, the data from patient NL coupled with
a reconsideration of previously reported cases of neglect suggest that visuospatial working memory may play an important role in and may offer the basis
for an account of unilateral representational neglect.
Acknowledgements. We thank L. Salvato, E. Brusini and physiotherapist A. Maffioli for
referring the clinical case. We thank also M. Laiacona who provided the stimuli used for
the word reading tests and E. De Renzi for his valuable comments on an earlier version of
our paper. We are also grateful to the University of Bergen, Norway, for providing office
and library facilities for the fourth author during the production of this manuscript.
Most of all, we are grateful to NL who consented to the lenghy testing in the clear
understanding that finding would not be of any help to him.

REFERENCES
ALBERT, M.L. A simple test of visual neglect. Neurology, 23: 658-664, 1973.
ANDERSON, B. Spared awareness for the left side of internal visual images in patients with left-sided
extrapersonal neglect. Neurology, 43: 213-216, 1993.
ARBOIX , A., JUNQUE, C., VENDRELL, P., and MARTf-VILALTA, J.L. Auditory ear extinction in lacunar
syndromes. Acta Neurologica Scandinavica, 81: 507-511 , 1990.
ARRIGONI, c., and DE RENZI, E. Constructional apraxia and hemipheric locus of lesion. Cortex, 1:
170-197, 1964.
ATKINSON, R.C., and SHIFFRIN, R.M. Human memory. A proposed system and its control processes.
In K.W. Spence and J.T. Spence (Eds.), The Phychology of Learning and Motivation. New York:
Academic Press, 1968, 2, pp. 89-105.

Representational neglect

23

ATKINSON, R.L., ATKINSON, R.C, SMITH, E.E., and BEN, D.J. Introduction to Psychology, 11th ed.
Orlando: Harcourt Brace College Publishers, 1993.
BADDELEY, A.D. Working Memory. Oxford: Oxford University Press, 1986.
BADDELEY, A.D. Human Memory: Theory and Practice. Hove: Lawrence Erlbaum Associates, 1990.
BADDELEY, A.D. , GRANT, S., WIGHT, E., and THOMSON, N. Imagery and visual working memory. In
P.M.A. Rabbitt and S. Doric (Eds.), Attention and Peiformance, V. London: Academic Press, 1975a,
pp. 205-217.
BADDELEY, A.D., and HITCH, G.J. Working memory. In G. Bower (Ed.), The Psychology of Learning
and Motivation. New York: Academic Press, 1974, vol. VIII, pp. 47-90.
BADDELEY, A.D., and LIEBERMAN, K. Spatial working memory. In R.S. Nickerson (Ed.), Attention and
Peiformance, VIII. Hillsdale, NJ: Lawrence Erlbaum Associates, 1980, pp. 521-539.
BADDELEY, A.D., THOMSON, N., and BUCHANAN, M. Word length and the structure of short memory.
Journal of Verbal Learning and Verbal Behavior, 14: 575-589, 1975b.
BARTOLOMEO, P., D'ERME, P., and GAINOTTI, G. The relationship between visuospatial and
representational neglect. Neurology, 44: 1710-1714, 1994.
BASSO, A., CAPITANI, E., and LAIACONA, M. Raven's coloured progressive matrices: normative values
on 305 adult normal controls. Functional Neurology, 2: 189-194, 1987.
BARBIERI, C, and DE RENZI, E. Patterns of neglect dissociations. Behavioural Neurology, 2: 13-24, 1989.
BARBUT, D., and GAZZANIGA, M.S. Disturbances in conceptual space involving language and speech.
Brain, 110: 1487-1496, 1987.
BAXTER, D.M., and WARRINGTON, E.K. Neglect dysgraphia. Journal of Neurology, Neurosurgery and
Psychiatry, 46: 1073-1078, 1983.
BERT!, A., and RIZZOLATTI, G. Visual processing without awareness: Evidence from unilateral neglect.
Journal of Cognitive Neuroscience, 4: 345-351, 1992.
BERTOLANI, L., DE RENZI, E., and FAGLIONI, P. Test di memoria non verbale di impiego diagnostico
in clinic a: taratura su soggetti normali. Archivio di Neurologia, Psicologia e Psichiatria, 54: 477486, 1993.
BESCHIN, N., CAZZANI, M., CUBELLI, R., DELLA SALA, S., and SPINAZZOLA, L. Ignoring left and far:
An investigation on tactile neglect. Neuropsychologia, 34: 41-49, 1996.
BISIACH, E., CAPITANI, E., LUZZATTI, C, and PERANI, D. Brain and conscious representation of outside
reality. Neuropsychologia, 4: 543-551, 1981.
BISIACH, E. Mental representation in unilateral neglect and related disorders: The twentieth Barlett
memorial lecture. Quarterly Journal of Experimental Psychology, 46: 435-461, 1993.
BISIACH, E., and LUZZATTI, C. Unilateral neglect of representational space. Cortex, 14: 129-133, 1978.
BISIACH, E., LUZZATTI, C., and PERANI, D. Unilateral neglect, representational schema and consciousness.
Brain, 102: 609-618, 1979.
BISIACH, E., and RUSCONI, M.L.D. Break-down of perceptual awareness in unilateral neglect. Cortex,
26: 643-649, 1990.
BISIACH, E., VALLAR, G., PERANI, D., PAPAGNO, C., and BERTI, A. Unawareness of disease following
lesions of the right hemisphere: anosognosia for hemiplegia and anosognosia for hemianopsia.
Neuropsychologia, 24: 471-482, 1986.
BRAIN, W.R. Visual disorientation with spatial reference to lesions to the right cerebral hemisphere.
Brain, 64: 244-272, 1941.
BROOKS, L.R. The suppression of visualization by reading. Quarterly Journal of Experimental
Psychology, 19: 289-299, 1967.
CARAMAZZA, A., and HILLIS, A. Spatial representation of words in the brain implied by studies of a
unilateral neglect patient. Nature, 346: 267-269, 1990a.
CARAMAZZA, A., and HILLIS, A. Levels of representation co-ordinate frames, and unilateral neglect.
Cognitive Neuropsychology, 7: 391-445, 1990b.
CHAMBERS, D., and REISBERG, D. Can mental images be ambiguous? Journal of Experimental
Psychology: Human Perception and Performance, Jl: 317-328, 1985.
CHAMBERS, D., and REISBERG, D. What an image depicts depends on what an image means. Cognitive
Psychology, 24: 145-174, 1992.
CHEDRU, F. Space representation in unilateral spatial neglect. Journal of Neurology, Neurosurgery and
Psychiatry, 39: 1057-1061, 1976.
CORNOLDI, C, LOGIE, R.H., BRANDIMONTE, M.A., KAUFMANN, G., and REISBERG, D. Stretching the
Imagination: Representation and Transformation in Mental Imagery. New York: Oxford University
Press, 1986.
COWAN, N. Evolving conceptions of memory storage, selective attention, and their mutual constraints
within the human information-processing system. Psychological Bulletin, 104: 163-191, 1988.
DELLA SALA, S., and LOGIE, R.H. When working memory does not work: The role of working memory
in neuropsychology. In F. Boller and H. Spinnler (Eds.), Handbook of Neuropsychology. Amsterdam:
Elsevier Publishers BV, 1993, 8, pp. 1-63.

24

Nicoletta Beschin and Others

DE RENZI , E., FAGLIONI, P., and SCOTII, G. Hemispheric contribution to exploration of space through
the visual and tactile modality. Cortex, 6: 191-203, 1970.
DE RENZI, E., GENTILINI , M., and PAITACINI, F. Auditory extinction following hemisphere damage.
Neuropsychologia, 22: 733-744, 1984.
DE RENZI, E., and SPIN NLER, H. Visual recognition in patients with unilateral cerebral disease. Journal
of Nervous and Mental Diseases, 142: 515-523, 1966.
DILLER, L., and WEINBERG, J. Hemi-inattention in rehabilitation: The evolution of a rational remediation
program. In E.A. Weinstein and R.P. Friedland (Eds.), Hemi-inattention and hemisphere
specialization, Advances in Neurology. New York: Raven Press, 18: 63-82, 1977.
01 PELLEGRlNO, G., and DE RENZI, E. An experimental investigation on the nature of extinction.
Neuropsychologia, 33: 153-170, 1995.
ELLIS, A., FLUDE, B., and YOUNG, A. "Neglect dyslexia" and the early visual processing of letters,
words, and non-words. Cognitive Neuropsychology, 4: 439-464, 1987.
ELLIS, A.x., DELLA SALA, S., and LOGIE, RH. The bailiwick of visuo-spatial working memory: Evidence
from unilateral spatial neglect. Cognitive Brain Research, 3: 71-78, 1996.
FARAH, M.J. The neurological basis of mental imagery: A componential analysis. Cognition, 18: 245272, 1984.
FARAH. M.J., HAMMOND, K.M. , LEVINE, D.N., and CALVANIO, R Visual and spatial mental imagery:
dissoci able systems of representation. Cognitive Psychology, 20: 439-462, 1988.
FARAH, M.1., MONHEIT, M.A., and WALLACE, M.A. Unconscious perception of "extinguished" visual
stimuli: Reassessing the evidence. Neuropsychologia, 29: 949-958, 1991.
FOLSTEIN, M.F., FOLSTEIN, S.E., and McHUGH, P.R. Mini-Mental State: A practical method for grading
the mental state of patients for the clinician. Journal of Psychiatric Research, 12: 189-198, 1975.
GAUTHIER, L. , DEHAUT, F. , and JOANEITE, Y. The Bells test: A quantitative and qualitative test for
visual neglect. International Journal of Clinical Neuropsychology, I: 49-54, 1989.
GROSS, RD. Psychology. The Science of Mind and Behaviour (2nd ed.). London: Hodder and Stoughton,
1992.
GROSSI, D., MODAFFERI, A., PELOSI, L., and TROJANO, L. One of the different roles of the cerebral
hemispheres in mental imagery: The "o'clock test" in two clinical cases. Brain and Cognition, 10:
18-27, 1989.
GROTE, e.L., PIERRE-LoUIS, S.J., SMITH, M.e., ROBERTS, R.1 ., and VARNEY, N.R Significance of
unilateral ear extinction on the dichotic listening test. Journal of Clinical and Experimental
Neuropsychology, 17: 1-18, 1995.
GUARIGLIA, C., PADOVANI, A., PANTANO, P., and PIZZAMIGLIO, L. Unilateral neglect restricted to visual
imagery. Nature, 364: 235~237, 1993.
HALLIGAN, P.W., and MARSHALL, J.C. Left visuo-spatial neglect: a meaningless entity? Cortex, 28:
525-535, 1992.
HANLEY, 1.R, PEARSON, N.A., and YOUNG, A.W. Impaired memory for new visual forms. Brain, ]]3:
1131-1148, 1990.
HANLEY, I.R, YOUNG, A.W., and PEARSON, N.A. Impairment of the visuo-spatial sketch pad. Quarterly
Journal of Experimental Psychology, 43: 101-125, 1991.
HASHER, L., and ZACKS, R.T. Working memory, comprehension and aging: A review and a new view.
The Psychology of Learning and Motivation, 22: 193-225, 1988.
HEcAEN, H.,TZORTZIS, e., and RONDOT, P. Loss of topographical memory with learning deficits. Cortex,
16: 525-542, 1980.
JUST, M., and CARPENTER, P. A capacity theory of comprehension: Individual differences in working
memory. Psychological Review, 99: 122-149, 1992.
MRTSOUN IS, L.D., and FINDLEY, L.J. Task specific visuospatial neglect related to density and salience
of stimuli. Cortex, 30: 647-659, 1994.
KINSBOURNE, M. A model for the mechanism of unilateral neglect of space. Transactions of the American
Neurological Association, 95: 143-146, 1970.
KOSSL YN, S.M. A cognitive neuroscience of visual cognition: Further developments. In R.H. Logie and
M. Denis (Eds.), Mental Images in Human Cognition. Amsterdam: Elsevier, 1991, pp. 352-381.
LOGIE, R.H. Visuo-spatial Working Memory. Hove, UK: Lawrence Erlbaum Associates, 1995.
LOGIE, R.H. The seven ages of working memory. In J.T.E. Richardson, RW. Engle, L. Hasher and
R.H. Logie (Eds.), Working Memory and Human Cognition. New York: Oxford University Press,
1996, 31-65.
MARCEL, A.J. Conscious and unconscious perception: Experiments on visual masking and word
recognition. Cognitive Psychology, IS: 197-237, 1983a.
MARCEL, A .I. Conscious and unconscious perception: An approach to relations between phenomenal
experience and perceptual processes. Cognitive Psychology, 15: 238-300, 1983b.
MARSHALL, I.e., and HALLIGAN, P.W. Blindsight and insight in visuo-spatial neglect. Nature, 336:
766-767, 1988.
MARSHALL, J.C., and HALLIGAN, P.W. Imagine only half of it. Nature, 364: 193-194, 1993.

Representational neglect

25

MASSIRONI, M., ANTONUCCI, G., PIZZAMIGLIO, L., VITALE, M.V., and ZOCCOLOTTI, P. The WundtJastrow illusion in the study of spatial hemi-inattention. Neuropsychologia, 26: 161-166, 1988.
MCGLINCHEy-BERROTH, R., MILBERG, W.P., VERFAELLIE, M., ALEXANDER, M., and KILDUFF, P.T.
Semantic priming in the neglected visual field: Evidence from lexical decision task. Cognitive
Neuropsychology, 10: 79-108, 1993.
MEASSO, G., CAVARZERAN, F., ZAPPALA, G., LEBOWITZ, B.D., CROOK, T.H., PIROZZOLO, FJ., et al.
The Mini-Mental State Examination: Normative study of Italian random sample. Developmental
Neuropsychology, 9: 77-85, 1993.
MILNER, B. Interhemispheric differences in the location of psychological processes in man. British
Medical Bulletin, 27: 272-277, 1971.
MOSCOVITCH, M., and BEHRMANN, M. Coding of spatial information in the somatosensory system:
Evidence from patients with neglect following parietal lobe damage. Journal of Cognitive
Neuroscience, 6: 151-155, 1994.
NOVELLI, G., PAPAGNO, c., CAPITANI, E., LAIACONA, M., CAPPA, S.F., and VALLAR, G. Tre test clinici
di memoria verbale a lungo termine. Taratura su soggetti normali. Archivio di Psicologia, Neurologia
e Psichiatria, 47: 278-295, 1986.
OGDEN, J.A. Controlesional neglect of constructed visual images in right and left brain-damaged patients.
Neuropsychologia, 23: 273-277, 1985.
PEARSON, D.G., LOGIE, R.H., and GREEN, C. Mental manipulation, visual working memory and executive
processes. Psychologie Betreige, 1996 (in press).
POSNER, M.L, WALKER, J.A., FRIEDRICH, FJ., and RAFAL, R.D. How do the parietal lobes direct covert
attention? Neuropsychologia, 25: 135-146, 1987.
QUINN, J.G. , and RALSTON, G.E. Movement and attention in visual working memory. Quarterly Journal
of Experimental Psychology, 38: 689-703, 1986.
QUINN, J.G., and MCCONNELL, J. The irrelevant picture effect in the visuo-spatial sketchpad. Quarterly
Journal of Experimental Psychology, 49: 200-215, 1996.
POPPELREUTER, W. Zur Psychologie und Pathologie des Optischen Wohmehmung. Zeitschrift fur die
Gesamte Neurologie und Psychiatrie, 83: 26-152, 1923.
REISBERG, D., and LOGIE, R.H. The in's and out's of visual working memory. Overcoming the limits on
learning from imagery. In M. Intons-Peterson, B. Roskos-Ewoldsen and R. Anderson (Eds.), Imagery,
Creativity, and Discovery: A Cognitive Approach. Amsterdam: Elsevier, 1993, pp. 39-76.
REY, A. L'examen psychologique dans les cas d'encephalopathie traumatique. Archives de Psycho logie,
28: 286-340, 1942.
REY, A. Reattivo della figura complessa di A. Rey. Firenze, Edizioni OS, 1959.
RICHARDSON, J.T.E., ENGLE, R.W., HASHER, L., and LOGIE, R.H. Working Memory and Human
Cognition. New York: Oxford University Press, 1996.
RIDDOCH, MJ., and HUMPHREYS, G.W. The effect of cueing on unilateral neglect. Neuropsychologia,
21: 589-599, 1983.
RrzzOLATTI, G., and BERTI, A. Neural mechanisms of spatial neglect. In I.H. Robertson and lC. Marshall
(Eds.), Unilateral Neglect: Clinical and Experimental Studies. Hove: Lawrence Erlbaum Associated:
1993, pp. 87-105.
ROEDIGER, H.L., and McDERMOTT, K.B. Implicit memory in normal human subjects. In F. Boller and
H. Spinnler (Eds.), Handbook of Neuropsychology, Vol. 8. Amsterdam: Elsevier Publishers BV,
1993, pp. 63-131.
SALWAY, A.F.S., and LOGIE, R.H. Visuo-spatial working memory, movement control and executive
demands. British Journal of Psychology, 86: 253-269, 1995.
SMYTH, M.M., and PENDLETON, L.R. Working memory for movements. Quarterly Journal of
Experimental Psychology, 41: 235-250, 1989.
SPINNLER, H., and TOGNONI, G. Standardizzazione e taratura italiana di test neuropsicologici. The Italian
Journal of Neurological Sciences, Suppl. 8, 6: 1-120, 1987.
STEIN, J.F. The representation of egocentric space in the posterior parietal cortex. Behavioral and Brain
Sciences, 15: 691-700, 1992.
SUNDERLAND, A. Cognitive factors in unilateral neglect. Unpublished doctoral dissertation. Brunei
University, 1984.
TURNBULL, O.H., BESCHlN, N., and DELLA SALA, S. Agnosia for object orientation: Implications for
theories of object recognition. Neuropsychologia, 1996 (in press).
UNGERLEIDER, L.G., and MISHKIN, M. Two cortical visual systems. In D.L. Ingle, RJ.W. Mansfield
and M.S. Goodale (Eds.), The Analysis of Visual Behavior. Cambridge, MA: MIT Press, 1982.
VALLAR, G. The anatomical basis of spatial hemineglect in humans. In LH. Robertson and J.C. Marshall
(Eds.), Unilateral Neglect: Clinical and Experimental Studies: Hove: Lawrence Erlbaum Associated,
1993, pp. 27-59.
VALLAR, G., and PERAN!, D. The anatomy of unilateral neglect after right-hemisphere stroke lesions.
A clinical/CT-Scan correlation study in man. Neuropsychologia, 24: 609-622, 1986.

26

Nicoletta Beschin and Others

V ALLAR, G. , RUSCONI, M .L. , BIGNAMINI, L. , GEMINIANI, G. , and PERANI, D. Anatomical correlates of
visual and tactile extinction in humans: A clinical CT Scan study. Journal of Neurology,
Neurosurgery and Psychiatry, 57: 464-470, 1994a.
VALLAR, G., RUSCONI, M.L. , and BISIACH, E. Awareness of controlesional information in unilateral
neglect: Effects of verbal cueing, tracing, and vestibular stimulation. In C. Umiltil, M. Moscovitch
(Eds.), Attention and Performance, xv. Cambridge: MIT Press, 1994b, pp. 377-391.
V ALLAER, G., STERZI, R., BOTTINI, G., CAPPA, S.F., and RUSCONI, M .L. Temporary remission of left
hemianestesia after vestibular stimulation. A sensory neglect phenomenon. Cortex, 26: 123-131,
1990.
WI LSON, B., COCKBURN, 1., and HALLIGAN, P.W . Behavioural Inattention Test. Titchfield, Hants: Thames
Valley Test Company, 1987.
ZEKI, S., and SHIPP, S. The functional logic of cortical connections. Nature, 335: 311-317, 1988.
ZINGERLE, H. Dber Storungen d er Wahremungdes eigenen Korpers bei organischen Gehirnerkrankungen
Monatschrift fur Psychiatrie und Neurologie, 34: 13-36, 1913 .
ZOCCOLOTTI, P., ANTONUCCI, G., JUDICA, A. , MONTENERO, P., PIZZA MIGLlO, L. , and RAZZANO, C.
Incidence and evolution of the he mineglect disorder in chronic patients with unilateral right brain
damage. International Journal of Neuroscience, 47: 209-216, 1989.
S. Della Sala or R.H. Logie. Dept. of Psychology, University of Aberdeen. King's College. AB24 2ub Aberdeen, UK.
E-mail: sergio@aberdeen.ac.uk;r.logie@aberdee n.ac.uk.

(Received 14 December 1995; accepted 10 May 1996)