c o r t e x x x x ( 2 0 1 1 ) 1 e6
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/cortex
Note
The visual attention span deficit in dyslexia is
visual and not verbal
Muriel Lobier a,*, Rachel Zoubrinetzky a,b and Sylviane Valdois a,c
a
Laboratoire de Psychologie et Neurocognition (UMR 5105 CNRS), Université Pierre Mendès France, Grenoble, France
Centre de Diagnostic des troubles du langage et des apprentissages, CHU de Grenoble, France
c
Centre National de la Recherche Scientifique (CNRS), France
b
article info
abstract
Article history:
The visual attention (VA) span deficit hypothesis of dyslexia posits that letter string deficits are
Received 2 February 2011
a consequence of impaired visual processing. Alternatively, some have interpreted this deficit
Reviewed 20 June 2011
as resulting from a visual-to-phonology code mapping impairment. This study aims to
Revised 11 July 2011
disambiguate between the two interpretations by investigating performance in a non-verbal
Accepted 8 September 2011
character string visual categorization task with verbal and non-verbal stimuli. Results show
Action editor Roberto Cubelli
that VA span ability predicts performance for the non-verbal visual processing task in normal
Published online xxx
reading children. Furthermore, VA span impaired dyslexic children are also impaired for the
categorization task independently of stimuli type. This supports the hypothesis that the
Keywords:
underlying impairment responsible for the VA span deficit is visual, not verbal.
ª 2011 Elsevier Srl. All rights reserved.
Reading
Developmental dyslexia
Visual attention
VA span
1.
Introduction
Developmental dyslexia is characterized by a severe reading
acquisition impairment in children of normal intelligence, free
of any neurological or psychiatric condition. It is widely
admitted that dyslexia is a consequence of a phonological
deficit (Vellutino et al., 2004; Ziegler and Goswami, 2005).
Nevertheless, as a significant proportion of dyslexic children
exhibit no phonological impairment (Bosse et al., 2007; White
et al., 2006), such a deficit cannot account for the full spectrum of the disorder. Concurrently, visual processing performance has been shown to contribute to reading performance in
typical readers (Kevan and Pammer, 2008; Kwon et al., 2007;
Pammer et al., 2005). Moreover, visuo-spatial attention
(Facoetti et al., 2008, 2006) and low level visual processing
(Boden and Giaschi, 2007) have been found to be impaired in
dyslexic readers. Multifactorial accounts of dyslexia (Menghini
et al., 2010) have opened new perspectives such as the existence of a visual rather than phonological system impairment
(Vidyasagar and Pammer, 2010).
Several studies have explored multi-element visual processing in dyslexic children using whole and partial report tasks
(Bosse et al., 2007; Valdois et al., 2003). Impaired performance on
these tasks was interpreted as evidence for a deficit in the
number of individual visual elements that can be processed
simultaneously, namely a visual attention (VA) span disorder.
* Corresponding author. Laboratoire de Psychologie et Neurocognition (UMR 5105 CNRS), Université Pierre Mendès-France, BP 47, 38040
Grenoble Cedex 9, France.
E-mail address: [email protected] (M. Lobier).
0010-9452/$ e see front matter ª 2011 Elsevier Srl. All rights reserved.
doi:10.1016/j.cortex.2011.09.003
Please cite this article in press as: Lobier M, et al., The visual attention span deficit in dyslexia is visual and not verbal, Cortex
(2011), doi:10.1016/j.cortex.2011.09.003
2
c o r t e x x x x ( 2 0 1 1 ) 1 e6
Research has shown that a subset of dyslexic children exhibits
such a deficit (Bosse et al., 2007). Within the framework of the
Multi Trace Memory (MTM) model of reading (Ans et al., 1998),
a VA span deficit can explain impaired reading acquisition by
preventing simultaneous processing of the constituent letters of
relevant orthographic units. The relevance of the VA span to
reading is highlighted behaviourally by the fact that VA span
performance is a reliable predictor of reading performance
independently of phonological processing in typically developing
(Bosse and Valdois, 2009) and dyslexic children (Bosse et al., 2007).
Even though the VA span disorder’s definition encompasses
all types of visual elements, behaviourally its hallmark is
impaired performance on letter-report tasks. These tasks have
two potential caveats: they necessitate a verbal response and
use verbal stimuli. Consequently it has been claimed (Hawelka
and Wimmer, 2008; Ziegler et al., 2010) that the impairment
does not follow from a visual processing deficit but from
impaired visual to phonological code mapping in line with the
phonological theory of dyslexia. In order to disambiguate from
these two interpretations, it is necessary to confront predictions of both hypotheses on available data of tasks requiring
verbal report and/or multi-element processing.
String visual processing has been studied extensively in
dyslexia, albeit with many different types of paradigms which
have led to contrasting results. Because of these numerous
paradigms, it is crucial to identify those that do, in fact,
investigate multi-element parallel visual processing. Not only do
appropriate tasks need to involve processing of several
elements, but paradigms also need to ensure that visual processing is parallel. Indeed studies that do not constrain stimulus presentation time (Hawelka and Wimmer, 2008; Pitchford
et al., 2009) are not relevant for the VA span hypothesis.
Performance of parallel visual processing can be assessed
independently of phonological ability by using tasks that require
no verbal report, tasks that use non-verbal stimuli or, ideally,
tasks that combine both. Poor performance for dyslexic readers
in tasks involving parallel processing of verbal stimuli but no
verbal response has previously been reported (Rutkowski et al.,
2003; Ziegler et al., 2010). While this data is in line with the
hypothesis of a purely visual processing deficit in dyslexia, it is
not sufficient to sideline an underlying phonological cause
insofar as the use of verbal material might automatically involve
phonological recoding and yield a phonologically-constrained
performance.
In order to completely disambiguate the role of visual and
phonological processing in dyslexia, visual processing has
been investigated using change detection tasks with nonverbal material. Not only were dyslexic readers impaired
(Jones et al., 2008; Pammer et al., 2004), but task performance
predicted reading performance in both children and adults
(Pammer et al., 2005, 2004). These results cannot be accounted
for by a phonological account of dyslexia, but are in line with
predictions of the VA span deficit hypothesis.
However, another symbol processing task has yielded
opposite results (Ziegler et al., 2010). Based on data from
a partial report identification task, the authors argued against
a visual deficit in dyslexia as the task yielded a deficit with
letters and digits but not with symbols.
The main goal of this study is to disambiguate which of
a parallel visual processing deficit or a phonological coding
deficit is the proximal cause of the VA span deficit in dyslexia.
We will investigate parallel visual processing in normal and
dyslexic children using non-verbal categorization and verbal
and non-verbal stimuli. Since we aim to specify the nature of
the VA span deficit, all dyslexic participants will exhibit such
a deficit. We aim to show (1) that performance on this nonverbal task relates to VA span performance in a large group of
typically developing children and (2) that VA span impaired
dyslexic children are impaired on this non-verbal task, regardless of character type.
2.
Methods
2.1.
Population
One hundred and nine typically-developing children and 14
dyslexic children took part in the experiment. The typical
Table 1 e Mean scores (SD) of age (chronological and reading) and reading performance for typically developing (TYP),
dyslexic (DYS) and control (CTRL) children.
Chronological age
Reading age
Reading
Regular word
Accuracy/20
Speed (wds/min)
Exception words
Accuracy/20
Speed (wds/min)
Pseudo-words
Accuracy/20
Speed (wds/min)
VA span
Global report
TYP (N ¼ 105)
DYS (N ¼ 14)
CRTL (N ¼ 14)
106.4 (14.6)
109.5 (21.5)
128.9 (14.6)
86.0 (5.9)
128.2 (11.1)
135.9 (18.7)
t(26) ¼ .13, n.s.
t(15.6) ¼ 9.54***
17.0 (2.3)
51.6 (22.7)
13.1 (3.8)
22.3 (8.4)
19.4 (.73)
79.9 (20.7)
Z ¼ 4.34***
t(17.1) ¼ 9.65***
10.9 (4.9)
44.2 (20.4)
7.9 (4.7)
21.0 (10.5)
16.1 (3.5)
71.1 (18.9)
Z ¼ 3.69***
t(20.3) ¼ 8.67***
15.7 (2.7)
39.7 (14.3)
10.6 (4.1)
19.7 (5.4)
17.5 (2.0)
56.8 (14.2)
Z ¼ 3.90***
t(16.6) ¼ 9.14***
80.7 (10.9)
64.0 (7.7)
89.6 (7.0)
t(26) ¼ 9.22***
t or Z
*** p < .001.
Please cite this article in press as: Lobier M, et al., The visual attention span deficit in dyslexia is visual and not verbal, Cortex
(2011), doi:10.1016/j.cortex.2011.09.003
c o r t e x x x x ( 2 0 1 1 ) 1 e6
children were native French speakers, had normal or corrected
to normal visual acuity and no reported learning disability. All
participants and their parents gave informed written consent
prior to the experiment. Characteristics are provided in Table 1.
Dyslexic children were recruited via the Dyslexia Unit of
the Grenoble University Hospital. Diagnosis of developmental dyslexia was established using both inventories and
testing procedures in accordance with the guidelines of the
ICD-10 classification of Mental and Behavioural disorders.
Dyslexic participants had normal IQ [full IQ superior to 85
on the WISC III or IV, or a score superior to the 25th
percentile on Raven’s Progressive Matrices (Raven et al.,
1998)]. The Alouette Reading Test (Lefavrais, 1965) was
used to estimate children’s reading age. Children were
diagnosed as dyslexics if their reading age was at least
18 months below their chronological age. Furthermore, all
dyslexic children were screened to ensure that individual
performance on the VA span report tasks (see task
description in Methods) was at least 1.5 standard deviations
(SDs) below age group norms.
An age-matched control group of 14 normal readers was
selected amongst the typically developing children. These
participants scored above the 15th percentile on all reading
measures and had no reported learning disability. None of them
exhibited a VA span deficit (Global Report Z-Score > 1). As
shown in Table 1, the two groups did not differ significantly in
chronological age [t(26) ¼ .13, n.s.], whereas the dyslexic group’s
reading age was significantly lower than the control group’s
[t(15.6) ¼ 9.54, p < .0001].
2.2.
Reading assessment
Participants’ reading skills were assessed using tasks of isolated word and pseudo-word reading (Jacquier-Roux et al.,
2002). Participants were given three lists of 20 items each
3
(20 regular words, 20 irregular words and 20 pseudo-words).
Both accuracy and reading speed were scored.
2.3.
VA span assessment
VA span performance was assessed using a five-consonant
global report task. Twenty random five letter-strings
(e.g., RHSDM; angular size ¼ 5.4 ) were built up from 10 consonants (B, P, T, F, L, M, D, S, R, H). Each letter was used 10 times
and appeared twice in each position. Strings contained no
repeated letters and never matched a real word skeleton (e.g.,
FLMBR for FLAMBER “burn”). Letters were presented in upper
case (Geneva, 24) in black on a white background. Distance
between adjacent letters was of .57 in order to minimize
crowding. At the beginning of each trial, a central fixation point
was presented for 1000 msec followed by a blank screen for
50 msec. A consonant-string was then displayed at the centre of
the screen for 200 msec. Children had to report verbally as many
letters as possible. Twenty experimental trials were preceded
by 10 training trials for which participants received feedback.
The dependent measure was the total number of letters accurately reported (identity not location) (maximal score ¼ 100).
2.4.
Experimental tasks
The categorization task was carried out using five different
character categories: two verbal (letters and digits) and three
non-verbal (Hiragana, pseudo-letters, and unfamiliar shapes).
Each category was made up of six different characters (see
Fig. 1). We calculated each set’s mean perimetric complexity
(Majaj et al., 2002) as the best measure of character identification efficiency (Pelli et al., 2006). Our stimuli categories
showed no significant difference in visual complexity. Categories were assigned a visual label made up of all the
Fig. 1 e Categorization task procedure and stimuli.
Please cite this article in press as: Lobier M, et al., The visual attention span deficit in dyslexia is visual and not verbal, Cortex
(2011), doi:10.1016/j.cortex.2011.09.003
4
c o r t e x x x x ( 2 0 1 1 ) 1 e6
category’s characters. All five categories and their characters
were presented to the participant for 1 min before testing.
A single element categorization task was designed to control
that participants had sufficient efficiency in identifying each
individual character’s category. Stimuli were the 30 characters
previously described, displayed once in each experimental block.
Each trial began with a black fixation cross displayed on a white
screen for 1000 msec, followed by a 50 msec blank screen. A
single character subtending a vertical angle of .7 was then displayed in the centre of the screen during 200 msec, followed by
a 500 msec mask. The answer screen (displayed next) required
the participant to click on the visual label of the character’s
category. Participants carried out 10 training trials with feedback,
and two blocks of 30 experimental trials with no feedback. The
dependent measure was the success rate across trials.
In the multi-element categorization task, participants had
to count the number of elements of a character string
belonging to a previously defined target category. Forty character strings were designed. Each string contained characters
from two categories, a target category and a distracter category. Ten category/distracter category combinations were
chosen so that each category was a target category twice and
a distracter category twice. Combinations indifferently
coupled verbal or non-verbal target categories with verbal or
non-verbal distracter categories. Overall, each category was
target for eight trials, with the number of target category
characters varying from 1 to 3. Positions of the target category
characters were counterbalanced across trials and trials were
randomly drawn. Character strings were displayed in black on
a white background. Each character subtended 1 vertical
angle and the distance between the centres of each character
was 1.4 in order to minimize crowding. Each trial began with
the target category label which was displayed until the
participant clicked on a button. A central fixation cross was
then displayed during 1000 msec, immediately followed by
a 50 msec blank screen. The character array was next displayed in the screen centre for 200 msec, immediately followed by a 500 msec mask. The answer screen displayed both
the target category as a reminder and three buttons (1, 2, and
3) on which the participant was required to click according to
the number of target category characters he had identified
(see Fig. 1). Participants carried out five training trials with
feedback before the 40 experimental trials with no feedback.
The dependent measure was the success rate across trials.
Scores for verbal (letters and digits) and non-verbal (Hiragana,
pseudo-letters and geometrical shapes) target categories were
also computed.
3.
Table 2 e Correlations (above the diagonal) and partial
correlations controlling for chronological age (below the
diagonal) between reading age, reading performance
(score and speed), VA span and categorization
performance (all and non-verbal) for normal reading
participants (N [ 105).
Reading Reading VA
age
speed span
Reading age
Reading speed
VA span
Categorization
Non-verbal cat
e
.87**
.64**
.29**
.30**
.89**
e
.69**
.28**
.31**
.64**
.69**
e
.31**
.29**
Cat
Non-verbal
cat
.43**
.40**
.35**
e
.92**
.41**
.41**
.33**
.93**
e
**pcorrected < .005.
developing population. Four children were excluded from
further analyses because of poor performance on the single
element categorization task. Single word reading speeds were
averaged across lists. To normalize distributions, reading age
and speed were log transformed. Table 2 provides Pearson’s
full and partial correlations when controlling for chronological age between these variables. A Bonferroni correction for
multiple correlations was applied ( ps ¼ .005). Performance in
the categorization task was moderately correlated with the
five letter-report task (.35) and with both reading measures
(.40e.43). Partial correlations controlling for chronological age
showed that categorization performance remained significantly correlated with all other measures (from .28 to .35). To
ensure that these correlations were not driven by verbal target
performance, we computed correlations for non-verbal target
performance. It was moderately correlated with global report
performance (.33) and with both reading measures (.41 for
both). Partial correlations controlling for chronological age
showed it remained significantly correlated with all other
measures (from .29 to .31).
Second, performance was compared between dyslexic and
control groups. Performance on the single element task was at
ceiling for both groups [mean DYS ¼ .98, SD ¼ .03; mean
CTRL ¼ .99, SD ¼ .01; Z(28) ¼ 1.53, n.s.]. For the multi-element
Results
Group comparisons showed that dyslexic readers performed
worse than control readers on all reading and VA span
measures, thus showing a mixed reading profile. Data that
met normality assumptions were analyzed using Student’s
t-tests; those that did not were analyzed using nonparametric ManneWhitney U tests. Results are detailed in
Table 1.
First, correlation analyses between categorization, VA span
and reading age and speed were run on the typically
Fig. 2 e Categorization performance (% correct) for dyslexic
and control participants, error bars represent the SD of the
mean.
Please cite this article in press as: Lobier M, et al., The visual attention span deficit in dyslexia is visual and not verbal, Cortex
(2011), doi:10.1016/j.cortex.2011.09.003
c o r t e x x x x ( 2 0 1 1 ) 1 e6
categorization task, accuracy scores for verbal (letters and
digits) and non-verbal targets were analyzed in a 2 2 analysis
of variance with Group (Dyslexics vs. Controls) as a betweensubjects factor and Type (Verbal vs. non-verbal) as a withinsubject factor (see Fig. 2). The main effect of Group
[F(1, 26) ¼ 10.9, p ¼ .003, h2 ¼ .295] was significant showing that
dyslexic children (mean verbal ¼ .62, SD ¼ .11; mean nonverbal ¼ .64, SD ¼ .15) performed worse than control children
(mean verbal ¼ .77, SD ¼ .09; mean non-verbal ¼ .81, SD ¼ .11).
More importantly, neither main Type effect [F(1, 26) ¼ 2.5,
p ¼ .12, h2 ¼ .09] nor Group Type interaction was significant
[F (1,26) ¼ .41, p ¼ .53, h2 ¼ .02].
To further disambiguate the effects of the distracter category on performance, we examined scores obtained for the
three category combinations that were entirely (i.e., both
targets and distracters) non alphanumeric. Results showed
that performance for entirely non alphanumeric character
strings was significantly lower for the dyslexic group [mean
DYS ¼ .72, SD ¼ .17; mean CTRL ¼ .86, SD ¼ .12; Z(28) ¼ 2.18,
p < .027] indicating that the dyslexic deficit endures even
when no letters or digits are present in the character string.
4.
Discussion
This study explored performance of both normal and VA span
impaired dyslexic children on a non-verbal task and both
verbal and non-verbal stimuli in order to disambiguate
between a verbal or visual account of the VA span deficit.
The first goal of our experiment was to show that the VA
span global report task has processes in common with the
non-verbal multi-element categorization task. Our results
show that high global report performance is associated with
high categorization performance, regardless of stimulus type,
consistent with a visual interpretation of the VA span. Indeed,
if verbal performance constrained VA span performance, no
links with the categorization task should have been found.
This pattern of findings held when age was controlled for. A
ceiling effect on the single element task made it impossible to
control for categorization performance, which should be
investigated in future research.
The second goal of our experiment was to show that the VA
span deficit in dyslexia could be extended to non-verbal tasks
and non-verbal stimuli. Since no visual to phonological code
mapping is involved in the categorization task, impaired
mapping will not lead to impaired performance. In contrast,
not only were dyslexic children impaired for alphanumeric
categories, but their deficit was present and of similar magnitude for non-verbal target categories. This pattern of results
cannot be reconciled with a phonological deficit account of
developmental dyslexia but is, however, compatible with
a visual processing deficit. A reduced number of visually processed elements would impact both the global report and
categorization task. Furthermore, not only was this deficit
present with both verbal and non-verbal targets, but it also
endured when trials with only non-verbal characters were
considered, indicating that it is present regardless of stimuli
type. Results show that VA span impaired dyslexic children
cannot process as many visual elements as their normal
reading peers, regardless of element type. Similar impaired
5
processing of verbal and non-verbal character strings is strong
evidence for a purely visual account of the VA span deficit.
Visual crowding (Whitney and Levi, 2011) is known to
influence character string processing and one could argue that
increased crowding in dyslexic children (Martelli et al., 2009)
could also account for their lower categorization performance.
Although spacing between characters in both categorization
and report tasks was set to a large enough value to prevent
crowding, further investigation is needed to assess a potential
crowding effect on dyslexic performance. In the same way,
potential relationships with attentional masking mechanisms
(Ruffino et al., 2010) have yet to be explored (but see Lallier et al.,
2010).
Furthermore, our results need to be balanced against those
of Ziegler et al. (2010) who found no symbol string deficit in
dyslexic children. In contrast with our study where performance for verbal and non-verbal stimuli was similar, performance for symbols was lower than for verbal stimuli. This is
consistent with previous results (Pelli et al., 2006) showing that
memory span constrains performance on tasks requiring
identification and recall of non-alphabetical stimuli. In that
case, symbol performance for normal reading and dyslexic
children would be equally low, as is the case in Ziegler et al.’s
(2010) results.
While further research remains necessary to exclude other
explanatory visual factors, this study provides strong
evidence that poor VA span performance in dyslexia stems
from a parallel visual processing deficit.
Acknowledgements
ML was funded by the ANR Blanc VASRA (ANR-07-BLAN0019-01) grant and the Region Rhône-Alpes Cluster 11.
references
Ans B, Carbonnel S, and Valdois S. A connectionist multiple-trace
memory model for polysyllabic word reading. Psychological
Review, 105(4): 678e723, 1998.
Boden C and Giaschi D. M-Stream deficits and reading-related
visual processes in developmental dyslexia. Psychological
Bulletin, 133(2): 346e366, 2007.
Bosse ML, Tainturier MJ, and Valdois S. Developmental dyslexia:
The visual attention span deficit hypothesis. Cognition, 104(2):
198e230, 2007.
Bosse ML and Valdois S. Influence of the visual attention span on
child reading performance: A cross-sectional study. Journal of
Research in Reading, 32(2): 230e253, 2009.
Facoetti A, Ruffino M, Peru A, Paganoni P, and Chelazzi L. Sluggish
engagement and disengagement of non-spatial attention in
dyslexic children. Cortex: A Journal Devoted to the Study of the
Nervous System and Behavior, 44(9): 1221e1233, 2008.
Facoetti A, Zorzi M, Cestnick L, Lorusso M, Molteni M, Paganoni P,
et al. The relationship between visuo-spatial attention and
nonword reading in developmental dyslexia. Cognitive
Neuropsychology, 23(6): 841e855, 2006.
Hawelka S and Wimmer H. Visual target detection is not impaired
in dyslexic readers. Vision Research, 48(6): 850e852, 2008.
Please cite this article in press as: Lobier M, et al., The visual attention span deficit in dyslexia is visual and not verbal, Cortex
(2011), doi:10.1016/j.cortex.2011.09.003
6
c o r t e x x x x ( 2 0 1 1 ) 1 e6
Jacquier-Roux M, Valdois S, and Zorman M. l’Odédys, un outil de
dépistage des dyslexies. Grenoble: Laboratoire cogni-sciences,
IUFM de Grenoble, 2002.
Jones MW, Branigan HP, and Kelly ML. Visual deficits in
developmental dyslexia: Relationships between non-linguistic
visual tasks and their contribution to components of reading.
Dyslexia (Chichester, England), 14(2): 95e115, 2008.
Kevan A and Pammer K. Making the link between dorsal
stream sensitivity and reading. NeuroReport, 19(4): 467e470,
2008.
Kwon M, Legge GE, and Dubbels BR. Developmental changes in
the visual span for reading. Vision Research, 47(22): 2889e2900,
2007.
Lallier M, Donnadieu S, Berger C, and Valdois S. A case study of
developmental phonological dyslexia: Is the attentional deficit
in the perception of rapid stimuli sequences amodal? Cortex,
46(2): 231e241, 2010.
Lefavrais P. Test de l’Alouette. Paris: Editions du centre de
psychologie appliquée, 1965.
Majaj NJ, Pelli DG, Kurshan P, and Palomares M. The role of spatial
frequency channels in letter identification. Vision Research,
42(9): 1165e1184, 2002.
Martelli M, Di Filippo G, Spinelli D, and Zoccolotti P. Crowding,
reading, and developmental dyslexia. Journal of Vision, 9(4): 1e18,
2009.
Menghini D, Finzi A, Benassi M, Bolzani R, Facoetti A,
Giovagnoli S, et al. Different underlying neurocognitive
deficits in developmental dyslexia: A comparative study.
Neuropsychologia, 48(4): 863e872, 2010.
Pammer K, Lavis R, Cooper C, Hansen PC, and Cornelissen PL.
Symbol-string sensitivity and adult performance in lexical
decision. Brain and Language, 94(3): 278e296, 2005.
Pammer K, Lavis R, Hansen P, and Cornelissen PL. Symbol-string
sensitivity and children’s reading. Brain and Language, 89(3):
601e610, 2004.
Pelli DG, Burns CW, Farell B, and Moore-Page DC. Feature
detection and letter identification. Vision Research, 46(28):
4646e4674, 2006.
Pitchford NJ, Ledgeway T, and Masterson J. Reduced orthographic
learning in dyslexic adult readers: Evidence from patterns of
letter search. Quarterly Journal of Experimental Psychology, 62(1):
99e113, 2009.
Raven JC, Raven JE, and Court JH. Progressive matrices. Oxford,
England: Oxford Psychologists Press, 1998.
Ruffino M, Trussardi A, Gori S, Finzi A, Giovagnoli S, Menghini D,
et al. Attentional engagement deficits in dyslexic children.
Neuropsychologia, 48(13): 3793e3801, 2010.
Rutkowski JS, Crewther DP, and Crewther SG. Change detection is
impaired in children with dyslexia. Journal of Vision, 3(1):
95e105, 2003.
Valdois S, Bosse ML, Ans B, Carbonnel S, Zorman M, David D, et al.
Phonological and visual processing deficits can dissociate in
developmental dyslexia: Evidence from two case studies.
Reading and Writing, 16(6): 541e572, 2003.
Vellutino FR, Fletcher JM, Snowling MJ, and Scanlon DM. Specific
reading disability (dyslexia): What have we learned in the past
four decades? Journal of Child Psychology and Psychiatry, and
Allied Disciplines, 45(1): 2e40, 2004.
Vidyasagar TR and Pammer K. Dyslexia: A deficit in visuo-spatial
attention, not in phonological processing. Trends in Cognitive
Sciences, 14(2): 57e63, 2010.
White S, Milne E, Rosen S, Hansen P, Swettenham J, Frith U, et al.
The role of sensorimotor impairments in dyslexia: A multiple
case study of dyslexic children. Developmental Science, 9(3):
237e255. discussion 265e239, 2006.
Whitney D and Levi DM. Visual crowding: A fundamental limit on
conscious perception and object recognition. Trends in
Cognitive Sciences, 15(4): 160e168, 2011.
Ziegler JC and Goswami U. Reading acquisition, developmental
dyslexia, and skilled reading across languages: A
psycholinguistic grain size theory. Psychological Bulletin, 131(1):
3e29, 2005.
Ziegler JC, Pech-Georgel C, Dufau S, and Grainger J. Rapid
processing of letters, digits and symbols: What purely visualattentional deficit in developmental dyslexia? Developmental
Science, 13(4): F8eF14, 2010.
Please cite this article in press as: Lobier M, et al., The visual attention span deficit in dyslexia is visual and not verbal, Cortex
(2011), doi:10.1016/j.cortex.2011.09.003