Content uploaded by Richard Kruk
Author content
All content in this area was uploaded by Richard Kruk on Nov 25, 2022
Content may be subject to copyright.
Visual processing characteristics of children
with Meares–Irlen syndrome
Richard Kruk
1
, Karen Sumbler
2
and Dale Willows
2
1
Department of Psychology, University of Manitoba, Winnipeg, MB, Canada R3T 2N2, and
2
Department of Human Development and Applied Psychology, Ontario Institute for Studies in
Education, University of Toronto, Toronto, ON, Canada
Abstract
The potential role of visual processing deficits in reading difficulty was brought to public attention by
claims that a large proportion of children with dyslexia suffer from a perceptual dysfunction currently
referred to as Meares–Irlen syndrome (MISViS). A previous study showing that visual perceptual
measures involving visual memory and discrimination predict independent variance in reading
achievement [J. Learn. Disabil. 28 (1995) 216] provides a basis to examine their relationships with
the diagnostic criteria of MISViS. This study examined these visual processing characteristics in 36
eight- to ten-year-old children, half of whom were experiencing reading difficulty. Children were
assessed for MISViS by Irlen screeners; approximately half of the participants in each group were
positively identified. Concurrent performance on standardized visual processing tests showed that
while a positive diagnosis of MISViS is not indicative of reading ability, nor in particular of a visual-
processing deficit subtype identified by Watson and Willows [J. Learn. Disabil. 28 (1995) 216],
MISViS can indicate visual processing difficulties potentially related to visual attention inefficiency.
Keywords: attention, dyslexia, Meares–Irlen syndrome, visual processing, visual stress
Introduction
There is a great deal of controversy surrounding the
aetiology and appropriate diagnostic procedures for
reading disabilities, or dyslexia. Although phonological
processing measures are key predictors of reading
success (e.g. Vellutino, 1987; Wagner and Torgesen,
1987), many additional factors, like rapid naming ability
(Bowers and Wolf, 1993) and visual perceptual abilities
(Rayner and Pollatsek, 1989; Watson and Willows,
1995; Stein, 2001; Watson et al., 2003), are associated
with reading acquisition. A growing body of literature
points to the existence of distinct patterns of abilities
and weaknesses that contribute to difficulties experi-
enced by children with dyslexia (Hooper and Willis,
1989; Watson and Willows, 1993, 1995). Common
among subtype groupings is a category for some form
of visual processing deficit (Boder, 1971; Watson et al.,
1983; Feagans et al., 1991). Reading difficulties have
been associated with a number of visual anomalies,
including perceptual deficits (Willows et al., 1993),
abnormal eye movements (Ciuffreda et al., 1983;
Stein, 1993), binocular vision dysfunctions (Garzia,
1993), attentional anomalies (Valdois et al., 2004; Sire-
teanu et al., 2005; Facoetti et al., 2006), and sensitivities
resulting in visual stress (Wilkins, 1993, 1995, 2003).
The connection between visual processes and dyslexia
has gained much publicity through the claims made by
Irlen (1983, 1991), who reported that many children
with dyslexia suffer from a perceptual dysfunction she
called Scotopic Sensitivity Syndrome. Currently referred
to as Meares–Irlen syndrome (Evans, 1997; here we use
the term Meares–Irlen syndrome/visual stress, MISViS,
given its relation to visual stress
1
) in recognition of the
work of both Irlen and Meares (1980), the condition is
treated by the use of proprietary coloured overlays in
the case of Irlen clinics (Irlen and Lass, 1989), and by
Received: 10 November 2006
Revised form: 31 August 2007
Accepted: 9 November 2007
Correspondence and reprint requests to: Richard Kruk.
Tel.: +1 (204) 474 7349; Fax: +1 (204) 474 7599.
E-mail address: krukr@cc.umanitoba.ca
1
We thank an anonymous reviewer for suggesting this acronym.
Ophthal. Physiol. Opt. 2008 28: 35–46
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists doi: 10.1111/j.1475-1313.2007.00532.x
precision overlays developed by Wilkins et al. (2005b).
Prevalence estimates vary depending on the criteria used
to define MISViS. A recent estimate of prevalence rates
between reader groups, using a strict criterion for
MISViS, indicates that it is present in 12.5% of average
readers and 31% of children with dyslexia, but the
difference did not reach statistical significance (Kriss
and Evans, 2005). Studies of MISViS in unselected
samples, using a 5% improvement in reading rate with
preferred overlays as a diagnostic criterion, show a
prevalence of 34% in adults (Evans and Joseph, 2002),
and 22–36% in children (Wilkins et al., 1996, 2001).
When the criterion is continued use of preferred overlays
10 months after initial assessment, the condition is
present in 24% of children (Jeanes et al., 1997).
Among children and adults with and without dys-
lexia, evidence of a link between MISViS and reading
has been found. Children with dyslexia co-occuring with
MISViS have been found to show improved reading
rates compared with dyslexic children without MISViS,
as well as matched non-dyslexic children with MISViS
(Singleton and Henderson, 2007), and similar patterns
have been identified with adult samples (Singleton and
Trotter, 2005).
According to Irlen, many individuals with MISViS
are highly sensitive to certain frequencies of the visible
spectrum, which results in discomfort under various
lighting conditions, inadequate background accommo-
dation when reading, difficulty with print resolution,
restricted span of focus, poor depth perception, and
difficulty with sustained reading. Accompanying phys-
ical symptoms can include fatigue, eyestrain and head-
aches.
Research reporting significant benefits from use of
Irlen tinted lenses for reading difficulty (e.g. Blaskey
et al., 1990; OÕConnor et al., 1990; Robinson and
Conway, 1990) has been criticized for methodological
flaws that weaken claims of beneficial effects (e.g.
Parker, 1990; Solan, 1990). However, work by Wilkins
and his collaborators, using strict research designs,
found that using precision tinted overlays and lenses
determined by their own diagnostic protocol for MIS-
ViS, showed improved reading speed in children and
adults (Lightstone et al., 1999; Bouldoukian et al., 2002;
Evans and Joseph, 2002; Waldie and Wilkins, 2004).
Among the several protocols that have been devel-
oped to diagnose individuals with MISViS, the Irlen
Reading Perceptual Scale (IRPS) consists of questions
evaluating subjective visual experience, and perfor-
mance on perceptual tasks. This is part of a complete
screening procedure conducted by the many Certified
Irlen Screeners who have been trained at Irlen Institutes
in California and Massachusetts. The IRPS provides a
basis for the diagnosis of MISViS and indicates the tints
that would be beneficial in ameliorating the symptoms
of MISViS and promoting success in reading acquisi-
tion. However, there is little in the scientific or clinical
literature to provide an evidence base for the protocol,
in comparison with other methods, such as those
developed by Wilkins and by Singleton and Trotter
(2005). Our choice of the IRPS was based on the
widespread use of the protocol, and by the need for
critical examination of the visual cognitive characteris-
tics of individuals who obtain a diagnosis of MISViS on
the basis of the IRPS. In comparison with the evidence-
based methods, the IRPS, as a proprietary instrument of
Irlen clinics, does suffer from the absence of published
clinical trials. However, other screening methods use
subjective ratings of perceptual distortions, like those
included in part of the IRPS, and have been found to be
meaningful in assessing MISViS, particularly if based on
immediate experience of perceptual distortion (Hollis
and Allen, 2006). Using an early version of the Irlen
protocol, an examination of the academic and optomet-
ric performance of children showed no significant
relationship between academic ability and degree of
MISViS, but children with more severe scores tended to
be individuals who scored poorly on optometric mea-
sures, and who benefited more from prescribed overlays
(Lopez et al., 1994). This finding provides some support
for the validity of the Irlen protocol.
The theoretical underpinnings of MISViS, based on
published reports from Irlen, are vague at best. In her
own literature Irlen writes Ôthe exact nature of the
mechanisms at play are not yet fully understoodÕ, but
that Ôit is a problem with how the nervous system
encodes and decodes visual informationÕ(Irlen, 2003).
As well, many of the symptoms do not seem to be
unique to MISViS. The visual symptoms attributed to
MISViS appear to overlap those associated with oculo-
motor problems of accommodation and convergence
(Stanley, 1990; Simmers et al., 2001) and conventional
optometric therapy has been shown to be successful in
elimination of these symptoms (Blaskey et al., 1990).
However, other optometric research has found little
evidence of co-occurring optometric conditions in indi-
viduals diagnosed with MISViS (Evans et al., 1996).
A widely accepted theoretical basis for MISViS, and
potentially its relation to reading difficulty, posits that
neural responses to specific colours of light result in high
levels of visual stress (Wilkins, 1995; Robinson and
Foreman, 1999; Wilkins et al., 1994, 2005a,b).
Wilkins (1995, 2003) focuses on the presence of visual
stress as a source of the types of difficulties described as
symptomatic of MISViS. This perspective assumes that
difficulties in perceiving text for people experiencing
visual stress are brought about by an extreme sensitivity
to specific colours of light, which produces an over-
activity of neural responding in the visual cortex leading
to the fatigue, headache, and other neurological symp-
36 Ophthal. Physiol. Opt. 2008 28: No. 1
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
toms that make it difficult for afflicted individuals to
focus on text. This explanation of the symptomology of
MISViS is compelling, as the treatment to ameliorate
visual stress related to reading is similar to that
described by Irlen: use of individually prescribed tinted
overlays or lenses that effectively filter out colours of
light that produce the experience of visual stress. Two
randomized controlled trials (Wilkins et al., 1994; Rob-
inson and Foreman, 1999) and recent experimental
work (Wilkins et al., 2005a,b) support the need for
individually and precisely prescribed coloured filters in
MISViS.
Regardless of the theoretical basis for MISViS, many
people with dyslexia in Great Britain, the United States,
Canada, Australia, and other countries, are currently
being treated for MISViS at considerable cost, but the
nature of the relationship between MISViS itself and
reading difficulty has not been established empirically.
Despite lingering disagreement about the optometric
characteristics of the subgroup of individuals with
dyslexia with this condition, there is a need to identify
the subgroupÕs perceptual and cognitive characteristics
empirically. If MISViS interferes with reading, then the
same effect could be manifested in poor performance on
visual perceptual and cognitive processing measures.
Such measures of visual processing characteristics have
been used to identify a subgroup of children experienc-
ing dyslexia involving a visual processing deficit (Wat-
son and Willows, 1995). These measures include
indicators of visual discrimination, visual memory,
visual-motor integration, pattern analysis and synthesis,
all of which were found to predict unique variance in
reading performance in the identified subgroup of
dyslexic readers. Although compelling evidence indi-
cates that MISViS and dyslexia constitute independent
conditions (Lopez et al., 1994; Kriss and Evans, 2005), it
is possible that children identified as having MISViS and
co-occurring dyslexia demonstrate the same visual
processing anomalies identified by Watson and Willows
(1995) in their visual processing subtype.
The potential of significant relationships among
MISViS, the visual processing measures outlined above,
and reading ability is partially indicated in a study of
adults showing that improved reading rate from colour
filtering of text is more likely if adults have co-occurring
MISViS and reading difficulty, compared to MISViS-
diagnosed adults without reading difficulty (Singleton
and Trotter, 2005). In addition, White et al. (2006)
showed that visual stress, over and above phonological
awareness, distinguished a subgroup of dyslexic chil-
dren, from other dyslexic and average readers. Further-
more, although the stimuli used in the visual processing
tests identified by Watson and Willows (1995) do not
involve the kinds of patterns such as square-wave
gratings that are ideally suited to evoke a strong visual
stress experience (Wilkins, 1993; Wilkins et al., 2004),
the high contrast patterns and bold line drawings,
including successive vertical strokes, used in the visual
processing measures could potentially contribute to the
experience of visual stress. If this is sufficient to evoke a
degree of visual stress in children with MISViS, then
significant relationships among reading ability, degree of
MISViS and performance on visual processing measures
are plausible. Evidence of less efficient visual processing
has been found in individuals who experience visual
discomfort, using a serial search visual attention task
without the presence of gratings (Conlon and Humph-
reys, 2001).
The objective of the present study was to determine if
children with dyslexia, diagnosed with MISViS, show
evidence of the visual processing characteristics identi-
fied by Watson and Willows (1995) and others (e.g.
Hooper and Willis, 1989; Feagans et al., 1991; Watson
and Willows, 1993). Systematic negative relationships,
with low scores on the visual processing measures
associated with high IRPS scores, would support the
utility of the IRPS, and by extension an MISViS
diagnosis, as an indicator of visual processing anomalies
in dyslexia that involve the encoding and decoding of
visual information, and by extension of the ability to
manipulate visual information in reading tasks.
Method
Participants
Children between the ages of 9 and 10 years with and
without dyslexia were recruited from four schools in
Toronto. There were 18 children in each of the two
reading ability groups. Children with dyslexia were
identified by school administrators, and were matched
on age and language ability with children in the average
reader group. Additional criteria for participation
included average to above average scores on measures
of receptive vocabulary, the Peabody Picture Vocabu-
lary Test – Revised (PPVT-R; Dunn and Dunn, 1981),
and expressive vocabulary, the Vocabulary subtest of
the Wechsler Intelligence Scale for Children – Third
Edition (WISC-3; Wechsler, 1991).
The criteria for inclusion in the dyslexic reader group
involved scoring at least 2 years below expected grade
level in word recognition, spelling and comprehension
abilities. A minimum 2-year discrepancy with the age-
and language-matched average-reader counterparts was
generally maintained on all three reading measures. The
Reading and Spelling subtests of the Wide Range
Achievement Test – Revised (WRAT-R; Jastak and
Wilkinson, 1984) were used to establish word recogni-
tion and spelling abilities respectively, and the Compre-
hension subtest of the Gates-MacGinitie Reading Test
Visual processing in Meares–Irlen syndrome: R. Kruk et al. 37
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
(GMRT; MacGinitie, 1980) was used in determining
comprehension abilities. All but three children spoke
English as a first language; these three were members of
the average reader group, and had been speaking
English for at least five years. These three participants
met all average reader group criteria for recognition,
spelling, comprehension and language abilities.
The Spelling subtest of the WRAT-R and the GMRT
were group administered in classrooms in participantsÕ
schools. Participants who met dyslexic or average reader
group criteria on both the WRAT-R Spelling test and
GMRT were given the WRAT-R Reading subtest,
PPVT-R and WISC-3 Vocabulary subtest in private
rooms in the schools.
Table 1 summarizes mean scores on participant char-
acteristics, which shows that the two groups were not
significantly different on age and spoken language
abilities. Group mean scores on the reading, spelling
and comprehension measures are also given in Table 1.
The dyslexic reader group scored significantly below the
average reader group on all reading measures, and
performed over 2 years below expected levels.
Measures
Irlen screeners, who were blind to the reading abilities of
the children, assessed participants for symptoms of
MISViS using the IRPS. Children were then adminis-
tered a battery of standardized measures shown to
represent a visual-processing-deficit factor (Watson and
Willows, 1993, 1995; Willows et al., 1993) made up of
measures of visual form discrimination, visual memory,
and pattern analysis and synthesis.
Irlen screening. The IRPS is a questionnaire that
includes self-report measures of reading difficulties and
eyestrain in the first part, and a second part adminis-
tered in a separate session that involves assessing visual
anomalies while performing perceptual tasks and select-
ing preferred overlays (Irlen, 1983). Two Irlen clinic
screeners, who were hired for the purposes of this study,
were involved in the MISViS assessment. One adminis-
tered the initial subjective questionnaire, and the other
conducted the perceptual portion of the IRPS and
carried out the overlay selection process. Hence, each
was masked to the reading abilities of the participants,
as well as to the outcomes of the partner screenerÕs
assessments. Testing was conducted in quiet rooms at
participantsÕschools.
The self-report part, which was a Short Intake form
that is typically used in such assessments, included
questions on the educational and reading history of the
participant, and questions on perceptual or physical
disturbances such as eyestrain, fatigue and headaches
while reading. Participants were asked informally about
need for assistance in the classroom for reading or other
subjects, and were also asked 16 specific questions from
the IRPS from Ôsection 1aÕ, on types of reading difficulty
experienced, and 16 questions from Ôsection 1bÕ,on
physical symptoms such as eyestrain and fatigue expe-
rienced while reading. Scores on each section were the
total number of items endorsed by the participant.
The second part of the IRPS involved four tasks
designed to detect perceptual difficulties reported to be
associated with MISViS symptoms, and the selection of
the preferred overlay tints (Irlen, 1983). During and
after each task, participants were queried on what they
saw. A larger number of perceptual anomalies reported
by participants resulted in a higher score on this part of
the IRPS, and indicated a greater severity of MISViS
symptoms. The tasks, presented in the following order
to each participant, were Box A: children counted the
number of simple line-divisions along two sides of a
cube; Box B: children counted number of simple line-
divisions using a more complex version of Box A with
three-dimensional line-divisions; Pumpkin: children
counted the number of symbols in a group on a
computerized illustration; and Musical Lines
2
: children
focused on a particular spot on a page displaying music
staves in order to determine if participants perceived any
movement of the music lines in the visual periphery.
During and after each task, participants were asked the
following types of questions: Did the lines (or symbols)
Table 1. Age, language and reading ability characteristics of dys-
lexic and average reader groups
Measure
Dyslexic
group
(n= 18)
Average
reader
group
(n= 18)
t-test
significance
Mean S.D. Mean S.D.
Age (years) 9.50 0.62 9.60 0.56 n.s.
PPVT-R
a
93.67 9.93 97.67 10.34 n.s.
WISC-3 Vocabulary
b
9.18 2.31 9.80 2.40 n.s.
WRAT-R Reading
c
2.22 0.50 5.37 1.24 <0.001
WRAT-R Spelling
c
2.30 0.50 5.50 1.03 <0.001
Comprehension
c
1.79 1.39 4.88 1.13 <0.001
a
PPVT-R result reported as standard score with mean 100 and
standard deviation 15.
b
WISC-3 subtest result reported as standard score with mean 10 and
standard deviation 3.
c
WRAT-R and comprehension results reported as grade equivalents.
Because participants were tested at the end of the school year,
expected grade equivalents for third and fourth graders are 3.9 and
4.9 respectively (mean = 4.45).
2
As a result of illness, the final task (Musical Lines) was not
administered to two children. According to the screeners, this did
not interfere with the assessment. For the purposes of data analysis the
averages of participantsÕscores on the three other tasks were used in
the place of their missing scores on the fourth task.
38 Ophthal. Physiol. Opt. 2008 28: No. 1
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
stay still, jiggle, dance, or move? Did the background
rise up or stay in place? Did your eyes feel rested, tired,
or strained?
Preferred overlays were selected by the Irlen clinician
in a standard manner by a process of elimination, based
on participant reports on overlay colours that produced
clearest, sharpest, or easiest to see images of letters on a
page of Dutch writing. Two different overlays were used
at the same time, each covering half the page. Partici-
pants were asked to identify the preferred overlay, after
which the less preferred one was removed, and a new one
introduced, selected from a prescribed set of overlays.
Once a single preferred overlay was chosen, combina-
tions of overlays were then introduced in the same
manner. The process was repeated until a final preferred
overlay, or combination of overlays was determined.
Detailed descriptions of the tint selection process can be
found in Lopez et al. (1994) and Irlen (1990).
Participants were then asked to name letters on the
page used to determine preferred overlays, with and
without the overlays. In addition, participants read
orally from an English passage (selected by the screener
from the participantsÕown books, or from screener-
supplied reading passages, based on the screenerÕs
judgement of the participantsÕreading abilities) with
and without the preferred overlays. The screener finally
observed letter naming and reading performance, as well
as subjective responses from the participants during the
task, with and without the overlays. Although all
information from the process was considered before a
final diagnosis for MISViS was made, the final deter-
minant of MISViS diagnosis was based on improved
perceptual and reading performance with the preferred
overlay. This criterion, of response to treatment as a
basis for diagnosis of MISViS rather than positive
symptomology from test data, has been identified as a
potential flaw that could give invalid results (Singleton
and Trotter, 2005) and potentially risk failing to identify
MISViS individuals who happen to fail to respond to
treatment. Interview information and scores on the
IRPS contributed to the evaluation of MISViS severity,
which ranged from low to severe, for those who showed
marked improvement in perceptual and reading perfor-
mance with their preferred overlays. Hence, MISViS
was indicated by improved reading accuracy as a result
of filtering with preferred tints compared with no-tint
reading performance, while intensity was determined by
results of self-report data on the IRPS.
The screening procedure identified a preferred overlay
colour or combination of colours for each participant,
regardless of whether the participant was diagnosed
with MISViS. This was typical, according to the
screeners, as most individuals have a preferred overlay
colour, even if they have no MISViS symptoms or
reading difficulties.
Visual processing measures. These tasks were used to
examine participantsÕvisual processing abilities, in order
to determine the significance of relationships between
MISViS diagnosis and the abilities that are character-
istic of a visual-processing-deficit subtype of reading
difficulty. This set of tests was selected from measures
found to have clustered on this subtype of reading
difficulty (Willows et al., 1993; Watson and Willows,
1995). All tests included stimuli that involved high-
contrast line drawings in black ink on white paper, with
spatial frequency dimensions of individual line strokes
that fell within the range of 2- to 12-cpd, consistent with
the spatial frequencies that were found to evoke the
experience of visual stress in individuals with MISViS
(Conlon et al., 2001; Wilkins et al., 2004). However,
because none of the stimuli included patterns approx-
imating the periodic white-black lines of square-wave
gratings, or lines of text, it cannot be assumed that the
stimuli evoked strong visual stress reactions.
The tests were administered in quiet rooms in partic-
ipantsÕschools. Significant correlations between MISViS
diagnosis and visual processing measures would dem-
onstrate that MISViS diagnosis predicts the visual
processing anomalies of this subtype of dyslexia. This
could be a consequence of visual stress being evoked by
the visual processing test stimuli, assuming that the
stimuli were sufficient in evoking visual stress. Alterna-
tively, visual encoding strategies, required by visual
processing task demands, could be poorly developed as
a byproduct of visual stress experiences in the MISViS
children.
The measures used were the Symbol Search (measur-
ing form discrimination and visual memory), Block
Design (pattern analysis and synthesis) and Coding
(symbol matching with attention, memory and motor
components) subtests of the WISC-3 (Wechsler, 1991);
four subtests of the Test of Visual Perception Skills
(TVPS; Gardner, 1982): i.e., Visual Discrimination
(measuring form analysis), Visual Memory (form anal-
ysis with a memory component), Visual/Spatial Rela-
tions (form and spatial/orientation discrimination), and
Visual Figure/Ground (embedded figure discrimina-
tion); and the Developmental Test of Visual-Motor
Integration (DVMI; Beery and Buktenica, 1981) that
involved copying geometric designs. The Digit Span
subtest of the WISC-3 was also administered. Although
it does not measure visual processes, it was included
because it distinguishes between children with and
without reading difficulty.
Results
The main question to be answered was: Is the degree of
MISViS associated with performance on the visual
processing (VP) tests for children of different reading
Visual processing in Meares–Irlen syndrome: R. Kruk et al. 39
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
abilities? A related question investigated how perfor-
mance on the two parts of the IRPS correlates with the
visual-processing measures. These questions were inves-
tigated by subjecting the results first to a series of
correlation analyses, followed by factorial analyses of
variance, examining performance differences among the
MISViS and reader groups. Because complete Irlen
assessment includes several subjective judgements by the
screeners, participantsÕnumerical scores on the IRPS
and the final MISViS diagnosis were examined for
potential differences in patterns of correlations with
reading and VP measures. Different patterns of corre-
lation were expected, based on different predictive
results that were found for the effects of colour filtering
on reading rate from subjective self-reports of previous
symptoms, compared to objective measures of current
symptoms (Hollis and Allen, 2006). As a further
determination of the predictive relationship between
MISViS diagnosis and reading ability, a series of
regression analyses on decoding, comprehension and
spelling outcome measures were conducted, using VP
and MISViS measures as predictors. If both types of
predictors account for unique variance in the outcomes,
then this would constitute evidence for different roles of
VP- and MISViS-related processes in reading perfor-
mance.
MISViS screening results
Participants were assigned to one of three groups
(MISViS, Questionable MISViS [QMISViS], and No
MISViS), based on the diagnosis of the two certified
Irlen screeners. The breakdown of participants into their
groups is shown in Figure 1. The 28 participants for
whom an evaluation of presence or absence of MISViS
was unambiguously obtained were rated on a scale of
severity of symptoms, which were given numeric
weightings as follows: No MISViS = 0, low = 1, low-
moderate = 1.5, moderate = 2, moderate-severe =
2.5, and severe = 3, indicating degree of MISViS
severity (MISD) The eight children with a diagnosis of
QMISViS could not be assigned unambiguously to
either the No MISViS or MISViS groups primarily
because of difficulties by some of the younger partici-
pants in articulating their perceptions. On the basis of
their total scores on the IRPS, the QMISViS partici-
pants (mean = 31.06, S.D. = 12.98) appeared to be
more similar to MISViS-diagnosed participants
(mean = 37.67, S.D. = 12.16) than to participants
identified as No MISViS (mean = 11.05, S.D. =
8.09). The percentage of participants with dyslexia
diagnosed with MISViS (45%) corresponds to the
approximate percentage of dyslexic readers reported
by Irlen to have MISViS (about 50%). The percentage
of average readers that were diagnosed with MISViS
was 55%, an unexpectedly high figure that surpasses the
7–15% anecdotal figure suggested by Irlen and the 20–
30% figure of published prevalence reports, but is
consistent with a previous report of no significant
difference in prevalence between reader groups (Kriss
and Evans, 2005).
10
8
6
Number of cases
4
2
0MISViS No MISViS
Dia
g
nosis
Questionable
MISViS
Figure 1. Breakdown of average (black bars) and dyslexic (white
bars) reader groups according to MISViS diagnosis type. Total
N = 36.
Table 2. Digit span, visual processing, and IRPS characteristics of
dyslexic and average reader groups
Measure
Dyslexic
reader
group
a
Average
reader group
t-test
significance
Mean S.D. Mean S.D.
Digit Span
b
8.00 2.22 10.33 2.35 t(34) = 3.06,
p< 0.004
Visual Processing Measures
b
Coding 10.17 3.81 11.61 3.11 n.s.
Block Design 10.11 2.97 10.78 3.56 n.s.
Symbol Search 10.94 2.84 12.28 3.29 n.s.
Visual Discrim. 13.50 3.22 13.44 3.76 n.s.
Visual Memory 9.22 2.60 10.67 3.11 n.s.
Visual Sp. Rel. 13.67 4.04 13.67 3.18 n.s.
Visual Fig. Gnd. 11.17 5.60 13.39 4.65 n.s.
DVMI 8.33 2.17 10.22 1.96 t(34) = 2.74,
p< 0.01
Meares–Irlen Syndrome Measures
MISD 1.31 1.13 1.40 1.06 n.s.
IRPS1a 5.67 3.15 4.72 3.51 n.s.
IRPS1b 3.83 3.05 3.75 2.88 n.s.
IRPS2 17.36 8.38 22.28 13.40 n.s.
a
n= 18 for each group in all analyses except for MISD. In the group
comparison of MISD scores, n= 13 for the dyslexic reader group
and n= 15 for the average reader group.
b
WISC-3 subtest (Digit Span, Coding, Block Design and Symbol
Search), TVPS subtest (Visual Discrimination, Visual Memory,
Visual Spatial Relations and Visual Figure Ground), and DVMI
results reported as standard scores with population mean 10 and
standard deviation 3.
40 Ophthal. Physiol. Opt. 2008 28: No. 1
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
Table 2 lists mean scores on the VP and IRPS
measures of the dyslexic and average reader groups.
The two reader groups were distinguished on the Digit
Span subtest of the WISC-3, which showed that,
consistent with the literature (e.g. Swanson et al.,
2006), the dyslexic group performed lower than the
average reader group on working memory ability.
Furthermore, the dyslexic group scored lower than the
average reader group on the DMVI, which indicates
that our dyslexic reader group demonstrates a charac-
teristic of the visual processing subtype of dyslexia
(Willows et al., 1993; Watson and Willows, 1995). None
of the IRPS measures, including MISD, distinguished
the dyslexic from the average reader groups. This is
consistent with previous literature showing that MISViS
and dyslexia are independent conditions (Kriss and
Evans, 2005). However, this independence might be
specific to average readers and dyslexic readers who do
not experience visual processing difficulties. The possi-
bility of a relationship between MISViS and visual
processing measures indicative of a visual processing
subtype of dyslexia is examined below.
Visual processing and IRPS measures
An initial analysis of the visual processing characteristics
of the MISViS groups was conducted using analysis of
variance (
ANOVAANOVA
). Because the diagnosis of MISViS is
based on several factors, including the subjective judge-
ment of the Irlen screener as well as improvement from
preferred overlays, subsequent correlational and regres-
sion analyses included the separate subjective as well as
objective measures used to make the MISViS diagnosis.
Hence, the following four sets of IRPS-scores were used
for the Pearson product–moment correlation analyses
with VP scores: MISD. Participants were given scores on
the basis of levels of severity of final diagnosis with the
complete procedure used by the screeners, as described
above, with scores ranging from 0 to 3. IRPS1a.
ParticipantsÕnumerical scores on the first part (section
1a) of the IRPS, which involved subjective self-ratings in
response to 16 questions on types of reading difficulties
experienced by the participant. Examples include blurred
print and losing place during reading. Scores ranged
from 1.0 to 12.0 for dyslexic readers, and from 0.5 to 11.0
for average readers. IRPS1b. ParticipantsÕscores in
section 1b of the IRPS, which involved evaluations of
reports of physical symptoms such as eyestrain and
headaches when reading, ranged from 0 to 13 for dyslexic
readers and from 0 to 10 for average readers. IRPS2.
This measure was based on participantsÕscores on the
last section of the IRPS, an objective index of perfor-
mance on perceptual tasks. It relied heavily on partici-
pantsÕself-reports during the tasks, and was subject to
potential differences in criteria in participant judgements
of what was perceived. Scores ranged from 1 to 33 for
dyslexic readers and from 0 to 40 for average readers.
If MISViS diagnosis is an indicator of the anomalous
visual processes identified by Watson and Willows
(1995) in their visual-processing-deficit subtype of dys-
lexia then negative relationships between extent of
MISViS sympotomology determined by each of the
above IRPS measures and scores on the VP tests that are
related to the visual processing deficit subtype should be
observed. The intercorrelations among the IRPS and VP
test scores for the entire sample are given in Table 3. The
results of the analyses on each IRPS measure are
discussed below. In addition, if the relationship is
specific to dyslexic children, then this negative relation-
ship should be observed in the dyslexic group but not
with the average reader group. This expectation was
tested with an
ANOVAANOVA
analysis of performance of
MISViS and reader groups on the VP measures.
MISViS groups. Between-group analyses of variance
(with MISViS Group and Reader Group as the
independent variables) were conducted on all VP and
reading-related measures. None of the VP measures
showed any significant effect of MISViS Group, or
interaction between MISViS and Reader groups, except
in the case of the DVMI in which a significant
interaction was found, F(2, 30) = 4.15, p= 0.026.
A Bonferroni-corrected post-hoc analysis of mean scores
showed that the no-MISViS dyslexic readers scored
worse than the no-MISViS average readers. In fact,
dyslexic readers with positive diagnosis of MISViS had
the same level of performance on the DVMI as all
average reader groups, regardless of MISViS status (see
Table 4). This result is contrary to the expected pattern,
and indicates that MISViS diagnosis is not an indicator
of a visual deficit subtype of dyslexia.
A significant MISViS group effect was also found on
WRAT-R Spelling, F(2, 30) = 7.39, p= 0.002, with
children diagnosed with MISViS and QMISViS dem-
onstrating poorer spelling performance (mean = 3.71,
S.D. = 1.32, and mean = 3.85, S.D. = 2.32 respec-
tively) than children without MISViS (mean = 4.28,
S.D. = 2.26). However, a significant interaction be-
tween Reader Group and MISViS Group, F(2,
30) = 12.50, p< 0.001 showed that the MISViS group
effect occurred among the average reader groups, and
not among the dyslexic reader groups. Post-hoc tests
showed that average readers diagnosed with MISViS
had poorer spelling abilities (mean = 4.74,
S.E. = 0.183) than average readers without MISViS
(mean = 6.38, S.E. = 0.259) or with QMISViS
(mean = 6.57, S.E. = 0.334). Dyslexic readers in the
three MISViS groups did not differ significantly on
WRAT-R Spelling scores; mean = 2.43, S.E. = 0.205
for those with MISViS, mean = 2.18, S.E. = 0.260 for
Visual processing in Meares–Irlen syndrome: R. Kruk et al. 41
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
those without MISViS, and mean = 2.22, S.E. = 0.260
for those with QMISViS.
MISD. The distribution of scores of the 28 partici-
pants who obtained unambiguous positive or negative
MISViS diagnoses was as follows: the five No MISViS
participants in each reader group were assigned scores
of 0.0. None of the participants who were given positive
MISViS diagnoses scored 1.0 (low), one MISViS-
participant in each reader group scored 1.5 (low-
moderate), five dyslexic and six average readers scored
2.0 (moderate), one dyslexic and three average readers
scored 2.5 (moderate-severe), and one dyslexic reader
scored 3.0 (severe).
Only one significant correlation, involving a moderate
negative correlation (r=)0.379, p< 0.05) between
MISD score and TVPS Visual Discrimination was
found. Low visual discrimination abilities are associated
with high scores on MISD. The characteristic visual
anomalies found by Watson and Willows (1995) for
their visual-processing-deficit subtype of dyslexia
include visual discrimination. However, performance
on the TVPS Visual Discrimination subtest captures
only a small degree of the range of visual processing
anomalies that typify this subgroup of dyslexic readers
(Willows et al., 1993; Watson and Willows, 1995).
IRPS1a. Individuals who reported having a greater
number of reading difficulties had lower scores on three
of the VP tests, indicated by low to moderate negative
correlations with WISC-3 Coding and Symbol Search
subtests, and the TVPS Visual Memory subtest (see
Table 3). These subtests have a substantial visual
memory component, indicating that visual problems
that the IRPS1a is detecting are likely to be related to
visual attention and to encoding visual information in
memory.
IRPS1b. Results showed that individuals who claimed
to have higher subjective degrees of eyestrain had lower
scores on two of the VP tests, indicated by negative
correlations with WISC-3 coding, and the TVPS Visual
Table 3. Intercorrelations among visual processing measures and MISViS grouping scores
1 2 34 5 6789 10111213141516
1. Coding –
2. Block 0.318 –
3. Symbol 0.627
***
0.429
**
–
4. Dig. Span )0.036 )0.075 0.142 –
5. Vis. Disc. 0.014 0.289 0.213 0.098 –
6. Vis. Mem. 0.371
*
0.440
**
0.395
*
0.374
*
0.329 –
7. Vis. Sp. 0.322 0.197 0.163 )0.072 0.449
**
0.258 –
8. Vis. Fig. 0.303 0.399
*
0.294 )0.073 0.325 0.248 0.214 –
9. DVMI )0.058 0.339
*
0.343
*
0.386
*
0.181 0.351
*
0.146 0.369
*
–
10. MISD
a
)0.199 )0.030 )0.048 0.236 )0.379
*
)0.239 )0.098 )0.270 0.187 –
11. IRPS1a )0.353
*
)0.155 )0.335
*
)0.038 0.065 )0.372
*
0.043 )0.160 )0.084 0.609
**
–
12. IRPS1b )0.369
*
)0.106 )0.277 0.104 )0.163 )0.321 )0.074 )0.362
*
)0.108 0.793
***
0.755
***
–
13. IRPS2 )0.246 )0.139 )0.124 0.305 )0.035 )0.168 0.008 )0.167 0.125 0.638
***
0.638
***
0.668
***
–
14. Reading 0.243 0.254 0.404
*
0.501
**
0.045 0.402
*
0.011 0.300 0.550
***
0.114 )0.194 )0.092 0.171 –
15. Spelling 0.305 0.114 0.346
*
0.476
**
)0.073 0.327 )0.012 0.182 0.474
**
)0.056 )0.319 )0.181 0.033 0.919
***
–
16. Comp. 0.152 0.374
*
0.398
*
0.357
*
0.004 0.281 0.005 0.380
*
0.430
**
0.097 )0.233 0.007 0.146 0.826
***
725
***
–
a
N= 28 for correlations involving MISD. All other correlations N= 36.
*Significant at the 0.05 level.
**Significant at the 0.01 level.
***Significant at the 0.001 level.
Table 4. Mean scores of the MISViS and reader groups on the
DVMI
MISViS group
Dyslexic
reader group
Average
reader group
t-test
significance
Mean S.E. Mean S.E.
No MISViS 6.00 0.81 10.67 1.04 t(6) = 3.29,
p= 0.017
Questionable
MISViS
8.20 0.81 10.40 0.81 n.s.
MISViS 9.88 0.64 10.00 0.57 n.s.
42 Ophthal. Physiol. Opt. 2008 28: No. 1
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
Figure/ground subtest (see Table 3). These tests measure
visual analysis and discrimination, and involve aspects
of visual attention and memory.
IRPS2. The objective perceptually based portion of
the IRPS showed no significant correlation with any of
the visual processing subtests.
IRPS and reading-related measures
A final set of correlation analyses was conducted
between reading skill measures and IRPS performance
outcomes. No significant correlations were found
between the reading measures (reading, spelling and
comprehension) and the four IRPS measures.
Visual processing and reading skill measures
Additional correlation analyses were conducted on the
VP measures and WRAT-R reading and spelling, and
the GMRT comprehension subtests to determine if the
same pattern of results found with IRPS measures
would be obtained with VP measures.
Nine of the reading-VP score pairings were significant
positive correlations (see Table 3). Lower reading,
spelling and comprehension scores tended to co-occur
with lower VP test scores, supporting the link between
visual processing deficits and reading difficulty in our
sample. Positive correlations between reading measures
and VP measures are consistent with the presence of a
visual processing deficit in our sample, indicated by
lower scores on the VP measures in participants who
scored low on the reading measures.
Word recognition ability, as measured by perfor-
mance on the WRAT-R reading subtest, was positively
correlated with Symbol Search and Digit Span subtests
of the WISC-3, the Visual Memory subtest of the TVPS,
and the DVMI. WRAT-R Spelling was correlated with
the Symbol Search and Digit Span subtests of the
WISC-3, and the DVMI; and the GMRT Comprehen-
sion measure was positively correlated with all measures
that were found to be correlated with the WRAT-R
Spelling subtest, as well as with the Block Design and
Visual Figure/ground subtests. This pattern of correla-
tions is consistent with the patterns identified by Watson
and Willows (1995) in their visual processing deficit
subtype.
MISViS and VP as predictors of reading and spelling
In order to clarify the roles that MISViS and VP
measures played in predicting reading and spelling
measures, regression analyses were conducted with
reading, spelling and comprehension as dependent
measures, and the IRPS (IRPS1a, IRPS1b, IRPS2)
and VP measures as predictors. For each of the
decoding (WRAT-Reading), spelling (WRAT-Spelling)
and comprehension (GMRT) outcome measures, none
of the IRPS measures were significant unique predictors.
This is consistent with MISViS being independent of
reading ability. Of the VP measures, DVMI (b= 0.433,
R
2
= 0.302, p< 0.01) and Coding (b= 0.280,
R
2
= 0.078, p< 0.05) emerged as a predictors of
unique variance in decoding. Digit Span was an addi-
tional predictor of unique variance in decoding
(b= 0.344, R
2
= 0.098, p< 0.05) consistent with
previous literature (e.g. Swanson et al., 2006).
With regard to spelling, the same three variables
accounted for unique variance: DVMI (b=
0.358, R
2
= 0.109, p< 0.05), WISC-3 Coding
(b= 0.338, R
2
= 0.104, p< 0.05) and Digit Span
(b= 0.350, R
2
= 0.226, p< 0.05). DVMI alone
was a predictor of unique variance in comprehension
(b= 0.430, R
2
= 0.185, p< 0.01). In a subsequent
exploratory analysis, vocabulary measures were
included in the set of potential predictors, and the
WISC Vocabulary subtest emerged as a predictor of
unique variance in comprehension (b= 0.337,
R
2
= 0.112, p< 0.05).
Discussion
This study explored the relationship between MISViS,
measured by the IRPS, and the visual processing
characteristics identified in a large proportion of chil-
dren with dyslexia (Watson and Willows, 1995). The
results of the correlation and
ANOVAANOVA
analyses showed
that children who experience difficulty reading, with
co-occurring symptoms of eyestrain and fatigue as
measured by subtests of the IRPS, demonstrate none
of the anomalous visual processing characteristics of the
visual-processing-deficit-subtype identified by Watson
and Willows (1995). Hence, in answer to the first
research question, the perceptual dysfunction involved
in MISViS is different from the perceptual dysfunction
identified as a visual processing deficit subtype of
dyslexia (Watson and Willows, 1995). Furthermore,
similar patterns of results were found for the first part of
the IRPS, but not for the second part of the IRPS.
The results, showing significant negative correlations
between the IRPS1a and IRPS1b measures and a
selection of VP measures, indicate that individuals
who report reading difficulties and symptoms of eye-
strain and fatigue experience less efficient visual pro-
cessing than those who do not report such symptoms.
The significant relationships were found on measures
that are related to encoding visual information, the
WISC-3 Coding and TVPS Visual Memory subtests, as
well as on measures involving visual spatial attention,
the WISC-3 Symbol Search, and TVPS Discrimination
Visual processing in Meares–Irlen syndrome: R. Kruk et al. 43
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
and Visual Figure Ground subtests. Although it cannot
be concluded that the correlations were functions of VP
stimuli evoking strong visual stress responses, this
explanation is plausible if a stress response were evoked
among individuals scoring high on the IRPS measures
that was sufficient to interfere with visual processing
required by the VP task demands. Furthermore, Conlon
et al. (1999) in their Experiment 3 reported a negative
relationship between WISC Coding and their measure of
MISViS, the Visual Discomfort Scale, which was similar
to our findings with the IRPS1a and IRPS1b. They
interpreted their result as indicating that their partici-
pantsÕability to attend to key aspects of the stimuli was
impaired by interference related to MISViS.
The IRPS2 results were not correlated with any of the
VP test scores. This is somewhat surprising as the IRPS2
was expected to predict a wide array of VP and reading
measures in dyslexic readers, particularly given the
finding by Hollis and Allen (2006) that a comparable
measure of immediate experience predicted changes in
reading rate performance in adults with co-occurring
MISViS and dyslexia, but not in adults with dyslexia
alone. Final diagnosis of MISViS severity based on the
complete Irlen screening procedure was linked only with
the measure indicative of visual discrimination ability.
The visual problems detected by higher IRPS scores
may constitute a distinct perceptual processing dysfunc-
tion, or may be a consequence of MISViS as described
by Wilkins (1993). Given the visual attention compo-
nents of the correlated visual ability measures, it would
seem that the IRPS measures in our sample are
predicting more than sensory aspects of visual pro-
cesses. The potential influence of MISViS on these
visual attention components may involve inability to
encode clear visual information, and inefficient spatial
attention processes, as consequences of symptoms
related to visual stress. Although the stimuli in general
had spatial frequency characteristics in the range
thought to be stress evoking in individuals identified
with visual stress (Wilkins, 1995; Conlon et al., 1998;
Conlon and Hine, 2000), they did not involve repeated
patterns of light and dark bars. Furthermore, no direct
measurement of visual stress experience was conducted
during the visual processing tests, and so we cannot
determine, with the present data, if the correlations
were related to a state of visual stress experience or of a
characteristic of MISViS experienced between states of
strong visual stress.
The evidence supports the presence of a relationship
between MISViS and perceptual processes, but not
reading performance. The regression results further
indicate that visual processing measures, particularly
the DVMI, predict reading and spelling performance,
but that none of the IRPS measures uniquely predict
reading performance. This underscores the indepen-
dence of MISViS and reading achievement. However,
the correlations between IRPS scores and VP measures
are consistent with at least two documented explana-
tions of the roles that MISViS might play in visual
processing, which involve visual spatial attention. The
relationships between IRPS and VP measures could
have come about as a result of a reduced efficiency in
processing visual information within an attentional
spotlight, or of a reduced ability to focus spatial
attention (Conlon and Humphreys, 2001). Although
distinguishing between these two alternatives is beyond
the scope of the data collected here, the results are in
harmony with the conclusions of Conlon and Humph-
reys (2001) that less efficient visual processing experi-
enced by individuals with MISViS is not restricted to
gratings, the ideal visual-stress-inducing stimuli, and
that the visual processing difficulties related to MISViS
can extend beyond early cortical levels.
Although MISViS and dyslexia are likely to be
distinct conditions, the co-occurrence of the two condi-
tions in children indicates different visual processing
characteristics than those found in children without
MISViS. The findings indicate that MISViS can coexist
with dyslexia as an independent condition, and that it is
not linked to any particular subtype. The results do not
support any link between MISViS and the visual deficit
subtype of dyslexia, and so the best treatment for people
with MISViS and dyslexia is to correct the MISViS, and
to encourage general education strategies, such as those
designed to strengthen phonological processes and
fluency in word recognition, to address the reading
disability.
Acknowledgements
This research was supported in part by a grant from
Rotary International.
References
Beery, K. E. and Buktenica, N. A. (1981) Developmental Test
of Visual-Motor Integration. Modern Curriculum Press,
Cleveland, OH.
Blaskey, P., Scheiman, M., Parisi, M., Ciner, E., Gallaway, M.
and Selznick, R. (1990) The effectiveness of Irlen filters for
improving reading performance: a pilot study. J. Learn.
Disabil. 23, 604–612.
Boder, E. (1971) Developmental dyslexia: prevailing diagnostic
concepts and a new diagnostic approach. In: Progress
in Learning Disabilities, Vol. 2 (eds H. R. Myklebust), Grune
& Stratton, New York, pp. 293–321.
Bouldoukian, J., Wilkins, A. J. and Evans, B. J. W. (2002)
Randomized controlled trial of the effect of coloured
overlays on the rate of reading of people with
specific learning difficulties. Ophthalmic Physiol. Opt. 22,
55–60.
44 Ophthal. Physiol. Opt. 2008 28: No. 1
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
Bowers, P. G. and Wolf, M. (1993) Theoretical links between
naming speed, precise timing mechanisms and orthographic
skill in dyslexia. Reading Writing Interdiscip. J. 5, 69–85.
Ciuffreda, K. J., Kenyon, R. V. and Stark, L. (1983) Saccadic
intrusions contributing to reading disability. Am. J. Optom.
Physiol. Opt. 62, 844–852.
Conlon, E. and Hine, T. (2000) The influence of pattern
interference on performance in migraine and visual discom-
fort groups. Cephalalgia 20, 708–713.
Conlon, E. and Humphreys, L. (2001) Visual search in migraine
and visual discomfort groups. Vision Res. 41, 3063–3068.
Conlon, E., Lovegrove, W., Hine, T., Chekaluk, E., Piatek, K.
and Hayes-Williams, K. (1998) The effects of visual
discomfort and pattern structure on visual search. Percep-
tion 27, 21–33.
Conlon, E., Lovegrove, W. J., Chekaluk, E. and Pattison, P. E.
(1999) Measuring visual discomfort. Vis. Cogn. 6, 637–663.
Conlon, E., Lovegrove, W., Barker, S. and Chekaluk, E.
(2001) Visual discomfort: the influence of spatial frequency.
Perception 30, 571–581.
Dunn, L. M. and Dunn, L. M. (1981) Peabody Picture
Vocabulary Test – Revised. American Guidance Service,
Circle Pines, MN.
Evans, B. J. W. (1997) Coloured filters and dyslexia: whatÕsin
a name? Dyslexia Rev. 9, 18–19.
Evans, B. J. W. and Joseph, F. (2002) The effect of coloured
filters on the rate of reading in an adult student population.
Ophthalmic Physiol. Opt. 22, 535–545.
Evans, B. J. W., Wilkins, A. J., Brown, J., Busby, A.,
Wingfield, A. E., Jeanes, R. and Bald, J. (1996) A
preliminary investigation into the aetiology of Meares–Irlen
Syndrome. Ophthalmic Physiol. Opt. 16, 286–296.
Facoetti, A., Zorzi, M., Cestnick, L., Lorusso, M. L., Molteni,
M., Paganoni, P., Umilta, C. and Mascetti, G. G. (2006) The
relationship between visuo-spatial attention and nonword
reading in developmental dyslexia. Cogn. Neuropsychol. 23,
841–855.
Feagans, L. V., Short, E. J. and Meltzer, L. J. (1991) Subtypes
of Learning Disabilities: Theoretical Perspectives and Re-
search. Lawrence Erlbaum Associates Inc., Hillsdale, NJ.
Gardner, M. F. (1982) Test of Visual Perceptual Skills (Non-
Motor) . Special Child Publications, Seattle, WA.
Garzia, R. P. (1993) Optometric factors in reading disability.
In: Visual Processes in Reading and Reading Disabilities (eds
D. M. Willows, R. S. Kruk and E. Corcos), Lawrence
Erlbaum Associates, Hillsdale, NJ, pp. 419–434.
Hollis, J. and Allen, P. M. (2006) Screening for Meares–Irlen
senstitivity in adults: Can assessment methods predict
changes in reading speed? Ophthalmic Physiol. Opt. 26,
566–571.
Hooper, S. and Willis, W. (1989) Learning Disability Subtyp-
ing. Springer-Verlag, New York.
Irlen, H. (1983) Successful treatment of learning disabilities.
Paper presented at The 91st Annual Convention of the
American Psychological Association, Annaheim, CA.
Irlen, H. (1990) Method and Apparatus of Treatment of
Symptoms of the Irlen Syndrome. United States patent
4961640. Retrieved 20 June 2007, from http://www.freepat-
entsonline.com/4961640.html
Irlen, H. (1991) Reading by the Colours: Overcoming Dyslexia
and Other Reading Disabilities by the Irlen Method, Avery,
Garden City Park, NY.
Irlen, H. (2003) Irlen Syndrome/Scotopic Sensitivity. Retrieved
28 August 2003, from http://www.irlen.com/index_SSS.html
Irlen, H. and Lass, M. J. (1989) Improving reading problems
due to symptoms of scotopic sensitivity syndrome using
Irlen lenses and overlays. Education 109, 413–417.
Jastak, S. and Wilkinson, G. S. (1984) Wide Range
Achievement Test – Revised. Jastak Associates, Wilming-
ton, DE.
Jeanes, R., Busby, A., Martin, J., Lewis, E., Stevenson, N.,
Pointon, D. and Wilkins, A. (1997) Prolonged use of
coloured overlays for classroom teaching. Br. J. Psychol.
88, 531–548.
Kriss, I. and Evans, B. J. W. (2005) The relationship between
dyslexia and Meares–Irlen Syndrome. J. Res. Reading 28,
350–364.
Lightstone, A., Lightstone, T. and Wilkins, A. (1999) Both
coloured overlays and coloured lenses can improve reading
fluency, but their optimal chromaticities differ. Ophthalmic
Physiol. Opt. 19, 279–285.
Lopez, R., Yolton, R. L., Kohl, P., Smith, D. L. and Saxerud,
M. H. (1994) Comparison of Irlen Scotopic Sensitivity test
results to academic and visual performance data. J. Am.
Optom. Assoc. 65, 705–714.
MacGinitie, W. H. (1980) Gates-MacGinitie Reading Tests –
Canadian Edition, Nelson, Toronto, ON.
Meares, O. (1980) Figure/ground, brightness contrast, and
reading disabilities. Visible Lang. 14, 13–29.
OÕConnor, P. D., Sofo, F., Kendall, L. and Olsen, G. (1990)
Reading disabilities and the effects of coloured filters.
J. Learn. Disabil. 23, 597–603.
Parker, R. M. (1990) Power, control, and validity in research.
J. Learn. Disabil. 23, 613–620.
Rayner, K. and Pollatsek, A. (1989) The Psychology of
Reading Prentice Hall, Englewood Cliffs, NJ.
Robinson, G. L. W. and Conway, R. N. F. (1990) The effects
of Irlen coloured lenses on studentsÕspecific reading skills
and their perception of ability: a 12-month validity study.
J. Learn. Disabil. 23, 589–596.
Robinson, G. L. and Foreman, P. J. (1999) Scotopic sensitivity/
Irlen Syndrome and the use of coloured filters: a long-term
placebo-controlled and masked study of reading achievement
and perception of ability. Percept. Mot. Skills 89, 83–113.
Simmers, A. J., Gray, L. S. and Wilkins, A. J. (2001) The
influence of tinted lenses upon ocular accommodation.
Vision Res. 41, 1229–1238.
Singleton, C. and Henderson, L. M. (2007) Computerized
screening for visual stress in children with dyslexia. Dyslexia
13, 130–151.
Singleton, C. and Trotter, S. (2005) Visual stress in adults with
and without dyslexia. J. Res. Reading 28, 365–378.
Sireteanu, R., Goertz, R., Bachert, I. and Wandert, T. (2005)
Children with developmental dyslexia show a left visual
‘‘minineglect’’. Vision Res. 45, 3075–3082.
Solan, H. A. (1990) An appraisal of the Irlen technique of
correcting reading disorders using tinted overlays and tinted
lenses. J. Learn. Disabil. 23, 621–626.
Visual processing in Meares–Irlen syndrome: R. Kruk et al. 45
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
Stanley, G. (1990) Rose-coloured spectacles: a cure for
dyslexia? Aust. Psychol. 25, 65–76.
Stein, J. F. (1993) Visuospatial perception in disabled readers.
In: Visual Processes in Reading and Reading Disabilities (eds
D. M. Willows, R. S. Kruk and E. Corcos), Lawrence
Erlbaum Associates, Hillsdale, NJ, pp. 331–346.
Stein, J. F. (2001) The magnocellular theory of dyslexia.
Dyslexia 7, 12–36.
Swanson, H. L., Howard, C. B. and Saez, L. (2006) Do
different components of working memory underlie different
subgroups of reading disabilities? J. Learn. Disabil. 39, 252–
269.
Valdois, S., Bosse, M. L. and Tainturier, M. J. (2004) The
cognitive deficits responsible for developmental dyslexia:
Review of evidence for a selective visual attention disorder.
Dyslexia 10, 339–363.
Vellutino, F. R. (1987) Dyslexia. Sci. Am. 256, 34–41.
Wagner, R. and Torgesen, J. (1987) The nature of phonolog-
ical processing and its causal role in the acquisition of
reading skills. Psychol. Bull. 101, 192–212.
Waldie, M. and Wilkins, A. (2004) How big does a coloured
overlay have to be? Ophthalmic Physiol. Opt. 24, 57–60.
Watson, C. and Willows, D. M. (1993) Evidence for a visual-
processing-deficit subtype among disabled readers. In:
Visual Processes in Reading and Reading Disabilities (eds
D. M. Willows, R. S. Kruk and E. Corcos), Lawrence
Erlbaum Associates, Hillsdale, NJ, pp. 287–309.
Watson, C. and Willows, D. M. (1995) Information-processing
patterns in specific reading disability. J. Learn. Disabil. 28,
216–231.
Watson, B. U., Goldgar, D. E. and Ryschon, K. L. (1983)
Subtypes of reading disability. J. Clin. Neuropsychol. 5, 377–
399.
Watson, C. S., Kidd, G. R., Horner, G., Connell, P. J.,
Lowther, A., Eddins, D. A., Krueger, G., Goss, D. A.,
Rainey, B. B., Gospel, M. D. and Watson, B. U.. (2003)
Sensory, cognitive, and linguistic factors in early academic
performance of elementary school children: The Benton-IU
project. J. Learn. Disabil. 36, 165–197.
Wechsler, D. (1991) Wechsler Intelligence Scale for Children,
3rd edn. Psychological Corporation, San Antonio, TX.
White, S., Milne, S., Rosen, S., Hansen, P., Swettenham, J.,
Frith, U. and Ramus, F. (2006) The role of sensorimotor
impairments in dyslexia: a multiple case study of dyslexic
children. Dev. Sci. 9, 237–255.
Wilkins, A. (1993) Reading and visual discomfort. In: Visual
Processes in Reading and Reading Disabilities (eds D. M.
Willows, R. S. Kruk and E. Corcos), Lawrence Erlbaum
Associates, Hillsdale, NJ, pp. 435–456.
Wilkins, A. (1995) Visual Stress. Oxford University Press,
Oxford.
Wilkins, A. (2003) Reading Through Colour: How Coloured
Filters can Reduce Reading Difficulty, Eye Strain and
Headaches. John Wiley & Sons, Chichester, UK.
Wilkins, A. J., Evans, B. J. W., Brown, J., Busby, A.,
Wingfield, A. E., Jeanes, R. and Bald, J. (1994) Double-
masked placebo-controlled trial of precision spectral filters
in children who use coloured overlays. Ophthalmic Physiol.
Opt. 14, 365–370.
Wilkins, A. J., Jeanes, R. J., Pumfreym, P. D. and Laskier, M.
(1996) Rate of Reading Test: Its reliability, and its validity in
the assessment of the effects of coloured overlays. Ophthal-
mic Physiol. Opt. 16, 491–497.
Wilkins, A. J., Lewis, E., Smith, F., Rowland, E. and Tweedie,
W. (2001) Coloured overlays and their benefit for reading.
J. Res. Reading 24, 41–64.
Wilkins, A., Huang, J. and Cao, Y. (2004) Visual stress theory
and its application to reading and reading tests. J. Res.
Reading 27, 152–162.
Wilkins, A. J., Sihra, N. and Myers, A. (2005a) Increasing
reading speed by using colours: issues concerning reliability
and specificity, and their theoretical and practical implica-
tions. Perception 34, 109–120.
Wilkins, A., Sihra, N. and Nimmo-Smith, I. (2005b) How
precise do precision tints have to be and how many are
necessary? Ophthalmic Physiol. Opt. 25, 269–276.
Willows, D. M., Corcos, E. and Kershner, J. R. (1993)
Perceptual and cognitive factors in disabled readersÕper-
ception and memory of unfamiliar visual symbols. In:
Studies in Visual Information Processing (eds S. Wright and
R. Groner), pp. 163–178. North-Holland Elsevier, Amster-
dam.
46 Ophthal. Physiol. Opt. 2008 28: No. 1
ª2008 The Authors. Journal compilation ª2008 The College of Optometrists
