Neuropsychologia 46 (2008) 2445–2462
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/neuropsychologia
Reviews and perspectives
The significance of visual information processing in reading: Insights
from hemianopic dyslexia
Susanne Schuetta,∗, Charles A. Heywooda, Robert W. Kentridgea, Josef Zihlb,c
aDepartment of Psychology, University of Durham, UK
bDepartment of Psychology, Neuropsychology, Ludwig-Maximilians-University Munich, Germany
cMax-Planck-Institute of Psychiatry, Germany
a r t i c l e i n f o
Received 26 August 2007
Received in revised form 18 April 2008
Accepted 25 April 2008
Available online 2 May 2008
a b s t r a c t
We present the first comprehensive review of research into hemianopic dyslexia since Mauthner’s orig-
inal description of 1881. We offer an explanation of the reading impairment in patients with unilateral
homonymous visual field disorders and clarify its functional and anatomical bases. The major focus of our
ing. An advanced understanding of the basis of hemianopic dyslexia and its rehabilitation also increases
our knowledge about normal reading and its underlying neural mechanisms. By drawing together vari-
ous sources of evidence we illustrate the significance of bottom-up and attentional top-down control of
visual information processing and saccadic eye-movements in reading. Reading depends critically on the
cortical–subcortical network subserving the integration of visual, attentional and oculomotor processes
involved in text processing.
© 2008 Elsevier Ltd. All rights reserved.
Reading: vision, attention, eye-movements, and language in (inter-)action .....................................................................
2.1.Eye-movements and visual information processing in reading ...........................................................................
2.2. The neural basis of text reading............................................................................................................
Hemianopic dyslexia: reading when the visual world shrinks ...................................................................................
3.1.Introducing cerebral visual field disorders.................................................................................................
3.2.Reading performance and eye-movements in hemianopic dyslexia......................................................................
3.2.2.Reading eye-movements .........................................................................................................
Reading without a parafovea: seeing only half the wor(l)d.......................................................................................
4.1.Word identification without a parafovea ..................................................................................................
4.2.Visual guidance of reading eye-movements without a parafovea ........................................................................
Looking beyond parafoveal visual field loss: is hemianopic dyslexia purely visually elicited? ..................................................
5.1.Hemianopic dyslexia and the question of spontaneous oculomotor adaptation..........................................................
5.2.Hemianopic dyslexia and its anatomical basis ............................................................................................
The rehabilitation of hemianopic dyslexia: re-learning eye-movement control in reading......................................................
Synopsis: insights from and into hemianopic dyslexia ...........................................................................................
∗Corresponding author at: Department of Psychology, University of Durham, Science Laboratories, South Road, Durham DH1 3LE, UK. Tel.: +44 191 3343268;
fax: +44 191 3343241.
E-mail address: email@example.com (S. Schuett).
0028-3932/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
Reading is a complex skill which can be disturbed at any of
its visual, lexical–semantic and phonological processing stages. A
wide variety of quantitatively and qualitatively different reading
disorders following brain injury has been identified (for reviews,
see Ellis & Young, 1996; Hillis & Caramazza, 1992; Shallice, 1988).
Acquired impairments of reading in subjects with previously well-
established reading skills immediately draw to mind the aphasic
reading disorders which involve disturbances of lexical and/or
post-lexical processes. These higher-level reading disorders (cen-
tral dyslexias) rank high in neuropsychology’s research agenda and
normal reading process.
Unfortunately, the acquired lower-level reading disorders,
which involve impairments of pre-lexical (visual) processes, have
been largely neglected. These so-called peripheral dyslexias arise
from disturbances at the more peripheral levels of text information
processing. Visual field disorders, deficits of visual acuity, spatial
contrast sensitivity and visual adaptation, disorders in visuospa-
tial perception, spatial restriction of the field of visual attention
(a prominent symptom of visual neglect and Balint’s syndrome),
visual agnosia, and visual illusions and hallucinations can all
impair reading at various levels of visual processing (Baylis, Driver,
Baylis, & Rafal, 1994; Behrmann, Moscovitch, Black, & Mozer, 1990;
Behrmann, Shomstein, Black, & Barton, 2001; De Luca, Spinelli, &
Zoccolotti, 1996; Hess, Zihl, Pointer, & Schmid, 1990; Zihl, 1989,
1995a; Zihl & Kerkhoff, 1990; Zihl & von Cramon, 1986). Although
the peripheral dyslexias have been attracting increasing attention
recently, the chief focus has been on the clinical syndromes of
neglect dyslexia and pure alexia or visual agnosia for letters.
Surprisingly, hemianopic dyslexia, the most elementary and
frequent peripheral dyslexia (present in ∼15% of patients in neuro-
logical rehabilitation centres, see Kerkhoff, 1999; Prosiegel, 1988),
is hardly considered in reviews or text books dealing with periph-
eral dyslexias (e.g. Ellis & Young, 1996; Riddoch, 1991; Shallice,
1988). It perhaps counts as the most important visual impair-
ment following brain injury affecting the patients’ occupational
and daily life as a pronounced visual handicap (Papageorgiou et
al., 2007; Zihl, 2000). Reading becomes so laborious that many
occupation, continuing employment may be at risk (Leff, Spitzyna,
Plant, & Wise, 2006). Hemianopic dyslexia (also called hemianopic
alexia) is an acquired reading disorder in which 80% of patients
with homonymous visual field defects affecting parafoveal (and
foveal) vision have severe reading difficulties despite intact lan-
guage functions (Zihl, 2000). In these patients, word identification
and the ability to plan and guide reading eye-movements is dis-
et al., 2007; Zihl, 1995a).
This article offers the first comprehensive review of research
into hemianopic dyslexia. We explain the nature of hemianopic
dyslexia and clarify its functional and anatomical bases. Further-
more, we consider what hemianopic dyslexia can tell us about
normal reading and its neural basis. In this manner we hope to
provide a coherent framework for future work.
Our review is organised into six sections. First, we give a brief
survey of the themes relevant for our critical examination of the
ing, eye-movement control, visuospatial attention and linguistic
processing (Section 2). In Section 3, we describe the features of
homonymous visual field disorders and review the findings from
hemianopic dyslexia research since Mauthner’s original descrip-
tion of 1881. In Section 4, we demonstrate the significance of
parafoveal vision for reading by discussing the effects of unilateral
oculomotor control in reading, both at the behavioural and neural
level. Examining the anatomy of hemianopic dyslexia in Section 5
shows that parafoveal visual field loss in itself cannot completely
satory treatment approach for rehabilitating hemianopic dyslexia,
which reveals important insights into the functional plasticity of
the visual, attentional and oculomotor systems involved in text
processing. In Section 7, we provide a synopsis of all sources of
evidence that demonstrates the important insights studying hemi-
anopic dyslexia generates into the normal reading process and its
of reading and eye-movement control.
2. Reading: vision, attention, eye-movements, and
language in (inter-)action
be possible (...) to conclude a priori that a hemianopia (...) must
impair reading” (p. 223) and regarded a detailed consideration of
theories of visual information processing and eye-movements in
normal reading as essential. However, he also firmly believed that
hemianopic dyslexia could not be explained as “merely a conse-
quence of (...) hemianopia” (p. 226). Thus, the basis of hemianopic
dyslexia may not be purely visual and we therefore consider the
visual, attentional, oculomotor and language processes involved in
normal reading and their underlying neural mechanisms.
2.1. Eye-movements and visual information processing in reading
Reading is the process of understanding written language. This
the speed of comprehension (Findlay & Gilchrist, 2003). The eyes
follow a typical scan path across the text, in the direction depend-
ing upon the language of the text (i.e. from left-to-right and from
top-to-bottom for Western cultures). Plotting eye position against
time reveals a staircase pattern as saccadic eye-movements regu-
words in a text are fixated, sometimes even twice (i.e. refixation;
15% of total fixations), many words are skipped; 2–3 letter words,
for instance, only receive a fixation about 25% of the time. On aver-
age, a fixation during reading lasts for about 200–250ms and is fol-
lowed by a saccade to some 7–9 characters forward (∼2–3◦). About
10–15% of our reading saccades are regressive. Towards the end
of a line of text, a large right-to-left slightly oblique saccadic eye-
the return-sweep depends upon line length (usually about 50 char-
acters, ∼17◦) (Rayner, 1998; Rayner & Pollatsek, 1989). In continual
information sampling, eye-movements may be coupled with head-
movements. As most studies of reading eye-movements immobi-
visual and lexical characteristics of the text information extracted
during a fixation (Rayner & Pollatsek, 1981). The region of effec-
tive processing during reading, the perceptual span, extends about
3–4 characters to the left and up to 15 characters to the right of
fixation (in left-to-right writing systems). As one degree of visual
angle encompasses about 3 characters for most normal text (Leff
et al., 2000), these values are equivalent to ∼1.3◦to the left and
5◦to the right of fixation (McConkie & Rayner, 1975, 1976; Rayner
& Bertera, 1979). Visual acuity falls symmetrically to either side of
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
ing in normal readers. During a fixation, readers extract visual information from the
foveal visual field (central white oval) and the parafoveal visual field (grey ellipse).
probably on the “h” of the fixated word “that” in normal readers (optimal viewing
foveal vision and the distribution of the perceptual span is there-
fore likely to reflect an attentional asymmetry in reading. Acuity
limitations determine only its right boundary. Discriminating fine
speed decrease sharply with increasing eccentricity in the hori-
zontal direction, and even more so in the vertical direction (Anstis,
Rayner, Well, Pollatsek, & Bertera, 1982; Underwood & McConkie,
1985). Beyond this, only coarse textual features can be discerned
up to the rightward boundary of the parafoveal visual field (Rayner,
1998). The range of letters that can be reliably identified without
moving the eyes, i.e. shifts of fixation, is called word identification
span or visual span. This range depends, of course, on print size;
larger fonts are more discriminable but, with increasing font size,
letter strings will fall further into the visual periphery with a con-
text processing, composed of the foveal and parafoveal visual field,
is illustrated in Fig. 1. Since the perceptual span’s spatial extent
exceeds the average-sized word at a given fixation and the mean
amplitude of reading saccades, text material is scanned in a highly
overlapping manner (Ikeda & Saida, 1978; Rayner & Bertera, 1979).
Foveal processing of fixated words enables lexical access and
hence word identification. Fixation duration is influenced by fac-
tors such as word frequency, predictability and age-of-acquisition
(Rayner, 1998). During successive saccades, foveal processing is
facilitated by information that has been extracted from the right
parafovea on the preceding fixation, i.e. the so-called parafoveal
preview benefit (Rayner, 1975; Rayner, White, Kambe, Miller, &
Liversedge, 2003). Such information includes that of word-length,
which is used for the selection of the to-be-fixated word and the
specification of the saccadic amplitude (Ducrot & Pynte, 2002;
Inhoff, Radach, Eiter, & Juhasz, 2003).
2.2. The neural basis of text reading
is sparse compared with what is known about the neural mecha-
nisms underlying single-word reading (for a recent review on word
identification, see Jobard, Crivello, & Tzourio-Mazoyer, 2003). The
tiation and maintenance of an oculomotor scanpath in addition to
word identification, has been investigated in only two studies (Leff
et al., 2000; Leff, Scott, Rothwell, & Wise, 2001). Reading involves
visual, attentional, oculomotor and language processes (Rayner &
Pollatsek, 1989), which are supported by large-scale neural net-
works (Mesulam, 1990). Distributed and coordinated processing
relying on multiple cortical and subcortical brain regions suggests
that white matter pathways connecting these regions play a cru-
Visual information is transmitted from the retinae to the pri-
mary visual (striate) cortex via the optic nerves, the optic chiasm,
the optic tracts, the lateral geniculate nucleus, and the optic radia-
tion (Gr¨ usser & Landis, 1991). The striate cortex (V1), the prestriate
visual area V2, the posterior parietal cortex and frontal eye fields,
as well as the supplementary eye fields and the dorsolateral pre-
frontal cortex form a network which integrates vision, attention
and eye-movements. Subcortical structures, particularly the supe-
rior colliculus and thalamus, also contribute to saccade control
(for a more detailed discussion, see Leigh & Zee, 2006; Pierrot-
Deseilligny, Rivaud, Gaymard, M¨ uri, & Vermersch, 1995). This
distributed neural system subserves the bottom-up (i.e. stimulus-
driven) and top-down (i.e. goal-directed) control of visual–spatial
attention and saccadic eye-movements via feedforward and feed-
back connections (Corbetta, 1998; Corbetta & Shulman, 2002).
Although “attention and ocular control did not evolve for read-
ing (...), reading is a special application of the attentional/ocular
control system” (Kliegl & Engbert, 2003, p. 492).
The primary visual cortex (V1) appears indispensable for visu-
There is evidence that the eyes are disparate on 40–50% of fix-
ations during reading (Kirkby et al., in press). It has therefore
been suggested that a single perceptual representation is achieved
through the visual integration of the two disparate retinal signals
at a very early stage in the visual pathway (Liversedge, Rayner,
White, Findlay, & McSorley, 2006). Word identification involves
the activation of left and right striate and ventral prestriate cortex
where foveal vision is represented. The guidance of reading eye-
movements requires the representation of right parafoveal vision
in the left primary visual cortex and neighbouring V2. The asym-
metric activation of left parafoveal V1/V2 during text reading has
been interpreted as physiological confirmation of the perceptual
tors (Leff et al., 2000). This top-down attentional modulation of
early visual information processing is mediated by fronto-parietal
activity (Kastner, Pinsk, De Weerd, Desimone, & Ungerleider, 1999;
Russell, Malhotra, & Husain, 2004) and results in the directing
of visual attention to the right of fixation during reading (Upton,
Hodgson, Plant, Wise, & Leff, 2003). Attentional processes facilitate
visual processing in the striate and extrastriate cortices (Martinez
et al., 2001) and in the ventral occipito-temporal stream (Mangun,
Hopfinger, Kussmaul, Flechter, & Heinze, 1997), which is crucially
involved in high-resolution, local processing of visual features and
object identification (Milner & Goodale, 2006). Thus, “attention
during reading acts early in the visual hierarchy” (Leff et al., 2000).
As words can be regarded as visual objects, the ventral stream
has been implicated in word processing and identification pro-
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
cesses (Poldrack, Desmond, Glover, & Gabrieli, 1998) which are
associated with an activation of the foveal part of the left and right
occipital cortex (V1/V2) (Brewer, Liu, Wade, & Wandell, 2005) and
the left posterior occipito-temporal junction in the inferior tempo-
ral gyrus (Leff, Crewes, et al., 2001). Word identification is also the
first stage of linguistic processing; its successful accomplishment
provides the basis for intact language comprehension as it makes
& Blythe, 2007). Left occipito-temporal activation might also be
mediated by top-down influences from the left-lateralized major
i.e. the posterior superior temporal gyri, implicated in lexical and
semantic processing, and the inferior frontal cortex, implicated in
syntactic processing (Binder et al., 1997).
The posterior parietal cortex (PPC) is crucial for the genera-
tion of a visuospatial representation (based on bottom-up visual
input from the parafoveal visual field) which then can be used by
cerned with visual information sampling from the top-down, i.e.
visuomotor integration (Andersen & Buneo, 2003; Zihl & Hebel,
1997). The projections from the parafoveal part of V1/V2 to pos-
terior parietal regions illustrate the significance of the parafoveal
visual field for the visual–spatial control of reading saccades.
The transformations carried out in the dorsal processing stream
mediate visuomotor control, and are thus an interface between
perception and action (eye-movements) (Milner & Goodale, 2006).
Bilateral activation of the PPC, with a greater signal on the left is
associated with efficient reading saccades from left-to-right, i.e.
into contralateral hemispace. Evidence suggests that it controls the
is required to read along each single line of text (Leff et al., 2000;
Leff, Scott, et al., 2001).
Bilateral activation of the frontal eye fields (FEF), with a greater
signal on the right is associated with the preparation of this
sensorimotor plan at the beginning of each new line and with
performing the return-sweep, which interrupts the oculomotor
scanpath and requires a change of the sensorimotor plan (ocu-
lomotor flexibility). FEF activation is minimal for the continued
generation of saccadic reading eye-movements along a line of
text. The FEF seem to be less important for visually guided sac-
cades but are crucial for intentional, voluntary generated saccades
irrespective of their direction (Leff et al., 2000; Leff, Scott, et
al., 2001). The FEF plays a key role in the top-down control of
oculomotor scanpaths that follow a previously learned rule (e.g.
reading direction imposed by the writing system). In addition,
the oculomotor aspects of eye-movement control interact with
cognitive processes underlying visual word identification, which
may also determine how long attention maintained at a spe-
cific position, i.e. the temporal aspect of saccade programming
(Heinzle, Hepp, & Martin, 2007). Higher-level linguistic processing
activities in the left anterior inferior prefrontal and left temporo-
parietal cortex may also influence the duration of a fixation
from the top-down (Posner, Abdullaev, McCandliss, & Sereno,
3. Hemianopic dyslexia: reading when the visual world
homonymous visual field defects on reading. His classic descrip-
tion marks the starting point of research into hemianopic dyslexia.
Wilbrand (1907) termed this reading impairment associated with
ing disorder” since hemianopia is the typical and most frequent
It is the “cardinal symptom” which dominates all postchiasmatic
visual pathway pathologies (Lenz, 1909).
3.1. Introducing cerebral visual field disorders
Homonymous visual field disorders account for about 20% of
are caused by injury to the postchiasmatic visual pathway, i.e. to
the optic tract, the lateral geniculate nucleus, the optic radiation,
or to the primary visual cortex (located at the calcarine sulcus)
(Zhang, Kedar, Lynn, Newman, & Biousse, 2006). For these patients
the “visual world shrinks” as vision is lost in both monocular hemi-
fields contralateral to the side of brain injury (Gr¨ usser & Landis,
1991, p. 136). Sufficient spontaneous recovery of the visual field
occurs rarely and, therefore, homonymous visual field deficits can
be regarded as chronic manifestations (Zihl & Kennard, 1996).
In addition, posterior cerebral artery infarctions, the most com-
Zhang et al., 2006; Zihl, 2000), are seldom restricted to calcarine
cortex only. Additional lesions to the occipital white matter, which
might affect fibre pathways connecting occipital, parietal, tempo-
ral and frontal cortical regions, as well as to the posterior thalamus
are the rule rather than the exception for these patients (Hebel
& von Cramon, 1987). As a consequence, the majority of patients
(about 70%) show persistent and severe impairments of reading
and visual exploration (for oculomotor scanning in hemianopia,
2002; Zihl, 1995b, 1999, 2000).
Visual field disorders can be measured quantitatively by peri-
metric techniques (see, e.g., Aulhorn & Harms, 1972) and are
classified according to the portion of the visual field affected. After
unilateral damage, the most common type is hemianopia, the loss
of vision in one hemifield (of both eyes), followed by quadranopia,
the loss of vision in one quadrant, and paracentral scotoma, a small
island-like field defect in the parafoveal visual field. Left-sided
bilateral brain injury, corresponding portions in both visual hemi-
fields are affected. The resulting disorders are analogously termed:
ranopia, and bilateral paracentral scotoma. The loss of vision in the
central visual field region is referred to as central scotoma. Uni-
lateral visual field disorders are much more common than those
resulting from bilateral brain injury (∼90% of patients with visual
vision can either be completely lost (anopia) or one or more visual
functions in the affected visual field can be reduced (amblyopia).
In cerebral amblyopia, light sensitivity is reduced whereas form
and/or colour vision is lost. The selective loss of colour vision is
referred to as achromatopsia (Heywood & Kentridge, 2003). Test-
ing visual functions like colour and form vision requires the use of
special targets and procedures in perimetric testing (see Aulhorn &
The extent of visual field sparing in the affected hemifield is
measured in degrees of visual angle from the fovea. In unilateral
is always spared. Macular sparing (visual sparing between 1 and
5◦to the left or right of fixation) is seldom and most likely results
tions (Zihl, 1989; Zihl & von Cramon, 1986). Approximately 75% of
a parafoveal visual field sparing of less than 4◦. Visual field spar-
ing (co-)determines the resulting functional visual impairment. As
a rule, patients with a smaller field sparing are more disabled,
especially with regard to visual functions that crucially depend on
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
Fig. 2. Schematic illustration of the visual field and perceptual span (comprising of foveal (central white oval) and parafoveal vision (grey ellipse)) for text processing in
patients with left- or right-sided unilateral homonymous parafoveal visual field loss (field sparing: ∼1◦) (affected binocular regions in black). (A) Hemianopia; (B1) upper
quadranopia; (B2) lower quadranopia; (C) paracentral scotoma. Note that the drawing is schematic and not drawn to scale. The cross-hairs indicating fixation position do not
resemble the actual initial fixation position, which would be probably located to the left of the optimal viewing position (i.e. left of the “h” in “that”) in right-sided field loss
(McDonald et al., 2006; Spitzyna et al., 2007); for left-sided field loss, the initial fixation position has not yet been investigated.
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
the parafoveal region, such as reading (Zihl, 1989, 2000). When
parafoveal visual field sparing is smaller than 4◦, 75% of patients
with left-sided field loss and as many as 92% of patients with right-
dyslexia (Zihl, 1994). Fig. 2 schematically illustrates the visual field
anopia, quadranopia, and paracentral scotoma. When visual field
exceeds 10◦(Zihl, 2000).
Diagnosing hemianopic dyslexia requires the presence of a
homonymous unilateral parafoveal visual field loss (as confirmed
by detailed perimetric testing). It is essential to exclude disor-
ders of visual acuity, spatial contrast sensitivity, visual adaptation,
disturbances of the anterior visual pathways or the oculomotor
system, macular disease (as revealed by ophthalmological exami-
processing of text material. Wilbrand (1907) clearly differentiated
dyslexia, reading is impaired despite intact lexical and post-lexical
processes (see also Best, 1917; Poppelreuter, 1917/1990).
A hemianopic reading impairment also must be clearly distin-
guished from pure alexia (letter-by-letter reading) (for the first
report, see Dejerine, 1891). Although pure alexia is usually accom-
panied by a right-sided hemianopia, the visual field defect is not
causally linked to it (for a collection of key articles, see Coltheart,
1998). Pure alexia seems to be associated with a serial encoding
of letters (Behrmann et al., 2001; Rayner & Johnson, 2005). The
diagnosis of hemianopic dyslexia also requires the absence of any
signs of visual–spatial neglect in standard tests. Left-sided hemi-
anopia and visual–spatial neglect often coexist and can be difficult
to disentangle (Walker, Findlay, Young, & Welch, 1991). Despite the
absence of neglect symptoms, however, patients may neverthe-
less exhibit neglect dyslexia (for the first report, see Brain, 1941).
Evidence suggests a clear double dissociation between neglect
symptoms and neglect dyslexia (for a review, see Haywood &
Coltheart, 2000; Riddoch, 1991). Recently, neglect dyslexia was
interpreted as a deficit of extracting visual information from the
left side of space (Behrmann, Black, McKeeff, & Barton, 2002).
Explaining neglect dyslexia in this manner may be reminiscent of
hemianopic dyslexia. Yet, both reading impairments are distinct
disorders and have to be differentiated.
3.2. Reading performance and eye-movements in hemianopic
Since Mauthner’s (1881) first description, several studies have
dealt with hemianopic dyslexia and a high degree of consensus
about its characteristics has been reached. It has consistently been
shown that a visual field defect “is a disturbing obstacle and,
depending on its location to the right or left of the fixation posi-
tion, unpleasantly manifests itself in different ways” (Wilbrand,
1907, p. 6, our translation). Yet, not only whether right- or left-
sided parafoveal vision is affected but also how much of it is spared
(co-)determines type and severity of the resulting reading impair-
ment in homonymous visual field disorders (Mackensen, 1962). It
to-right text reading in patients with unilateral left- or right-sided
unilateral homonymous hemianopia (LH, RH).
3.2.1. Reading performance
In hemianopic dyslexia research reading speed (correctly read
the standard measures of reading performance, are recorded while
patients engage in reading aloud standardised texts, as quickly and
Examples of oral reading in patients with left-sided (A) or right-sided (B) unilat-
eral homonymous parafoveal visual field loss (field sparing: ∼2◦) ([abc] indicates
visual omission errors; [pause] indicates reading interruptions; guessing errors are
The trees were in leaf, and the rumps of the tourist buses were thick and fat
in the traffic, and all the farmers wanted fertilizer admixes rather than
storehouse insulation when Sixsmith finally made his call. In the interim,
Alistair had convinced himself of the following: before returning his
aggrieved letter, Sixsmith had steamed it open and then resealed it. During
this period also, Alistair had grimly got engaged to Hazel. But the call came
[The] trees were in leaf, and [the] rumps of the tourist [b]uses [pause] buses
were thick and fat [in] the traffic, and all the [pause] farmers wanted
fertilizer [ad]mixes rather than [store]house insulation when [Six]smith
finally made his call. In the interim, Alistair had convinced [him]self [pause]
himself of the following: before [re]turning his [ag]grieved [pause] aggrieved
letter, [Six]smith had steamed it open and then [re]sealed [pause] [re]sealed
it. During this period also, Alistair had grimly got [en]gaged [pause] engaged
to Hazel. But the call came
The trees were in [leaf], and the rumps of [the] tourist buses were thick and
fat [in] the traffic, and all [the] farmers want[ed] fertile [pause] wanted to be
fertile [pause] admix[es] [pause] admixture [pause] rather than store[house]
[pause] insulation when Six[smith] [pause] finally made his call. In the
interim, Ali[stair] [pause] Alistair had convince[ed] to convince[d] [pause]
himself of the following: before return[ing] the return[ing] of his aggrieved
letter, Six[smith] had steamed it open and then resealed it. During this
period also, Ali[stair] had grimly got engage[d] to Hazel [pause] an
engagement with Hazel. But the call came
Note. The text is taken from Amis (1994). Figure is adapted from Zihl (2000, p. 72).
accurately as possible. These texts are easy to comprehend and let-
ter size, spacing between lines, words and letters are maintained
as optimal for reading. Reading speed is significantly reduced in
both LH and RH, in comparison with age-matched normal read-
ers. Slowness of reading is the distinctive attribute of hemianopic
dyslexia, and applies not only to text but also to reading single
Zihl et al., 1984; Zihl & von Cramon, 1986). Reading time increases
with each additional letter, especially in patients with small visual
field sparing. Yet the effect is not as pronounced as in letter-by-
letter reading (Behrmann et al., 2001; Leff, Crewes, et al., 2001;
Rayner & Johnson, 2005).
Patients make only relatively few reading errors and are,
therefore, often overlooked in neuropsychological examinations.
Nevertheless, reading errors in hemianopic dyslexia do occur and
can be characterised as visual omissions of letters, syllables, and
even short words. Patients also make meaningful completions of
their end. As a result, errors are introduced by guesses. Patients do
not show letter-by-letter reading or spelling errors. Reading errors
are visually related to the actual word being read and consistently
affect just one side of the word, i.e. the side of the blind field (Zihl,
dyslexia are presented in Table 1.
Oral reading performance (reading speed and errors) consider-
ably differs between left- and right-sided parafoveal visual field
defects. Patients with a left-sided defect require about twice as
much reading time as normal readers. An average reading speed
of 78wpm was measured in a sample of left-sided hemianopic
patients whereas the corresponding figure for normal subjects
(N) was 174wpm (Zihl, 2000). Reading errors mainly consisted of
omissions of prefixes and small words, especially at the begin-
ning of lines (∼4 errors, Zihl, 1995a) (see Table 1(A)). In patients
with a right-sided defect reading speed was only ∼56wpm (Zihl,
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
2000). They also made three times as many errors as patients with
left-sided field loss (∼13 errors, Zihl, 1995a). These errors can be
characterised as omissions and substitutions of suffixes and small
words, especially at the end of lines (see Table 1(B)).
The reading impairment as defined by reading rate and num-
ber of errors is not only related to the side but also to the severity
of the parafoveal visual field loss. Reading time and errors increase
for both left- and right-sided parafoveal visual field loss but is
more pronounced in right-sided field loss (LH: <3◦: 53wpm, >5◦:
124wpm; RH: <3◦: 43wpm, >5◦: 98wpm) (Zihl, 2000).
3.2.2. Reading eye-movements
The first formal electro-oculographic investigations of hemi-
anopic dyslexia were carried by Remond, Lesevre, and Gabersek
(1957) (cited in Ciuffreda, 1994), followed by Mackensen (1962)
and Gassel and Williams (1963a). Mackensen (1962) viewed read-
ing as a sensorimotor ability and therefore regarded the study of
eye-movements in hemianopic dyslexia as indispensable. His eye-
movement recordings revealed a dramatic increase in the number
of fixations and saccades. Gassel and Williams (1963a) observed
similar irregularities in a larger sample of patients with unilateral
homonymous hemianopia. The severe alteration of the oculomotor
reading pattern is the most objective behavioural manifestation of
the reading impairment in hemianopic dyslexia.
Detailed eye-movement analyses have provided a comprehen-
sive understanding of the global temporal and spatial oculomotor
measures associated with text processing during silent reading
in hemianopic dyslexia (Zihl, 1995a, 2000). Overall, prolonged
fixation durations (LH: 310ms; RH: 410ms; N: 250ms), smaller
87; N: 56), and a much higher percentage of refixations (LH: 37%;
RH: 44%; N: 15%) have been reported (Zihl, 2000). The increased
number and duration of fixations and especially the increased like-
in hemianopic dyslexia (McDonald et al., 2006; Zihl, 1995a).
Although word-based analyses of text reading are standard in
experimental reading research (Rayner, 1998), it is only recently
that local spatial and temporal oculomotor measures have been
obtained in patients with right-sided hemianopia. The initial land-
ing position for longer words moves from the centre towards the
beginning and small words are less likely to be skipped (RH: 22%;
the total fixation time are about twice as long as in normal readers
(McDonald et al., 2006; see also Spitzyna et al., 2007). Like experi-
mental reading research all hemianopic dyslexia research has been
based on monocular eye-movement recordings. Binocular record-
and the mechanisms underlying the formation of a single percep-
tual representation from disparate retinal signals (Liversedge et al.,
The differential effects of left- and right-sided parafoveal visual
field loss on reading eye-movement patterns were reported by
Mauthner (1881). Patients with left-sided hemianopia showed dif-
ficulties to find the beginning of a new line. When compared to the
impairments associated with right-sided hemianopia, Mauthner
regarded these difficulties as negligible. In left-sided hemianopia,
to him the “more pleasant” disorder, “only the words which have
already been read disappear, and looking ahead at the upcoming
words is not disturbed” (p. 370, our translation) whereas in right-
sided hemianopia “the despair is enormous that from the point of
ing; hence nothing can be read ahead” (Mauthner, 1881, p. 370, our
translation). Wilbrand (1907) described a considerable uncertainty
and hesitation about where to move the eyes next in patients with
right-sided parafoveal visual field loss; to him, it looked as if their
eyes were stuck at the currently fixated word (see also Best, 1917).
The differences between left- and right-sided parafoveal visual
field loss in the majority of global eye-movement parameters are
illustrated in Fig. 3. In left-sided field visual field loss, the oculo-
motor reading pattern differs from the typical staircase pattern
of normal readers (see Fig. 3(A)) although overall reading perfor-
mance can be regarded as slowed yet, more or less, fluent reading.
The return-sweep appears fragmented. It is reduced to half of its
smaller leftward saccades and show a higher percentage of repeti-
tions of saccades and fixations to the left (Zihl, 2000) (see Fig. 3(B)).
A right-sided visual field defect, on the contrary, impairs shifting
the gaze systematically from left-to-right while the return-sweep
remains unaffected. The staircase-like oculomotor reading pattern
is severely deteriorated and replaced by many small and irregu-
lar saccadic eye-movements to the right (see Fig. 3(C): ellipse).
The amplitude of rightward saccades is significantly reduced and
the total number of saccades increases (Zihl, 2000). Numerous
leftward-directed regressions occur (see Fig. 3(C): small arrow)
and fixation durations are considerably prolonged (up to 1.5s, Zihl,
1995a) (see Fig. 3(C): ovals). These findings have been confirmed
and replicated elsewhere (De Luca et al., 1996; Eber, Metz-Lutz,
Bataillard, & Collard, 1987; Leff et al., 2000; McDonald et al., 2006;
Schoepf & Zangemeister, 1993; Spitzyna et al., 2007; Trauzettel-
Klosinski & Brendler, 1998).
The degree of visual field sparing also clearly contributes to the
irregularities of the oculomotor pattern in terms of an inverse rela-
tionship. Patients with only 1–2◦of field sparing show the most
disturbed oculomotor reading pattern, in particular when the right
hemifield is affected. Patients with a right-sided defect and 5◦of
defect of the same extent who show a close to normal reading eye-
& Brendler, 1998; Zihl, 1995a, 2000).
If at least 3◦to the left and 5◦to the right of fixation are
spared in homonymous visual field loss, reading is nearly unim-
paired (Mackensen, 1962; Trauzettel-Klosinski & Brendler, 1998;
Zihl, 1995a, 2000). Yet, how much visual field sparing is required
for reading to be unimpaired might be better described in number-
of-characters than in degrees-of-visual-angle. It is the number of
characters (in relation to print size) that determines saccade size
in reading (Morrison & Rayner, 1981; O’Regan, 1983) and sets the
spatial boundary of the perceptual and word identification span.
Describing visual sparing in terms of number of characters rather
than visual angle may also explain better Wilbrand’s (1907) finding
that the reading impairment of his patients with a paracentral sco-
toma was more pronounced for small print, despite normal visual
acuity. Print size determines the number of letters which can be
identified at a glance; the smaller the print, the smaller is the word
identification span (Anstis, 1974). The values given above as critical
visual field requirement for unimpaired reading in patients with
visual field defects hold for text that has 3 characters per degree.
Thus, according to hemianopic dyslexia research the perceptual
span extends 15 characters to the right and 9 to the left of fixa-
tion, confirming the asymmetry as well as the right boundary of
the perceptual span in normal readers (McConkie & Rayner, 1975,
1976; Rayner & Bertera, 1979).
The left boundary may vary depending on whether readers
mer case return-sweeps have to be performed, which may require
9 rather than 3–4 characters to the left of fixation. One may specu-
late that the perceptual span flexibly adapts to the changed reading
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
Fig. 3. Infra-red eye-movement recordings in normal readers (A) and in patients with left- (B) or right-sided (C) unilateral homonymous hemianopia (field sparing: ∼1◦).
For illustration purposes, eye-movement patterns for reading five lines (x-axis: time period of recording; y-axis: horizontal extension of line from left to right) are shown in
relation to the visual field and perceptual span for text processing (B and C: affected binocular regions in black). Downward arrows indicate moving the eyes from the end
to the beginning of a new line (which is disturbed in B (crossed arrow) as indicated by ellipses); upward arrows indicate moving the eyes from the beginning to the end of
a line (which is disturbed in C (crossed arrow) as indicated by ellipses). Ovals indicate prolonged fixations, small arrows indicate regressions. Eye-movement recordings are
adapted from Zihl (1995a).
tion. This would also explain the finding that the first word is often
omitted in left-sided parafoveal visual field loss (Wilbrand, 1907).
Most studies investigated reading in patients with hemianopia
and, therefore, less is known about the distinctive effects of a
quadranopia or paracentral scotoma (see Fig. 2: B1/2, C). Nev-
ertheless, there are some data available suggesting that the
Fig. 4. Infra-red eye-movement recordings in patients with left- (A) or right-sided (B) unilateral homonymous hemianopia (field sparing: ∼5◦). Eye-movement patterns
during reading of five lines are shown (x-axis: time period of recording; y-axis: horizontal extension of line from left to right). Note the more or less regular oculomotor
reading pattern in A in comparison to the distorted oculomotor pattern in B (ovals indicate regressions and prolonged fixations). Eye-movement recordings are adapted from
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
characteristic reading difficulties are present in all patients with
unilateral homonymous visual field disorders if parafoveal vision
is affected by brain injury (Mackensen, 1962; Wilbrand, 1907;
Zihl, 2000). Although seemingly small and negligible, paracen-
tral scotomata can disturb reading substantially: Wilbrand’s (1907)
patients reported a “notorious impediment to their usual speed
when gliding along the lines of text” (p. 1, our translation) (see also
Mackensen, 1962). Yet investigations of single cases suggest that
reading speed is higher and the number of fixations and refixations
much smaller than in patients with a hemianopia. A quadranopia
seems to affect reading performance and oculomotor parameters
even less than paracentral scotoma (Zihl, 2000).
Furthermore, it seems that no complete loss of vision (anopia)
is necessary for a reading impairment to emerge. Amblyopic forms
of unilateral homonymous visual field disorders can cause hemi-
anopic dyslexia if the residual visual field for form vision is smaller
than 4–5◦. Since text processing requires the visual discrimination
of forms (letters), the reading impairment in homonymous hemi-
amblyopia is almost identical with the impairment in hemi-anopia
(Wilbrand, 1907; Zihl, 2000). Hemianopic dyslexia is quite com-
subtle hemianopic dyslexia in right-sided unilateral homonymous
quadrant-amblyopia has been reported in detail. The threshold for
luminance detection was elevated and form vision (especially for
increased. Minor abnormalities in the oculomotor reading pattern
were found (Habekost & Starrfelt, 2006).
4. Reading without a parafovea: seeing only half the wor(l)d
Despite a growing literature on reading impairments in uni-
lateral homonymous visual field defects and relatively consistent
Mauthner (1881) and Wilbrand (1907) suggested that it is the loss
of the parafoveal visual field in unilateral homonymous visual field
disorders which causes hemianopic dyslexia. The discussion of the
effects of parafoveal visual field loss on word identification and
oculomotor control in reading, both at the behavioural and neural
level, demonstrates the significance of parafoveal vision for read-
that explaining hemianopic dyslexia as merely a functional con-
sequence of parafoveal visual field loss may not provide the full
4.1. Word identification without a parafovea
The activation of the left occipito-temporal junction associ-
ated with word identification processes is still present in patients
with right-sided homonymous hemianopia. Even patients with a
very small visual sparing show this activation although the nec-
essary input from left striate cortex representing right foveal and
parafoveal vision is missing (Leff, Crewes, et al., 2001). In contrast
to pure alexia, the left occipito-temporal junction as well as its
afferents from left and right striate cortex is spared in hemianopic
dyslexia (Leff et al., 2006). Hence, information from the intact con-
must be transferred to the left occipito-temporal junction via the
splenium of the corpus callosum. Intact afferent connections from
the right occipital cortex to the left occipito-temporal junction
appear sufficient to support word identification in patients whose
right parafoveal vision is compromised (Leff, Crewes, et al., 2001).
Word identification through this indirect route, however, can
be regarded as less efficient. In left-to-right readers, words are pro-
cessed and identified best in the right foveal/parafoveal visual field
2000; Nazir, Ben-Boutayab, Decoppet, Deutsch, & Frost, 2004). This
may also explain the finding that patients with larger right-sided
visual field sparing and patients with left-sided visual field loss (i.e.
right-sided injury) are less impaired in word identification (Upton
et al., 2003).
In most patients with unilateral homonymous parafoveal visual
field loss the perceptual and visual span may be no more than
3–4 characters. Yet the more letters can be identified at a single
itates faster reading. The visual span imposes a limit on reading
speed and is also referred to as the “sensory bottleneck” in reading
(Legge et al., 2007). If this bottleneck is additionally restricted by
parafoveal visual field loss, forward saccades become smaller and
many more saccades have to be made to extract the same amount
of text information for correct word identification. This effect is
most pronounced when reading longer words (Leff, Crewes, et al.,
2001; McDonald et al., 2006). Converging evidence stems from low
vision readers whose visual and perceptual span is restricted by
foveal processing difficulties due to macular disease (Chung, Legge,
& Cheung, 2004; Crossland & Rubin, 2006; Legge, Ahn, Klitz, &
Parafoveal visual field loss prevents that the beginning and
end of a word are simultaneously visually apprehended. Especially
longer words are never seen as a whole and parallel letter pro-
cessing, which is required for efficient lexical word identification
of half-seen words are encoded and forwarded to higher-level lin-
guistic processing units. Often, the visual input is insufficient to
activate corresponding representations in the mental lexicon. If
an incompletely encoded word makes sense and activates a lexi-
cal representation, visual omission errors emerge. Guessing errors
occur if the predictive value of the incomplete percept is used for
a meaningful completion of the word (Zihl, 2000). For instance,
words which can be misread by omission or substitution of the first
letter (e.g. peach: each or beach) increase the likelihood of errors in
left-sided parafoveal visual field loss (Ellis, Flude, & Young, 1987).
Patients seem to over-rely on higher-level linguistic processes
to compensate for the missing visual information when trying to
word (comprehension) rather than inspecting the entire word first
(visual apprehension) is the preferred strategy. Higher-level lin-
guistic processes come into play too early which disrupts further
acquisition and processing of text information located in the blind
hemifield. Overall, processing words when the parafoveal visual
field is compromised requires more time. Difficulties in word iden-
tification, which also affect language comprehension, are reflected
by longer fixation durations. As regressions occur as attempts to
correct linguistic processing difficulties (Rayner & Pollatsek, 1989),
Hence, for processing text information, patients make use of
their residual foveal/parafoveal vision and linguistic processes.
Reading in parafoveal visual field loss can, therefore, still be
regarded as non-random visual information sampling (see also
McDonald et al., 2006). Further evidence stems from a small sam-
ple of patients with a left- or right-sided homonymous hemianopia
(De Luca et al., 1996). Patients identified high-frequency words
much quicker than low-frequency words. Reading passages con-
taining low-frequency words was associated with an increased
number of saccades and regressions, longer fixation durations and
smaller saccadic amplitudes. Words embedded in a textual con-
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
text were identified quicker than words in isolation (contextual
constraints). Meaningful words were inspected and read quicker
errors than is reading meaningful text material as much less facili-
Poppelreuter’s (1917/1990) patients with a right-sided paracentral
scotoma showed a pronounced reading impairment when con-
fronted with meaningless or Latin text but “read familiar text (...)
like a normal” (p. 224). High contextual constraint (as determined
by word predictability) facilitates word skipping, reduces fixation
& Rayner, 1981; Pynte & Kennedy, 2006; Rayner, 1998; Rayner &
Right-sided parafoveal visual field loss affects not only pro-
cessing the foveal word but also impairs preprocessing of the
to-be-identified word located in the parafovea. During a fixa-
tion, readers process information from the fovea and parafovea;
attentional top-down processes facilitate processing of the foveal
text information first and the attentional focus then shifts to
the parafoveal visual field (Reichle, Rayner, & Pollatsek, 2003).
Parafoveal preprocessing is indispensable for maintaining fast and
fluent reading (Blanchard, Pollatsek, & Rayner, 1989; Inhoff, 1987;
view benefit in right-sided parafoveal visual field loss contributes
increase in fixation duration and number of fixations (McDonald et
a single fixation and, therefore, a larger proportion of words have
to be fixated. Furthermore, guiding reading saccades towards the
centre of the to-be-fixated word where word processing is opti-
mal (O’Regan & L´ evy-Schoen, 1987; Vitu, O’Regan, & Mittau, 1990)
becomes increasingly difficult as it requires right parafoveal word-
length information. The initial fixation position wanders towards
the beginning of the word and thus further away from the optimal
viewing location. The resulting difficulties in word processing are
reflected by longer fixation durations and an increased number of
refixations (McDonald et al., 2006).
4.2. Visual guidance of reading eye-movements without a
Parafoveal visual field loss disturbs the integration of visual
and motor process: “Successive gaze-shifts from left to right (...)
are no longer in the order dictated by the visual information, but
occur irregularly” (Poppelreuter, 1917/1990, p. 224). Visual infor-
mation extraction from the parafoveal (and peripheral) visual field
regions that provides the basis for a top-down control of visual
attention and eye-movements in space and further local process-
ing of fine details is impaired (Hochstein & Ahissar, 2002; Juan &
Walsh, 2003; Upton et al., 2003). Word- and line-length as well
as page boundary information may be represented at higher levels
and form a coordinate system containing the relative spatial loca-
tion of word-objects (Kennedy, Brooks, Flynn, & Prophet, 2003).
This spatial coordinate system enables the attentional selection
of the to-be-identified word. Saccades are computed accordingly
detail are initiated (Deubel, O’Regan, & Radach, 2000; Deubel &
Schneider, 1996; McConkie & Zola, 1987).
duces a less pronounced reading impairment than a hemianopia
may confirm such an assumption. He speculated that the lines
above and below the paracentral scotoma may be used for visual
guidance of reading eye-movements. A hemianopia, on the con-
trary, prevents the extraction of this visual information (compare
Fig. 2: A vs. C). Although the information below and above the
current line does not affect oculomotor control in normal read-
ers (Pollatsek, Raney, LaGasse, & Rayner, 1993), it may alleviate the
reading impairment in patients with a paracentral scotoma.
A functional neuroimaging study (PET) investigated reading
eye-movements in three patients with right-sided homonymous
hemianopia and complete loss of right parafoveal vision (Leff
et al., 2000). Eye-movement recordings of text reading revealed
abnormal oculomotor reading parameters and reading speed was
severely reduced. Instead of the left-lateralised PPC and right-
lateralised FEF activation observed in normal readers, PPC was
symmetrically activated and FEF activation was left-lateralised.
Interestingly, a patient with a right-sided homonymous hemi-
anopia that spared parafoveal vision showed the normal patterns
of activation. His reading speed was in the range of age-matched
controls and the oculomotor reading pattern was, despite a slight
increase in the number of rightward saccades, more or less nor-
mal. Hence, the extent of the visual field defect seems to determine
the level of functioning of the neural systems (PPC, FEF) subserv-
ing eye-movement control during text reading. Based on these
results hemianopic dyslexia was interpreted as a disconnection of
the motor systems involved in planning and guiding reading sac-
cades from the representation of right parafoveal vision in the left
striate and prestriate cortex (Leff et al., 2000).
striate cortex, left-lateralised PPC activation) seems to be of greater
importance for controlling oculomotor activities along a line of
text than the right hemisphere (Leff et al., 2000; Leff, Scott, et
al., 2001). The observation that right-sided parafoveal visual field
loss (left-sided injury) impairs left-to-right reading more severely
than a left-sided field loss (right-sided injury) is in line with
this finding. However, one might question a fundamental, hard-
Reading-related brain activation and its lateralisation appears to
be functionally determined as indicated by investigations of the
neural basis of reading across writing systems (Al-Hamouri et al.,
2005; Bolger, Perfetti, & Schneider, 2005; Skoyles, 1988). Evidence
suggests that cultural differences in writing systems are reflected
by differential activation across the neural network(s) mediat-
ing reading-related processes (Schlaggar & McCandliss, 2007). The
influence of reading direction on text information processing and
related eye-movements might be mediated by a top-down control
which determines the dynamics of visuospatial attention alloca-
tion, i.e. the size and location of the perceptual span (Osaka, 2003).
The reversed asymmetry of the perceptual span in right-to-left
1981) supports this assumption.
Converging evidence stems from a case study that reports a
skilled bilingual reader with a left-sided hemianopia who had
pronounced reading difficulties in his mother tongue Hebrew
(right-to-left reading) but not in his second language English (left-
to-right reading) (Leker & Biran, 1999; Mohamed, Elsherbiny, &
Goulding, 2000). That the asymmetry of the perceptual span in
bilinguals flexibly adapts according to the reading direction of the
language which is currently being read is in line with this study
differences in reading impairment between left- and right-sided
parafoveal visual field loss might be functional: in right-to-left
writing systems a “right-sided hemianopia appears to be more
desirable” (Mauthner, 1881, p. 370, our translation). Functional
neuroimaging (and behavioural) studies of hemianopic dyslexia
in right-to-left writing systems would be very illuminating in this
regard. Comparing patients with right-sided parafoveal visual field
loss in left-to-right writing systems with patients showing left-
sided parafoveal visual field loss in right-to-left writing systems
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
of the cortical structures involved in text processing and related
Planning and guiding the return-sweep is associated with right
visual field (McConkie & Zola, 1987). Left-sided parafoveal visual
line and, therefore, affects the visual guidance of the return-sweep.
The observation that overly long lines disrupt the return-sweeps
of normal readers supports this assumption (Gassel & Williams,
1963a; Rayner & Pollatsek, 1989). Gassel and Williams (1963a)
reported that the return-sweep of their left-sided hemianopic
patients improved after reading a few lines. In contrast to left-to-
right reading saccades, the return-sweep’s saccadic target, i.e. the
first word of the next line, is almost always at a fixed horizontal
position (most print text is left justified). After some practice with
a text the spatial coordinates of the left text boundary might be
represented within a higher-level framework, which may mitigate
the effects of a left-sided parafoveal visual field loss on the visual
guidance of the return-sweep.
5. Looking beyond parafoveal visual field loss: is
hemianopic dyslexia purely visually elicited?
Poppelreuter (1917/1990) pointed out that “the impairment
“the disturbance of the co-ordination of the reading gaze-shifts”
(p. 224) associated with hemianopic dyslexia may not be solely
visually elicited. Examining adaptation processes in homonymous
parafoveal visual field loss and the anatomical conditions that are
responsible for the severe and long-lasting reading impairments in
patients with hemianopic dyslexia will show that parafoveal visual
field loss is a necessary yet not a sufficient condition that causes
5.1. Hemianopic dyslexia and the question of spontaneous
Moving masks and window studies with normal readers may
suggest that hemianopic dyslexia is purely visually elicited. Visual
masks or windows occluding either the foveal or parafoveal visual
field produce reading impairments in normal readers similar to
those caused by homonymous visual field disorders (Cummings &
McConkie & Rayner, 1975, 1976; Rayner & Bertera, 1979; Rayner,
Inhoff, Morrison, Sowiaczek, & Bertera, 1981).
Reading using parafoveal and peripheral vision, i.e. “reading
without a fovea” (Rayner & Bertera, 1979), is almost impossible
(see also Fine & Rubin, 1999a, 1999b, 1999c; Rayner et al., 1981)
as is found by patients with a central scotoma (Teuber, Battersby, &
Bender, 1960). Two single cases have been reported where reading
speed was as low as 3 and 12 words per minute (see Zihl, 2000,
pp. 151–164). Having to rely exclusively on foveal vision (reading
without both parafoveas) also makes reading difficult. A one-letter
moving window forces normal readers into letter-by-letter read-
ing (Rayner & Bertera, 1979; Rayner et al., 1981), similar to the
reading-style of pure alexics (Johnson & Rayner, 2007; Rayner &
Johnson, 2005). The ‘natural’ counterparts of these experimental
moving windows are bilateral homonymous visual field disorders,
found the distinctive reading impairments of left- and right-sided
parafoveal visual field loss appear in combination in such patients.
Reading performance is worst in patients with a bilateral hemi-
anopia (tunnel vision) as their residual vision may be nothing else
than a one-letter moving window (Zihl, 2000). Reading without a
parafovea may be less difficult. Yet, obliterating the left or right
parafoveal visual field in normal readers produces reading impair-
ments similar to hemianopic dyslexia (Cummings & Rubin, 1992;
Ikeda & Saida, 1978; McConkie & Rayner, 1975, 1976; Rayner et al.,
1981; Rayner, Liversedge, & White, 2006).
One must not forget that if an artificial visual field defect
is imposed the resulting reading impairments are not as
severe and long-lasting as in hemianopic dyslexia. Normal sub-
jects seem to adapt quickly to visual field loss (Poppelreuter,
1917/1990), although interindividual differences may be substan-
tial (Zangemeister & Utz, 2002). Furthermore, not all patients with
unilateral homonymous parafoveal visual field loss necessarily
show impaired reading. Adequate reading performance was found
in 16% (out of 50 cases) about six weeks after brain injury (Zihl,
1995a), and in 29% (out of 35 cases) when followed over a period
of three years (Gassel & Williams, 1963a). Despite the prevailing
as the concomitant abnormalities of the oculomotor parameters
sided homonymous hemianopia with only 0.5◦visual field sparing.
tion of the extent to which hemianopic dyslexia has a purely visual
From his investigations Mackensen (1962) concluded that the
severity of the reading impairment is not only determined by the
presence of parafoveal visual field loss but also by whether and
how well one has learnt to compensate for the visual defect. To
overcome their visual impairment the most obvious solution for
patients seems to be using appropriate eye-movement strategies.
Patients consistently shift their gaze, thus their visual field bor-
der, into the area corresponding to their blind hemifield, thereby
bringing obscured visual information briefly into the seeing field.
It was Poppelreuter (1917/1990) who first reported spontaneous
oculomotor compensation in visual field loss.
There is a consistent set of compensatory oculomotor strategies
to which patients resort in order to cope with their lost part of the
visual field. Targets located in the blind hemifield are approached
with a safe-but-slow staircase strategy (series of small stepwise,
hypometric saccadic eye-movements) especially if the target is
unpredictable (Meienberg, Zangemeister, Rosenberg, Hoyt, & Stark,
1981). Most patients resort to such strategy, which is, however,
time-consuming, laborious and simply insufficient to effectively
compensate for parafoveal visual field loss (see also Poppelreuter,
1917/1990; Williams & Gassel, 1962; Zihl, 2000). They also employ
this careful, safe-but-slow staircase strategy in reading (“begin-
ning and end of line detective”)—their reading rate is considerably
reduced, and the number of errors is increased, in comparison with
acterised by top-down guided, predictive overshooting saccades
in the direction of the blind field (“blind hemifield overshooting”)
(Meienberg et al., 1981; Zangemeister, Oechsner, & Freska, 1995;
Zangemeister & Utz, 2002).
Such spontaneous adaptative strategies are, however, rarely
found (Schoepf & Zangemeister, 1993). A common observation
is rather that patients with homonymous visual field loss shift
their head towards the affected side (Zihl, 2000). As head move-
ments normally follow and depend on saccadic eye-movements
(Uemura, Arai, & Shimazaki, 1980), reversing this normal physio-
logical sequence to compensate for visual field loss is maladaptive
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
tomas regain normal reading performance despite only 1–2◦of
persist (Zihl, 2000). So, parafoveal visual field loss in itself cannot
completely account for hemianopic dyslexia.
Consequently, there must be specific requirements for the abil-
ity to develop a compensatory eye-movement strategy with time
(see also Kennard, 2002). Unquestionably, effective compensation
implies some (implicit) knowledge of how to compensate (Zihl,
2000). Furthermore, learning to cope with a homonymous visual
field loss and developing spontaneous compensatory strategies
should require some time: “the complicated processes of com-
pensation (...) can come to light as only slowly and gradually
acquired improvements” (Poppelreuter, 1917/1990, p. 239). Evi-
dence suggests that reading performance and the concomitant
eye-movement parameters can improve with time after the onset
of visual field loss (Gassel & Williams, 1963a). However, patients
seem to either start very early spontaneously compensating for
their parafoveal field loss or they do not regain normal reading per-
formance even several weeks or months after their initial visual
field loss (Zihl, 1995a). Thus, patients can be classified into two
categories according to whether or not they develop spontaneous
The decisive factor seems to be whether injury to the postchi-
asmatic visual pathway is accompanied by additional injury to the
occipital white matter, occipitoparietal structures, or the posterior
thalamus. Patients in which these structures are spared show very
efficient spontaneous oculomotor compensation, notwithstanding
very small degrees of visual field sparing. Even after posterior cere-
bral artery infarction extra-striate injury is the rule rather than
the exception (Hebel & von Cramon, 1987), which may explain
the high percentage of patients showing little or no spontaneous
the fact that these structures and their reciprocal connections are
assumed to be part of a cortical–subcortical network subserving
the bottom-up and top-down control of visual–spatial attention
and related saccadic eye-movements (Corbetta, 1998; Corbetta &
taneous compensatory oculomotor strategies. If the structures and
interconnecting callosal fibre pathways of this network are spared,
parafoveal visual field loss can be mitigated by a specific set of
top-down controlled intentional saccades into the blind hemifield.
As a consequence, the regular eye-movement pattern required for
effective text processing can be regained (Zihl, 1995a).
5.2. Hemianopic dyslexia and its anatomical basis
From our knowledge of the anatomy of the retino-striate visual
pathway, we can infer the anatomical loci in which damage gave
rise to a specific pattern of visual field loss. Injury to central,
i.e. postchiasmatic portions of the pathway leads to character-
istic homonymous visual field defects, which can be predicted
from the retinotopic organization of the pathway. Most commonly,
lesions are located in the optic radiations and the striate cortex
anopia with and without macular sparing. Injury to the posterior
part of the optic radiations and the striate cortex results in con-
gruous homonymous visual field defects, i.e. they share the same
location, extent and shape in the two monocular visual fields.
Incongruous and incomplete defects typically occur in cases with
injury to the anterior parts of the postchiasmatic pathway (optic
tract, lateral geniculate body, and the anterior part of optic radia-
tion) (Harrington, 1976; Zihl & von Cramon, 1986).
However, although no empirical data are available, it appears
reasonable to assume that for hemianopic dyslexia to emerge the
locus of damage along the postchiasmatic visual pathway is not
decisive and has no differential effects on the resulting reading
impairment. Hemianopic dyslexia can be caused either by injury
to the optic tract, the lateral geniculate body, the optic radiation, or
tional damage to the fibre pathways and/or structures constituting
the neural network subserving the bottom-up and attentional
top-down control of visual information processing and saccadic
eye-movements in reading (Zihl, 1995a). Injury to the primary
visual cortex (or its geniculostriate afferents) in itself (Leff et al.,
2006) cannot completely account for hemianopic dyslexia.
anopic dyslexia (Zihl, 1995a). In a sample of 50 patients with left-
or right-sided homonymous hemianopia, reading performance of
tion of primary visual cortex) only (16% of patients) was close
to normal (155wpm; ∼89% of normal reading performance, N:
175wpm), and sufficient for their occupational and daily life. In
cases with larger lesions involving the striate cortex and partially
the occipital white matter (44% of patients) a moderate reading
impairment was found (108wpm; ∼62% of normal reading perfor-
mance). Extensive unilateral injuries involving the occipital white
matter (in 26% of patients) and the posterior thalamus (in 14% of
patients) resulted in a severe and long-lasting reading impairment
performance) (Zihl, 1995a).
Reports of three single cases with right-sided homonymous
hemianopia (visual field sparing: 2◦) further confirm and illustrate
(Zihl, 1995a). Despite showing the same visual field defect and the
same field sparing, these patients differed greatly with regard to
their reading speed (A: 120wpm; B: 82wpm; C: 32wpm). In addi-
tion, they did not differ with regard to age (A: 50 years; B: 46 years;
C: 46 years) nor time since lesion (A: 8 weeks; B: 9 weeks; C: 14
weeks). Although patient C had the longest time since lesion, and
the most severe reading impairment. A comparison of their under-
reading performance, the lesion was restricted to calcarine cortex;
in patient B, who showed a moderate reading impairment, occip-
ital white matter was, in addition, partly affected; in patient C,
ital white matter and the posterior thalamus were affected. The
homonymous hemianopia (LH; 2◦) (Zihl, 1995a).
The pronounced differences in reading speed between right-
and left-sided parafoveal visual field losses seem to diminish if the
extent and site of lesions is controlled for when making the com-
parison. The lesions of patient A (RH) and D (LH; age: 46 years,
time since lesion: 7 weeks) were restricted to calcarine cortex, and
ilar reading speeds (A: 120wpm; D: 105wpm); Patient B (RH) and
E (LH; age: 52 years, time since lesion: 8 weeks) had both partly
occipital white matter involvement and showed a similar reduc-
tion in reading speed (B: 82wpm; E: 87wpm). Patient C (RH) and
F (LH; age: 58 years, time since lesion: 9 weeks) showed both
extensive occipital white matter involvement and a severe reading
impairment. Yet, the greater reduction of reading speed in patient
C (32wpm; F: 68wpm) cannot be fully explained by the difference
tional involvement of damage to the posterior thalamus in patient
C has to be taken into account (Zihl, 1995a).
etal and frontal lobes, and the limbic neocortex are involved in the
visual guidance of eye-movements (Ogren, Mateer, & Wyler, 1984;
Robinson & Petersen, 1992; Zihl & von Cramon, 1979). Injury to
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
the occipital white matter might damage the fibre pathways which
connect the visual areas of the brain to motor areas for the visual
control of eye-movements. In addition to the cortico-cortical fibre
connections between visual, parietal and frontal areas, the subcor-
tical pathways connecting visual cortical areas and pontine cells,
which also receive input from the superior colliculus, may also be
Injury to the striate cortex or its geniculostriate afferents, the
occipital white matter comprising subcortical–cortical recipro-
cal connections, and/or the posterior thalamus, causes parafoveal
visual field loss. These injuries may impair, to varying degrees, the
bottom-up and attentional top-down control of visual informa-
tion processing in the fovea and parafovea and the eye-movements
involved in reading. Lesions that are confined to calcarine cor-
tex result in parafoveal visual field loss which may disturb visual
information processing and bottom-up oculomotor control (Leff
et al., 2000). However, restricted calcarine cortex lesions preserve
the neural network that mediates efficient visual processing and
top-down control can facilitate visual information processing and
the guidance of eye-movements into the blind field via feedback
connections. The interactive flow of activation between V1/V2 and
parietal as well as frontal cortical regions via feedforward (bottom-
up) and feedback connections (attentional top-down modulation
of V1/V2) supports such view (Foxe & Simpson, 2002; Hochstein &
Ahissar, 2002; Juan & Walsh, 2003). Where top-down attentional
mechanisms are intact, an oculomotor strategy can be developed
and adjusted to compile a complete percept of each word, even
though each fixation provides only an incomplete view. Sponta-
neous oculomotor adaptation efficiently substitutes the lost visual
more or less intact (Zihl, 1995a, 2000).
Evidence on the anatomical basis of hemianopic dyslexia
allows us to conclude that this reading impairment is more than
purely visually determined. Hemianopic dyslexia is not caused
by parafoveal visual field loss resulting from unilateral postchias-
abnormalities in reading and related impairments of text process-
ing require widespread damage to the distributed neural network
subserving the bottom-up and attentional top-down control of
visual information processing and saccadic eye-movements in
reading. In contrast, patients with sparing of the structures belong-
ing to this neural network usually compensate for their parafoveal
visual field loss and show a close to normal reading performance.
The high frequency of combined striate/white matter lesions in
patients with homonymous visual field defects (Hebel & von
Cramon, 1987) nevertheless justifies the further usage of the term
hemianopic dyslexia to characterize this special type of reading
impairment (Zihl, 1995a).
Our current knowledge of the anatomical basis of hemianopic
dyslexia is based on an analysis of CT and MRI scans (Zihl, 1995a).
These methods may underestimate the extent of lesions. An uni-
lateral lesion to the optic radiation or striate cortex might change
glucose metabolism in the intact ipsilateral thalamus and visual
association areas as revealed by PET studies (Bosley et al., 1985).
Such ‘remote’ effects are interpreted as interruption of the fibre
pathways interconnecting both structures, which leads to a deac-
tivation of the primary intact structure (Gr¨ usser & Landis, 1991).
These effects have to be differentiated from primary lesion sites
for a valid interpretation of behavioural deficits and for devel-
oping a model of the functional organisation of the processes
underlying complex behaviour such as reading (Zihl, 1995a). Con-
sequently, we may (re-)interpret the effects on reading-related PPC
and FEF activation patterns in patients with right-sided homony-
left V1/V2 damage. It is also possible that fibres connecting cortical
visual areas with parietal and frontal areas may have been affected
in these patients.
6. The rehabilitation of hemianopic dyslexia: re-learning
eye-movement control in reading
Poppelreuter (1917/1990) was the first who systematically
attempted training patients with hemianopic dyslexia to learn, or
re-learn, oculomotor control in reading. He showed convincingly
that in patients with a lost parafoveal visual field “relearning of
reading was successful” (p. 249). As Poppelreuter (1917/1990) said,
the main goal for patients is “to make the preserved paracentral
portion (...) a field for reading” and to move “the location of the
position of the clearest vision further to the right or to the left” (p.
248). He taught his patients to use a wooden reading stick which
they moved successively from word to word of the text they read
off a board. Patients with a right-sided visual field loss were asked
to place the reading stick at the end of the word that is currently
being read, patients with a left-sided defect had to place it at the
beginning of words. Patients therefore learn to shift their atten-
tion and gaze intentionally into their blind field. After a few weeks
of training, hesitant reading gave way to regular reading with cor-
hemianopic dyslexia, see Mackensen, 1962; Zihl, 2000), and diffi-
culties in identifying words and text comprehension were reduced.
Reading speed increased and errors were reduced (Poppelreuter,
Gassel and Williams (1963b) also found that the refinement and
employment of attentional and gaze-shifts are the basis for ocu-
lomotor compensation in patients with homonymous visual field
defects. To regain reading performance, patients have to intention-
ally shift their gaze further than they can actually see, i.e. into their
again; they learn “to keep the ‘blind side’ in sight” (Zangemeister
& Oechsner, 1999, p. 89). Well-timed gaze shifts can re-establish
the temporal and spatial coherence of successive extracted parts of
visual information, which leads to the experience of seeing words
at one glance again (Zihl, 1995a). Intentionally shifting attention
and gaze so as to perceive each word as a whole before reading
it is of particular importance in ameliorating word processing and
identification difficulties. It is crucial that patients learn to visually
apprehend before comprehend a word (Zihl, 2000).
Although Poppelreuter’s (1917/1990) wooden reading stick has
not stood the test of time, the rationale behind his quirky pro-
cedure is still valid. It has survived in form of a compensatory
treatment approach for rehabilitating patients with hemianopic
dyslexia using, instead of a wooden stick, an electronic reading
aid with gliding text (Zihl, 1995a; Zihl et al., 1984). An alternative
yet more flexible and efficient treatment method is the PC-based,
tachistoscopic presentation of text material (Zihl, 2000). Regular
and systematic massed practice allows new oculomotor strategies
to be consolidated into flexible oculomotor reading routines (Ofen-
Noy, Dudai, & Karni, 2003). Over-learning gradually leads to the
‘automatization’ of this strategy and hence comfortable reading
(see also B¨ ackman, 1999).
Eye-movement recordings after only a few training sessions
(about 10–15 sessions, 45min) reveal more or less normal oculo-
motor reading patterns and reading performance in the majority of
cases (Zihl, 1995a, 2000). Overall, patients make fewer and shorter
The amplitude of rightward saccades increases especially in right-
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
Fig. 5. Infra-red eye-movement recordings before and after treatment in patients with left- (A) or right-sided (B) unilateral homonymous hemianopia (field sparing: ∼3◦).
Eye-movement patterns during reading of five lines are shown (x-axis: time period of recording; y-axis: horizontal extension of line from left to right). (A) Note the prolonged
fixations and regressions (ovals) as well as the interrupted return sweep (ellipse) before and the normalisation of the oculomotor pattern after treatment. (B) Note the
distorted left-to-right oculomotor reading pattern (prolonged fixations, smaller forward saccades, and regressions (ellipses)) before and its normalisation after treatment.
Eye-movement recordings are adapted from Zihl (1995a).
leftward saccades (return-sweeps) (see Fig. 5) (Zihl, 1995a, 2000).
Reading fluency is regained; reading speed increases (before treat-
ment: 76wpm (LH), 53wpm (RH); after treatment: 113wpm (LH),
96wpm (RH) (Zihl, 1995a)) and fewer errors are made. Follow-
up assessments show that these treatment effects remain stable
(Kerkhoff et al., 1992; Spitzyna et al., 2007; Zihl, 1995a, 2000).
After treatment, reading performance in hemianopic dyslexia is
markedly improved although parafoveal vision is still lost. Re-
organizing oculomotor control is decisive for making our ‘optical
instruments’ useful once again for reading (Gassel & Williams,
Most patients benefit from systematic oculomotor practice.
Patients with right-sided parafoveal visual field loss, however,
similar left-sided field loss. Even then, they still show a poorer out-
come in comparison with patients with left-sided field loss (Zihl,
1995a, 2000). The differential or “asymmetrical” effect of left- and
right-sided parafoveal visual field loss on rehabilitation outcome
appears to be specific to reading. When scanning a visual scene
visual field defects as in reading (Tant et al., 2002; Zihl, 1995b,
1999). Furthermore, oculomotor scanning performance, in con-
trast to reading performance, is not associated with the degree
of visual field sparing in such a way that the smaller the sparing,
the more impaired is oculomotor scanning. Also the location of
the visual defect within the visual field is much more important
in determining the resulting reading impairment than the scan-
ning impairment (Poppelreuter, 1917/1990; Zihl, 1995b). Wilbrand
(1907) reported that small paracentral scotomas only posed an
impediment to reading but not to exploring surroundings in his
patients. Furthermore, spontaneous oculomotor adaptation to a
homonymous visual field disorder in visual exploration is more
In addition, both abilities seem to require specific training for their
improvement and there appears to be no obvious transfer effect
between both domains. One may speculate that the control of the
oculomotor scanpath for reading is mediated by different neural
networks than the scanpath for visual exploration, although both
networks probably overlap (Zihl, 1995a, 1995b, 2000).
The effect of top-down text processing strategies on inter- and
intraindividual variation in reading ability might be marginal in
normal readers (see O’Regan, 1992). Yet, differences in factors such
as self-control may be crucial when a new reading strategy has to
field loss more quickly than ‘cautious’ word-by-word readers do
(although possibly at the expense of omitting words or syllables).
Rehabilitation of hemianopic dyslexia depends on perceptual
and oculomotor (i.e. procedural) learning processes. Learning and
consolidation of new oculomotor reading strategies are top-down
guided and modulated by attention. PPC function may play an
important role in mediating these learning processes. Right PPC is
crucial for perceptual learning and attention, and practice-related
decrease of activation has been observed for the practice of visual
search tasks (Walsh, Ashbridge, & Cowey, 1998; Walsh, Ellison,
Ashbridge, & Cowey, 1999).
That patients with additional extensive injuries to the occip-
ital white matter and/or to occipitoparietal regions require the
dyslexia and demonstrates the importance of intact functional
connections between the cortical visual areas and the areas that
supposedly mediate the treatment effect.
Interestingly, patients with normal visual fields but posterior
parietal damage reported difficulties in finding their way through
lines of text on a page (Zihl & Hebel, 1997). Comparing this read-
ing impairment to the reading difficulties of patients who have a
similar posterior parietal involvement but an accompanying uni-
lateral homonymous parafoveal visual field loss could illuminate
the relative contributions of attentional posterior parietal and sen-
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
sory striate cortex functions to reading and also to learning new
oculomotor reading strategies.
tigation of brain representation of visually guided saccades in a
small group of patients with pure striate cortex lesions resulting
in right- or left-sided homonymous hemianopia. These patients
showed no impairments of visual exploration or reading. Making
saccades to targets presented in the intact and compromised hemi-
field was associated with a bilateral activation of the frontal and
Increased activation in patients was found in the posterior parietal
cortex of the unaffected hemisphere, i.e. contralateral to the side
of the intact hemifield, suggesting that visual field defects after
striate lesions are associated with changes in the fronto-parietal
network underlying the cortical control of saccades. Whether this
activation represents a neural correlate of (spontaneous and/or
training-induced) oculomotor compensatory processes needs fur-
ther study (Nelles et al., 2007).
Mirror reading provides further insights into the involvement
of parietal and also frontal mechanisms in the rehabilitation of
hemianopic dyslexia. The acquisition of mirror-reading skill in
normal subjects is associated with changes in activation patterns
of posterior brain regions and stronger activation in the parietal
associative cortex and the frontal eye fields. After training when
reading strategies have been learned successfully and become
routine, a practice-related decrease of activation in prefrontal
and posterior parietal areas is observed (Kassubek, Schmidtke,
Kimmig, L¨ ucking, & Greenlee, 2001; Poldrack et al., 1998; Poldrack
& Gabrieli, 2001). Prefrontal cortex activity is critical for procedu-
ral learning (Beldarrain, Grafman, Pascal-Leone, & Garcia-Monco,
1999; Jueptner et al., 1997; Miller & Cohen, 2001) and the FEF in
particular are involved in intentional, voluntary generated atten-
tional and eye-movement shifts according to a rule (Heinzle et al.,
Many assumptions about the underlying mechanisms of the
resulting improvement in rehabilitation of hemianopic dyslexia
must remain speculative without evidence from functional neu-
roimaging. Nevertheless, the finding that the lost parafoveal visual
field region can be successfully substituted by spontaneous or
ticity of the visual, attentional and oculomotor systems and their
underlying neural mechanisms involved in text reading. Read-
ing eye-movements can be controlled either from bottom-up
(parafoveal visual field) or via an attentional top-down text pro-
7. Synopsis: insights from and into hemianopic dyslexia
A great deal has been learnt about hemianopic dyslexia since
it was first reported by Mauthner in 1881. Studying patients with
hemianopic dyslexia offers important insights into the normal
ing theories and models of reading and eye-movement control.
Hemianopic dyslexia is a special type of reading impairment that is
caused by injury to systems subserving the bottom-up and atten-
tional top-down control of visual information processing in the
foveal and parafoveal visual field and saccadic eye-movements
involved in reading. The anatomical basis of hemianopic dyslexia
lostriate afferents compromising the representation of parafoveal
reading impairments lies elsewhere. It is in the fibre pathways
that reciprocally connect the visual areas of the brain to the pari-
etal, frontal, and temporal areas, as well as to the subcortical areas
involved in the control of visuospatial attention and the guidance
of the scanpath in text processing.
Hemianopic dyslexia provides valuable neuropsychological
insights into the neural mechanisms essential for normal read-
ing. It shows that the visual field is not only a sensory surface or
passive information intake zone but “as much a measure of the
attention (...) as of the anatomical substrate” (Williams & Gassel,
1962, p. 243). Visual information processing in reading requires
guistic processes, facilitates visual processing at the early stages
of the reading process for word identification and eye-movement
eye-movement control, visuospatial attention, and linguistic pro-
brain regions supported by large-scale neural networks.
Hemianopic dyslexia shows that parafoveal vision is crucially
involved in reading although it is not absolutely essential. It is
crucially involved insofar as it subserves word processing and
identification and the visual guidance of reading eye-movements.
Obliterating parafoveal vision, either by injury to the striate cortex
or its geniculostriate afferents or by experimental masks in normal
subjects impairs text processing and alters the oculomotor read-
ing scanpath from bottom-up. Furthermore, the side and extent
of the artificial or natural visual field defect determine, together
with the functional demands of the writing system and the read-
ing task per se, the quality and severity of the resulting reading
impairments. Purely visually elicited impairments are, however,
can be adjusted to re-establish sufficient visual information pro-
cessing for reading to proceed in a regular fashion, although each
fixation still only provides an incomplete view. An attentional
top-down control of visual sampling can successfully ‘substitute’
parafoveal vision. The representation of parafoveal vision in striate
and prestriate cortex may not be essential to reading in so far as
its (artificial or brain injury-related) loss can be compensated for.
Parafoveal visual field loss is a necessary yet not sufficient compo-
nent for the emergence of hemianopic dyslexia.
Successful spontaneous and training-induced oculomotor com-
pensation for parafoveal visual field loss in reading suggests that
there is a discrepancy between involvement and absolute neces-
sity of the cortical and subcortical areas involved in reading. This
discrepancy demonstrates the functional plasticity of the visual,
attentional and oculomotor systems involved in reading and “may
reflect significant functional reserve and plasticity within the
cortical network as a whole” (Leigh & Kennard, 2004, p. 474).
Oculomotor adaptation to parafoveal visual field loss in reading
reciprocal connections to visual areas. These systems and their
motor integration and the attentional top-down modulation of
visual information processing which are required for reading. This
network is therefore not only involved but necessary for normal
reading to occur. It consists of visual cortical, parietal (esp. PPC)
and frontal (esp. FEF) areas.
If, in addition to unilateral homonymous parafoveal visual field
functioning of its components, are affected by brain injury, hemi-
anopic dyslexia results. The level of functioning of this network
determines the extent to which the residual visual field can be
utilised via a top-down attentional strategy for word identification
dyslexia demonstrates the importance to reading of white matter
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
pathways reciprocally connecting the foveal/parafoveal parts of V1
with the parietal, frontal, and temporal cortices and the subcortical
areas involved in saccade control. Despite different contributions
of parietal and frontal areas to the control of saccadic activity, both
areas and their close cooperation are essential in sampling the
visual world in reading. Hemianopic dyslexia may be interpreted
best as a visual–attentional–oculomotor–network disorder.
man Academic Exchange Service (DAAD). This work was supported
the German Federal Ministry of Education and Research (BMBF;
Project: Brain plasticity and Perceptual learning, Sub-project B).
We wish to thank the reviewers, Keith Rayner and Alex Leff, and
the editor, Morris Moscovitch, for their helpful comments on an
earlier version of this manuscript.
Al-Hamouri, F., Maestu, F., del Rio, D., Fernandez, S., Campo, P., Capilla, A., et al.
(2005). Brain dynamics of Arabic reading: A magnetoencephalographic study.
Neuroreport, 16, 1861–1864.
Amis, M. (1994). Career move. In G. Gordon & D. Hughes (Eds.), The Minerva book of
short stories (p. 14). London: Mandarin Paperbacks.
Andersen, R. A., & Buneo, C. A. (2003). Sensorimotor integration in posterior parietal
cortex. Advances in Neurology, 93, 159–177.
Anstis, S. M. (1974). A chart demonstrating variations in acuity with retinal position.
Vision Research, 14, 579–582.
Aulhorn, E., & Harms, H. (1972). Visual perimetry. In D. Jameson & L. M. Hurvich
B¨ ackman,¨O. (1999). A theoretical reading perspective on training methods for low
vision patients. Visual Impairment Research, 1, 85–94.
Baylis, G. C., Driver, J., Baylis, L. L., & Rafal, R. D. (1994). Reading of letters and words
in a patient with Balint’s syndrome. Neuropsychologia, 32, 1273–1286.
of word reading in hemispatial neglect. Physiology and Behavior, 77, 613–619.
Behrmann, M., Moscovitch, M., Black, S. E., & Mozer, M. (1990). Perceptual and con-
ceptual mechanisms in neglect dyslexia. Brain, 113, 1163–1183.
Behrmann, M., Shomstein, S. S., Black, S. E., & Barton, J. J. S. (2001). The eye
movements of pure alexic patients during reading and nonreading tasks. Neu-
ropsychologia, 39, 983–1002.
Beldarrain, M. G., Grafman, J., Pascal-Leone, A., & Garcia-Monco, J. C. (1999). Pro-
cedural learning is impaired in patients with prefrontal lesions. Neurology, 52,
Ben-Shachar, M., Dougherty, R. F., & Wandell, B. A. (2007). White matter pathways
in reading. Current Opinion in Neurobiology, 17, 258–270.
and psychic blindness after brain injury]. Graefe’s Archiv f¨ ur Ophthalmologie, 93,
Binder, J. R., Frost, J. A., Hammeke, T. A., Cox, R. W., Rao, S. M., & Prieto, T. (1997).
Human brain language areas identified by functional magnetic resonance imag-
ing. The Journal of Neuroscience, 17, 353–362.
Binder, J. R., & Mohr, J. P. (1992). The topography of callosal reading pathways. Brain,
Binder, K. S., Pollatsek, A., & Rayner, K. (1999). Extraction of information to the left of
and Performance, 25, 1162–1172.
information in reading. Perception & Psychophysics, 46, 85–94.
Bolger, D. J., Perfetti, C. A., & Schneider, W. (2005). Cross-cultural effect on the brain
revisited: Universal structures plus writing system variation. Human Brain Map-
ping, 25, 92–104.
Bosley, T. M., Rosenquist, A. C., Kushner, M., Burke, A., Stein, A., & Dann, R. (1985).
Ischemic lesions of the occipital cortex and optic radiations: Positron emission
tomography. Neurology, 35, 470–484.
hemisphere. Brain, 64, 244–272.
Brewer, A. A., Liu, J., Wade, A. R., & Wandell, B. A. (2005). Visual field maps and
stimulus selectivity in human ventral occipital cortex. Nature Neuroscience, 8,
Brysbaert, M., & Nazir, T. A. (2005). Visual constraints in written word recognition:
Chung, S. T. L., Legge, G. E., & Cheung, S. (2004). Letter-recognition and reading
speed in peripheral vision benefit from perceptual learning. Vision Research, 44,
Ciuffreda, K. J. (1994). Reading eye movements in patients with oculomotor dis-
turbances. In J. Ygge & G. Lennerstrand (Eds.), Eye movements in reading (pp.
163–188). Oxford: Pergamon.
Corbetta, M. (1998). Frontoparietal cortical networks for directing attention and the
eye to visual locations: Identical, independent, or overlapping neural systems?
Proceedings of the National Academy of Sciences, 95, 831–838.
Corbetta, M., & Shulman, G. L. (2002). Control of goal-directed and stimulus-driven
attention in the brain. Nature Reviews Neuroscience, 3, 201–215.
Crossland, M. D., & Rubin, G. S. (2006). Eye movements and reading in macular
disease: Further support for the shrinking perceptual span hypothesis. Vision
Research, 46, 590–597.
Cummings, R. W., & Rubin, G. S. (1992). Reading speed and saccadic eye movements
with an artificial paracentral scotoma. Investigative Ophthalmology and Visual
Science, 33, 1418.
De Luca, M., Spinelli, D., & Zoccolotti, P. (1996). Eye movement patterns in reading as
Dejerine, J. (1891). Sur un cas de c´ ecit´ e verbale avec agraphie, suivi d’autopsie [On a
case of word blindness with agraphia, followed by an autopsy]. Comptes Rendus
Hebdomadaires des S´ eances et M´ emoires de la Soci´ et´ e de Biologie, Ninth series, 3,
Deubel, H., O’Regan, J. K., & Radach, R. (2000). Attention, information processing,
and eye movement control. In A. Kennedy, R. Radach, D. Heller, & J. Pynte (Eds.),
Reading as a perceptual process (pp. 355–374). Amsterdam: Elsevier.
Evidence for a common attentional mechanism. Vision Research, 36, 1827–1837.
Ducrot, S., & Pynte, J. (2002). What determines the eyes’ landing position in words?
Perception & Psychophysics, 64, 1130–1144.
Eber, A. M., Metz-Lutz, M. N., Bataillard, M., & Collard, M. (1987). Reading eye
movements of patients with homonymous hemianopia. In J. K. O’Regan & A.
L´ evy-Schoen (Eds.), Eye movements: From physiology to cognition (pp. 544–545).
Ehrlich, S. E., & Rayner, K. (1981). Contextual effects on word perception and eye
movements during reading. Journal of Verbal Learning and Verbal Behavior, 20,
Ellis, A. W., & Young, A. W. (1996). Human cognitive neuropsychology. Hove, UK:
Findlay, J. M., & Gilchrist, I. D. (2003). Active vision: The psychology of looking and
seeing. Oxford: Oxford University Press.
Fine, E. M., & Rubin, G. S. (1999a). The effects of simulated cataract on reading with
normal vision and simulated central scotoma. Vision Research, 39, 4274–4285.
Fine, E. M., & Rubin, G. S. (1999b). Reading with central field loss: Number of letters
masked is more important than the size of the mask in degrees. Vision Research,
Fine, E. M., & Rubin, G. S. (1999c). Reading with simulated scotomas: Attending to
the right is better than attending to the left. Vision Research, 39, 1039–1048.
Foxe, J. J., & Simpson, G. V. (2002). Flow of activation from V1 to frontal cortex in
humans. A framework for defining “early” visual processing. Experimental Brain
Research, 142, 139–150.
Gassel, M. M., & Williams, D. (1963a). Visual function in patients with homonymous
hemianopia. Part II. Oculomotor mechanisms. Brain, 86, 1–36.
Gassel, M. M., & Williams, D. (1963b). Visual function in patients with homonymous
hemianopia. Part III. The completion phenomenon; insight and attitude to the
defect; and visual functional efficiency. Brain, 86, 229–260.
Glickstein, M. (2000). How are visual areas of the brain connected to motor areas
for the sensory guidance of movement? Trends in Neuroscience, 23, 613–617.
Gr¨ usser, O.-J., & Landis, T. (1991). Visual agnosias and other disturbances of visual
perception and cognition. In O.-J. Gr¨ usser & T. Landis (Eds.), Visual agnosias.
Basingstoke, England: Macmillan.
Habekost, T., & Starrfelt, R. (2006). Alexia and quadrant-amblyopia: Reading disabil-
ity after a minor visual field deficit. Neuropsychologia, 44, 2465–2476.
Haywood, M., & Coltheart, M. (2000). Neglect dyslexia and the early stages of visual
word recognition. Neurocase, 6, 33–44.
Hebel, N., & von Cramon, D. (1987). Der Posteriorinfarkt [Posterior infarction].
Fortschritte der Neurologie in der Psychiatrie, 55, 37–53.
Heinzle, J., Hepp, K., & Martin, K. A. C. (2007). A microcircuit model of the frontal eye
fields. Journal of Neuroscience, 27, 9341–9353.
Hess, R. F., Zihl, J., Pointer, J. S., & Schmid, C. (1990). The contrast sensitivity deficit
in cases with cerebral lesions. Clinical Vision Sciences, 5, 203–215.
Heywood, C. A., & Kentridge, R. W. (2003). Achromatopsia, color vision and cortex.
Neurological Clinics of North America, 21, 483–500.
Hillis, A. E., & Caramazza, A. (1992). The reading process and its disorders. In D. I.
Margolin (Ed.), Cognitive neuropsychology in clinical practice (pp. 229–261). New
York: Oxford University Press.
Hochstein, S., & Ahissar, M. (2002). View from the top: Hierarchies and reverse
hierarchies in the visual system. Neuron, 36, 791–804.
Inhoff, A. W. (1987). Parafoveal word perception during eye fixations in reading:
Effects of visual salience and word structure. In M. Coltheart (Ed.), Attention
& performance XII: The psychology of reading (pp. 403–418). London: Lawrence
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
parafoveal processing of spatial and linguistic information during eye fixations
in reading. Quarterly Journal of Experimental Psychology, 56, 803–827.
Jobard, G., Crivello, F., & Tzourio-Mazoyer, N. (2003). Evaluation of the dual route
theory of reading: A metaanalysis of 35 neuroimaging studies. Neuroimage, 20,
Johnson, R. L., & Rayner, K. (2007). Top-down and bottom-up effects in pure alexia:
Evidence from eye movements. Neuropsychologia, 45, 2246–2257.
Juan, C.-H., & Walsh, V. (2003). Feedback to V1: A reverse hierarchy in vision. Exper-
imental Brain Research, 150, 259–263.
Journal of Neurophysiology, 77, 1313–1324.
Kassubek, J., Schmidtke, K., Kimmig, H., L¨ ucking, C. H., & Greenlee, M. W. (2001).
Changes in cortical activation during mirror reading before and after training:
An fMRI study of procedural learning. Cognitive Brain Research, 10, 207–217.
Kastner, S., Pinsk, M. A., De Weerd, P., Desimone, R., & Ungerleider, L. G. (1999).
Increased activity in human visual cortex during directed attention in the
absence of visual stimulation. Neuron, 22, 751–761.
Kennard, C. (2002). Scanpaths: The path to understanding abnormal cognitive pro-
cessing in neurological disease. Annals of the New York Academy of Sciences, 956,
Kennedy, A., Brooks, R., Flynn, L.-A., & Prophet, C. (2003). The reader’s spatial code.
In J. Hy¨ on¨ a, R. Radach, & H. Deubel (Eds.), The mind’s eye: Cognitive and applied
aspects of eye movement research (pp. 193–212). Oxford: Elsevier.
Kerkhoff, G. (1999). Restorative and compensatory therapy approaches in cere-
bral blindness: A review. Restorative Neurology and Neuroscience, 15, 255–
Kerkhoff, G., M¨ un?inger, G., Eberle-Strauss, G., & St¨ ogerer, E. (1992). Rehabilita-
tion of hemianopic alexia in patients with postgeniculate visual field disorders.
Neuropsychological Rehabilitation, 2, 21–42.
Kirkby, J. A., Webster, L. A. D., Blythe, H. I., & Liversedge, S. P. (in press). Binocular
coordination during reading and non-reading tasks. Psychological Bulletin.
Kliegl, R., & Engbert, R. (2003). How tight is the link between lexical processing and
saccade programs? Behavioral and Brain Sciences, 26, 491–492.
Lee, C. (1999). Eye and head coordination in reading: Roles of head movement and
cognitive control. Vision Research, 39, 3761–3768.
Leff, A. P. (2004). A historical review of the representation of the visual field in pri-
mary visual cortex with special reference to the neural mechanisms underlying
macular sparing. Brain and Language, 88, 268–278.
Leff, A. P., Crewes, H., Plant, G. T., Scott, S. K., Kennard, C., & Wise, R. J. S. (2001). The
functional anatomy of single-word reading in patients with hemianopic and
pure alexia. Brain, 124, 510–521.
Leff, A. P., Scott, S. K., Crewes, H., Hodgson, T. L., Cowey, A., Howard, D., et al. (2000).
Impaired reading in patients with right hemianopia. Annals of Neurology, 47,
reading saccades: A repetitive transcranial magnetic stimulation study. Cerebral
Cortex, 11, 918–923.
Leff, A. P., Spitzyna, G. A., Plant, G. T., & Wise, R. J. S. (2006). Structural anatomy of
pure and hemianopic alexia. Journal of Neurology, Neurosurgery and Psychiatry,
Legge, G. E., Ahn, S. J., Klitz, T. S., & Luebker, A. (1997). Psychophysics of reading. XVI.
The visual span in normal and low vision. Vision Research, 37, 1999–2010.
Legge, G. E., Cheung, S., Yu, D., Chung, S. T. L., Lee, H.-W., & Owens, D. P. (2007). The
case for the visual span as a sensory bottleneck in reading. Journal of Vision, 7,
Leigh, R. J., & Kennard, C. (2004). Using saccades as a research tool in the clinical
neurosciences. Brain, 127, 460–477.
Leigh, R. J., & Zee, D. S. (2006). The neurology of eye movements (4th ed.). New York:
Oxford University Press.
Leker, R. R., & Biran, I. (1999). Unidirectional dyslexia in a polyglot. Journal of Neurol-
ogy, Neurosurgery and Psychiatry, 66, 517–519.
Ber¨ ucksichtigung ihrer Ergebnisse f¨ ur die Anatomie und Physiologie [On the pathol-
ogy of the cerebral visual pathway with special consideration of its results for
anatomy and physiology]. Leipzig, Germany: Verlag von Wilhelm Engelmann.
Liversedge, S. P., & Blythe, H. I. (2007). Lexical and sublexical influences on eye
movements during reading. Language and Linguistics Compass, 1, 17–31.
coordination of the eyes during reading. Current Biology, 16, 1726–1729.
Mackensen, G. (1962). Die Untersuchung der Lesef¨ ahigkeit als klinische Funk-
tionspr¨ ufung [Examining the ability to read as clinical functional analysis].
Fortschritte in der Augenheilkunde, 12, 344–379.
Mangun, G. R., Hopfinger, J., Kussmaul, C. L., Flechter, E., & Heinze, H. J. (1997).
Covariations in ERP and PET measures of spatial selective attention in human
extrastriate cortex. Human Brain Mapping, 5, 273–279.
Martinez, A., DiRusso, F., Anllo-Vento, L., Sereno, M. L., Buxton, R. B., & Hillyard, S. A.
selection processes in striate and extrastriate visual areas. Vision Research, 41,
Mauthner, L. (1881). Gehirn und Auge [Brain and eye]. Wiesbaden, Germany:
McConkie, G. W., & Rayner, K. (1975). The span of the effective stimulus during a
fixation in reading. Perception & Psychophysics, 17, 578–586.
cerebralen Sehbahn unterbesonderer
McConkie, G. W., & Rayner, K. (1976). Asymmetry of the perceptual span in reading.
Bulletin of the Psychonomic Society, 8, 365–368.
McConkie, G. W., & Zola, D. (1987). Visual attention during eye fixations while read-
(pp. 385–401). Hove, UK: Lawrence Erlbaum.
McDonald, S. A., Spitzyna, G., Shillcock, R., Wise, R. J. S., & Leff, A. P. (2006). Patients
with hemianopic alexia adopt an inefficient eye movement strategy when read-
ing text. Brain, 129, 158–167.
Meienberg, O., Zangemeister, W. H., Rosenberg, M., Hoyt, W. F., & Stark, L. (1981).
Saccadic eye movement strategies in patients with homonymous hemianopia.
Annals of Neurology, 9, 537–544.
Mesulam, M.-M. (1990). Large-scale neurocognitive networks and distributed pro-
cessing for attention, language, and memory. Annals of neurology, 28, 597–613.
Annual Review of Neuroscience, 24, 167–202.
Mohamed, M. D., Elsherbiny, S. M., & Goulding, P. J. (2000). Unidirectional dyslexia
in a polyglot. Journal of Neurology, Neurosurgery and Psychiatry, 68, 537.
spaces and not visual angle. Perception & Psychophysics, 30, 395–396.
Nazir, T. A., Ben-Boutayab, N., Decoppet, N., Deutsch, A., & Frost, R. (2004). Read-
ing habits, perceptual learning, and recognition of printed words. Brain and
Language, 88, 294–311.
Nelles, G., De Greiff, A., Pscherer, A., Stude, P., Forsting, M., Hufnagel, A., et al. (2007).
Saccade induced cortical activation in patients with post-stroke visual field
defects. Journal of Neurology, 254, 1432–1459.
O’Regan, J. K. (1983). Elementary perceptual and eye movement control processes
in reading. In K. Rayner (Ed.), Eye movements in reading: Perceptual and language
processes (pp. 121–140). New York: Academic Press.
O’Regan, J. K. (1992). Optimal viewing position in words and the strategy-tactics
cognition: Scene perception and reading (pp. 333–354). New York: Springer.
O’Regan, J. K., & L´ evy-Schoen, A. (1987). Eye-movement strategy and tactics in word
psychology of reading (pp. 363–383). Hove, England: Lawrence Erlbaum.
Ofen-Noy, N., Dudai, Y., & Karni, A. (2003). Skill learning in mirror reading:
How repetition determines acquisition. Cognitive Brain Research, 17, 507–
Ogren, M. P., Mateer, C. A., & Wyler, A. R. (1984). Alterations in visually related
eye movements following left pulvinar damage in man. Neuropsychologia, 22,
Osaka, N. (2003). On the perceptual and neural correlates of reading models. Behav-
ioral and Brain Sciences, 26, 495–496.
Pambakian, A. L. M., Wooding, D. S., Patel, N., Morland, A. B., Kennard, C., & Mannan,
S. K. (2000). Scanning the visual world: A study of patients with homonymous
hemianopia. Journal of Neurology, Neurosurgery and Psychiatry, 69, 751–759.
Papageorgiou, E., Hardiess, G., Schaeffel, F., Wiethoelter, H., Karnath, H.-O., Mallot,
H., et al. (2007). Assessment of vision-related quality of life in patients with
homonymous visual field defects. Graefe’s Archive for Clinical and Experimental
Ophthalmology, 245, 1749–1758.
Pierrot-Deseilligny, C., Rivaud, S., Gaymard, B., M¨ uri, R., & Vermersch, A. I. (1995).
Cortical control of saccades. Annals of Neurology, 37, 557–567.
Poldrack, R. A., Desmond, J. E., Glover, G. H., & Gabrieli, J. D. E. (1998). The neural
basis of visual skill learning: An fMRI study of mirror reading. Cerebral Cortex, 8,
Poldrack, R. A., & Gabrieli, J. D. E. (2001). Characterizing the neural mechanisms
of skill learning and repetition priming: Evidence from mirror reading. Brain,
Pollatsek, A., Bolozky, S., Well, A. D., & Rayner, K. (1981). Asymmetries in the percep-
tual span for Israeli readers. Brain and Language, 14, 174–180.
Pollatsek, A., Raney, G. E., LaGasse, L., & Rayner, K. (1993). The use of informa-
tion below fixation in reading and visual search. Perception & Psychophysics, 40,
Poppelreuter, W. (1917/1990). Disturbances of lower and higher visual capacities
Posner, M. I., Abdullaev, Y. G., McCandliss, B. D., & Sereno, S. C. (1999). Anatomy,
circuitry and plasticity of world reading. In J. Everatt (Ed.), Reading and dyslexia:
Visual and attentional processes (pp. 137–162). London: Routledge.
Powell, H. W., Parker, G. J., Alexander, D. C., Symms, M. R., Boulby, P. A., Wheeler-
Kingshott, C. A., et al. (2006). Hemispheric asymmetries in language-related
pathways: A combined funtional MRI and tractography study. Neuroimage, 32,
Prosiegel, M. (1988). Beschreibung einer Patientenstichprobe einer neuropsy-
chologischen Rehabilitationsklinik [Description of a patient sample of a
neuropsychological rehabilitation clinic]. In D. von Cramon & J. Zihl (Eds.), Neu-
Pynte, J., & Kennedy, A. (2006). An influence over eye movements in reading exerted
from beyond the level of the word: Evidence from reading English and French.
Vision Research, 46, 3786–3801.
2462 Download full-text
S. Schuett et al. / Neuropsychologia 46 (2008) 2445–2462
Rayner, K. (1975). The perceptual span and peripheral cues in reading. Cognitive
Psychology, 7, 65–81.
Rayner, K. (1979). Eye guidance in reading: Fixation location within words. Percep-
tion, 8, 21–30.
Rayner, K. (1998). Eye movements in reading and information processing: 20 years
of research. Psychological Bulletin, 124, 372–422.
Rayner, K., & Bertera, J. H. (1979). Reading without a fovea. Science, 206, 468–469.
Rayner, K., Inhoff, A. W., Morrison, R. E., Sowiaczek, M. L., & Bertera, J. H. (1981).
Masking foveal and parafoveal vision during eye fixations in reading. Journal of
Experimental Psychology: Human Perception and Performance, 7, 167–179.
Rayner, K., & Johnson, R. L. (2005). Letter-by-letter acquired dyslexia is due to the
serial encoding of letters. Psychological Science, 16, 530–534.
Rayner, K., Liversedge, S. P., & White, S. J. (2006). Eye movements when reading
disappearing text: The importance of the word to the right of fixation. Vision
Research, 46, 310–323.
Rayner, K., & Pollatsek, A. (1981). Eye movement control during reading: Evidence
for direct control. Quarterly Journal of Experimental Psychology, 33A, 351–373.
Rayner, K., & Pollatsek, A. (1989). The psychology of reading. Hillsdale, NJ: Lawrence
Rayner, K., & Well, A. D. (1996). Effects of contextual constraint on eye movements
in reading: A further examination. Psychonomic Bulletin & Review, 3, 504–509.
Rayner, K., Well, A. D., Pollatsek, A., & Bertera, J. H. (1982). The availability of useful
information to the right of fixation in reading. Perception & Psychophysics, 31,
Rayner, K., White, S. J., Kambe, G., Miller, B., & Liversedge, S. P. (2003). On the pro-
cessing of meaning from parafoveal vision during eye fixations in reading. In J.
Hy¨ on¨ a,R.Radach,&H.Deubel(Eds.),Themind’seye:Cognitiveandappliedaspects
of eye movement research (pp. 213–234). Oxford: Elsevier.
Reichle, E. D., Rayner, K., & Pollatsek, A. (2003). The E–Z Reader model of eye-
movement control in reading: Comparisons to other models. Behavioral and
Brain Sciences, 26, 445–526.
Remond, A., Lesevre, N., & Gabersek, V. (1957). Approche d’une semeiologie electro-
graphique du regard [Towards an electrographic semiology of the gaze]. Revue
Neurologique, 96, 536–546.
Riddoch, M. J. (Ed.). (1991). Neglect and the peripheral dyslexias. East Sussex, UK:
Robinson, D. L., & Petersen, S. E. (1992). The pulvinar and visual salience. Trends in
Neuroscience, 15, 127–132.
Russell, C., Malhotra, P., & Husain, M. (2004). Attention modulates the visual field in
healthy observers and parietal patients. Neuroreport, 15, 2189–2193.
Schlaggar, B. L., & McCandliss, B. D. (2007). Development of neural systems for
reading. Annual Review of Neuroscience, 30, 475–503.
gies to the status of adaptation in patients with hemianopic visual field defects.
Annals of the New York Academy of Sciences, 682, 404–408.
Shallice, T. (1988). From neuropsychology to mental structure. Cambridge, England:
Cambridge University Press.
Skoyles, J. R. (1988). Right hemisphere literacy in the ancient world. In D. de Ker-
ckhove & C. J. Lumsden (Eds.), The alphabet and the brain: The lateralization of
writing (pp. 363–380). Berlin: Springer.
Spitzyna, G. A., Wise, R. J. S., McDonald, S. A., Plant, G. T., Kidd, D., Crewes, H., et al.
(2007). Optokinetic therapy improves text reading in patients with hemianopic
alexia: A controlled trial. Neurology, 68, 1922–1930.
Tant, M. L. M., Cornelissen, F. W., Kooijman, A. C., & Brouwer, W. H. (2002). Hemi-
anopic visual field defects elicit hemianopic scanning. Vision Research, 42,
Teuber, H.-L., Battersby, W. S., & Bender, M. B. (1960). Visual field defects after pene-
trating missile wounds of the brain. Cambridge, MA: Harvard University Press.
Trauzettel-Klosinski, S., & Brendler, K. (1998). Eye movements in reading with hemi-
anopic field defects: The significance of clinical parameters. Graefe’s Archive for
Clinical and Experimental Ophthalmology, 236, 91–102.
Uemura, T., Arai, Y., & Shimazaki, C. (1980). Eye-head coordination during lateral
gaze in normal subjects. Acta Otolaryngica, 90, 420–424.
Underwood, N. R., & McConkie, G. W. (1985). Perceptual span for letter distinctions
during reading. Reading Research Quarterly, 20, 153–162.
Upton, N. J., Hodgson, T. L., Plant, G. T., Wise, R. J. S., & Leff, A. P. (2003). Bottom-up”
and “top-down” effects on reading saccades: A case study. Journal of Neurology,
Neurosurgery and Psychiatry, 74, 1423–1428.
Vitu, F., O’Regan, J. K., & Mittau, M. (1990). Optimal landing position in reading
isolated words and continuous text. Perception & Psychophysics, 47, 583–600.
Walker, R., Findlay, J. M., Young, A. W., & Welch, J. (1991). Disentangling neglect and
hemianopia. Neuropsychologia, 29, 1019–1027.
Walsh, V., Ashbridge, E., & Cowey, A. (1998). Cortical plasticity in perceptual learn-
ing demonstrated by transcranial magnetic stimulation. Neuropsychologia, 36,
Walsh, V., Ellison, A., Ashbridge, E., & Cowey, A. (1999). The role of the parietal cor-
tex in visual attention—hemispheric asymmetries and the effects of learning: A
magnetic stimulation study. Neuropsychologia, 37, 245–251.
Wilbrand, H. (1907).¨Uber die makul¨ ar-hemianopische Lesest¨ orung und die v. Mon-
reading disorder and the v. Monakowian projection of the macula on the visual
sphere]. Klinische Monatsbl¨ atter f¨ ur Augenheilkunde, 45, 1–39.
Williams, D., & Gassel, M. M. (1962). Visual functions in homonymous hemianopia.
Part I. The visual fields. Brain, 85, 175–250.
Zangemeister, W. H., & Oechsner, U. (1999). Adaptation to visual field defects with
virtual reality scotoma in healthy subjects. In W. Becker, H. Deubel, & T. Mergner
(Eds.), Current oculomotor research (pp. 89–92). New York: Kluwer.
Zangemeister, W. H., Oechsner, U., & Freska, C. (1995). Short-term adaptation of eye
of human scanpath. Optometry and Vision Science, 72, 467–477.
of Cognition, 21, 281–303.
Zhang, X., Kedar, S., Lynn, M. J., Newman, N. J., & Biousse, V. (2006). Homony-
mous hemianopias: Clinical–anatomic correlations in 904 cases. Neurology, 66,
Zihl, J. (1989). Cerebral disturbances of elementary visual functions. In J. W. Brown
(Ed.), Neuropsychology of visual perception (pp. 35–58). Hillsdale, NJ: Lawrence
Research and new development in rehabilitation (pp. 287–295). Amsterdam: IOS
Zihl, J. (1995a). Eye movement patterns in hemianopic dyslexia. Brain, 118, 891–912.
Neuropsychologia, 33, 287–303.
Zihl, J. (1999). Oculomotor scanning performance in subjects with homonymous
visual field disorders. Visual Impairment Research, 1, 23–31.
Zihl, J. (2000). Rehabilitation of visual disorders after brain injury. Hove, UK: Psychol-
Zihl, J. (2003). Recovery and rehabilitation of cerebral visual disorders. In M. Fahle
& M. W. Greenlee (Eds.), The neuropsychology of vision (pp. 319–338). Oxford:
Oxford University Press.
posterior parietal or frontal lobe damage. Neuropsychologia, 35, 893–906.
Zihl, J., & Kennard, C. (1996). Disorders of higher visual function. In T. Brandt, L.
R. Caplan, J. Dichgans, H. C. Diener, & C. Kennard (Eds.), Neurological disorders:
Course and treatment (pp. 201–212). San Diego, CA: Academic Press.
Zihl, J., & Kerkhoff, G. (1990). Foveal photopic and scotopic adaptation in patients
with brain damage. Clinical Vision Sciences, 5, 185–195.
Zihl, J., Krischer, C. C., & Mei?en, R. (1984). Die hemianopische Lesest¨ orung und ihre
Behandlung [Hemianopic dyslexia and its treatment]. Nervenarzt, 55, 317–323.
Zihl, J., & von Cramon, D. (1979). The contribution of the ‘second’ visual system to
directed visual attention in man. Brain, 102, 835–856.
Zihl, J., & von Cramon, D. (1986). Zerebrale Sehst¨ orungen [Cerebral visual disorders].