PRISM ADAPTATION IMPROVES CHRONIC VISUAL AND
HAPTIC NEGLECT: A SINGLE CASE STUDY
Robert D. McIntosh1, Yves Rossetti2 and A. David Milner1
(1Department of Psychology, University of Durham, UK; 2Espace et Action, INSERM
Unité 543, Bron, France)

ABSTRACT
Visuomotor adaptation to rightward displacing optical prisms is known to induce
temporary improvements in the symptoms of left visual neglect. We report a 74 year-old
woman with severe and chronic neglect of nine months duration, who underwent three
weekly sessions of prism adaptation. Substantial improvements were obtained on tests of
visual neglect (cancellation, copying and bisection). Improvement was also observed on a
spatial judgement task, with no explicit visual component, in which CS was required to
locate the centre of a haptically explored circle. These observations confirm that brief
periods of prism exposure can benefit even chronic neglect disorders. Moreover, the
improvement observed on the haptic task supports the belief that this procedure can
influence higher levels of spatial representation.
Key words: haptic, neglect, prism adaptation, rehabilitation

INTRODUCTION
Hemispatial neglect is a common syndrome in which a patient fails to report
or respond to stimulation from the side of space opposite a brain lesion (usually
of the right hemisphere), where these symptoms are not due to primary sensory
or motor deficits (Heilman, 1979). The syndrome has received intensive study
over the last two decades for a number of reasons. First, it provides a fascinating
window on the organisation of spatial representation in the brain. The fact that
neglect is widely heterogeneous, manifesting many dissociable subtypes, implies
some degree of functional independence between its different symptoms. For
instance, neglect may dissociate according to the sensory modality of
stimulation, the spatial range of its influence or the spatial reference frame
affected (see Vallar, 1998, for a review). Neglect may even be observed in
mental imagery tasks, in which patients show a paucity of report for details on
the left side of mentally evoked visual scenes (Bisiach and Luzzatti, 1978).
A second compelling reason for seeking to understand the neglect syndrome
is with a view to its rehabilitation. Many patients recover from the acute
manifestations of neglect relatively rapidly, but a significant proportion develop
a chronic disorder (31.5% of patients with acute neglect still exhibit deficits at 3
months post-stroke: Cassidy et al., 1998). Several studies have found visual
neglect to be a negative prognostic factor for functional outcome after stroke
(e.g. Denes et al., 1982; Cherney et al., 2001). Moreover, the application of
specific training regimes targeting neglect has been found to have concurrent
Cortex, (2002) 38, 309-320

310

Robert D. McIntosh and Others

benefits for functional and motor recovery (Paolucci et al., 1996). Several
sensory manipulations have now been shown to alleviate the symptoms of
neglect transiently (see Vallar et al., 1997, for a review). However, longer
lasting gains have proved elusive and many treatments that influence the clinical
symptoms of neglect fail to generalise to everyday behaviours.
Recently, a new treatment for neglect was introduced by Rossetti et al.
(1998). The procedure involves visuomotor adaptation to a rightward prismatic
shift of the visual field (i.e. away from the neglected side). Rossetti et al.
reported a reduction in neglect amongst six patients across a range of clinical
measures. The treatment is remarkable for several reasons. First, large gains may
follow a brief period of prism exposure (3-5 minutes). Second, the effects
generalise widely across clinical measures of neglect, and subsequent studies
have demonstrated that the effects may generalise to wheelchair navigation
(Rossetti et al., 1999b), postural control (Tilikete et al., 2001) and neglect of
mental imagery (Rode et al., 1999, 2001). Finally, the effects are persistent.
Rossetti et al. (1998) measured benefits at two-hours post-adaptation which, if
anything, were stronger than those observed immediately. More recently, three
patients have been reported in whom benefits were observable after four days
(Pisella et al., 2002). These observations suggest that visuomotor adaptation
stimulates a reorganisation of the neural representation of space that develops
autonomously after removal of the prisms, though the mechanisms of change
remain obscure (see Farnè et al., 2002, for discussion of possible mechanisms).
Research into the prism adaptation treatment for neglect is in its infancy and
many basic empirical questions present themselves. For instance, how
generalised are the effects across tasks and across patients? What is the typical
time course of the changes induced? Can cumulative benefits be achieved and
how specific are the effects with respect to the various symptoms of neglect?
We present a longitudinal case study of a patient (CS) with neglect of nine
months’ duration that may help shed some light upon these issues. We report
that, despite the chronicity and severity of CS’s neglect, a reliable influence of
prism adaptation was obtained across a range of clinical tasks. We also provide
a novel demonstration that the effects of prism adaptation are not limited to
tasks with an explicit visual component, thus supporting the notion that neglect
may affect higher levels of spatial representation. The results are discussed with
regard to the rehabilitative potential of this procedure.
CASE REPORT
CS is a right-handed woman with nine years of formal education and professional
secretarial qualifications. She is unmarried and had worked as a secretary until her
retirement. At the age of 73, CS suffered a haemorrhagic stroke in the territory of the right
middle cerebral artery. An MRI examination performed at two days post-stroke showed a
large lesion in the right parietal lobe extending to postero-lateral occipital cortex, with
sparing of medial occipital cortex. Her presenting clinical symptoms were left
hemianaesthesia, left hemianopia to confrontation, left hemiplegia (with some movement in
the upper limb), rightward cephalic deviation and severe left visual neglect. Formal
assessment of CS’s visual fields was performed by Humphrey perimetry at 14 months poststroke. This examination confirmed that her left hemianopia was complete and
homonymous.

Prism adaptation improves chronic neglect

311

An initial neuropsychological evaluation was performed at 7 days post-stroke at which
time CS was alert and well oriented to time, place and person. Bedside testing found total
extinction of left auditory and tactile stimuli and also of relatively leftward stimuli within
the spared right visual field. At this time, formal assessments using visual materials proved
impractical since, even when stimuli were placed on the extreme right side of her
peripersonal space, CS would respond only to the rightmost part of the sheet. However, on
three verbal sub-tests of the WAIS-R (information, vocabulary and digit span), CS
performed adequately, achieving age-scaled scores of 10, 10 and 5 respectively (all within
two standard deviations of the normal mean).
Formal assessment of CS’s visual neglect was made at 9 weeks post-stroke, by which
time she had recovered sufficiently to attend to visual material presented directly in front
of her on a tabletop. CS was administered five standard screening tests for neglect: line
cancellation (Albert, 1973), star cancellation (Wilson et al., 1987), scene copying (Gainotti
et al., 1972), representational drawing (Wilson et al., 1987) and line bisection (see Materials
and Methods for test details). Her performance is shown in Table I.
At six months post-stroke, now ambulatory, CS was discharged from hospital and
initially attempted to make the transition to home living. However, her severe neglect
proved a major obstacle to independent functioning, impeding many daily activities and
causing her to have frequent collisions with objects on the left side. During the next month,
she was transferred to a home for the elderly where the studies reported here were
conducted. CS was 74 years old at the time of these studies.

EXPERIMENTAL STUDY
A longitudinal study was carried out in four sessions, spaced at weekly
intervals, at approximately nine months post-stroke (39-42 weeks). In each
session, a neglect test battery was administered. In order of administration, the
battery comprised star cancellation, scene copying, line bisection, haptic circle
centring and poem reading (see below for test details). In weeks 40, 41 and 42,
CS was adapted to a rightward optical shift of 10° (see below for details). The
event sequence for the four sessions was as follows:
Week 39: Pre-test1
Week 40: Pre-test → prism adaptation → 2 hour post-test
Week 41: Pre-test → prism adaptation → immediate post-test2
Week 42: Pre-test → prism adaptation → 2 hour post-test

TABLE I

CS’ Neglect Screening Performance at 9 Weeks Post-Stroke
Line
cancellation
L lines omitted
R lines omitted
100%
0%

Star cancellation
Scene copying
Drawing
cancellation
(5 items)
(3 items)
L stars omitted
Items copied
R stars omitted (symmetrical items) Symmetrical items
100%
41%

2 (1)
see Figure 2b

3

Line bisection
10 × 20cm lines
Mean error
(SD)
+ 59.8 mm
(17.4)

1Due to an oversight, the scene copying task was not performed at week 39.
2The planned two-hour post-test in week 41 had to be cancelled. An immediate post test was instead performed with

a reduced battery that omitted the most time consuming test (haptic circle centring). To compensate for this disruption,
a final session was added at week 42 with the event sequence matched to that of week 40.

312

Robert D. McIntosh and Others

MATERIALS AND METHODS
Test Details
All tasks were untimed and all stimuli were presented at body midline.
Star cancellation was taken from the Behavioural Inattention Test (Wilson et al., 1987).
The A4 stimulus sheet contained 56 targets (small stars) pseudo-randomly interspersed with
distracter items. The targets actually fell into six columns, with two additional targets
located centrally. The experimenter indicated clearly the full extent of the sheet and crossed
out the two central targets. CS was then asked to cancel the remaining small stars. The
number of targets omitted in each lateral half of the sheet was counted.
Scene copying was adapted from Gainotti et al. (1972). CS was asked to copy a simple
scene made up of five items (see Figure 2). Performance was scored in two ways: the
number of items copied and the number of items symmetrically depicted.
In the line bisection task, CS was presented with 10 black horizontal lines, each 20 cm
long and 2 mm thick and each centred on an A4 sheet of white paper. Rightward errors
were scored as positive values and leftward errors as negative.
The haptic circle centring task employed a 30 cm diameter circular groove cut centrally
into a 1 cm thick square of transparent perspex (37.5 × 37.5 cm). CS was blindfolded and
the stimulus was placed in front of her with its centre at a radial depth of about 25 cm
from her body. She was invited to explore the stimulus using her right hand, and was given
up to five practice trials before each trial block. In each experimental trial, CS’s right index
finger was placed in the circumference groove at one of four points: North, East, South or
West with respect to CS (she was not informed that these four points were used). She was
then required to make one full (clockwise or anticlockwise) exploration of the circumference
groove. While she did this, the experimenter placed a finger at the original starting point so
as to block further movement of CS’s finger when the circumference had been fully
explored. When CS felt her exploration blocked in this way, she was asked to move her
fingertip to the centre of the circle and then to bring it to rest. The experimenter recorded
(to the nearest 5 mm) the lateral and radial coordinates of the response from a scale attached
to the underside of the stimulus. Rightward and distal errors were scored with positive
values; leftward and proximal errors were scored with negative values. Each time the task
was administered, CS performed 48 trials divided into two blocks with a short break
between blocks. Within each block, start position (North, East, South, West) and direction
of exploration (clockwise, anticlockwise) were counterbalanced. At no time was CS allowed
to see the stimulus circle.
The above task was designed as an approximate haptic analogue of the visual line
bisection task. Previously, the tactile line bisection task has been used for this purpose in the
study of neglect patients and normals. However, studies of unilaterally brain damaged patients
have failed to detect any evidence of neglect phenomena using the tactile line bisection task
(Fujii et al., 1991; Hjaltason et al., 1993; McIntosh, 1999). This insensitivity of tactile line
bisection to neglect may be due partly to the adoption of counting strategies by subjects
performing this task, a tendency that has been noted by several authors (Hatta and
Yamamoyto, 1986; Levander et al., 1993; McIntosh, 1999). The haptic circle centring task is
a bi-dimensional bisection task that precludes the use of counting strategies and thus promotes
the adoption of spatial strategies. In support of this claim, significant left neglect (abnormally
large rightward errors) has been observed on this task amongst neglect patients in whom
tactile line bisection failed to identify any significant lateral biases (McIntosh, 1999).
Finally, a poem reading task was administered. CS was presented with eight stanzas
(32 lines) of “The Hunting of the Snark” by Lewis Carroll (1939), a poem with which she
was unfamiliar. The choice of a poem that features occasional “portmanteau” words (i.e.
pronounceable nonwords) was incidental, although it was anticipated that the presence of
such nonwords might make CS’s reading errors less apparent to her. The text was presented
in 16 point arial bold lowercase font and was justified centrally. The poem was presented
on a laptop screen in the first session and on A4 paper in subsequent sessions. A different
section of the poem was presented each time (mean total of 256 words). CS was asked to
read the poem aloud.

Prism adaptation improves chronic neglect

313

Following the division of neglect dyslexic errors advocated by Young et al. (1991),
errors were scored separately for whole words omitted and for words that were attempted
but misread. Conservative criteria were applied to ensure that errors that were not a
consequence of left neglect were not counted. For errors of word omission, the criterion
adopted was that that the entire word was omitted and that no word lying to the left of the
omitted word on the same line of text was attempted. Misreading errors for attempted words
were determined according the criterion introduced by Ellis et al. (1987): that the target
and error words were identical to the right of an identifiable “neglect point” in each word,
but had no letters in common to the left of their neglect points. For each administration of
the poem reading test, the number of neglect word omissions was expressed as a percentage
of the total number of words presented. The number of neglect misreading errors was
expressed as a percentage of the total number of words attempted.
Prism Adaptation Procedure3
The adaptation procedure followed the method of Rossetti et al. (1998). CS wore a pair
of goggles fitted with wide-field point-to-point prismatic lenses creating a rightward optical
shift of 10°. Exposure consisted of 50 fast pointing movements made to visual targets
presented 10° to left and right of the body midline, with 25 movements to each in pseudorandom order. A shelf was held under CS’s chin to prevent her viewing her hand at its
starting position, but allowing her to see the targets and her terminal errors. The exposure
period lasted for between three and five minutes.

RESULTS
For all of the following tasks, the most pertinent comparisons are between
pre- and post-prism performance in the two matched sessions, conducted at
weeks 40 and 42. For completeness, the immediate post-adaptation results from
week 41 are also included in Figures 1-5.
Star Cancellation
Figure 1 shows the percentage of total targets omitted at each administration of
the star cancellation test (black dots indicate omissions in the right half of the
sheet). The overall effect of prism adaptation was statistically evaluated by pooling
the pre-prism data across weeks 40 and 42 and similarly pooling the post-prism
data. A Mann-Whitney U test confirmed that a significantly greater number of
targets were cancelled following prism adaptation (z = 2.77, p < 0.001). Similarly,
a significant effect of week was confirmed by comparing the pooled pre- and postprism data of week 40 with those of week 42 (z = 3.64, p < 0.001).
Scene Copying
Figure 2 illustrates CS’s performance of the scene copying test. No example
from week 39 is available (see Materials and Methods). However, CS had
attempted the task four times previously and all of her copies are shown here,
3An anonymous referee has pointed out that the pre-prism assessments should ideally have been preceded by a “sham

adaptation” procedure, in which the above pointing task is performed with no prismatic deviation. This modification,
which would control for any beneficial effects of the pointing component of the adaptation procedure, is recommended
for future case studies.

314

Robert D. McIntosh and Others

Fig. 1 – Percentage of total targets omitted at each session (black dots indicate omissions in the
right half of the sheet). A clear improvement of performance is seen after prism adaptation. (*The
post-test at week 41 was immediate, rather than two hours post-adaptation.)

illustrating the stability of her performance over this period. A dramatic
improvement was stimulated by the first period of prism adaptation (week 40).
This was largely maintained across sessions, even though novel stimuli were
presented in subsequent weeks.
Line Bisection
Figure 3 shows CS’s mean directional error at each administration of the line
bisection test. The raw data from the matched sessions at weeks 40 and 42 failed to
pass Levene’s test for homogeneity of error variances [F (3, 36) = 3.54, p < 0.05].
Homogeneity of variance was obtained by adding a constant (+ 5) to the raw values
to render them universally positive and squaring the result. The transformed data
were entered into a two-way ANOVA by week (40, 42) and prism (pre, post). The
main effect of prism was highly significant [F (1, 36) = 22.48, p < 0.001] and the
effect of week narrowly reached significance [F (1, 36) = 4.11, p = 0.05].
Haptic Circle Centring
Figure 4 shows CS’s mean lateral and radial circle centring errors. For lateral
and radial errors separately, two-way ANOVAs by week (40, 42) and prism
(pre, post) were conducted. For lateral errors, there was a highly significant
effect of prism [F (1, 188) = 8.58, p < 0.005]. For radial errors, there were no
significant effects.

Prism adaptation improves chronic neglect

315

Fig. 2 – CS’s scene copying performance at nine time points post-stroke with periods of prism
exposure indicated. The three stimulus scenes employed are shown in the top row, with CS’s copies
below: (a) Week 1 (with the page presented in right hemispace); (b) Week 9; (c) Week 22; (d) Week
40 (pre-test); (e) Week 40 (two hour post-test); (f) Week 41 (pre-test); (g) Week 41 (immediate posttest) (h) Week 42 (pre-test); (i) Week 42 (two hour post-test). Pre-adaptation performance was stable
but a dramatic improvement was stimulated by the first period of prism adaptation. This was largely
maintained in subsequent sessions even though novel stimuli were presented.

Poem Reading
Figure 5 shows CS’s neglect dyslexic errors at each administration of the
poem reading test. CS’s neglect dyslexia was relatively mild and expressed
primarily as whole word omissions, with 8.0% of words being omitted overall.
Neglect misreading errors (shown as black dots in Figure 5) were produced on
only 1.7% of attempted words overall. Prism adaptation had no consistent effect
on CS’s neglect dyslexia for either class of error.
DISCUSSION
The present case study, though necessarily limited in scope, provides several
novel findings. First, it is clear that a prism adaptation treatment of five minutes or
less can substantially improve the symptoms of neglect even in a severe case of

316

Robert D. McIntosh and Others

Fig. 3 – CS’s mean bisection errors (± S.E.) at each session. A reliable reduction of rightward
errors occurred following prism adaptation in weeks 40 and 42. (*The post-test at week 41 was
immediate, rather than two hours post-adaptation.)

nine months’ chronicity. The immediacy, strength and generality of CS’s
improvement suggest that there may be no upper limit to the time after stroke at
which prism adaptation can be effective. In support of this, it may be noted that the
group study of Rossetti et al. (1998) included patients up to 14 months post-stroke.
A second striking aspect of the present results is that prism adaptation
induced a clear improvement of neglect on a spatial task with no explicit visual
component. The haptic circle centring task was designed as a two-dimensional
analogue of visual line bisection. However, unlike in the visual task, the haptic
exploration of the circular stimulus was standardised to ensure that the
circumference of the circle was fully explored before the midpoint judgement
was made. Thus, the improvement of neglect observed on this task cannot be
attributed to an improvement of haptic exploration. It is unclear whether the
reduction of rightward errors was due to an alteration of CS’s ability to mentally
represent (or mentally explore a representation of) the stimulus circle, or to a
reduction of a rightward manual response bias. The former possibility, however,
is fully consistent with the findings that prism adaptation improves neglect on
imagery tasks with no manual component (Rode et al., 1999, 2001). These
cross-modal effects strongly support the view that prism adaptation induces
changes not only in visuomotor coordination but also at higher cognitive levels
of spatial representation (Rossetti et al., 1998, 1999b; Rode et al., 1999, 2001).
The generality of the influence of adaptation across tasks and sensory
modalities suggests a recalibration of a high-level representation of space, and
signals great promise for the therapeutic use of the prism treatment. Crucially,
the benefits seem to generalise to activities and spatial scales directly relevant to

Prism adaptation improves chronic neglect

317

Fig. 4 – The upper graph shows CS’s mean lateral circle centring errors (± S.E.) at each session.
A reliable reduction in rightward errors occurred after prism adaptation in weeks 40 and 42. The
lower graph shows CS’s mean radial errors (± S.E.), with no reliable influence of prism adaptation.
(*No post-test was performed for this task at week 41.)

real world settings. For instance, adaptation has positive effects on the ability of
wheelchair-bound neglect patients to navigate around a hospital (Rossetti et al.,
1999a). In the case of CS, there is anecdotal evidence that similar daily benefits
were obtained. At the time that the present study commenced, nursing home
staff would not allow CS to walk unaccompanied around the building due to the
frequency with which she collided with obstacles on her left side (e.g. door
frames, furniture). By week 42, the perceived danger of such collisions had

318

Robert D. McIntosh and Others

Fig. 5 – CS’s neglect dyslexic errors at each session. Bars indicate word omissions (black dots
indicate misreadings). There was no consistent influence of prism adaptation on either class of error.
(*The post-test at week 41 was immediate, rather than two hours post-adaptation.)

reduced to the point that CS was now allowed to walk around the building
alone. Unfortunately, when CS was visited again ten weeks later, she had
returned to her former accident-prone status, indicating that the gains observed
were not permanent.
In light of the generality of the effects of prism adaptation, it is surprising
that CS’s reading errors were apparently unaffected. It is tempting to relate this
result to previous reports of dissociations of neglect dyslexia from other clinical
symptoms of neglect (see Ellis et al., 1993). However, CS’s performance should
not be too readily interpreted in this manner. First, the variability of performance
apparent in Figure 5 is rather high relative to the mild level of neglect dyslexia
exhibited overall. It is possible that subtle effects of the prism treatment might
have been masked by this background noise. Second, a recent group study of
single word reading in neglect has demonstrated robust improvements in
response to prism adaptation, indicating that neglect dyslexic misreading errors
are not generally resistant to this intervention (Farnè et al., 2002). Similarly, text
reading was among the tests that were performed better following prism
exposure in the original group study of Rossetti et al. (1998). It is worth noting,
however, that Rossetti et al did not distinguish between neglect dyslexic errors
of word omissions and errors of misreading of single words. Given evidence that
these classes of neglect dyslexic errors may have different underlying causes
(Costello and Warrington, 1987; Young et al., 1991), it remains possible that
they may be differentially susceptible to prismatic intervention.
Finally, it is noted that CS’s data are consistent with a persistent influence of

Prism adaptation improves chronic neglect

319

prism exposure. The retention of gains over time was most obvious for the scene
copying task, which CS performed almost perfectly on each occasion following
the first prism treatment, even though novel stimuli were presented (the scene
presented in week 42 was particularly intricate). Significant reductions of neglect
across sessions were also found for line bisection and star cancellation, and there
was a qualitative improvement across sessions for the haptic task. It must be
conceded, however, that these effects are difficult to ascribe unambiguously to
treatment effects, as distinct from possible effects of task repetition or
spontaneous fluctuations of neglect severity. This caveat applies particularly to
the star cancellation, line bisection and haptic circle centring tasks, as only two
pre-treatment baseline measurements were obtained for these tasks (weeks 39
and 40) and qualitative improvements between baseline measurements were seen
in all cases. Accordingly, whilst the present data are consistent with the
conclusion that a single period of prism adaptation may yield measurable
benefits at one week, this conclusion must be tentative. Moreover, whilst
previous studies have found beneficial effects on line bisection up to four days
after adaptation in individual neglect patients (Pisella et al., 2002; Rossetti et al.,
1999a), a recent group study found no evidence of gains on a single-word
reading task at one week post-adaptation (Farnè at al., 2002). Clearly, the typical
time-course of the effects of prism adaptation in neglect, and its possible taskdependency, are topics requiring detailed further study.
CONCLUSIONS
This case study confirms the ameliorative power of the prism adaptation
treatment for neglect, even when the disorder is severe and chronic. These effects
are temporary, but remarkably long-lived in relation to the brevity of the prism
exposure. The effects also generalise widely across tasks and the present study
provides a first demonstration that prism adaptation can have an impact on haptic
spatial judgements. This supports the view that the treatment stimulates high-level
changes at the level of spatial representation rather than being limited to low-level
sensorimotor re-mappings. It remains to be seen whether prism adaptation is
beneficial for all varieties of neglect, and whether more intensive schedules of
exposure can induce more permanent changes in spatial awareness. In particular, it
will be of paramount importance to determine whether prism adaptation, in tandem
with other therapies, can improve long-term outcome for patients with neglect.
Acknowledgements. We wish to thank CS for her patience and humour, all staff at the
nursing home for their co-operation and Richard Weir for artistic assistance. This work was
supported by grants from the Wellcome Trust (048060) and the Leverhulme Trust
(F/00128/C) to ADM, and grants from the Wellcome Trust (010343) and INSERM
(4P012E) to YR.
REFERENCES
ALBERT M. A simple test of visual neglect. Neurology, 43: 213-216, 1973.
BISIACH E and LUZZATTI C. Unilateral neglect of representational space. Cortex, 14: 129-133, 1978.
CARROLL L. The Complete Works of Lewis Carroll. New York: Nonesuch Press, 1939.
CASSIDY TP, LEWIS S and GRAY CS. Recovery from visuospatial neglect in stroke patients. Journal of
Neurology, Neurosurgery and Psychiatry, 64: 555-557, 1998.

320

Robert D. McIntosh and Others

CHERNEY LR, HALPER AS, KWASNICA CM, HARVEY RL and ZHANG M. Recovery of functional status
after right hemisphere stroke: Relationship with unilateral neglect. Archives of Physical Medicine
and Rehabilitation, 82: 322-328, 2001.
COSTELLO A DE L and WARRINGTON EK. The dissociation of visuospatial neglect and neglect dyslexia.
Journal of Neurology, Neurosurgery, and Psychiatry, 50: 1110-1116, 1987.
DENES G, SEMENZA C, STOPPA E and LIS A. Unilateral spatial neglect and recovery from hemiplegia: A
follow-up study. Brain, 105: 543-553, 1982.
ELLIS AW, FLUDE BM and YOUNG AW. “Neglect dyslexia” and the early visual processing of letters in
words and nonwords. Cognitive Neuropsychology, 4: 439-464, 1987.
ELLIS AW, YOUNG AW and FLUDE BM. Neglect and visual language. In IH Robertson and JC Marshall
(Eds.), Unilateral Neglect: Clinical and Experimental Studies. Hove, UK: Lawrence Erlbaum Associates,
1993, pp. 233-255.
FARN È A, ROSSETTI Y, TONIOLO S and LÀDAVAS E. Ameliorating neglect with prism adaptation: Visuomanual and visuo-verbal measures. Neuropsychologia, 40: 718-729, 2002.
FUJI T, FUKATSU R, KIMURA I, SASO S and KOGURE K. Unilateral neglect in visual and tactile modalities.
Cortex, 27: 339-343, 1991.
GAINOTTI G, MESSERLI P and TISSOT R. Qualitative analysis of spatial neglect in relation to laterality of
cerebral lesions. Journal of Neurology, Neurosurgery, and Psychiatry, 35: 545-550, 1972.
HATTA T and YAMAMOYTO M. Hemispheric asymmetries in a tactile bisection task: effects of hemispace
of presentation. Neuropsychologia, 24: 265-269, 1986.
HEILMAN KM. Neglect and related disorders. In KM Heilman and E Valenstein (Eds), Clinical
Neuropsychology. New York: Oxford University Press, 1979.
HJALTASON H, CANEMAN G and TEGNÉR R. Visual and tactile rod bisection in unilateral neglect. Cortex,
29: 583-588, 1993.
LEVANDER M, TEGNÉR R and CANEMAN G. Tactile line bisection in normal subjects. Perceptual and
Motor Skills, 76: 831-836, 1993.
MCINTOSH RD. Unilateral Neglect: Visual and Manual. Unpublished PhD thesis, Glasgow Caledonian
University, 1999.
PAOLUCCI S, ANTONUCCI G, GUARIGLIA C, MAGNOTTI L, PIZZAMIGLIO L and ZOCCOLOTTI P. Facilitatory
effect of neglect rehabilitation on the recovery of left hemiplegic stroke patients: A cross-over study.
Journal of Neurology, 243: 308-314, 1996.
PISELLA L, RODE G, FARNÈ A, BOISSON D and ROSSETTI Y. Dissociated long lasting improvements of
straight-ahead pointing and line bisection tasks in two hemineglect patients. Neuropsychologia; 40:
327-334, 2002.
RODE G, ROSSETTI Y, LI L and BOISSON D. Improvement of mental imagery after prism exposure in
neglect: A case study. Behavioural Neurology, 11: 251-258, 1999.
RODE G, ROSSETTI Y and BOISSON D. Prism adaptation improves representational neglect.
Neuropsychologia, 39: 1250-1254, 2001.
ROSSETTI Y, RODE G, PISELLA L, FARNÈ A, LI L, BOISSON D and PERENIN M-T. Prism adaptation to a
rightward optical deviation rehabilitates left hemispatial neglect. Nature, 395: 166-169, 1998.
ROSSETTI Y, RODE G, PISELLA L and BOISSON D. Plasticité sensori-motrice et récupération fonctionnelle:
les effets thérapeutiques de l’adaptation prismatique sur la négligence spatiale unilatérale.
Médecine/Sciences, 15: 239-245, 1999a.
ROSSETTI Y, RODE G, PISELLA L, FARN È A, LING L and BOISSON D. Sensorimotor plasticity and
cognition: Prism adaptation can affect various levels of space representation. In M Grealy and JA
Thomson (Eds), Studies in Perception and Action. Hove, UK: Lawrence Erlbaum Associates,
1999b, pp. 265-269.
TILIKETE C, RODE G, ROSSETTI Y, PICHON J, LI L and BOISSON D. Prism adaptation to a rightward optical
deviation improves postural imbalance in left-hemiparetic patients. Current Biology, 11: 1-5, 2001.
VALLAR G. Spatial hemineglect in humans. Trends in Cognitive Sciences, 2: 87-97, 1998.
VALLAR G, GUARIGLIA C and RUSCONI ML. Modulation of the neglect syndrome by sensory stimulation. In P
Thier and HO Karnath (Eds), Parietal Lobe Contribution to Orientation in 3D Space. Heidelberg: Springer,
1997, pp. 555-579.
WILSON B, COCKBURN J and HALLIGAN P. Behavioural Inattention Test. Titchfield, Hampshire: Thames
Valley Test Co., 1987.
YOUNG AW, NEWCOMBE F and ELLIS AW. Different impairments contribute to neglect dyslexia.
Cognitive Neuropsychology, 8: 177-191, 1991.
RD McIntosh, Department of Psychology, University of Durham, Science Laboratories, South Road, DH1 3LE, UK. E-mail:
r.d.mcintosh@dur.ac.uk

(Received 15 April 2001; reviewed 22 July 2001; revised 18 October 2001; accepted 19
October 2001; Action Editor Roberto Cubelli)