VISUAL COGNITION, 1999, 6 (6), 637–663
Measuring Visual Discomfort
Elizabeth G. Conlon
Griffith University, Australia
William J. Lovegrove
Griffith University, Australia
Eugene Chekaluk
Macquarie University, Australia
Philippa E. Pattison
University of Melbourne, Australia
A two-parameter Rasch Rating Scale model was developed to measure visual
discomfort. Initially it was found that participants reporting frequent severe
headache, reading difficulties of a visual nature, and short effective reading times
experienced more severe visual discomfort. The validity of this measurement instrument was tested in four experiments. In Experiments 1 and 2 reports of unpleasant somatic and perceptual side-effects or ratings of unpleasantness were
obtained for low, moderate, and high scorers on the Visual Discomfort Scale. It
was found that those with higher scores reported a greater number of unpleasant
side-effects and rated square-wave patterns across the spatial frequency range,
and a letter pattern as more unpleasant and distorted to view than those with low
scores on the scale. In Experiments 3 and 4 these subjective findings were extended to performance. Efficiency was measured using a copying and reading
task. It was found that those obtaining high scores on the Visual Discomfort Scale
performed with less efficiency than others. It was concluded that experience of
unpleasant somatic and perceptual side-effects from pattern viewing produce
performance difficulties in individuals who we sensitive to pattern. The Visual
Discomfort Scale is a reliable and valid measure for testing the extent of these
difficulties.
Requests for reprints should be addressed to Dr. E. Conlon, School of Applied Psychology,
Griffith University, Gold Coast, PMB 50, Gold Coast Mail Centre, Queensland, 9726, Australia.
Email: [email protected]
Ó 1999 Psychology Press Ltd
638
CONLON ET AL.
It has been known for over one hundred years that repetitive striped patterns
may lead to visual side-effects, but it is only relatively recently that the nature of
such patterns and the consequences of the induced effects have been systematically studied. There are two aims of this paper. The first is to combine the
results of this recent research in order to produce a non-intrusive measurement
tool that will help to identify and classify those individuals who may suffer
from such visual effects, and the second is to empirically test these individuals
on a number of subjective and performance tasks.
Photosensitive epileptic patients show specific psychophysiological
responses to patterns of a certain type. Using EEG recordings, Wilkins, Binnie,
and Darby (1980) measured the strength of paroxysmal activity at the site of the
visual cortex while presenting a number of spatial patterns for a short duration.
The degree of paroxysmal activity depended on variables such as the spatial
frequency of the pattern (maximal for square waves between 1 and 4c/deg), the
duty cycle of the pattern (maximal for 50%), the area of the pattern (maximal
for large areas), the mode of viewing (maximal for binocular), and the nature of
the pattern (maximal for repetitive stripes rather than checkerboards or plaids).
In addition to this, the degree of activity was not dependent upon the exact area
of the retina stimulated by the pattern, but rather the degree of cortical stimulation present. These findings (reviewed by Wilkins, 1995) led Wilkins et al.,
(1984) to see whether the same stimulus pattern parameters would lead to
reports of side-effects in a group in the general population that was not epileptic. The stimulus parameters previously found to induce greatest paroxysmal
activity in photosensitive epileptic patients also induced the greatest number of
side-effects in this group. The side-effects experienced were both somatic (e.g.
sore or tired eyes) and visual/perceptual (e.g. flicker or appearance of colour
even though the patterns were monochromatic). This collection of side-effects
was termed visual discomfort by Wilkins et al. (1984). More recently this same
collection of symptoms has been described as visual stress (Wilkins, 1995). In
this paper the term visual discomfort will be used.
Wilkins et al. (1984) also obtained two groups of individuals, one reporting
high headache susceptibility and a second reporting symptomatology of
migraine. In their first experiment, measurement of differences in side-effects
from pattern viewing between high and low headache susceptible groups was
largest with presentation of a 3c/deg square-wave, with no differences between
groups obtained for lower and higher spatial frequencies. This effect at 3c/deg
was replicated in a group with characteristics of migraine (Marcus & Soso,
1989). In a further study the migraine group was shown to be more likely to
report symptoms of visual discomfort with exposure to 3c/deg square-waves
(Wilkins et al., 1984).
VISUAL DISCOMFORT
639
It may be argued that to reduce unpleasant side-effects susceptible individuals should avoid prolonged exposure to square-wave gratings that have the
characteristics previously described. Wilkins and Nimmo-Smith (1984, 1987),
however, drew an interesting parallel between the spatial characteristics of a
square-wave grating and the pattern formed by a page of text. The spatial frequency of a text page is obtained by treating the interline spacing as the bright
bar and the lines of text minus the ascenders and descenders as the dark bar of a
square-wave (Wilkins & Nimmo-Smith, 1987). Simply observing text produced similar effects in a control group as did observation of a square-wave. In
a study using a group of volunteers reporting eye-strain and headache when
reading, reducing the amount of pattern available to three lines reduced the
unpleasant effects reported. Effective reading time also increased, leading to a
suggestion that reducing the unpleasant arrangement of text may increase performance efficiency when reading in sensitive individuals. A recent letter identification study also found that more sensitive individuals were less task
efficient (Conlon et al., 1998). From these findings two things were concluded.
First, unpleasant effects could be induced from observation of striped patterns
either in square-wave or text form, and second, individual differences in sensitivity determine the number of effects reported.
The mechanism by which visual side-effects arise when inspecting gratings
or text has been the subject of some speculation. According to Wilkins (1986) a
pre-existing visual system hypersensitivity of striate cortex cells, which are
responsive to specific spatial frequencies and orientations, may produce
unpleasant side-effects. Presentation of repetitive striped patterns in the intermediate spatial frequency range may result in a spread of excitation in a localized area of the visual system. This is physically manifested as reports of
unpleasant somatic effects and the induction of a number of different illusory
features. In participants who do not experience these side-effects, according to
Wilkins, there is present an inhibitory mechanism that prevents this spread of
excitation. A failure of this mechanism, then, leads to the experience of perceptual side-effects associated with visual discomfort (Meldrum & Wilkins,
1984). Evidence for this speculation has come from a variety of sources. Physiologically based data has linked illusions of spatial frequency, orientation, and
motion to a release of inhibition of specific cell groups (called “disinhibition”)
(Georgeson, 1976, 1980, 1985). This notion is supported psychophysically by
measuring threshold changes following adaptation at specific spatial frequencies with prolonged adaptation at one spatial frequency. This results in inhibition to similar spatial frequencies, but facilitation, at spatial frequencies 2–3
octaves higher and lower than the adapting spatial frequency (Dealy &
Tolhurst, 1972; Tolhurst & Barfield, 1978). The use of evoked potentials has
also confirmed the psychophysical data obtained earlier (Suter, Armstrong,
Suter, & Powers, 1991).
640
CONLON ET AL.
Visual discomfort is not the only term that has been used to describe a set of
symptoms associated with reading text. Irlen (1991) coined the term scotopic
sensitivity syndrome to describe a collection of symptoms described by individuals experiencing difficulty when reading. Difficulty with accurate copy writing (e.g. Hannell et al., 1989; Whiting, 1988), poor concentration, and slower
performance skills (Irlen, 1993, 1991; Whiting, 1988) have been described as
characteristic of this syndrome, but little objective supporting evidence is
available. Although Irlen did not undertake systematic investigation of the
stimulus patterns that induce the symptoms, descriptions of unpleasant sideeffects were noted to be similar to those of visual discomfort by Wilkins, Peck,
and Jordon (1991). At a more general level, perceptual side-effects when reading have long been recognized in the reading literature (see e.g. Jordon, 1972;
Meares, 1980). This collection of symptoms have more recently been labelled
Meares–Irlen syndrome by Evans et al. (1996).
Two aspects of Wilkins’ investigation of visual discomfort that have
received scant attention are the way addition of the number of side-effects
reported are viewed as an index of pattern sensitivity and the range of performance difficulties that may be manifest with oversensitivity to pattern. As previously outlined, there is a report of one group of participants that report eye
strain and headache when reading (Wilkins & Nimmo-Smith, 1984), but no
systematic investigation of reading-induced symptomatology. On the other
hand, Meares–Irlen syndrome is identified using the Irlen Differential Perceptual Schedule (IDPS), which is an instrument that assesses somatic, perceptual,
and performance difficulties on a number of tasks, with a major focus on reading. Unfortunately, there is no published reliability and validity data on the
IDPS as a measuring instrument.
The initial aim of this paper is to report the development of a unidimensional
test scale to measure the visual anomaly based on Wilkins’ and Irlen’s conceptualizations of visual discomfort and Meares–Irlen syndrome. The instrument
is to be limited to predicting somatic, perceptual, and performance difficulties
when processing text, but should be applicable to processing difficulties with
other stimuli. If increasing reports of difficulty are an indicator of increasing
sensitivity the probability that an individual will experience visual discomfort
will increase as a greater number of positive responses are obtained to items
designed to measure this trait. Additionally, four experiments will be reported
that measure somatic and perceptual side-effects and performance difficulties
based on scores obtained on the measuring instrument, referred to as the Visual
Discomfort Scale. If the instrument measures both Wilkins’ and Irlen’s conceptualizations of this visual anomaly, those with higher scores should report
greater somatic and perceptual difficulty in pattern observation tasks and perform more poorly on a copying task and reading test than those with lower
scores.
VISUAL DISCOMFORT
641
DEVELOPMENT OF THE MEASUREMENT TOOL OF
VISUAL DISCOMFORT
Method
Participants . The sample included 518 volunteers. These were school
leavers applying for admission to the University of Wollongong in 1991, firstyear students enrolled in the Nursing Course at the University of Wollongong
in 1991, and a small number of students from Sydney metropolitan universities.
Ages ranged from 16 to 48 years.
Stimuli and Materials. Items in the questionnaire addressed possible
somatic, perceptual, and performance difficulties experienced with exposure to
different light sources or when reading. These were obtained from previous
investigations of Irlen (1983, 1991), Meares (1980), and Wilkins et al. (1984),
Additional items eliciting information concerning reports of the frequency of
experience of severe headache, maximum effective reading times, and experience of reading difficulties of a visual nature were included. All items used are
reported in Table 1.
Procedure. Four hundred copies of the questionnaire were posted to prospective students who applied for accommodation through the University of
Wollongong between November 1990 and January 1991. Of the 400 issued,
300 were returned. Participants used the instructions attached to the questionnaire as a guide to answering questions. For the nursing students and those from
Sydney metropolitan universities the questionnaire was administered individually or in a group setting.
Respondents used a 4-point continuous scale that extended from “Never” to
“it happens almost always”. The words “often” and “occasionally” separated
these extremes. “Occasionally” was classified as once or twice a year and
“often” as every few weeks. “Almost always” described the event occurring
frequently. Respondents were instructed to mark the most appropriate point
between the extremes on the scale (see Table 1). An equal distance separated
each of the sections of the scale and any response placed between two categories was scored in the category closest to the response. Half the responses
placed halfway between categories were coded in the lower category and the
other half in the higher category. This occurred in less than 5% of cases. Following examination of the response patterns, the scale was recoded into a 3point system because of the small number of responses in the “almost always”
category. A zero response indicated “it never occurs”, a 1-point response represented “occasional difficulty”, and a 2-point response represented a difficulty
that occurs “often to always”. The Rasch rating scale model was generated
using Titan, an interactive test analysis program (Adams & Toon, 1996).
TABLE 1
Items and Original Response Scale Used on the Visual Discomfort Scale
Response categories: Visual Discomfort Scale
0 = Event never occurs
1 = Occasionally. A couple of times a year
2 = Often. Every few weeks
3 = Almost always
|Never|.............................|Occasionally|..............................|Often|............................|Almost always|
Items used
(1) Do your eyes every feel watery, red, sore, strained, tired, dry, gritty, or do you rub them a lot,
when viewing a striped pattern?
(2) Do your eyes every feel watery, red, sore, strained, tired, dry or gritty, after you have been
reading a newspaper or magazine with clear print?
(3) Do your eyes every feel watery, red, sore, strained, tired, dry or gritty, when working under
fluorescent lights?
(4) How often do you get a headache when working under fluorescent light?
(5) Do you ever get a headache from reading a newspaper or magazine with clear print.
(6) When reading, do you ever unintentionally reread the same words in a line of text?
(7) Do you have to use a pencil or your finger to keep from losing your place when reading a page
of text in a novel or magazine?
(8) When reading do you ever unintentionally reread the same line?
(9) When reading do you ever have to squint to keep the words on a page of clear text from going
blurry or out of focus?
(10) When reading, do the words on a page of clear text ever appear to fade into the background
then reappear?
(11) Do the letters on a page of clear text ever go blurry when you are reading?
(12) Do the letters on a page ever appear as a double image when you are reading?
(13) When reading, do the words on the page ever begin to move or float?
(14) When reading, do you ever have difficulty keeping the words on the page of clear text in
focus?
(15) When you are reading a page that consists of black print on white letters, does the background
ever appear to overtake the letters making them hard to read?
(16) When reading black print on a white background, do you ever have to move the page around,
or continually blink to avoid glare which seems to come from the background?
(17) Do you ever have difficulty seeing more than one or two words on a line in focus?
(18) Do you ever have difficulty reading the words on a page because they begin to flicker or
shimmer?
(19) When reading under fluorescent lights or in bright sunlight, does the glare from bright white
glossy pages cause you to continually move the page around so that you can see the words clearly?
(20) Do you have to move your eyes around the page, or continually blink or rub your eyes to keep
the text easy to see when you are reading?
(21) Does the white background behind the text ever appear to move, flicker, or shimmer making
the letters hard to read?
(22) When reading, do the words or letters in the words ever appear to spread apart?
(23) As a result of any of the above difficulties, do you find reading a slow task?
642
VISUAL DISCOMFORT
643
Results and Discussion
A two-parameter Rasch rating scale measurement technique was used to generate a unidimensional logistic test scale. This model is based on simultaneous
estimation of item and person parameters, together with a discrimination index
for each item. The measurement scale produces a profile of visual discomfort
based on the response pattern to all items. If the model fits the data, participants
with high overall scores on the scale should produce a response profile based on
the item position ordered from least to most indicative of visual discomfort.
These measures are tested using reliability measures and t-tests based on the
weighted and unweighted residuals from the differences between observed and
expected scores (Adams & Toon, 1996). If the model fits the data, the t statistics
have a value close to zero and mean squares and standard deviation close to one
(Masters, Adams, and Wilson, 1990).
The logic behind this model is that items can be ordered in a cumulative fashion based on item difficulty or “intensity”. The number of positive responses to
each item is used as a measure of the item’s intensity, represented by the item
threshold. When a two-parameter model is generated, two thresholds are produced for each item. The first threshold occurs at a position on the scale where
there is a 50% chance that a score of zero or one will be obtained for the item.
The second threshold occurs at the point on the scale where there is a 50%
chance that the response to the item will be one or two. In both cases a score
above the threshold value is allocated to the higher response category. When
more than one response category is used, the difference in magnitude between
thresholds is equal for each (Masters et al., 1990). Item thresholds ordered from
lowest to highest are shown on the right-hand side of Figure 1.
To test the fit of individual items, each participant’s obtained and expected
score for each item is computed. The expected score is obtained from each participant’s total score on the scale. A low total logit score should produce positive responses for items with low thresholds, such as the performance difficulty
items. Negative responses should be obtained for items with high thresholds
such as those concerning perceptual difficulty. Item fit is estimated from the
difference between the observed and expected score for each item. The infit
mean square estimates were 0.97 and the t statistic –0.46. The outfit mean
square was 0.92 and outfit t statistics –0.53. This is a good fit to the model. The
standard deviations of the estimates were larger than expected with the infit
standard deviation 2.72 and the outfit standard deviation 2.25. The findings
demonstrate that the spread of the items produced greater variability than
expected. The estimate of internal consistency for all items was 0.91, an acceptable level of reliability (Sattler, 1986).
On the left-hand side of Figure 1 is a representation of the person parameter.
Each participant’s total score is ordered on the scale from lowest to highest. The
higher the obtained score the greater the probability of a positive response to a
FIG. 1. The Rasch Rating Scale Model. On the left-hand side of the figure is the person parameter. The
higher the score, the more likely observers are to report positive responses to items on the right-hand side
of the scale. On the right-hand side the thresholds are presented individually for all items. A suffix of 1
following the item number represents the threshold at which there is a 50% probability that the item will
be responded to with a 0 or 1. The suffix of 2 following the item number represents the point at which
there is a 50% probability that observers will respond with either a 1 or 2 to an individual item. The
higher the score appears on the scale the less likely it is that a positive response will be obtained to the
item.
644
VISUAL DISCOMFORT
645
greater number of items. For example, a participant with a scaled score of –1.0
logits has a 50% chance of responding zero or one to the item, “experience of
visual fatigue when reading”. For all items that appear below this item on the
scale a positive response is expected. In this case the participant may experience some performance difficulty and sensitivity to light but may not
experience severe visual discomfort. The overall fit statistics for the person
parameter were obtained by comparing each participant’s expected item
response pattern with their observed response pattern on the total scale. The
expected response pattern is obtained from each participant’s total logit score.
For the person parameter an acceptable fit to the model was obtained. The mean
squares for the estimates were 0.98 for the infit mean square and 0.92 for the
outfit mean square. The infit and outfits ts were both –0.07. The infit t standard
deviation was 1.20 and the outfit t standard deviation 0.92. This acceptable fit is
also demonstrated in the estimate of the person reliability index, which was
0.90 (Adams & Toon, 1996).
The discriminability of each item is estimated from the item mean squares. If
this is up to 30% above or below the expected value of one, the item produces an
adequate fit to the model (Adams & Toon, 1996). Three items have a poor fit.
Two of these received more positive responses than predicted. Both items concerned focusing difficulties when reading. The more frequent than expected
response pattern indicates that these items may be indicators of more generalized difficulties with refractive error rather than visual discomfort per se. The
item “the need to continually move the page around when working in bright
light in order to see the words clearly” received a smaller number of positive
responses than expected. The total score for the item predicted a greater proportion of positive responses than found. All three items we considered poor
discriminators of visual discomfort.
The test scale demonstrates that visual discomfort can be measured on a single dimension. As a participant’s score on the scale increases so does the probability of positive responses to items with higher thresholds. In this way
increasing visual discomfort is represented by an ordered expected response
pattern. The acceptable fit of the parameters demonstrates that this form of scaling may be the most accurate way to produce a testable model.
Wilkins et al. (1984) reported that individuals with a high headache susceptibility report a greater number of unpleasant side-effects from viewing a 3
c/deg square-wave than those with a low headache susceptibility. Additionally,
reports of somatic and perceptual difficulty from viewing a page of text have
been reported to produce short effective reading times (Irlen, 1991; Wilkins &
Nimmo-Smith, 1984). Irlen (1991) argued that these forms of visual difficulty
constitute a specific form of reading difficulty. The relationship between participants’ scores on the Visual Discomfort Scale and the previously discussed
measures was investigated using reports of frequency of severe headache,
646
CONLON ET AL.
reading difficulty induced from the type of overt manifestation outlined, and
maximum effective reading times.
Total and partial visual discomfort scores were computed for each participant. The total score was obtained by summing each participant’s score on the
items used to develop the scale. When obtaining the partial score, items
describing headache and unpleasant somatic side-effects were removed from
the total, leaving reports of performance and perceptual difficulty. This score
was generated to remove any influence of somatic difficulty when evaluating
the role of headaches and maximum effective reading times on reports of visual
discomfort. A square root transformation was performed on these scores to normalize the variances and obey the assumptions of the analyses conducted.
Individuals reporting no severe headache were separated from those reporting experience of severe headache. These scores were cross-classified with
those reporting no reading difficulty, occasional reading difficulty, and regular
reading difficulty. Two factorial analyses of variance revealed a significant
main effect for severe headache when the total, F(1, 508) = 17.52, p < .05, and
partial, F(1, 509) = 10.27, p < .05, visual discomfort scores were used. Significant differences were also found for reading difficulty with the total, F(2,
508) = 127.9, p < .05, and partial F(2, 509) = 128.44, p < .05, scores. No interaction was obtained for either score. Participants reporting experience of severe
headache, or substantial reading difficulty, are likely to have high scores on the
Visual Discomfort Scale (Table 2). This finding supports previous findings
(Marcus & Soso, 1989; Wilkins et al., 1984) that individuals susceptible to
severe headache are likely to report symptoms of visual discomfort.
Further analyses investigated the effect of the amount of time the individual
could effectively read without taking a break on visual discomfort scores.
TABLE 2
Analysis of Reports of Severe Headache by Reports of Visually
Induced Reading Difficulty for the Total Visual Discomfort Score
and the Partial Visual Discomfort Score (N = 515)
Variable
No headache
No reading difficulty
Some reading difficulty
Regular reading difficulty
Severe headache
No reading difficulty
Some reading difficulty
Regular reading difficulty
N
Visual Discomfort
————————————–
Total Score Partial Score
—————
——————
Mean (SD)
Mean (SD)
252
105
31
6.35 (1.2)
15.36 (1.0)
25.9 (1.0)
4.7 (1.2)
12.32 (0.8)
21.52 (0.6)
69
34
23
10.49 (1.0)
19.27 (1.2)
30.91 (0.5)
7.67 (1.0)
15.13 (1.0)
24.01 (0.5)
VISUAL DISCOMFORT
647
Those reporting maximum reading times of less than 1 hour, of reading times
between 1 and 2 hours, and of reading times of 2 hours or more were separated
into three groups. The analyses for the total, F(2, 510) = 117.26, p < .05), and
partial, F(2, 511) = 111.75, p < .05), visual discomfort scores revealed that the
three groups differed significantly from one another (Table 3). The shorter the
effective reading time the higher the mean group score on the Visual Discomfort Scale when items concerning experience of somatic difficulty and headache were included or excluded from the analyses. These results demonstrate
that subjective reports of effective reading time can provide an indication of
visual difficulty when reading, as previously reported by Irlen (1991) as a
symptom of Meares–Irlen syndrome.
Together with the subjective reports of reading difficulty and headache
behaviour shown previously, a relationship between scores on the Visual Discomfort Scale and reports of unpleasant perceptual and somatic side-effects
induced from prolonged pattern viewing was expected. This was investigated
in Experiment 1. Reports of specific types of unpleasant somatic and perceptual
side-effects induced from exposure to a 2.5c/deg square-wave were obtained.
The 2.5c/deg square-wave was used as this is in the range of greatest sensitivity,
as reported previously by Wilkins (1995). It was predicted that those with
higher scores on the Visual Discomfort Scale would report a greater number of
unpleasant side-effects from the pattern observation task.
EXPERIMENT 1
Method
Participants. One hundred and seventy-seven naive volunteers, consisting
of University students, staff, or visitors to the University of Wollongong on
Open Day participated. Any questionnaire respondent with a history of epilepsy or with knowledge of an uncorrected visual anomaly was excluded from
the study. All participants were informed of the possibility that unpleasant
TABLE 3
Analysis of Reports of Effective Reading Times for the Total Visual
Discomfort Score and the Partial Visual Discomfort Score (N = 515)
Maximum Effective Reading Time
2 hours +
1 to 2 hours
Less than hour
N
278
77
158
Visual Discomfort
—————————————–
Total Score
Partial Score
—————
——————
Mean (SD)
Mean (SD)
6.81 (1.2)
13.69 (1.2)
19.36 (1.7)
5.29 (1.2)
10.37 (1.0)
15.29 (1.4)
648
CONLON ET AL.
somatic side-effects may occur. It was emphasized that if difficulties arose the
observation task was to be terminated.
Stimuli and Materials. A 2.5c/deg horizontal square-wave grating with a
Michelson contrast of 0.8 was used as the observation stimulus. Dimensions of
the pattern were 32cm wide by 25cm high. This subtended 15.6° horizontally
and 12.3° vertically. A response sheet containing possible illusory features and
unpleasant somatic side-effects was used. The methodology employed was
similar to that of Wilkins et al. (1984), where measures of different types of illusory features induced were obtained with a checklist of illusions. Participants
indicated which of the following perceptual effects occurred: “blurring, shimmering, flickering, bending of lines, appearance of patterns and colour (red,
yellow, green, blue)” (Wilkins et al., 1984, p. 993). Unpleasant somatic sideeffects of tired eyes, feelings of nausea, dizziness and headache were also
included. The checklist used is shown in Table 4.
Procedure. Participants scanned the horizontally presented 2.5c/deg high
contrast square-wave for 60sec. Scanning was used to avoid formation of afterimages. The viewing distance of 114cm was controlled with a chin rest. Room
2
lighting was maintained at a photopic luminance level of 150cd/m .
Reports of unpleasant somatic side-effects and perceived illusory features
induced while viewing the pattern were obtained. Observed side-effects were
recorded on the checklist. Side-effects perceived but not included on the
TABLE 4
Checklist for Reports of Illusions and Unpleasant Somatic Side-Effects, Experiment 1
(N = 179)
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
Colour
Red blue Green yellow
Movement
Swirling movement
Shimmer
Thicker lines at a different angle
Thinner lines at a different angle
Please describe the dominant pattern that you see ____________________________________
Don’t see any pattern
Lattice*
Thin lines at same angle*
Thick lines at same angle*
Makes me feel dizzy
Makes me feel nauseous
Makes my eyes feel tired
Makes my head ache
Causes no difficulty
*Patterns added following inspection of written descriptions.
VISUAL DISCOMFORT
649
checklist were added and any pattern generated was drawn. On completion of
this part of the task the experimenter viewed each participant’s response sheet
and, if this was nuclear, further explanation was elicited.
A response was scored as one if the effect was reported. On the basis of
Wilkins et al’s. (1984) index of severity, two variables were generated: “pattern
sensitivity” was composed of all reported perceptual illusions and “somatic
strain” included all unpleasant somatic side-effects. Together these variables
provide an overall measure of sensitivity.
Participants were separated into groups on the basis of scores obtained on
the Visual Discomfort Scale. The low visual discomfort group contained all
participants with a score of less than 35%, the moderate visual discomfort
group consisted of those with a score between 35% and 69.5%, and the severe
visual discomfort group obtained a score of 70% or greater on the scale. All
analyses obeyed the assumptions of the analysis of variance and were conducted using SPSS.
Results and Discussion
The most frequently reported side-effects were movement, shimmer, the
appearance of colour, and tired eyes. Spatial illusions, headache, and dizziness
were reported least frequently.
Analysis of the variable “pattern sensitivity” demonstrated that there was a
significant difference between groups on scores of the Visual Discomfort
Scale, (F(2, 174) = 3.631, p < .05. Tukey’s post hoc analysis revealed that the
number of perceptual illusions reported for the moderate and severe visual discomfort groups did not differ from one another but was significantly greater
than the number reported for the low visual discomfort group (Table 5). A significant between-groups difference was also found for reports of “somatic
strain”, (F(2, 174) = 9.45, p < .05. Post-hoc analyses revealed that no differences were found in the number of effects reported by the moderate and severe
visual discomfort groups, but both groups reported significantly more symptoms of “somatic strain” than the low visual discomfort group. These findings
support the prediction that higher scorers on the Visual Discomfort Scale report
a greater number of unpleasant side-effects from viewing the high contrast
2.5c/deg square-wave.
The correlations between individual scores on the Visual Discomfort Scale
2
and the number of perceptual illusions was 0.19, R = 3.6%, F(1, 175) = 6.41,
2
p < .05, and the number of somatic side-effects reported was 0.35, R =
12.25%, F(1, 175) = 24.51, p < .05. These results demonstrate that subjective
reports of either type of side-effect on their own may not be a good indicator of
visual discomfort as measured by this scale. These findings may result from the
prolonged nature of the task. Wilkins et al. (1984) used an observation period of
about 5–10sec. This study used 60sec, long enough for the threshold of
650
CONLON ET AL.
TABLE 5
Number of Unpleasant Somatic and Perceptual Side-effects
Reported From Viewing a 2.5c/deg Square-wave Pattern in
Experiment 1 (N = 177)
Visual Discomfort Groups
————————————————————————–
Low (n = 94)
Moderate (n = 62) Severe (n = 21)
Mean (SD)
Mean (SD)
Mean (SD)
Total illusions
Somatic difficulty
3.16 (0.17)
0.80 (0.08)
3.73* (0.20)
1.27* (0.12)
4.00* (0.30)
1.48* (0.13)
*Groups that did not differ from one another but differed from the low Visual
Discomfort group.
stimulation for induction of illusory side-effects and somatic difficulty to be
exceeded for many less sensitive participants. Additionally, it is unclear that
the number of side-effects of both types reported is comparable to a measure of
magnitude or severity of difficulty. The subjective nature of the task may have
produced difficulty in delineating individual illusory features; for example,
some participants may have reported experience of flicker and shimmer, with
others reporting only one effect.
A further pattern observation task was conducted in Experiment 2. Using a
rating scale, the magnitude of somatic and perceptual difficulty induced from
short exposures to a number of striped repetitive patterns was obtained.
EXPERIMENT 2
The stimuli used were square-wave patterns with spatial frequencies between 1
and 16c/deg, and a letter pattern. In line with Wilkins’ earlier findings it was
predicted that the moderate to severe visual discomfort group would rate 2–4c/
deg patterns and the letter pattern as more somatically and perceptually
unpleasant to view than the low visual discomfort group. Additionally, there
would be no between group differences in reports of either type of unpleasant
side-effects at higher or lower spatial frequencies.
Method
Participants. There were 11 naive volunteers with normal or corrected to
normal visual acuity in each of two groups. Groups were classified on the basis
of scores obtained on the Visual Discomfort Scale. The low visual discomfort
group consisted of participants scoring less than 35%, and the moderate to
severe visual discomfort group consisted of those scoring in excess of 50%.
Mean score for the low visual discomfort group was 19.18% (SD = 10.26) and
for the moderate to high visual discomfort group 76.64% (SD = 18.45).
VISUAL DISCOMFORT
651
Stimuli and Apparatus. The square-wave patterns were 1, 2, 4, 8, 12, and
16c/deg square-wave gratings with a Michelson contrast of 0.8. The striped
patterns measured 8cm vertically and horizontally, each subtending a visual
angle of 4°. The letter pattern (shown in Figure 2) had the same physical dimensions as the 1c/deg square-wave grating with the dark bar replaced by a series of
the capital letter “E”, presented in its four possible orthogonal orientations from
vertical. Each of these patterns was centred on a white card measuring
15 × 10cm. These patterns were smaller than that used in Experiment 1 and
more similar to those used in Wilkins et al. (1984, Exp. 1).
The stimuli were mounted on a magnetic white board and placed at a viewing distance of 114cm. Care was taken to ensure that patterns of similar spatial
frequency were not placed together. Arranging presentation location was done
in a quasi-random fashion for each participant. The study was conducted in a
windowless room with fluorescent lights on at all times.
Procedure. Participants were instructed to scan each of the patterns for
3sec. Each stimulus pattern was ranked for the magnitude of induced somatic
difficulty and perceptual distortion on a rating scale between 0 and 5. A rating
of 0 indicated no difficulty, and a rating of 5 difficulty so great that the pattern
was too unpleasant or distorted to view. Ratings between these points were
indicative of increasing difficulty. For example, a rating of 3 for distortion ratings indicated that the pattern was beginning to move and colour could be seen.
When a rating of 3 was make for somatic difficulty ratings the participant
reported that they felt some discomfort when attempting to focus. The scale
FIG. 2.
Letter pattern used as an observation stimulus in Experiment 2.
652
CONLON ET AL.
was continuous and the rating was obtained by measuring the actual distance
from 0 on the scale. Reports of the predominant illusory features induced from
the most distorted patterns were also obtained.
Results and Discussion
Analysis of all stimulus patterns revealed a significant main effect for group,
F(1, 20) = 20.865, p < .05. The moderate to severe visual discomfort group
rated all spatial frequencies and the letter pattern as more somatically unpleasant and perceptually distorted to view than the low visual discomfort group.
There was no effect of rating type, perceptual or somatic, but there was a significant main effect for spatial frequency/letter pattern, F(6, 120) = 44.24, p <
.05. No significant interactions were found. Inspection of Figure 3 shows that
the groups responded similarly to all the stimulus patterns, but differed in the
severity of response.
A priori contrast analysis combined the ratings of somatic and perceptual
difficulty and different groups to investigate differences across spatial frequencies. As spatial frequency increased from 1 to 12c/deg, ratings of unpleasant
somatic and perceptual side-effects also increased. Significant differences
were found between 1 and 2c/deg, F(1, 21) = 18.42, p < .05, 2 and 4c/deg, F(1,
21) = 44.77, p < .05, 4 and 8c/deg, F(1, 21) = 25.53, p < .05, and 8 and 12c/
deg, F(1, 21) = 9.92, p < .05. No differences were found between 12 and 16c/
deg. When the letter pattern was considered, it was rated as more unpleasant
than the 1c/deg pattern, F(1, 21) = 6.74, p < .05, and significantly less unpleasant than all the spatial frequencies above 2c/deg.
Relationships between individual scores on the Visual Discomfort Scale and
ratings of distortion for the individual patterns produced significant moderate
or strong correlations (Table 6). The total distortion rating was computed by
obtaining the mean of each participant’s ratings of somantic and perceptual
distortion.
These findings demonstrate that the moderate to severe visual discomfort
group experience greater somatic and perceptual difficulty with exposure to
repetitive striped patterns across the range of spatial frequencies tested, and for
the letter pattern. As the magnitude of reported difficulty increases, so do scores
on the Visual Discomfort Scale. Although there may be some effect of
TABLE 6
Correlations between Scores on the Visual Discomfort Scale and Overall Ratings of
Distortion (N = 22)
Spatial Frequency (c/deg)
Correlation
2
R
1
2
4
8
12
16
Letter
0.590
35%
0.750
56%
0.773
59.7%
0.675
45.6%
0.712
50.7%
0.636
40.4%
0.700
49%
All correlations were significant at .05.
VISUAL DISCOMFORT
653
FIG. 3. Experiment 2: Ratings of distortion for the moderate to severe visual discomfort group and
low visual discomfort group for the different spatial frequencies and letter pattern (N = 22).
suggestibility due the nature of visual discomfort these findings cannot be
entirely attributed to suggestibility from completing the Visual Discomfort
Scale, as items do not address issues concerning striped repetitive patterns, but
distortion generated from a text page. Experiments 3 and 4, which are performance task, address this issue.
Wilkins et al. (1984) found that individuals with a high headache susceptibility reported a greater total number of unpleasant side-effects than individuals with a low headache susceptibility from 5 to 10sec exposures to a 3c/deg
square-wave. With the 3sec exposure duration used in this study, ratings of
greater difficulty were obtained at all spatial frequencies, and for the letter
pattern for the moderate to severe visual discomfort group.
654
CONLON ET AL.
In this experiment both groups reported greatest distortion at the highest spatial frequencies tested. This finding does not replicate earlier reports of greatest
distortion experienced at 3c/deg. Two explanations are tenable here. First,
Wilkins et al. (1984) performed a paired comparisons task to directly compare
unpleasantness at individual spatial frequencies in a control group. That methodology, which found the 3c/deg pattern was most unpleasant, directly compared spatial frequencies, whereas this study investigated each spatial
frequency separately. (A study with a methodology similar to Wilkins’ is currently being undertaken in this laboratory.) Second, a previous study compared
groups of participants with high and low headache susceptibility on frequency
of illusory features reported with exposure to spatial frequencies up to 16c/deg.
That study found no differences between the groups on reports of effects above
3c/deg; however, inspection of the data reveals that both groups reported a substantial number of effects in the higher spatial frequency range (see Wilkins et
al., 1984, p. 997). That study used either reports of no illusions, or one, or
greater to determine group differences, not the magnitude of difficulty which
was the focus of this study. The magnitude of effects may simply be greater for
the moderate to severe visual discomfort group across the spatial frequency
range.
Additional support for greater perceptual difficulty experienced by the moderate to severe visual discomfort group was found in the types of perceptual
side-effects reported. These included complaints of wavy movement, lines
merging, blur and focus difficulties, temporal instability, and figure–ground
difficulty. Some illusory features of pattern, the same as the orientation illusions reported to occur at a cortical level (Georgeson, 1976), were reported
across the spatial frequency range. This subjective finding supports the notion
that the threshold of stimulation of induction of illusory features was exceeded
for the moderate to severe visual discomfort group. The low visual discomfort
group predominantly reported effects of wavy movement, lines merging, blur
and focus difficulties at 12 and 16c/deg, with few reports of perceptual difficulty at lower spatial frequencies or for the letter pattern. No illusions of pattern
were reported in this group.
Support for the text page producing unpleasant side-effects was found with
the moderate to severe visual discomfort group reporting greater difficulty
when viewing the letter pattern. This adds support to Wilkins and NimmoSmith’s (1987) earlier findings. Although greater difficulty was reported in
Experiments 1 and 2 by those with moderate to severe symptoms of visual discomfort, both studies have relied on subjective judgements of difficulty. In the
following two experiments, performance measures were obtained to determine
if objective, as well as subjective differences occur between groups on the basis
of scores on the Visual Discomfort Scale.
VISUAL DISCOMFORT
655
EXPERIMENT 3
In this experiment the Coding B (Digit symbol) sub-test from the Wechsler
Intelligence Scale for Children—Revised (WISC-R) was used. In this task participants must correctly identify the relevant symbol and provide a visual representation of that symbol by drawing it. The grid pattern within which the
symbols must be placed forms a square-wave-like pattern that may induce
somatic and perceptual difficulty in participants experiencing visual discomfort. If sensitivity to pattern produces poorer task efficiency, performance
should be poorer with increasing scores on the Visual Discomfort Scale. This
test is generally used as a measure of attention and concentration (Kaufman,
1979), which are the difficulties Irlen (1991) argues occur in individuals with
Meares–Irlen syndrome.
Method
Participants . There were 39 naive University student volunteers with normal or corrected to normal visual acuity. There were 13 participants in each of
the low, moderate, and severe visual discomfort group. Groups were classified
as in Experiment 1.
Stimuli. The Coding B sub-test from the WISC-R with black symbols presented on a white background.
Procedure. Participants were tested individually using the standardized
instructions from the WISC-R (Wechsler, 1974, p. 100–101). Several participants from the severe visual discomfort group reported some perceptual distortion from the heavy black lines from which the boxes were constructed. They
also complained of easily losing their place on the page. Three participants
from this group used their finger as an aid to minimize the difficulty.
Responses were scored using the WISC-R instructions. An item was classified as correct if it was clearly identifiable, even if corrected or imperfectly
drawn (Wechsler, 1974). For some participants in the low visual discomfort
group the task was completed in less than the allotted time. In this case the highest score of 93 was assigned. Conversion to standardized scores was not undertaken, as the aim was to obtain information on absolute differences in scores on
the speeded task. All analyses satisfied the relevant assumptions.
Results and Discussion
The analysis of variance produced a significant difference between groups,
F(2, 36) = 42.20, p < .05. The performance score of the low visual discomfort
group (X = 86.69) was significantly higher than that of the moderate visual
656
CONLON ET AL.
discomfort group (X = 72.2), F(1, 36) = 3.37, p < .05, whose performance was
significantly better than that of the severe visual discomfort group (X = 64.15),
F(1, 36) = 11.04, p < .05. A bivariate correlation analysis produced a significant negative linear relationship between scores on the performance measure
and total scores on the Visual Discomfort Scale, F(1, 37) = 82.408, p < .05. A
Pearson’s correlation, r, of – 0.83 accounted for 69% of the variance. This relationship is shown in Figure 4. The power coefficient was 0.99 at the 0.05 level
of significance (Howell, 1996) demonstrating the robust nature of the
relationship.
The results of this experiment show, that performance efficiency is reduced
with increasing scores on the Visual Discomfort Scale for this task. Using a pattern with a repetitive global structure, the interference induced may have
reduced participants’ ability to attend to the relevant aspects of the display,
reducing performance efficiency. This task is generally considered to measure
attention and concentration and when WISC-R items are factor analysed, the
task is considered to be a factor describing general problems with
distractability (Kaufman, 1979; Sattler, 1986). These are skills that Irlen (1991)
FIG. 4. Experiment 3: Scatter diagram for total score for the copying task by total score on the Visual
Discomfort Scale (N = 39).
VISUAL DISCOMFORT
657
argues are poor in those reporting somatic and perceptual difficulties with reading. This experiment supports these observations in an objective fashion.
This task, however, differs from reading. Identification of single elements
from a number of elements has been the focus. Attaching meaning to the stimulus has not been required. Meares–Irlen syndrome was initially associated with
difficulty reading (Irlen, 1993; Whiting, 1985). Reading performance of individuals experiencing visual discomfort has not been assessed. Experiment 4
obtained a measure of reading rate and comprehension for low, moderate, and
severe visual discomfort groups.
EXPERIMENT 4
The Neale Analysis of reading ability is a standardized test of oral reading rate,
accuracy, and comprehension (Neale, 1988). The extension passage was developed as part of the diagnostic tutor, as an assessment tool for mature readers.
The linguistic complexity of the passage has been matched to that of material
used in later secondary school and tertiary education (Neale, 1988). This part of
the test has not been standardized. It is predicted that the reading rate of the
severe visual discomfort group will be slower than that of other groups. If the
visual inefficiency induced in participants experiencing severe visual discomfort reduces the capacity of participants to cognitively process the text passage,
comprehension for this group should also be reduced. The moderate visual discomfort group should also perform the task with less speed and poorer comprehension than the low visual discomfort group.
Method
Participants . Participants were 24 naive University student volunteers
with normal or corrected to normal visual acuity who had successfully completed at least one year of university study. Them were eight participants in
each of the low, moderate, and severe visual discomfort groups who were classified into groups on the basis of scores on the Visual Discomfort Scale, as previously specified.
Materials. The extension passage from the Neale Test of Reading Ability
was used. This passage is 327 words in length, spread over two pages. The text
is double spaced and presented as black print on a bright white glossy background. Eight comprehension questions followed reading of the text.
Procedure. Testing was conducted individually. An explanation of the
test was provided with participants being informed that they would be required
to read orally a passage of text and that it would be timed. It was emphasized
to all participants that understanding the context of the passage was more
658
CONLON ET AL.
important than reading speed, as a number of comprehension questions would
be asked immediately following completion of the passage.
Reading rate in words per minute was obtained for all participants. The total
number of comprehension questions answered correctly was converted to
percentage correct scores. All analyses satisfied the required statistical
assumptions.
Results and Discussion
A one-way analysis of variance showed that, as predicted, reading rate differed
between groups, F(2, 21) = 13.708, p < .05. No significant differences were
found between the low (X = 143.9 words/minute) and moderate (X = 127.8
words/minute) visual discomfort groups. The combined low and moderatevisual discomfort group read the text passage significantly more quickly than
the severe (X = 92.42 words/minute) visual discomfort group, F(1, 21) =
25.049, p < .05. The faster reading speed of the low and moderate visual discomfort groups cannot be attributed to a speed comprehension trade-off as no
significant differences were found between groups on this measure, with all
groups scoring about 50% of questions correctly. This result demonstrates that
reading efficiency is reduced for the severe visual discomfort group. Although
comprehension is the same as that found for other groups the severe visual discomfort group require longer reading time to attain the same level of understanding of the text.
A bivariate correlation analysis revealed a significant moderate positive
relationship between reading rate and visual discomfort score, F(1, 22) =
22.067, p < .05. Pearson’s correlation (r) of .71 accounted for 50% of the variance. This is shown in Figure 5. The robust nature of the relationship is
reflected in its power coefficient of .93 at the .05 level of significance (Howell,
1997). Inspection of Figure 5 shows that as a group the moderate visual discomfort group read a little faster than expected and the severe visual discomfort
group a little slower than expected. This finding is similar to that found in a
letter identification task reported previously (Conlon et al., 1998). There it was
argued that the moderate visual discomfort group experienced some somatic
difficulty when performing short duration tasks but this does not extend to
performance. The severe visual discomfort group experience somatic and
performance difficulty. Figure 5 also shows this performance difficulty.
Tinker (1963) reported that reading speed is a measure of the legibility of the
text. In this case the interference induced by the global pattern may have
reduced legibility, and as a result increased reading time for the severe visual
discomfort group. A negative effect on comprehension was not found suggesting that poorer reading efficiency is being measured in the severe visual discomfort group, not a form of visual reading disability, which has previously
been suggested (Irlen, 1983). When the moderate visual discomfort group is
VISUAL DISCOMFORT
659
FIG. 5. Experiment 4: Reading rate in words per minute by total score on the Visual Discomfort Scale
(N = 24).
considered, the text format, been double spaced, may have been insufficient to
produce enough discomfort to produce poorer performance. Wilkins et al.
(1984) have shown that increasing the duty cycle, which is the same as double
spacing the text, results in a reduction in reports of unpleasant side-effects. This
manipulation may have been sufficient to reduce difficulty in the moderate
group for this task.
GENERAL DISCUSSION
Visual discomfort has been described as a collection of unpleasant somatic and
perceptual side-effects resulting from extreme sensitivity to some forms of
bright light and repetitive patterns (Wilkins et al., 1984). This study bus shown
that using a Rasch rating scale technique visual discomfort can be measured on
a continuum, with increasing scores indicative of more severe visual discomfort. The validity of this scale has been demonstrated by showing that increasing reports of unpleasant subjective side-effects from viewing repetitive striped
patterns (e.g. Wilkins, 1995) and reduced efficiency when reading (Irlen, 1991)
660
CONLON ET AL.
and performing a complex cognitive task are related to increasing symptoms of
visual discomfort. Three important issues arise from these findings. These are:
(1) the relationship between reports of unpleasant side-effects from viewing
striped repetitive patterns and repetitive patterns containing cognitively meaningful stimuli such as a text page (2) how subjective reports of difficulty we
related to poorer performance; and (3) the role of factors such as general ability
and suggestibility in inducing greater reports of difficulty and poorer
performance.
The similarity of the spatial characteristics of the square-wave, a repetitive
text page, and the illusory characteristics induced by them supports Wilkins
and Nimmo-Smith’s (1984, 1987) hypothesis that perceptual illusions and
unpleasant somatic side-effects are induced from the global spatial frequency
characteristics of the viewed pattern. In Experiment 2, higher scorers on the
Visual Discomfort Scale reported more difficulty when viewing patterns of all
spatial frequencies and the letter pattern than those with low scores. In a previous paper (Conlon et al., 1998) it was argued that this form of pattern sensitivity
acts to distract attention from the salient aspects of the visual scene. This would
result in a reduction in visual attention to the local or letter components of the
pattern and is induced from the unpleasant somatic and perceptual side-effects
from the repetitive striped pattern.
The extent to which this form of pattern sensitivity produces performance
difficulty is related to the degree of individual sensitivity experienced and the
cognitive demands of visual processing tasks. This was shown initially with
participants reporting shorter effective reading times and reading difficulties of
a visual nature as scores on the Visual Discomfort Scale increased. From an
objective point of view, in Experiments 3 and 4 it was demonstrated that the
severe visual discomfort group performed with less efficiency than the other
groups. The moderate visual discomfort group performed with less efficiency
than the low visual discomfort group in Experiment 3 only, where a greater
combination of cognitive skills were required for efficient performance. Cognitive difficulty, for example, difficulty combining salient features in an appropriate way when attention is overloaded or distracted (Treisman, 1982) or a
reduction in the efficiency of short-term memory may occur because of the
greater cognitive processing effort required to process basic visual information
(Oakhill & Garnham, 1988) This notion is supported on the one hand by a study
of Dickinson and Rabbitt (1991) where reading comprehension was measured
for distorted text. In that study comprehension was reduced with more difficult
text. In parallel, greater somatic and perceptual difficulty experienced by the
severe visual discomfort group can explain reduced performance efficiency in
both performance studies undertaken in this work. On the other hand the moderate visual discomfort group, who reported less severe somatic and perceptual
difficulty may require higher task demands before performance difficulty
occurs. The copying task required greater visual attention than the reading task,
VISUAL DISCOMFORT
661
where comprehension cues were available, accounting for the poorer performance of that group in Experiment 3. Increasing the cognitive difficulty of the
text passage or the unpleasantness of the pattern, by using single-spaced text
may increase the problems in this group.
In addition to the reduction in comprehension found with increasing task difficulty, Dickinson and Rabbitt (1991) reported that text distortion had a smaller
effect on individuals of higher general ability. It is possible that higher scorers
on the Visual Discomfort Scale may simply have a lower general ability than
those with lower scores, accounting for the slower reading rate and poorer
copying ability. The general ability and performance scores of three members
of the severe visual discomfort group are presented in Table 7. All have average
or above average general ability as measured by the WISC-R, showing that
poorer performance on the above tasks cannot be explained by lower overall
general ability scores. Interestingly two participants had significantly lower
performance IQ than verbal IQ. This finding suggests that these individuals did
not have a specific reading disability either, giving greater support to the notion
that high scorers on the Visual Discomfort Scale perform visual tasks less efficiently. The reading rate and scores on the copying task for these participant’s
are substantially slower than those obtained for the same tasks for those in the
low visual discomfort group.
The role of suggestibility cannot be completely ruled out of reports of
unpleasant side-effects obtained in Experiment 1. The finding that performance difficulties corresponds to these subjective reports for those with high
scores on the Visual Discomfort Scale gives greater credibility to those results.
Further research will investigate the nature of basic visual processing differences that occur from exposure to these square-wave-like repetitive patterns
and will be performed to shed light on the origin of visual discomfort.
TABLE 7
Individual General Ability Data and Scores on Performance Subtests for Three High Scorers on the Visual Discomfort Scale
General Ability
——————————
Verbal Performance
(%)
(%)
104
140
120
115
135
101
Reading Rate
(Words per minute)
Digit Symbol
——————
Total/93
Visual
Discomfort
Score
109
66
78
59
60
59
91
99
96
662
CONLON ET AL.
REFERENCES
Adams, R.J., & Toon, K.S. (1996). Quest: The interactive test analysis system version 2.1 [Computer program]. Camberwell, Victoria, Australia: Australian Council for Educational Research.
Conlon, E.G., Hine, T., Lovegrove, W.J., Chekaluk, E., Piatek, K., & Hayes-Williams,
K. (1998). The influence of visual discomfort and pattern structure on attention in a visual
search task. Perception, 27, 21–33.
Dealy, R.S., & Tolhurst, D.J. (1972). Is spatial adaptation an after-effectof prolonged inhibition.
Journal of Physiology, 242, 261–270.
Dickinson, C.M., & Rabbitt, P.M.A. (1991). Simulated visual impairment: Effects on text comprehension and reading speed. Clinical Visual Sciences, 4, 301–308.
Evans, B.J.W., Wilkins, A.J., Brown, J., Busby, A., Wingfield, A., Jeanes, R., & Bald,
J. (1996). A preliminary investigation into the aetiology of Meares-Irlen syndrome. Ophthalmic and Physiological Optics, 16, 286–296.
Georgeson, M.A. (1976). Psychophysical hallucinations of orientation and spatial frequency.
Perception, 5, 99–111.
Georgeson, M.A. (1980). The perceived spatial frequency, contrast and orientation of illusory
gratings. Perception, 9, 695–712.
Georgeson, M.A. (1985). Inferring cortical organisation from subjective visual patterns. In J.
Rose & V.G. Dobson (Eds), Models of the visual cortex (pp. 223–232). New York: John Wiley
& Sons.
Hannell, G., Gole, G.A., Dibden, G.A., Rooney, K.F., Pidgeon, K.J., & McGlinchey, N.D.
(1989). Reading improvement with tinted lenses: A report of two cases. Clinical & Experimental Optometry, 72(5), 170–176.
Howell, D.C. (1997). Statistical methods for psychology. Boston: PWS Publishing.
Irlen, H. (1983). Successful treatment of learning disabilities.Paper presented at the 91st annual
convention of the American Psychological Association, Anaheim, California.
Irlen, H. (1991). Reading by colors. New York: Avery Publishing Co.
Jordon, D.R. (1972). Dyslexia in the classroom. Chicago: Bell & Howell.
Kaufman, A.S. (1979). Intelligence testing with the Wisc-R. New York: Wiley.
Marcus, D.A., & Soso, M.J. (1989). Migraine and striped induced visual discomfort.Archives in
Neurology, 46, 1129–1132.
Masters, G.N., Adams, R.A., & Wilson, M. (1990). Charting student progress. In T. Husen &
T.N. Postlewaite (Eds), International encyclopaedia of education: Research and studies
(Suppl. Vol. 2). Oxford, UK: Pergamon Press.
Meares, O. (1980). Figure/ground, brightness, contrast and reading disabilities. Visible Language, XIV, 13–29.
Meldrum, B.S., & Wilkins, A.J. (1984). Photosensitive epilepsy in man and the baboon: Integration of pharmacological and psychophysical evidence. In P.A. Schwartzfroin & H.V. Wheal
(Eds), Electrophysiology of epilepsy. London: Academic Press.
Neale, M.J. (1988). Neale analysis of reading ability—revised. Camberwell, Victoria, Australia: Australian Council for Educational Research.
Oakhill, J., & Garnham, A. (1988). Becoming a skilled reader. Oxford, UK: Basil Blackwell.
Sattler, J.M. (1986). Assessment of children (3rd edn.). San Diego, CA: Jerome Sattler.
Suter, S., Armstrong, C.A., Suter, P.S., & Powers, J.C. (1991). Spatial frequency tuned attenuation and enhancement of the steady-state VEP by grating adaptation. Vision Research, 31,
1167–1175.
Tinker, M.A. (1963). Legibility of print. Iowa City, IA: Iowa State University Press.
Tolhurst, D.J., & Barfield. (1978). Interactions between spatial frequency channels. Vision Research, 18, 951–958.
VISUAL DISCOMFORT
663
Treisman, A. (1982). Perceptual grouping and attention in visual search for features and for
objects. Journal of Experimental Psychology: Human Perception and Performance, 8, 194–214.
Wechsler, D. (1974). Wechsler intelligence scale for children—revised. New York: Psychological Corporation.
Whiting, P. (1985). How difficult can reading be? New insight into reading problems. Journal of
English Teacher’s Association, 49, 49–55.
Whiting, P.R. (1988). Improvement in reading and other skills, using Irlen coloured lenses. Australian Journal of Remedial Education, 20, 49–55.
Wilkins, A.J. (1986). What is visual discomfort? Trends in Neuroscience, 9, 343–346.
Wilkins, A.J. (1995). Visual stress. Oxford, UK: Oxford University Press.
Wilkins, A.J., Binnie, C.D., & Darby, C.E. (1980). Visually induced seizures. Progress in
Neurobiology, 15, 85–117.
Wilkins, A.J., & Nimmo-Smith, J. (1984). On the reduction of eye-strain when reading. Ophthalmology and Physiology Optics, 4, 53–59.
Wilkins, A.J., & Nimmo-Smith, J. (1987). The clarity and comfort of printed text. Ergonomics,
30, 1705–1720.
Wilkins, A.J., Nimmo-Smith, I., Tait, A., McManus, C., Sala, S.G., Tilley, A., Arnold, K., Barrie,
M., & Scotts, S. (1984). A neurological basis for visual discomfort. Brain, 107, 989–1017.
Wilkins, A.J., Peck, A., & Jordon, B. (1991). Visual discomfort in the classroom. Child Language: Teaching and Therapy, 7, 326–330.
Manuscript received 1 October 1997
Revised manuscript received 22 July 1998