A case study of developmental phonological dyslexia: Is the attentional deficit in the perception of rapid stimuli sequences amodal?

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1

ava i lab le at www.sc ienced i rec t . com

journa l homepage : www. e lsev ier . com/ loca te / cor tex

Research report

A case study of developmental phonological dyslexia: Is theattentional deficit in the perception of rapid stimulisequences amodal?

Marie Lalliera,d,*, Sophie Donnadieua,c, Carole Bergera,c and Sylviane Valdoisa,b,d

aLaboratoire de Psychologie et NeuroCognition (UMR 5105 CNRS), Universite Pierre Mendes France, Grenoble, FrancebCentre National de la Recherche Scientifique, FrancecUniversite de Savoie, Chambery, FrancedUniversite Pierre Mendes-France, Grenoble, France

a r t i c l e i n f o

Article history:

Received 6 May 2008

Reviewed 30 October 2008

Revised 29 December 2008

Accepted 26 March 2009

Action editor Roberto Cubelli

Published online 17 April 2009

Keywords:

Visual attention span

Auditory attention

Sluggish attentional shifting

Temporal processing

Dyslexia

* Corresponding author. Laboratoire de PsychCentrale BP 47, 38040 Grenoble Cedex 9, Fra

E-mail address: sylviane.valdois@upmf-g0010-9452/$ – see front matter ª 2009 Elsevidoi:10.1016/j.cortex.2009.03.014

a b s t r a c t

The attentional blink (AB) refers to a decrease in accuracy that occurs when participants

are required to detect the second of two rapidly sequential targets displayed randomly in

a stream of distracters. Dyslexic individuals have been shown to exhibit a prolonged AB in

the visual modality, interpreted as evidence of sluggish attentional shifting (SAS). However,

the amodal SAS theory predicts that the disorder should further extend to the auditory

modality, then resulting in a phonological disorder as typically found in developmental

dyslexia. Otherwise, it has been demonstrated that a visual attention (VA) span deficit

contributes to the poor reading outcome of dyslexic individuals, independently of their

phonological skills. The present study assesses the amodality assumption of the SAS

theory together with questioning its relation with the VA span deficit. For this purpose,

visual and auditory ABs were explored in a well compensated young adult, LL, who exhibits

a pure phonological dyslexia characterised by poor pseudo-word processing and poor

phonological skills but preserved VA span. The investigation revealed two different kinds

of deficits in LL. Her AB was prolonged and marginally deeper in the visual modality

whereas a primarily deeper in amplitude and a subtle prolonged AB was found in the

auditory modality. The atypical performance patterns of LL in both modalities suggest that

her perceptual attention disorder is amodal as predicted by the SAS theory. This amodal

disorder was here reported in a dyslexic participant with a phonological disorder, well in

accordance with the hypothesis that sluggish auditory attention shifting contributes to

difficulties in phoneme awareness and literacy acquisition. Furthermore, prolonged VA

blink was observed in the absence of VA span disorder, thus suggesting that visual

attentional shifting and VA span might be distinct mechanisms, contributing indepen-

dently to reading acquisition and developmental dyslexia.

ª 2009 Elsevier Srl. All rights reserved.

ologie et Neuro-Cognitionnce.renoble.fr (M. Lallier).er Srl. All rights reserved

(UMR5105 CNRS), Universite Pierre Mendes France, 1251 Avenue

.

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1232

1. Introduction

1 The term ‘perceptual attention’ here refers to those atten-tional processes (here automatic shifting) which directly influ-ence perceptual processing, thus allowing information to be moreefficiently (here more rapidly) processed.

Developmental dyslexia is a specific disorder of learning to

read, despite normal intelligence and adequate literacy

instruction. The phonological deficit hypothesis (Frith, 1997;

Snowling, 2000; Vellutino et al., 2004) assumes that an under-

lying phonological impairment is the core deficit in develop-

mental dyslexia. Dyslexic individuals have been shown to

exhibit phonological processing difficulties when engaged in

tasks of phoneme awareness, phonological learning or non-

word repetition (Gathercole and Baddeley, 1990; Pennington

et al., 1990; Brady and Shankweiler, 1991; Frith, 1995; Mody

et al., 1997). However, they further suffer from various sensory

deficits which are not directly related to reading skills but

might be responsible for their phonological impairment. In

particular, a temporal processing deficit has been proposed as

a more basic problem yielding to the phonological impairment

observed in developmental dyslexia (Farmer and Klein, 1995).

This hypothesis is supported by a number of studies showing

evidence for impaired brief stimuli temporal processing within

the auditory and visual modalities in dyslexic people. In the

auditory domain, the temporal deficit hypothesis predicts that

dyslexic individuals exhibit deficits in the perception of

sequences of rapidly changing acoustic stimuli, typically when

the interstimulus interval (ISI) is short (Tallal, 1980; Hari and

Kiesila, 1996; Tallal et al., 1998; Helenius et al., 1999). Similar

results have been reported in the visual modality within the

framework of the magnocellular theory of developmental

dyslexia (Livingstone et al., 1991; Lovegrove, 1993; Stein and

Walsh, 1997; Stein, 2003). However, the deficits mentioned

have been raised into question in both the auditory (Nittrouer,

1999; Marshall et al., 2001; Chiappe et al., 2002) and the visual

modality (Spinelli et al., 1997; Ben-Yehudah et al., 2001; Amitay

et al., 2002; Bretherton and Holmes, 2003). In addition, studies

have shown that dyslexic participants are not impaired in pure

auditory temporal processing tasks such as detection of inter-

aural phase modulation (Witton et al., 1997) or inter-aural

temporal cues (Hari et al., 1999). These findings led Hari and

Renvall (2001) to propose that the temporal deficit in devel-

opmental dyslexia may be held up by a more crucial

dysfunction, which they assumed to be sluggish attentional

shifting (SAS). According to this hypothesis, when dyslexic

individuals deal with rapid stimuli sequences, their automatic

attention system cannot disengage fast enough from one item

to the next, yielding degraded processing. SAS is supposed to

distort cortical networks, more specifically those which

support phonological representations. Thus, the phonological

deficit in developmental dyslexia would be the consequence of

an attentional deficit. Hari and Renvall (2001) moreover

postulate that SAS affects all sensory modalities. Reviewing

a series of studies they argued that dyslexic participants

showed impaired performance in both modalities (Hari and

Kiesila, 1996; Hari et al., 1999, 2001; Helenius et al., 1999). Their

conclusion however, was that of an amodal disorder, based on

evidence from different modality specific experimental para-

digms and from different groups of dyslexic participants. Such

a claim would require demonstrating that the same dyslexic

individuals have impaired performance on attentional shifting

tasks in different modalities.

Although they did not straightforwardly address the role of

attentional processing, several studies focused on the amodal

temporal perceptual deficit hypothesis (Laasonen et al., 2001;

Meyler and Breznitz, 2005). Their results suggest that the

disorder is not modality specific but rather extends over

several modalities as expected by Hari and Renvall (2001).

Only one study directly addressed the amodality assumption

of the SAS theory. Using the same lateralized cued detection

task paradigm in vision and audition, Facoetti et al. (2005),

concluded that visual and auditory attentional captures are

both sluggish in children with developmental dyslexia.

However, the disorder was observed in dyslexic children

whose phonological skills were not assessed. Indeed, as it

would have been expected by the SAS theory, those children

should have exhibited an additional phonological disorder.

While both visual and auditory attention deficits have been

reported in developmental dyslexia as potentially responsible

for the phonological disorder exhibited by dyslexic readers,

another kind of visual attention (VA) disorder – referred to as

the VA span deficit hypothesis (Bosse et al., 2007)– has been

pointed out as typically dissociating from phonological

problems. The VA span disorder was defined as a limitation in

the number of discrete visual elements that can be processed

in parallel in a multi-element display. With respect to reading,

this disorder results in a reduction of the number of letters

processed simultaneously in a letter string. VA span disorders

have been reported in case studies of dyslexic children who

exhibited no phonological problem (Valdois et al., 2003; Dubois

et al., 2007). Group studies provided additional support that

phonological and VA span deficits typically dissociate in

dyslexic individuals (Bosse et al., 2007; Prado et al., 2007;

Lassus-Sangosse et al., 2008). Bosse et al. (2007) demonstrated

that a non-trivial number of dyslexic children exhibited

a single VA span disorder that related to reading performance

independently of the child phonological skills. Further

evidence for the independent contribution of VA span abilities

to reading performance was provided by a study conducted on

large samples of typically developing children from first to fifth

grade (Bosse and Valdois, 2009). However, large sample studies

further showed that some dyslexic readers exhibited a double

disorder (poor phonological skills together with poor VA span

abilities). Because VA span abilities were not assessed in the

previous studies conducted within the SAS framework and

because SAS abilities were not assessed in those studies which

addressed VA span abilities, the question remains whether

SAS and poor VA span are independent or related disorders.

Evidence for amodal perceptual attention disorders1 in

dyslexic individuals who exhibit no VA span disorder but

impaired phonological skills would strengthen the SAS theory

which establishes a theoretical link between perceptual

attention disorders and phonological problems.

The present study evaluates the amodality assumption of

the SAS theory through a single-case report. The participant,

LL, is an exceptionally pure case of developmental phono-

logical dyslexia. Although dyslexic individuals as a group

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1 233

typically exhibit phonological problems, only a subgroup

demonstrates selectively poor performance in pseudo-word

reading and poor phonological skills (Sprenger-Charolles

et al., 2000; White et al., 2006). In French, Bosse et al. (2007)

found that only 26% of their dyslexic participants showed poor

phoneme awareness, and the phonological disorder was the

only underlying disorder in only 19% of the participants.

Moreover, most previous studies on developmental phono-

logical dyslexia did not take into account the VA span capac-

ities of their dyslexic participants. In contrast, our participant

was recruited because of showing a prototypical reading and

spelling pattern of developmental phonological dyslexia and

a single underlying phonological disorder together with

demonstrating preserved VA span.

The dyslexic participant was administered similar atten-

tional blink (AB) tasks in vision and audition, The AB is

typically attributed to an inability to rapidly reallocate

attentional resources from a first target (T1) to a second

target (T2) in a rapid serial presentation stream. The AB

prevents any adequate processing of T2 during 400–500 msec

after T1 presentation in the visual modality (e.g., Broadbent

and Broadbent, 1987; Raymond et al., 1992, 1995; Duncan

et al., 1994; Shapiro et al., 1994; Chun and Potter, 1995;

Jolicoeur and Dell’ Acqua, 1998). As a result, the rapid serial

presentation task is a useful paradigm providing important

pieces of information about how attentional processes

deploy in the temporal domain. Using this method, several

studies have shown that dyslexic subjects exhibit a pro-

longed visual AB – lasting in average 600–800 msec – as

compared to normal readers (Hari et al., 1999; Visser et al.,

2004; Buchholz and Davies, 2007; Facoetti et al., 2008). This

finding suggests that T1 captures attentional resources for

longer time in dyslexic participants than control partici-

pants, in the visual modality. However, although researchers

have brought evidence for an AB in the auditory modality as

well (Duncan et al., 1997; Mondor, 1998; Arnell and Jolicoeur,

1999; Arnell and Larson, 2002; Tremblay et al., 2005; Vachon

and Tremblay, 2005, 2006; Shen and Mondor, 2006), no study

evaluated whether developmental dyslexia was further

characterised by an atypical AB in the auditory modality.

Such a disorder is however, predicted by the SAS theory and

should be all the more expected in the auditory modality

that the auditory – not the visual – SAS is supposed to

straightforwardly account for the phonological disorder in

developmental dyslexia.

In the current paper, we assessed the amodality deficit

hypothesis postulated by the SAS theory by administering

two similar rapid serial presentation tasks in the visual

and the auditory modalities. This study directly addresses

the validity of the SAS theory since for the first time the AB

is assessed in the auditory modality in a young adult

dyslexic participant, LL, who exhibits a phonological

disorder and poor pseudo-word reading. This prototypical

phonological dyslexic participant was further chosen to

have no VA span disorder. In line with the SAS theory,

we expected that impaired performance in the visual AB

task – revealing impaired visual attentional sequential

processing – should be found in LL despite preserved VA

span abilities (thus, good visual attentional simultaneous

processing).

2. Material and method

2.1. Participants

2.1.1. Control groupThirty young adults, with French as native language, were

included as controls for the evaluation of the AB in the visual

and auditory modalities. They were undergraduate students

in Psychology at either the University of Savoie (Chambery) or

the University Pierre Mendes-France (Grenoble). All were

skilled readers and none reported any learning impairment

for reading or spelling. They had normal or corrected-to-

normal vision and normal hearing and no history of neuro-

logical or psychiatric disorders.

2.1.2. The dyslexic participant: LLThe dyslexic adult who took part in this study, LL, was a 35

years old female, right-handed native French speaker. She

had no history of neurological disorder and her anatomical

magnetic resonance imaging was normal. Her hearing was

good and she reported no severe problems in speech language

development. LL had a mild myopia corrected by glasses. She

was free from any history of psychiatric illness, or any medical

treatment. She received conventional reading instruction

when attending primary school. LL never repeated a grade but

she was considered as a slow reader at school. She never

received neither special help nor remediation for her diffi-

culties in reading. She was research assistant at the time of

testing but naıve with respect to the purpose and background

of the current research. On the ‘‘Alouette Reading Test’’

(Lefavrais, 1965), LL achieved a reading age of 10 years 11

months, demonstrating persistence of severe reading

difficulties.

In the absence of standardized tests for the neuro-

psychological assessment of young adults in French, our

dyslexic participant, LL, was given original tasks designed

for the purpose of the current study. Her reading and

spelling abilities were assessed together with her phono-

logical skills and her VA span abilities. Her performance on

the Raven Standard Progressive Matrices (Raven, 1960),

verbal short-term memory, reading, spelling, phonological

and VA span tasks was compared to that of 20 chronological

age matched skilled readers. For each task, modified t-tests

(Crawford and Howell, 1998) were carried out to compare the

performance of LL with that of the control group. Scores of

the dyslexic participant are provided in Table 1 together

with the mean scores (and standard deviations) of the

control group.

Results show that LL’s non-verbal IQ was normal. Her

verbal short-term memory performance did not differ signif-

icantly from that of skilled readers in both forward and

backward digit spans, despite a trend for poor working

memory skills. With respect to reading, LL performed within

the normal range for both irregular word and pseudo-word

reading accuracy. She read both types of items more slowly

than the control readers but the difference only reached

significance for the pseudo-words [tmodified(19)¼�2.33,

p< .05]. She otherwise demonstrated very good knowledge of

grapheme-phoneme conversion rules.

Table 1 – Performance of LL, in reading, spelling,phonological tasks and VA span tasks as comparedto chronological age controls (n [ 20).

LL Controls

Performance[T value]

Mean (SD)

Raven matrices (raw score) 54 [�.05] 54.20 (3.8)

Verbal short-term memory

Digit span forward 6 [�.09] 6.58 (1.16)

Digit span backward 3 [�1.35] 5.41 (1.73)

Reading

Irregular words (/40)

Accuracy 38 [þ.06] 36.80 (1.94)

Rate 32 [�1.29] 23.75 (6.20)

Pseudo-words (/40)

Accuracy 38 [þ.06] 37.90 (1.62)

Rate 56 [�2.33]* 37.75 (7.64)

GPC knowledge (/51) 51 [þ1.14] 48.50 (2.13)

Spelling

Consistent words (/40) 37 [þ1.33] 36.35 (1.90)

Inconsistent words (/40) 22 [�1.43] 29.55 (5.53)

Pseudo-word (/40) 33 [�2.42]* 37.45 (2.4)

Phonological processing

Pseudo-word repetition (/92) 83 [�3.13]* 89.45 (2.01)

Phoneme segmentation (/28) 14 [�1.62] 21.90 (4.74)

Spoonerisms (/10) 1 [�5.22]* 8.50 (1.40)

VA span

Global report – 5-letters (/100) 58 [�.92] 73.10 (16.01)

Global report – 6-letters (/144) 115 [�.30] 119.75 (15.49)

Partial report – 6-letters (/72) 48 [�1.40] 54.20 (7.17)

Note: GPC: grapheme–phoneme conversions. The asterisk (*) indi-

cates a significant difference at p< .05 between LL and the control

group.

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1234

LL was further asked to spell 80 words – 40 regular words

(e.g., /kameRa/ / camera; /dRam/ / drame) and 40 irregular

words including one or several phonemes associated to

infrequent graphemes (e.g., /arne/ / harnais, /solanel/ /

solennel; /kwan/ / couenne) – and 40 legal (2–4 syllables long)

pseudo-words. As shown in Table 1, LL’s spelling performance

was similar to that of skilled readers for the real words

whatever their regularity but she performed below the norm

in pseudo-word spelling [tmodified(19)¼�2.42, p< .05].

Her phonological abilities were assessed using three orig-

inal tasks of pseudo-word repetition, phoneme segmentation

and spoonerisms. In the pseudo-word repetition task, she was

asked to repeat 92 pseudo-words which were uttered in turn

by the experimenter. The 1–4-syllable-long pseudo-words

were derived from real words by substituting the vowels

(e.g., catastrophe/katastrOf/ / cotastrephe/kotastRef/); most

pseudo-words included a consonant cluster. In the phoneme

segmentation task, LL had to segment all the phonemes from

a real word and produce the resulting phonemes series (e.g.,

/golf/ / /g/þ /o/þ /l/þ /f/). The segmentation task included

28 words which were pronounced by the experimenter. The

spoonerism task required exchanging the first phonemes of

two heard words (e.g., mouton /muto/ – tulipe /tylip/ / /tuto/

– /mylip/). The test comprised 10 series of 2 words made up of

4.8 phonemes on average (range 4–6). LL performed poorly on

all three phonological tasks but her performance was signifi-

cantly lower than the controls on the spoonerism and pseudo-

word repetition tasks only (for both t-values, ps< .05).

The participant undertook three tasks of global and partial

letter-string report inspired from those previously used by

our team to assess VA span abilities in control and dyslexic

children (Valdois et al., 2003; Bosse et al., 2007; Prado et al.,

2007; Lassus-Sangosse et al., 2008; Bosse and Valdois, 2009).

The tasks were adapted for the assessment of young adults

by either reducing presentation time or increasing letter-

string length. In all three tasks, stimuli were random conso-

nant strings presented in upper-case in black on a white

background. At the start of each trial, a central fixation point

was displayed for 1000 msec followed by a blank screen for

50 msec. A letter string was then briefly presented horizon-

tally, centred on the fixation point. In the 5-letters global

report task, the 20 letter strings (e.g., R H S D M) were dis-

played for 100 msec, immediately followed by a mask (a

series of 5 asterisks). In the 6-letters global report task, the 24

letter strings (e.g., B F D S L R) were displayed for 200 msec

without mask. In the two tasks of global report, the partici-

pants had to recall as many letters as possible (identity not

location) immediately after their presentation. In the partial

report task, 72 6-letter strings were displayed for 200 msec. At

the offset of the letter string, the cue – a vertical bar –

appeared for 50 msec under one of the letters. The partici-

pants had to report the cued letter only. In each task, scores

correspond to the total number of accurately (identity not

location) reported letters. LL was found to score within the

normal range on all three VA span tasks (for all t-values,

ps> .05).

To summarize, neuropsychological assessment revealed

that LL’s performance in reading and spelling was charac-

terised by slowed pseudo-word reading rate and poor

pseudo-word spelling. Her performance pattern suggests

selective impairment of the analytic spelling procedure and

some weakness of the analytic reading procedure, as typi-

cally reported in well compensated phonological dyslexics.

Her reading pattern is typical of developmental phonological

dyslexia in more transparent languages than English in

which pseudo-word reading speed is typically reported

as impaired whereas reading accuracy is preserved (see

Wimmer, 1993 for German and Sprenger-Charolles et al.,

2000 for French; or Share, 2008 for a critical review).

Assessment further revealed poor pseudo-word repetition

and poor phoneme awareness skills, thus reflecting an

underlying phonological disorder. In contrast, LL showed

normal simultaneous processing of letter strings, thus sug-

gesting preserved VA span abilities. Overall, LL is a very pure

case of developmental phonological dyslexia, characterised

by poor pseudo-word reading and spelling performance and

an underlying phonological disorder. More importantly, LL

does not suffer from any VA span disorder making her case

particularly relevant to study the impact of phonological

disorder on amodal sequential attentional processing in

developmental dyslexia.

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1 235

2.2. Apparatus and stimuli

As is usual in the experimental condition, the AB was

measured with a dual task procedure consisting of two

targets: the first one (T1) to be identified and the second (T2) to

be detected. A single condition was further administered in

which the participants only had to detect T2. It served as

a control condition to ensure that the participants were able to

correctly process T2 when it was not preceded by the identi-

fication of another target.

2.2.1. The Visual AB taskThe stimuli were black or red digits (1–9) – 12 Arial font sub-

tending approximately .7� � .7�of visual angle – presented on

a grey (192, 192, 192) background. The digits were displayed in

rapid serial visual presentations in the same location at the

centre of the screen. The target to be identified (T1) was

the only red digit in the stream and it was either a ‘‘1’’ or a ‘‘5’’.

The target to be detected (T2) was a ‘‘0’’ and it was black as the

distracters. Thus, T1 differed qualitatively from the set of

distracters by a colour change. Each digit was exposed for

40 msec with an ISI of 60 msec, yielding a stimulus rate

presentation of 10 digits per second. Stimuli were presented

using E-prime software on a PC computer running Windows

2000. The computer screen was a 17-in. and had a refresh rate

of 85 Hz. The viewing distance was set to 60 cm.

2.2.2. Auditory AB taskRapid serial auditory stimuli presentations consisted of

a sequence of sounds digitally edited to 16-bit resolution at

a sampling rate of 44 kHz using Sound Forge 8.0. Twenty-two

pure tones were used as distracters and they were ranging

from 293 Hz to 1588 Hz (293, 320, 348, 380, 415, 452, 472, 493,

515, 561, 612, 668, 728, 794, 866, 944, 1030, 1123, 1224, 1335,

1456, and 1588 Hz). A higher-pitched tone of 4000 Hz was used

as T1 target. This tone was either a complex tone (square tone)

or a pure tone. Thus, T1 noticeably differed qualitatively from

the set of distracters (higher in pitch). T2 was a pure tone of

600 Hz belonging to the distracters’ frequency range but it was

Fig. 1 – Procedure of the rapid serial visual

delivered at a higher amplitude level (12% of 65 dB level higher

than the other stimuli). All tones lasted 40 msec (including

5 msec linear onset/offset amplitude ramps in order to

prevent onset and offset clicks); they were separated by silent

gaps of 60 msec, yielding a stimulus rate presentation of 10

tones per second. Stimuli were presented using E-prime soft-

ware on a PC computer running the rapid serial auditory

sequences binaurally through headphones (earthquake, TS

800) at approximately 65 dB sound pressure level.

2.3. Procedure

Fig. 1 illustrates the procedure for the visual (see Fig. 1A) and

the auditory modality (see Fig. 1B). Each participant completed

two blocks of 96 trials in each modality. Each block corre-

sponded to a single condition (96 trials) in one modality fol-

lowed by a dual condition (96 trials) in the same modality. In

the single condition, no T1 target was displayed and the

participants only had to decide whether T2 occurred in the

stream. In the dual condition, T1 was always present and T2

occurred in half of the trials. Participants were instructed to

attend to and report T1 (red digit number ‘‘1’’ or ‘‘5’’ for the

visual modality and the pure or complex tones for the auditory

modality) while judging whether T2 occurred or not (the black

digit number ‘‘0’’ for the visual modality and the louder sound

for the auditory modality). Each trial consisted of either 15, 19

or 23 items (digits or tones). In the dual condition, T1 always

appeared within the stream (‘‘real’’ position of T1). The single

condition was identical to the dual condition except that T1

was omitted and replaced by a black digit or a pure tone

(‘‘virtual’’ position of T1). The position of T1 was randomly

permuted within trials so that it appeared an equal number of

times in positions 7, 11 or 15. The stimuli on a given trial were

randomly generated by the computer under the constraint

that the same digit or tone could not appear in the previous

four positions. On a given dual condition trial, T1 was

randomly chosen between the two digits 1 and 5 for the visual

task and between the pure and square tones for the auditory

task. For both the single and the dual conditions, eight lags

(A) and auditory (B) presentation tasks.

Table 2 – Percentage of T2 detection (present trials only) inthe single conditions as a function of the virtual T1–T2 lagforthecontrolgroup (n [ 27)andthedyslexicparticipantLL.

Single condition: virtual lag

1 2 3 4 5 6 7 8

Visual task

Mean 95.1 97.5 96.3 93.8 97.5 98.1 93.2 54.9

SD 8.9 5.9 6.9 12.1 5.9 5.2 10.4 17.5

LL 100 100 100 100 83* 100 83 83

Auditory task

Mean 79.0 84.6 82.7 83.3 86.4 90.7 91.4 92.0

SD 16.7 18.1 17.8 23.6 19.8 15.9 12.3 14.6

LL 100 100 100 100 100 100 100 100

Note: the asterisk (*) indicates a significant difference at p< .05

between LL and the control group (n¼ 27).

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1236

between T1 and T2 positions, from lag 1 (no intervening items,

Stimulus Onset Asynchrony (SOA)¼ 100 msec) to lag 8

(SOA¼ 800 msec), were crossed with the three serial ‘‘real’’ or

‘‘virtual’’ positions of T1, for a total of 96 trials, with 12 trials at

each lag. T2 absent and T2 present trials were randomized.

Thus, the experimental design of each block was as follows: 48

sequences [3 (pre-T1 fillers)� 8 (lag)� 2 (T2 presence)] pre-

sented twice. One practice block of 15 trials (where a feedback

was given) was followed by the single block. The order of the

tasks (visual or auditory task) was counterbalanced between

subjects. For the auditory task a learning phase was con-

ducted for each participant before beginning the test. First,

a T2 learning phase was conducted to ensure that the partic-

ipants differentiated the louder tone from the distracters. This

training phase corresponded to 14 single condition trials. This

block continued until the participant reached 75% accuracy.

After this first training phase the participants were trained to

identify the three different auditory targets (T2: the ‘‘louder’’

tone, the first T1 target: the ‘‘4000 Hz pure tone’’ and the

second T1 target: the ‘‘4000 Hz square tone’’).

The experiment was conducted in a dimly lit room. The

experimenter initiated each trial by pressing the space bar on

the computer keyboard. Then, a cross lasting 500 msec

appeared at the centre of the monitor screen for fixation. One

hundred milliseconds after the offset of the fixation cross the

stream of stimuli were presented successively at the centre of

the screen (visual task) or binaurally through the headphones

(auditory task). At the end of the trial, the experimenter had to

press the space bar on the computer keyboard to continue the

experiment. Immediately after each trial, participants were

instructed to report aloud whether T2 was present or not

(‘‘yes’’ answer or ‘‘no’’ answer). In the dual condition, they had

to name T1. The experimenter entered the responses on the

computer; no feedback was given.

3. Results

3.1. Single conditions

For the two single conditions (visual and auditory), mean

target detection accuracy was calculated for each virtual lag

for the dyslexic participant and for the control group. Three

control subjects were excluded because of their poor ability to

detect T2 (with a score 2 standard deviations below the

average in either the visual or the auditory modality). Table 2

provides T2 detection accuracy scores (for the present trials)

as a function of lag for both the control group and the dyslexic

participant in the visual and the auditory modality.

First, performance was high in the single conditions for

both the control group (mean accuracy of 91% and 86% in the

visual and auditory tasks, respectively) and LL (94% and 100%).

The participants thus demonstrated very good abilities for

identifying a single target in both the auditory and visual rapid

serial presentation streams. The decline of performance

observed at lag 8 for the control group in the visual condition

might reflect the drop of sustained attentional resources,

leading them to omit the target when it appeared at the end of

the stream. For each single condition (visual and auditory),

modified t-tests (Crawford and Howell, 1998) were carried out

to compare the performance of LL with that of the control

group at each lag. LL performed as well as the controls (all

ps> .05) in both the visual and the auditory modality, except

for virtual lag 5 ( p< .05) in the visual condition. A ceiling effect

characterised LL performance in the auditory modality and

only one control participant performed as well as LL, in this

task. As LL largely conformed to the control group perfor-

mance, we chose to consider performance of the control group

as the baseline on which to compare LL’s scores in the dual

conditions.

3.2. Dual conditions: AB assessment

There is evidence for an AB when the detection of T2 is

significantly lower in the dual condition (in which subjects

have to identify T1 first) than in the single condition according

to the position of T2 (lag) in the stream. Thus, a significant

condition by lag interaction is the signature of the AB.

Statistical analyses were conducted on those trials for which

T1 was accurately identified. This corresponded to 98% of the

trials in each modality for the control participants. The

dyslexic participant accurately identified 99% T1 targets in

vision and 100% in audition. Table 3 provides T2 detection

accuracy rates (when T1 was well reported) as a function of lag

for the control group and the dyslexic participant.

3.2.1. The control groupFor each modality, an Analysis of Variance (ANOVA) with

condition (single, dual) and lag (1, 2, 3, 4, 5, 6, 7, 8) as within-

subjects factors was carried out on T2 detection rate. In

both the visual and the auditory modalities, significant

main effects of lag [vision: F(7,182)¼ 29.37, p< .001; audition:

F(7,182)¼ 13.54, p< .01] and condition [vision: F(1,26)¼ 15.32,

p< .01; audition: F(1,26)¼ 11.43, p< .01] were found. T2 detec-

tion was easier in the single condition than in the dual condi-

tion for the visual (90.8% vs. 83.7%) and the auditory (86.3% vs.

79.8%) modalities, suggesting greater difficulty when two

targets had to be processed. In the visual modality, the condi-

tion by lag interaction was significant [F(7,182)¼ 16.30,

p< .001]. Post-hoc comparison analysis (Newman–Keuls)

showed that the control group performed significantly lower

Table 3 – Percentage of T2 detection (when T1 was wellreported) in the dual conditions as a function of lag for thecontrol group and the dyslexic participant LL.

Dual condition: lag

1 2 3 4 5 6 7 8

Visual task

Mean 82.1 67.0 79.9 83.5 92.3 95.1 96.7 73.5

SD 19.2 29.9 17.2 15.4 14.4 8.9 8.4 19.1

LL 50* 33* 50* 67* 83* 80* 100 67

Auditory task

Mean 59.4 73.1 74.1 79.3 85.6 90.9 86.3 89.6

SD 25.6 23.6 23.9 24.3 22.6 13.4 17.6 15.7

LL 17* 17* 83 100 100 100 83 100

Note: digits in italic indicate lags for which an AB was observed in

the controls. The asterisk (*) indicates a significant difference at

p< .05 between the dual condition in LL and the single condition in

the control group (n¼ 27).

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1 237

in the dual than in the single condition at lags 1, 2, 3 and 4

(all ps< .05). Control participants thus exhibit a visual AB width

for approximately 400 msec following T1. In the auditory

modality, a significant condition by lag interaction was also

found [F(7,182)¼ 3.11, p< .01]. Post-hoc comparison analyses

(Newman–Keuls) showed that the condition effect was signif-

icant at lags 1 and 2 (all ps< .05) corresponding to an auditory

AB width of approximately 200 msec.

3.2.2. Case LLIt is first noteworthy, as shown on Fig. 2, that LL curve patterns

differ from those of the controls in the dual condition for both

modalities, even if the difference is stronger in the auditory

modality in which LL shows a sharp drop of performance with

lag. For each modality and each lag, modified t-tests were

carried out to compare LL’s performance in the dual condition

with performance of the controls in the single condition. In the

visual modality, LL’s performance was impaired at lags 1, 2, 3,

4, 5, and 6 (all ps< .05). She only reached control performance

Fig. 2 – Visual (A) and auditory (B) AB in the control group (n [ 27

White dots are for the single condition, black dots for the dual

depicted for each lag in both conditions.

at lag 7 ( p> .05). This indicates that LL exhibits a visual AB

prolonged up to 600 msec after T1 apparition, thus 200 msec

longer than for the controls. In the auditory modality, LL’s

performance was impaired at lags 1 and 2 (both ps< .05) but

she performed as the controls from lag 3 upwards. This result

indicates that LL exhibits an auditory AB lasting 200 msec after

T1 apparition, thus similar in length to the controls. However,

when focusing on the sequential processing of two successive

lags for the control group and for LL separately, data suggest

that LL’s performance is characterised by a longer auditory AB

as a priori expected in the framework of the SAS theory. In the

control group, a statistically significant contrast was found

between lag 1 and lag 2 ( p< .05) whereas no such difference

was found later on [between lags 2 and 3, lags 3 and 4, lags 4 and

5, lags 5 and 6, lags 6 and 7 or lag 7 and 8 (all ps> .05)]. These

results suggest that performance of the control group signifi-

cantly improved between lag 1 and lag 2 only, thus reflecting

optimal processing of the second target from a 200 msec gap. In

contrast, LL’s performance was very low at both lag 1 and lag 2

(17%). Her performance did not improve at lag 2 but strongly

increased (17%–80%) from lag 2 to lag 3. Therefore, LL exhibited

optimal processing of the second target from a 300 msec gap

only, thus 100 msec later than for the controls. This later result

might reflect prolonged auditory AB in LL as compared to the

control group.

Finally, we analysedand comparedthe AB amplitude (costof

the dual condition) in the controls and in LL to assess in what

extent the identification of T1 had a negative impact on T2

detection. Modified t-tests were used to compare differences in

performance between the single and dual conditions in LL and

the control group at each lag and for each modality. In the visual

modality, no significant difference was found between LL and

the controls at any lag (all ps> .05). The analysis however,

revealed a tendency for deeper in amplitude visual AB, at lag 1

[tmodified(26)¼ 1.89, p¼ .07], and lag 3 [tmodified(26)¼ 2.03,

p¼ .053] in LL. In the auditory modality, the AB amplitude was

significantly deeper in LL than in the controls at lags 1 and 2

(both ps< .05). It thus clearly appears more difficult for LL than

the controls to process a second target that was displayed

during the blink period, in the auditory domain.

) and in the phonological dyslexic patient, LL (dotted lines).

condition. For the control group, standard error bars are

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1238

4. Discussion

As expected, the control group in the current study was blind

to the second to-be-detected visual target when it was pre-

sented within 400 msec after the first to-be-identified target.

The estimated width of their AB thus conforms to that typi-

cally reported in the visual modality (from 400 msec to

600 msec; Raymond et al., 1992; Hari et al., 1999). In the same

way, an AB of 200 msec was found in the auditory modality,

well in accordance with previous findings (Shen and Mondor,

2006). As previously reported in dyslexic individuals (in adults:

Hari et al., 1999, or Buchholz and Davies, 2007; and in children:

Visser et al., 2004, or Facoetti et al., 2008), LL visual AB was

200 msec longer than in the controls, suggesting prolonged

visual attentional dwell time (Duncan et al., 1994; Hari et al.,

1999; Ward et al., 1996), thus a difficulty to disengage her

attention from the first to-be-identified target.

This prolonged visual AB suggests a difficulty to process

rapidly presented visual material in a participant, LL, who

otherwise exhibits phonological processing difficulties which

might be expected to more straightforwardly relate to an

auditory processing disorder. This finding is however,

consistent with Lum et al. (2007)’s study showing a prolonged

and deeper visual AB in adolescents with specific language

impairment. Such a co-occurrence of language/phonological

problems and prolonged visual AB would be expected only in

the context of an amodal SAS, affecting both the visual and

the auditory modalities. Accordingly, the SAS amodality

hypothesis (Hari and Renvall, 2001) predicts that LL should

further exhibit an atypical AB in the auditory modality. As

expected, we found that LL’s performance was far from

normal in the auditory modality but it differed from that

observed in the visual modality and was primarily charac-

terised by a deeper in amplitude rather than prolonged AB.

However, as in vision, her auditory deficit obviously resulted

from the task constraint to process two successive rapidly

presented auditory targets since LL demonstrates top-level

abilities when asked to detect T2 targets in the absence of T1.

Thus, deeper amplitude of the auditory AB in LL obviously

corresponds to the cost of the dual condition compared to the

single condition on T2 detection performance.

This study is the first report of auditory AB data in devel-

opmental dyslexia and for the first time the AB was here

investigated in both the auditory and the visual modality.

Although atypical AB patterns were found in both modalities

in LL, these patterns substantially differed from one modality

to the other, thus questioning whether performance reflected

similar SAS in both modalities. At this point, we have to take

into account that the visual and the auditory modalities

strongly differ in temporal processing and that the AB tasks

when used in these two modalities did not necessarily give rise

to the same results, even in control individuals. The auditory

system has a much better temporal resolution than the visual

system. This is a first general issue which could have yielded

differences in the expression of the visual and auditory AB

deficits in LL (Vachon and Tremblay, 2008). For example, it

might be assumed that our lag duration (100 msec) was too

long to highlight a temporal deficit in the ‘rapid’ auditory

system. Thus, even in a pathological context, a prolonged AB

might not have been reflected by a T2 deficit spreading up to

later lags. Perhaps it is that more sensitive variables should be

looked at to highlight such a deficit in the auditory modality.

Moreover, a shorter AB is typically reported in the auditory

modality when pure tones are used (Shen and Mondor, 2006;

Vachon and Tremblay, 2005, 2006) as compared to the visual

modality, and our own findings reveal a lag 1 sparing effect in

the visual modality but not in the auditory modality for the

controls as for LL. This latter asymmetry has been already

noted in the literature (Tremblay et al., 2005; Shen and

Mondor, 2006; Vachon and Tremblay, 2006). Lag 1 sparing

refers to the finding that T2 visual detection is only slightly

impaired when it is temporally adjacent to T1 (at lag 1) as

compared to the maximally impaired T2 detection at lag 2

when T1 and T2 are separated by an intervening distracter.

This result has been explained in terms of sluggish attentional

gate that opens for T1 but then closes slowly, thus allowing T2

to gain access to processing when immediately following T1.

However, even in the visual modality, the lag 1 sparing effect

is abolished under some specific experimental conditions. In

their review of AB studies, Visser et al. (1999) found that lag 1

sparing never occurred when T1 and T2 were presented in

different spatial locations (see also, Visser et al., 2004). In our

study and for the visual modality, T1 and T2 at lag 1 appeared

at the same spatial location so that they fell into the same

visual attentional capture, thus yielding simultaneous pro-

cessing of the two targets. Tremblay et al. (2005) proposed that

the same rule applies to the auditory modality except that the

spatial distance in vision would correspond to the pitch

distance in audition. Accordingly, a lack of lag 1 sparing in the

auditory modality might have been caused by the cost of

reallocating attention to different frequency ranges (from

4000 Hz for T1 to 600 Hz for T2) in a very short time (100 msec).

Accordingly, the auditory AB task might have required

sequential processing of the targets from lag 1 due to rapid

shifting from a high-pitch tone to a low-pitch tone instead of

simultaneous processing as in the visual modality.

Obviously, our statistical results primarily reveal a deficit

in depth in LL’s auditory AB. However, when focusing on

sequential processing differences between successive lags for

the control group and for LL, LL was found to exhibit optimal

processing of the second auditory targets from lag 3, thus

100 msec later than the controls. Her performance thus

suggests both deeper and prolonged AB in the auditory

modality.

Interestingly, this sequential deficit was also highlighted

by Helenius et al. (1999) in young adults in an auditory stream

segregation task, and interpreted as evidence for an auditory

attentional shifting deficit by Hari and Renvall (2001) them-

selves. The auditory stream segregation task evaluated the

speed that participants could disengage their attentional

focus from one stimulus, in order to process the next different

one. For long SOAs, sequences of alternating high (1000 Hz)

and low pitch (400 Hz) tones are perceived as ‘‘connected’’,

whereas at short SOAs, they are perceived as segregating into

two streams of different frequency (a high one and a low one).

The authors found that the streams segregated at 210 msec for

the dyslexic participants whereas the tone sequence was

perceived as connected down to 130 msec in skilled readers.

Hari and Renvall (2001) interpreted these findings as showing

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1 239

that dyslexic participants could not disengage their atten-

tional focus as fast as control participants in the auditory

modality. They claimed that aberrant segregation of sound

streams in dyslexic subjects was evidence for prolonged

attentional dwell time. Actually, we can assume that LL’s

auditory AB amplitude deficit at lag 1 and lag 2 might have

been caused by a lower auditory segregation threshold as

compared to skilled readers. If the auditory AB amplitude

indeed reflects differences of segregation thresholds in LL and

the controls, then higher AB amplitude should be observed at

lag 1 as compared to the subsequent lags in the controls.

Consequently, lower performance on T2 detection at lag 1

might correspond to a segregation threshold superior to 100

but inferior to 200 msec. This is consistent with Helenius et al.

(1999)’s study in which segregation threshold was estimated

around 130 msec. In the same line of reasoning, LL segregation

threshold should occur between 200 and 300 msec here again

in accordance with Helenius et al.’s findings who found that

the dyslexic auditory segregation threshold occurred for

a SOA of 210 msec. If the AB amplitude in the auditory

modality is another manifestation of prolonged attentional

dwell time then the observation of prolonged duration of the

AB in the visual modality and higher amplitude in the auditory

modality might be different illustrations of SAS in our dyslexic

participant.

Moreover, the engagement/disengagement deficit postu-

lated by the SAS theory of Hari and Renvall could be related to

the opening/closing of the attentional gate in the model of Li

et al. (2004). These authors suggest that a slower re-opening of

the gate could lead to a deeper AB reflecting limited attention

capacity. It seems reasonable to hypothesise that limited

capacity as well as slower re-opening of the gate would

prevent efficient shifting from one target to the other. In both

modalities, the observed pattern suggests an attention

disorder. Moreover, Cousineau et al. (2006) proposed a new

analysis framework of the AB phenomenon. This analysis

permits to characterize the AB according to four parameters

(lag 1 sparing, width – i.e., duration – depth and amplitude).

The authors showed that the width and depth parameters

tightly correlated. Thus, considering only one dimension (i.e.,

width) apart from the others (i.e., depth) when trying to study

the AB might be too simplistic. In the current paper, a wider

visual AB in LL was associated with a marginally significant

deeper visual AB on two of the lags at which the control group

exhibited an AB (lag 1 and lag 3). Similarly, although LL’s

auditory AB was primarily deeper in amplitude, her

improvement of successive target processing was delayed,

thus suggesting prolonged AB in the auditory modality as well.

As such, the overall findings suggest that similar mechanisms

were involved in both modalities, thus providing support to

the SAS theory.

Overall, the current findings suggest that LL exhibits

a perceptual attention disorder characterised by prolonged

attentional dwell time in both the auditory and visual

modalities. It is further noteworthy that evidence for visual

attentional sequential deficit was present in the absence of

any visual simultaneous attentional processing disorder, thus

in the context of preserved VA span abilities. These findings

are in line with previous data suggesting that sequential and

simultaneous visual processing do dissociate in dyslexic

individuals (Lassus-Sangosse et al., 2008). Moreover, VA span

disorders have been previously reported in dyslexic individ-

uals who were free from phonological problems (Bosse et al.,

2007). It has also been demonstrated that VA span abilities

primarily contribute to irregular word reading in typically

developing children (Bosse and Valdois, 2009). At the theo-

retical level and by reference to the multitrace memory model

of reading (Ans et al., 1998), a VA span disorder is expected to

primarily affect the acquisition of specific orthographic

knowledge (Valdois et al., 2004). A reduced VA span would

thus primarily result in poor irregular word reading perfor-

mance. The current data extend previous findings in sug-

gesting that poor VA span and SAS are distinct disorders.

However, a definite conclusion about the independence of

these two disorders based on a single-case study would

certainly be premature and, future studies on larger samples

of dyslexic children would be required before concluding.

SAS characterised by poor spatial (Facoetti et al., 2006) and

non-spatial (Facoetti et al., 2008) visual attentional orienting

abilities has been reported in dyslexic individuals with

phonological problems. This disorder was found to predict

pseudo-word reading performance after controlling for

phonological skills (Facoetti et al., 2006). Accordingly, Facoetti

et al. hypothesised that visuo-spatial attention was involved in

the analytic procedure of reading independently of phonolog-

ical mechanisms (see Perry et al., 2007 for a theoretical

account). In line with this account, SAS in the visual modality

was found to co-occur with selectively poor pseudo-word

reading in our dyslexic participant. The current data thus

suggest that two kinds of attention disorder might indepen-

dently contribute to reading acquisition. Thus, sluggish visual

attentional shifting would more specifically relate to analytic

processing and pseudo-word reading through graphemic

parsing. In contrary Bosse and Valdois (2009) recently showed

VA span would primarily contribute to global processing and

irregular word reading in allowing the whole sequence of input

words to be processed simultaneously. However, evidence for

a direct link between sluggish visual attentional shifting and

pseudo-word reading is only supported by some data and

additional studies are required to firmly establish this rela-

tionship while excluding the additional and well documented

impact of the phonological disorder on pseudo-word reading.

In addition, a theoretical link has been established between VA

span abilities and pseudo-word reading (Valdois et al., 2004,

2006). This link, which dissociates from phonological abilities,

is supported by empirical data from groups of normally devel-

oping (Bosse and Valdois, 2009) and dyslexic children (Bosse

et al., 2007). Future studies are needed to clarify the relevant

impact of VA span and visual attentional shifting abilities on

pseudo-word reading.

5. Conclusion

For the first time the magnitude of the AB was simultaneously

measured in the visual and the auditory modalities in a young

adult phonological dyslexic participant, who further showed

normal VA span abilities. Our purpose was to assess the

amodal hypothesis of the SAS theory of dyslexia, proposed by

Hari and Renvall (2001), together with investigating whether

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1240

visual attentional shifting and VA span abilities are indepen-

dent mechanisms. The investigation revealed two different

kinds of deficits – a prolonged AB in the visual modality and an

increase in amplitude AB in the auditory modality – that were

interpreted as reflecting prolonged attentional dwell time in

both modalities. Accordingly, the current findings are direct

evidence for an amodal disorder affecting rapid processing of

stimulus sequences in developmental dyslexia as postulated

by the SAS theory. Furthermore, this amodal disorder was

reported in a dyslexic participant with a phonological disorder,

in accordance with the SAS theory hypothesis that sluggish

shifting of auditory attention might contribute to difficulties in

achieving phoneme awareness and in the acquisition of

literacy. Moreover the visual AB deficit was found indepen-

dently of any VA span disorder, suggesting that these two types

of attentional processing can dissociate and might indepen-

dently contribute to developmental dyslexia. Additional

studies conducted on homogeneous groups of dyslexic indi-

viduals with and without phonological or VA span problems

are required to better define the scope of the SAS theory.

Acknowledgements

This research was supported by the CNRS (Centre National de

la Recherche Scientifique). Marie Lallier is chercheur-boursier

of the French Research Ministry. We thank LL and all the

control participants for their help and participation in this

research. We are grateful to Francois Vachon and Nicholas

Peatfield for their valuable advice and helpful comments

during the preparation of this article.

r e f e r e n c e s

Amitay S, Ben-Yehudah G, Banai K, and Ahissar M. Disabledreaders suffer from visual and auditory impairments but notfrom a specific magnocellular deficit. Brain, 125: 2272–2285,2002.

Ans B, Carbonnel S, and Valdois S. A connectionist multi-tracememory model of polysyllabic word reading. PsychologicalReview, 105: 678–723, 1998.

Arnell K and Jolicoeur P. The attentional blink across stimulusmodalities: evidence for central processing limitations. Journalof Experimental Psychology: Human Perception and Performance, 25:105–147, 1999.

Arnell K and Larson JM. Cross-modality attentional blinkswithout preparatory task-set switching. Psychonomic Bulletin &Review, 9: 497–506, 2002.

Ben-Yehudah G, Sackett E, Malchi-Ginzberg L, and Ahissar M.Impaired temporal contrast sensitivity in dyslexics is specificto retain-and-compare paradigms. Brain, 124: 1381–1395, 2001.

Bosse ML, Tainturier MJ, and Valdois S. Developmental dyslexia:The visual attention span deficit hypothesis. Cognition, 104:198–230, 2007.

Bosse ML and Valdois S. Influence of the visual attention span onchild reading performance: A cross-sectional study. Journal ofResearch in Reading, 32: 230–253, 2009.

Brady SA and Shankweiler D. Phonological Processes in Literacy: ATribute To Isabelle Y. Liberman. Hillsdale, New Jersey: LawrenceErlbaum Associates Publishers, 1991.

Bretherton L and Holmes VM. The relationship between auditorytemporal processing, phonemic awareness, and readingdisability. Journal of Experimental Child Psychology, 84: 218–243,2003.

Broadbent DE and Broadbent MH. From detection to identification:Response to multiple targets in rapid serial visual presentation.Perception & Psychophysics, 42: 105–113, 1987.

Buchholz J and Davies AA. Attentional blink deficits observed indyslexia depend on task demands. Vision Research, 47: 1292–1302, 2007.

Chiappe P, Stringer R, Siegel LS, and Stanovich KE. Why thetiming deficit hypothesis does not explain reading disabilitiyin adults. Reading and Writing: An Interdisciplinary Journal, 15:73–107, 2002.

Chun MM and Potter MC. A two-stage model for multiple targetdetection in rapid serial visual presentation. Journal ofExperimental Psychology: Human Perception and Performance, 21:109–127, 1995.

Crawford JR and Howell DC. Comparing an individual’s test scoreagainst norms derived from small samples. The ClinicalNeurospychologist, 12: 482–486, 1998.

Cousineau D, Charbonneau D, and Jolicoeur P. Parameterizing theattentional blink effect. Canadian Journal of ExperimentalPsychology, 60: 175–189, 2006.

Dubois M, Lafaye de Micheaux P, Noel MP, and Valdois S. Pre-orthographical constraints on visual word recognition:Evidence from a case study of developmental surface dyslexia.Cognitive Neuropsychology, 24: 623–660, 2007.

Duncan J, Ward R, and Shapiro KL. Direct measurement ofattentional dwell time in human vision. Nature, 369: 313–315,1994.

Duncan J, Martens S, and Ward R. Restricted attentional capacitywithin but not between sensory modalities. Nature, 387:808–810, 1997.

Facoetti A, Lorusso ML, Cattaneo C, Galli R, and Molteni M. Visualand auditory attentional captures are both sluggish inchildren with developmental dyslexia. Acta NeurobiologiaeExperimentalis, 65: 61–72, 2005.

Facoetti A, Ruffino M, Peru A, Paganoni P, and Chelazzi L. Sluggishengagement and disengagement of non-spatial attention indyslexic children. Cortex, 44: 1221–1233, 2008.

Facoetti A, Zorzi M, Cestnick L, Lorusso ML, Molteni M,Paganoni P, et al. The relationship between visuo-spatialattention and nonword reading in developmental dyslexia.Cognitive Neuropsychology, 23: 841–855, 2006.

Farmer ME and Klein RM. The evidence for a temporal processingdeficit-linked to dyslexia. Psychonomic Bulletin & Review, 2:469–493, 1995.

Frith U. Beneath the surface of developmental dyslexia. InPetterson KE, Marshall JC, and Coltheart M (Eds), SurfaceDyslexia. London: Routledge and Kegan Paul, 1995.

Frith U. Brain, mind and behaviour in dyslexia. In Hulme C andSnowling M (Eds), Dyslexia: Biology, Cognition and Intervention.London: Whurr Publishers, 1997.

Gathercole SE and Baddeley AD. Phonological memory deficits inlanguage disordered children: Is there a causal connection?Journal of Memory and Language, 29: 336–360, 1990.

Hari R and Kiesila P. Deficit of temporal auditory processing indyslexic adults. Neuroscience Letters, 205: 138–140, 1996.

Hari R, Valta M, and Uutela K. Prolonged attentional dwell time indyslexic adults. Neuroscience Letters, 271: 202–204, 1999.

Hari R and Renvall H. Impaired processing of rapid stimulussequences in dyslexia. Trends in Cognitive Sciences, 5: 525–532,2001.

Hari R, Renvall H, and Tanskanen T. Left minineglect in dyslexicadults. Brain, 124: 1373–1380, 2001.

Helenius P, Uutela K, and Hari R. Auditory stream segregation indyslexic adults. Brain, 122: 907–913, 1999.

c o r t e x 4 6 ( 2 0 1 0 ) 2 3 1 – 2 4 1 241

Jolicoeur P and Dell’ Acqua R. The demonstration of short-termconsolidation. Cognitive Psychology, 36: 138–202, 1998.

Laasonen M, Service E, and Virsu V. Temporal order andprocessing acuity of visual, auditory, and tactile perception indevelopmentally dyslexic young adults. Cognitive, Affective andBehavioral Neuroscience, 1: 394–410, 2001.

Lassus-Sangosse D, N’Guyen-Morel MA, and Valdois S. Sequentialor simultaneous visual processing deficit in developmentaldyslexia. Vision Research, 48: 979–988, 2008.

Lefavrais P. Description, definition et mesure de la dyslexie:Utilisation du test ‘‘l’Alouette’’. [Describing, defining andmeasuring dyslexia using the ‘‘Alouette’’ test]. Revue dePsychologie Appliquee, 15: 33–44, 1965.

Livingstone MS, Rosen GD, Drislane FW, and Galaburda AM.Physiological and anatomical evidence for a magnocellulardeficit in developmental dyslexia. Proceedings of the NationalAcademy of Science, 88: 7943–7947, 1991.

Lovegrove W. Weakness in the transient visual system: A causalfactor in dyslexia? Annals of the New York Academy of Sciences,682: 57–69, 1993.

Lum JAG, Conti-Ramsden G, and Lindell AK. The attentional blinkreveals sluggish attentional shifting in adolescents with specificlanguage impairment. Brain and Cognition, 63: 287–295, 2007.

Marshall J, Snowling M, and Bailey CE. Rapid auditory processingand phonological ability in normal readers and readers withdyslexia. Journal of Speech, Language and Hearing Research, 4:925–940, 2001.

Meyler A and Breznitz Z. Visual, auditory and cross-modalprocessing of linguistic and nonlinguistic temporal patternsamong adult dyslexic readers. Dyslexia, 11: 93–115, 2005.

Mody M, Studdert-Kennedy M, and Brady S. Speech perceptiondeficits in poor readers: Auditory processing or phonologicalcoding? Journal of Experimental Child Psychology, 64: 199–231,1997.

Mondor TA. A transient processing deficit following selection of anauditory target. Psychonomic Bulletin & Review, 5: 305–311, 1998.

Nittrouer S. Do temporal processing deficits cause phonologicalprocessing problems? Journal of Speech Language and HearingResearch, 42: 925–942, 1999.

Pennington BF, Van Orden GC, Smith SD, Green PA, andHaith MM. Phonological processing skills and deficits in adultsdyslexics. Child Development, 61: 1753–1778, 1990.

Perry C, Ziegler JC, and Zorzi M. Nested incremental modelling ofthe development of computational theories: The CDPþ modelof reading aloud. Psychological Review, 114: 273–315, 2007.

Prado C, Dubois M, and Valdois S. Eye movements in readingaloud and visual search in developmental dyslexia: Impact ofthe VA span. Vision Research, 47: 2521–2530, 2007.

Raven JC. Standard Progressive Matrices: Sets A, B, C, D and E.London: HK Lewis, 1960.

Raymond JE, Shapiro KL, and Arnell KM. Temporary suppressionof visual processing in the RSVP task: An attentional blink?Journal of Experimental Psychology: Human Perception andPerformance, 18: 849–860, 1992.

Raymond JE, Shapiro KL, and Arnell KM. Similarity determinesthe attentional blink. Journal of Experimental Psychology: HumanPerception and Performance, 21: 653–662, 1995.

Shapiro KL, Raymond JE, and Arnell KM. Attention to visualpattern information produces the attentional blink in rapidserial visual presentation. Journal of Experimental Psychology:Human Perception and Performance, 20: 357–371, 1994.

Share DL. On the anglocentricities of current reading researchand practice. Psychological Bulletin, 134: 584–615, 2008.

Shen D and Mondor TA. Effect of distractor sounds on the auditoryattentional blink. Perception and Psychophysics, 68: 228–243, 2006.

Snowling M. Dyslexia. Oxford: Blackwell, 2000.Spinelli D, Angelelli P, De Luca M, Di Pace E, Judica A, and

Zoccolotti P. Developmental surface dyslexia is not associatedwith deficits in the transient visual system. NeuroReport, 8:1807–1812, 1997.

Sprenger-Charolles L, Cole P, Lacert PH, and Serniclaes W. Onsubtypes of developmental dyslexia: Evidence for processingtime and accuracy scores. Canadian Journal of ExperimentalPsychology, 54: 87–103, 2000.

Stein J and Walsh V. To see but not to read; the magnocellulartheory of dyslexia. Trends in Neuroscience, 20: 147–152, 1997.

Stein J. Visual motion sensitivity and reading. Neurospychologia,41: 1785–1793, 2003.

Tallal P. Auditory temporal perception, phonics, and readingdisabilities in children. Brain and Language, 9: 182–198, 1980.

Tallal P, Merzenich M, Miller S, and Jenkins W. Languagelearning impairments: Integrating basic science, technology,and remediation. Experimental Brain Research, 123: 210–219,1998.

Tremblay S, Vachon F, and Jones D. Attentional and perceptualsources of the auditory attentional blink. Perception &Psychophysics, 67: 195–208, 2005.

Vachon F and Tremblay S. Auditory attentional blink:Masking the second target is necessary, delayed masking issufficient. Canadian Journal of Experimental Psychology, 59:279–286, 2005.

Vachon F and Tremblay S. Delayed masking and auditoryattentional blink: A test for retrieval competition andbottlenecks models. Experimental Psychology, 53: 182–190, 2006.

Vachon F and Tremblay S. Modality specific and amodal sourcesof interference in the attentional blink. Perception &Psychophysics, 70: 1000–1015, 2008.

Valdois S, Bosse ML, Ans B, Zorman M, Carbonnel S, David D, et al.Phonological and visual processing deficits are dissociated indevelopmental dyslexia: Evidence from two case studies.Reading and Writing, 16: 543–572, 2003.

Valdois S, Bosse ML, and Tainturier MJ. Cognitive correlates ofdevelopmental dyslexia: Review of evidence for a selectivevisual attentional deficit. Dyslexia, 10: 1–25, 2004.

Valdois S, Juphard A, Baciu M, Ans B, Peyrin C, Segebarth C, et al.Differential length effect in reading and lexical decision:Convergent evidence from behavioural data, connectionistsimulations and functional MRI. Brain Research, 1085: 149–162,2006.

Vellutino FR, Fletcher JM, Snowling MJ, and Scanlon DM. Specificreading disability (dyslexia): What have we learned in the pastfour decades? Journal of Child Psychology and Psychiatry, 45: 2–40,2004.

Visser TAW, Bischof WF, and Di Lollo V. Attentional switching inspatial and non spatial domains: Evidence from theattentional blink. Psychological Bulletin, 125: 458–469, 1999.

Visser TAW, Boden C, and Giaschi DE. Children with dyslexia:Evidence for visual attention deficits in perception of rapidsequences of objects. Vision Research, 44: 2521–2535, 2004.

Ward R, Duncan J, and Shapiro K. The slow time-course of visualattention. Cognitive Psychology, 30: 79–109, 1996.

White S, Milne E, Rosen S, Hansen P, Swettenham J, Frith U, et al.The role of sensorimotor impairments in dyslexia: A multiplecase study of dyslexic children. Developmental Science, 9:237–255, 2006.

Wimmer H. Characteristics of developmental dyslexia in a regularwriting system. Applied Psycholinguistics, 14: 1–33, 1993.

Witton C, Richardson A, Griffiths TD, Rees A, and Green GGR.Temporal pattern analysis in dyslexia. British Journal ofAudiology, 31: 100–101, 1997.