Available online at www.sciencedirect.com
Vision Research 48 (2008) 979–988 www.elsevier.com/locate/visres
Sequential or simultaneous visual processing deficit in developmental dyslexia? Delphine Lassus-Sangosse a,b,*, Marie-Ange N’guyen-Morel b, Sylviane Valdois a,* b
a Laboratoire de Psychologie et de Neuro-Cognition (UMR 5105 CNRS), Universite´ Pierre Mende`s France, BP 47, 38040 Grenoble, Cedex 9, France Centre de Diagnostic des troubles du langage et des apprentissages, Poˆle Couple-Enfant, CHU de Grenoble, BP 217, 38040 LA TRONCHE, Cedex 9, France
Received 26 July 2007; received in revised form 24 January 2008
Abstract The ability of dyslexic children with or without phonological problems to process simultaneous and sequential visual information was assessed using two tasks requiring the oral report of simultaneously or sequentially displayed letter-strings. The two groups were found to exhibit a simultaneous visual processing deficit but preserved serial processing skills. However, the impairment in simultaneous processing was larger in the dyslexic group with no phonological disorder. Although sequential and simultaneous processing skills both related to reading performance, simultaneous processing alone significantly contributed to reading speed and accuracy. These findings suggest that a simultaneous processing disorder might contribute to developmental dyslexia. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Reading; Visual attention span; Developmental dyslexia; Parallel processing; Sequential processing
1. Introduction The most broadly accepted cognitive theory of dyslexia, the phonological theory, asserts that developmental dyslexia results from a core phonological deficit (Bishop & Snowling, 2004; Frith, 1997; Snowling, 2001; Vellutino, Fletcher, Snowling, & Scanlon, 2004; Ziegler & Goswami, 2005). According to this theory, deficits in the representation and use of phonological information would result in poor phoneme awareness and weak grapheme–phoneme recoding skills (Griffiths & Snowling, 2002; Ramus & Szenkovits, 2008; Vellutino et al., 2004). In support of this hypothesis, deficits in phoneme awareness (the ability to segment and manipulate the constituent sounds of the oral language) have been consistently reported in the dyslexic population (Blachman, 2000; Snowling, 2001; Vellutino et al., 2004, for a review) as well as difficulties in pseudoword reading and spelling. The causal role of phonological *
Corresponding authors. E-mail addresses:
[email protected] (D. Lassus-Sangosse),
[email protected] (S. Valdois). 0042-6989/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2008.01.025
deficits in developmental dyslexia is further supported by studies on reading development. Indeed, results from typically developing children demonstrated a reciprocal causal relationship between phoneme awareness and reading acquisition (Castles & Coltheart, 2004). Longitudinal studies showed that phoneme awareness predicts subsequent reading performance and training studies revealed that instruction in phoneme awareness facilitates the process of reading acquisition (Ehri et al., 2001). While the phonological deficit and its impact on reading are relatively well understood; the potential impact of visual deficits on the reading performance of dyslexic children remains an open and debated question. First introduced by Morgan (1896) and Hinshelwood (1917), the hypothesis of a deficiency in visual processing as the cause of developmental dyslexia came back at the core of the debate with the magnocellular hypothesis. Indeed, a series of studies provided evidence for a deficit in low level visual processing which was interpreted as reflecting a magnocellular impairment in developmental dyslexia (Livingstone, Rosen, Drislane, & Galaburda, 1991; Stein & Walsh, 1997; see Stein, 2003 for a review).
980
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
Although validity of the visual magnocellular theory is hotly debated (Amitay, Ben-Yehudah, Banai, & Ahissar, 2002; Ben-Yehudah, Sackett, Malchi-Ginzberg, & Ahissar, 2001; Ramus et al., 2003; Skottun, 2000), similar low level processing disorders were evidenced in the auditory modality (Laasonen, Service, & Virsu, 2001; McAnally & Stein, 1996; Renvall & Hari, 2002; Witton et al., 1998). In both modalities, performance was characterised by selective difficulties in fast temporal processing (Breznitz & Meyler, 2003; Stein & Walsh, 1997; Tallal, 1980). Deficits in rapid stimulus sequence processing have been interpreted within the sluggish attentional shifting theory framework (Hari & Renvall, 2001). According to this theory, the attention of dyslexics once engaged cannot easily disengage in all sensory modalities. Evidence for sluggish attentional capture and prolonged attentional dwell time were in particular provided in the visual modality through tasks of attentional blink (Hari, Valta, & Uutela, 1999; Visser, Boden, & Giaschi, 2004) and visual precedence (Hari & Renvall, 2001) that involve allocating attention to rapidlysequential stimuli. In the auditory modality, the deficit in sequential processing was found not only for rapid stimuli (Facoetti, Lorusso, Catteaneo, Galli, & Molteni, 2005; Renvall & Hari, 2002) but also when the stimulus onset asynchrony was larger. It was thus concluded that dyslexic children suffered from a more general sequential processing disorder (Amitay et al., 2002; Ben-Yehudah & Ahissar, 2004; Ben-Yehudah et al., 2001; Conlon, Sanders, & Zapart, 2004; Ram-Tsur, Faust, & Zivotofsky, 2006). When found, the auditory temporal processing disorder was typically reported in dyslexic participants with a phonological deficit (Boets, Ghesquie`re, Wieringen, & Wouters, 2007; Boets, Wouters, Van Wieringen, & Ghesquiere, 2006; De Martino, Espesser, Rey, & Habib, 2001; Talcott et al., 1999; Tallal, 1980). A number of findings further suggest that low level visual processing skills differ in different subgroups of developmental dyslexia (see however, Ben-Yehudah & Ahissar, 2004; Ben-Yehudah et al., 2001). Borsting et al. (1996) showed in the visual modality that reduced sensitivity to low spatial frequency was only found in those dyslexic children with phonological problems (see also Slaghuis & Ryan, 1999, 2006 for evidence on motion detection). In return, low level visual processing was reported as intact in dyslexic children with preserved phonological skills (Spinelli et al., 1997). When cross-modality temporal processing was examined in different subgroups of dyslexia, deficits were found in the phonological dyslexic group only (Cestnick, 2001). McAnally, Castles, and Stuart (2000) argued that the auditory deficit alone was responsible for the reading disorder; the visual deficits being viewed as just another manifestation of a general sequential processing deficit without direct impact on reading performance. Based on an account of general sensory deficits as affecting reading via speech perception and phoneme awareness impairments, the visual and auditory deficits should tend to co-occur in the dyslexic individuals showing an associated phonological disorder. Accordingly,
visual sequential processing problems might more specifically characterise a subgroup of dyslexic children with associated phonological problems. Alternatively, deficits in processing strings of visual elements presented simultaneously have been reported in developmental dyslexia (Bednarek et al., 2004; Bosse, Tainturier, & Valdois, 2007; Hawelka, Huber, & Wimmer, 2006; Hawelka & Wimmer, 2005; Helenius, Tarkiainen, Cornelissen, Hansen, & Salmelin, 1999; Pammer, Lavis, Hansen, & Cornelissen, 2004; Pammer, Lavis, Cooper, Hansen, & Cornelissen, 2005; Prado, Dubois, & Valdois, 2007; Valdois, Bosse, & Tainturier, 2004; Valdois et al., 2003). Hawelka and Wimmer (2005, 2006) showed that the dyslexic participants with a multi-element processing disorder were not impaired in coherent motion detection and visual precedence tasks, and that their multi-element processing deficit was independent from their phoneme awareness skills. This study thus suggests that at least some form of visual processing deficit can occur in developmental dyslexia without associated magnocellular or phonological impairments. This deficit might rather affect the simultaneous processing of visual information. According to Bosse et al. (2007), impaired multi-element processing reflects deficits in allocating attention across letter or symbol strings, thus limiting the number of elements that can be processed in parallel during reading. By reference to the connectionist multitrace memory (MTM) model of reading (Ans, Carbonnel, & Valdois, 1998), they interpreted this disorder as reflecting a visual attention span reduction. The visual attention (VA) span was defined as the number of distinct visual elements that can be simultaneously processed in a multi-element array. Bosse et al. (2007) identified different subgroups of dyslexic children characterised by either a single phonological disorder or a single VA span deficit. A third group showed a double deficit, i.e. impaired phonological and VA span skills. The VA span deficit was found to account for the reading performance of dyslexic children, independently of their phonological skills. Moreover, a majority of the dyslexic participants exhibited a single disorder, thus providing additional support for the hypothesis that the phonological and VA span deficits contribute independently to developmental dyslexia. A simultaneous visual processing disorder might thus primarily characterize a subgroup of dyslexic children without associated phonological problems. Only a few studies on developmental dyslexia directly compared sequential and simultaneous processing skills in the visual modality. They typically concluded to a deficit in sequential processing with preservation of simultaneous processing skills (Ben-Yehudah & Ahissar, 2004; BenYehudah et al., 2001; Ram-Tsur et al., 2006). In their study, Ben-Yehudah and Ahissar (2004) examined dyslexics’ spatial frequency discrimination under simultaneous versus sequential presentations. The stimuli were displayed for 250 ms in the simultaneous condition and at different ISIs in the sequential condition. The dyslexic participants performed as controls in the simultaneous condition but
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
they were impaired in the two sequential conditions using ISIs of 250 and 500 ms. The authors argued that the majority of the dyslexic individuals were impaired in sequential processing but showed preserved simultaneous processing skills. Results of this study did not reveal differences according to the reading subtype. They however showed that a sequential processing disorder could be evidenced at time intervals of hundreds rather than tens of milliseconds, thus far longer than the intervals typically used in fast temporal processing studies. Only very rarely were deficits reported in parallel processing in the absence of serial processing problems. Yap and Van der Leij (1993) however reported that dyslexic children were impaired in matching tasks in which two visual digits were displayed simultaneously for 200 ms but performed at the level of chronological-age-matched controls when the two visual digits were displayed one after the other. The present study aimed at assessing whether the ability to process simultaneous and sequential visual information varied in different subtypes of developmental dyslexia. In line with previous findings, a selective sequential processing disorder was expected in the subgroup of dyslexic participants with associated phonological problems. In contrast, impaired simultaneous processing skills, but preserved serial processing, was expected in those dyslexic children with no associated phonological problems. 2. Method 2.1. Population Thirty-nine native French speakers participated in this experiment: 26 dyslexic children, with or without phonological problems, and 13 chronological age matched controls. The dyslexic participants were selected from a larger sample of 52 dyslexic children who were recruited at the ‘‘Centre de diagnostic des troubles du Langage et des apprentissages” (Diagnostic Centre for Specific Language and Learning Disorders) of the Pediatric Department of the Hospital of Grenoble. The 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. All the dyslexic participants had a normal IQ (full IQ superior to 85 on the WISCIII or WISC IV, or a score superior to the 25th percentile on the Raven’s Progressive Matrices; Raven, Court, & Raven, 1998). They attended school regularly and were free from neurological or psychiatric illness, or any medical treatment. All had normal or corrected to normal visual acuity and a normal hearing level. Children underwent clinical examination before the experiment to evaluate their reading skills and their phonological abilities. Neuropsychological data were analysed at the individual level. The Alouette Reading Test (Lefavrais, 1967) was used to estimate the reading age of each child. The children were diagnosed as dyslexics if their reading age was at least 18 months lower than expected according to their chronological age and if they scored below the 10th percentile on tests of either word or pseudo-word reading (speed or accuracy). The dyslexic participants were then selected and divided into two subgroups of 13 children each. Characteristics of the subgroups are provided in the following sections.
2.2. Neuropsychological screening 2.2.1. Phonological assessment The dyslexic children were submitted to three phoneme awareness tasks (phoneme deletion, phoneme segmentation and acronyms)
981
- Phoneme deletion. The participants were asked to delete the first sound of a word and produce the resulting pseudo-word (e.g., /uti/‘‘outil” ? /ti/;/plakar/‘‘placard” ? /lakar/). In the 20 word list, seven words began with a vocalic phoneme corresponding to a multiple letter grapheme so that deletion of the first letter (instead of the first phoneme) yielded incorrect responses; nine words began with a consonantal cluster, four with a singleton. - Phoneme segmentation. Fifteen words were presented auditorily to the participants who had to successively sound out each of the word constituent phonemes. The words were 4.2 phonemes long on average (range 3–6). - Acronyms. Two words were successively pronounced by the experimenter (one word per second). The participants had to extract the first phoneme of each word and blend them to produce a syllable. The ten word couples were taken from the BELEC Battery (Mousty, Leybaert, Alegria, Content, & Morais, 1994). The 26 dyslexic participants were selected and assigned to one of the two dyslexic groups according to their performance on the phonological tasks. Each child belonging to the phonological group exhibited a deficit (scoring at the 10th percentile or below) on at least one of the phoneme awareness tasks. Actually, all the children but three were below the tenth percentile on at least two of these tasks. In contrast, only those children performing within the normal range on each of the three phoneme awareness tasks were assigned to the nonphonological group. At the group level, ANOVAs were performed to compare the two dyslexic groups (see Table 1). Performance on all three phoneme awareness tasks was significantly lower in the phonological group as compared to the nonphonological group.
2.3. Reading skills assessment As shown on Table 2, the two dyslexic groups did not differ significantly in either chronological age (phonological group: CA = 129.92 months, SD = 21.7 months; nonphonological group: CA = 128.69 months, SD = 15.81 months; F < 1) or reading level (phonological group: RL = 88.00 months, SD = 4.76 months; nonphonological group: RL = 89.23 months, SD = 5.36 months; F < 1). Their chronological age was similar to that of the control group (CAC = 128.92 months, SD = 5.07 months; F < 1) but their reading level was significantly lower (CAC = 128.69 months, SD = 6.33 months; phonological group: F(1, 36) = 423.55; p < .0001; nonphonological group: F(1, 36) = 398.45; p < .0001). The participants’ reading skills were further assessed using tasks of isolated word and pseudo-word reading, taken from the ODEDYS neuropsychological battery (Jacquier-Roux, Valdois, & Zorman, 2002). The participants were given six lists of 20 items each for a total of 40 regular words, 40 irregular words and 40 pseudo-words. The word lists were matched for letter and syllable length, grammatical class and frequency. The 40 pseudo-words were legal pseudo-words without lexical neighbours. The participants were instructed to read aloud each of the six lists of 20 items as quickly and as accurately as possible. Both accuracy and reading rate were taken into account. As a group (see Table 2), the dyslexic participants performed worse than normally achieving readers on all the reading measures. Two analyses of variance were carried out with group as between-subjects factor and items as within-subjects factor, one for accuracy, and the other for reading rate. The analysis showed significant main effects of group (F(2, 36) = 28.07; p < .0001) and items (F(2, 72) = 51.09; p < .0001) on reading accuracy and a significant group by items interaction (F(4, 72) = 8.46; p < .00002). Accuracy performance was significantly lower on all types of items (all ps < .0001) for the dyslexic participants as compared to the control children. The two dyslexic groups performed similarly in irregular word (F(1, 36) = 2.68; p = .11) and pseudo-word reading accuracy (F(1, 36) = 1.02; p = .32) but the phonological group performance in regular word reading accuracy was significantly lower as compared to the nonphonological group (F(1, 36) = 7.77; p = 0.008).
982
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
Table 1 Performance of the two dyslexic groups in phoneme awareness Phoneme awareness
Phonological dyslexics
Nonphonological dyslexics
Significance
Phoneme deletion (score/20) (Z score) Phoneme segmentation (score/15) (Z score) Acronyms (score/10) (Z score)
11.08 (4.19) ( 1.39) 4.77 (4.32) ( 0.79) 4.5 (1.09) ( 1.77)
16.92 (4.65) (0.15) 10.38 (4.49) (0.70) 7.85 (2.24) ( 0.12)
F(1, 24) = 19.47; p < .001 F(1, 4) = 14.76; p < .001 F(1, 24) = 46.32; p < .001
Table 2 Chronological age, reading age and reading performance of the groups of dyslexic children (with and without phonological problems) and control participants Phonological dyslexics,a mean (SD)
Nonphonological dyslexics,b mean (SD)
Chronological age controls, mean (SD)
Chronological age (months) Reading age (months)
129.92 (21.70) 88.00 (4.76)
128.69 (15.81) 89.23 (5.36)
128.92 (5.07) 128.69 (6.33)a,b
Reading Regular words Score (max = 40) Speed (sec)
29.92 (5.17)b 88.08 (22.37)
34.31(4.59)a 91.77 (33.68)
39.46 (0.66)a,b 30.85 (5.57)a,b
Exception words Score (max = 40) Speed (sec)
18.23 (6.64) 113.23 (33.57)
22.85 (10.33) 121.08 (62.48)
37.69 (1.97)a,b 33.77 (4.78)a,b
Pseudo words Score (max = 40) Speed (sec)
24.62 (6.17) 105.85 (19.61)
26.69 (6.28) 117.31 (41.81)
36.23 (2.24)a,b 43.08 (7.45)a,b
a b
Means that the difference is significant compared to the phonological dyslexic group. That it is significant by comparison to the nonphonological group.
Reading rate analysis showed significant main effects of group (F(2, 36) = 25.9; p < .0001) and items (F(2, 72) =19.81; p < .0001) and a significant group by items interaction (F(4, 72) = 3.01; p = .023). The dyslexic participants read all types of items significantly more slowly than the controls (all ps < .0001). None of the differences in reading rate between the two, phonological and nonphonological, groups of dyslexic children was significant (all p > .05). Overall, the dyslexic children exhibited a severe delay in reading acquisition (around 40 months). The two dyslexic groups did not differ significantly in reading performance. They were severely impaired in both reading accuracy and reading rate on all types of items. Unexpectedly, the dyslexic children without associated phonological problems were found to exhibit a pseudo-word reading impairment as severe as in the group of dyslexic children with phonological problems.
were presented in upper case (Geneva, 24) in black on a white background. The distance between adjacent letters was of 0.57° in order to minimise lateral masking.
2.4. Experimental tasks
2.4.2. Sequential processing: the sequential report task As in global report, the sequential report task requires reporting as many letters as possible from a series of 5 consonants but the letters were here displayed sequentially, one after the other (see Fig. 1b). The strings were constructed following the same constraints as in global report.
2.4.1. Simultaneous processing assessment: the global report task The global letter-report task was designed to estimate the number of distinct letters that could be extracted in parallel from a brief visual display (i.e., the VA span). This task, used in a number of our previous studies (Bosse et al., 2007; Prado et al., 2007; Valdois et al., 2003, 2004), requires the report of all of the letters of briefly displayed 5-consonant strings. 2.4.1.1. Stimuli. Twenty random five letter-strings (e.g., RHSDM; angular size = 5.4°) were constructed 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. The strings contained no repeated letters. Two subsequent letters never corresponded to a French grapheme (e.g., PH, TH) or a frequent bigram in French (e.g., TR, PL, BR). The 5-consonant string never matched the skeleton of a real word (e.g., FLMBR for FLAMBER ‘‘burn”). Letters
2.4.1.2. Procedure. At the beginning of each trial, a central fixation point was presented for 1000 ms followed by a blank screen for 50 ms. Then, a letter-string was displayed at the centre of the screen for 200 ms, a duration which corresponds to the mean duration of fixations in reading, long enough for an extended glimpse, yet too short for a useful eye movement (Fig. 1a). Children had to name verbally as many letters as possible. The 20 experimental trials were preceded of 10 training trials for which participants received feedback. No feedback was given during the 20 experimental trials. The dependent measure was the number of letters accurately reported (identity not location) across the 20 trials (maximal score = 100).
2.4.2.1. Procedure. A central fixation point was presented for 1000 ms followed by a blank screen for 50 ms. Then 5 letters were successively displayed one at a time at the centre of the screen. Each letter was displayed for 200 ms (ISI = 0). Participants were asked to report as many letters as possible, without any order or time constraints. They started naming letters at the end of the sequential display. Ten training trials were proposed at the beginning of the task, for which participants received feedback. No feedback was given during the 20 experimental trials. The dependent measure was the number of letters accurately reported (identity not location) across the 20 trials (maximal score = 100).
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
983
1000 ms
50 ms
R 200 ms
H
1000 ms 200 ms 50 ms
S 200 ms
BLMSR
D 200 ms
200 ms Until the next trial
Time
M 200 ms
Time Until de next trial
Fig. 1. The simultaneous (a) and sequential (b) report procedures.
3. Results 3.1. Performance in simultaneous and sequential processing Results are reported in Fig. 2. An analysis of variance was carried out with group as between-subjects factor and task as within-subjects factor. The analysis showed significant main effects of group (F(2, 36) = 8.47; p < .001) and task (F(1, 36) = 37.32; p < .0001). Significantly more letters were reported in the simultaneous than in the sequential task. The group task interaction was significant (F(2, 36) = 9.81; p < .0004). Performance was similar in the three groups (F < 1) for sequential processing but it differed significantly according to the group in the simultaneous processing task (F(2, 36) = 16.07; p < .0001). In this latter condition, the proportion of accurately reported letters was significantly lower for the nonphonological group of dyslexic children as compared to both CA controls (88.23 vs. 67.85; F(1, 36) = 32.12; p < .0001) and the phonological dyslexic group (77.69 vs. 67.85; F(1, 36) =
7.49; p < .01). Against our expectations however, the phonological group also reported fewer letters than the CA control group (88.23 vs. 77.69; F(1, 36) = 8.59; p < .006) in the simultaneous condition of letter report. 3.2. Relationship with reading performance We were here interested in investigating whether performance of the normally developing and dyslexic children in the simultaneous and sequential letter report tasks related to their reading performance. Correlation analyses were first conducted on the measures of chronological age, reading age, reading accuracy and reading rate for the different types of items and the two simultaneous and sequential letter-report conditions. Table 3 provides Pearson’s correlations between these different tasks for the whole participants (N = 39). As shown on Table 3, strong correlations (from .51 to .95) were found between the different measures of reading performance (accuracy and rate). Most reading measures
100
Number of correct responses
90 80 70 60 50 Simultaneous processing Sequential processing
40 30 20 10 0
Phonological dyslexics CA controls Non phonological dyslexics GROUPS
Fig. 2. Number of letters accurately reported by each group in the simultaneous and sequential report conditions.
984
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
Table 3 Pearson’s correlations between chronological age, reading age, reading performance (score and speed), and the two conditions of simultaneous and sequential report in the dyslexic and control participants
Chronological age Reading age Reading RW score RW speed IW score IW speed PW score PW speed
Reading age
RW score
0.12
0.09 0.71***
RW speed
IW score
IW speed
PW score
PW speed
0.23 0.86***
0.25 0.83***
0.29 0.79***
0.03 0.72***
0.14 0.83***
0.18 0.64***
0.32* 0.32*
0.64***
0.83*** 0.83***
0.55*** 0.95*** 0.81***
0.82*** 0.71*** 0.79*** 0.63***
0.51*** 0.93*** 0.68*** 0.90*** 0.63***
0.42** 0.60*** 0.59*** 0.57*** 0.55*** 0.52***
0.47** 0.32* 0.48** 0.33* 0.32* 0.17
Report tasks SIMP *
**
SIMP
SEQP
0.50** ***
p < .05; p < .01; p < .001. RW: regular word; IW: irregular word; PW: pseudo-word; SIMP: simultaneous processing; SEQP: sequential processing.
(accuracy and rate) significantly correlated with performance in both the sequential and simultaneous letter report tasks but these two later tasks further correlated with one another. Chronological age significantly related to none of the reading measures but it was found to correlate with performance in the sequential letter-report task. Partial correlation analyses were thus computed to assess whether performance in simultaneous processing still contributed to reading performance after adjusting for sequential processing skills and chronological age. As the reading measures corresponding to the different types of items (words or pseudo-words) strongly correlated (all Pearson’s coefficient superior at .80), two composite measures of reading accuracy and reading rate were calculated (by averaging the individual scores). The analysis revealed that simultaneous processing skills accounted for 21% (t(35) = 2.94; p < .006) and 33% (t(35) = 3.66; p < .0009) of unique variance in reading accuracy and reading rate, respectively. In contrast, performance in sequential processing was not found to significantly contribute to reading performance after controlling for the participants’ age and simultaneous processing skills. 4. Discussion The aim of the current study was to assess whether a simultaneous or sequential visual processing disorder characterised different subgroups of dyslexic children. A sequential processing disorder was mostly expected to occur with phonological problems in developmental dyslexia whereas a simultaneous processing disorder would primarily characterise the dyslexic children with good phonological skills. First, we found that the dyslexic children performed as the control participants in the sequential processing condition whatever their phonological skills. This result goes against previous findings suggesting a deficit in the rapid
sequential processing of visual information in developmental dyslexia. In line with the fast temporal deficit hypothesis, dyslexic readers were consistently described as exhibiting a temporal processing disorder for short interstimulus intervals (for a review: Hari & Renvall, 2001). The present findings challenge this view in showing that our dyslexic participants exhibited good sequential processing skills despite stimulus-intervals of zero milliseconds. It is however noteworthy that stimulus onset asynchrony (SOA) was typically shorter in the studies concerned with rapid temporal processing information as compared to the present one. For example, SOAs around 100 ms are typically used in tasks of Attentional Blink against the 200 ms here used. However, a number of other studies have reported sequential processing deficits despite using longer presentation durations. Ram-Tsur et al. (2006) showed a temporal processing deficit in dyslexic adults for visual items that were displayed for 500 ms (at different SOAs, from 530 ms to more than a second). In the same way, Ben-Yehudah and Ahissar (2004) concluded to a sequential processing disorder in developmental dyslexia using a discrimination task in which two gratings were presented for 250 ms (at different SOAs from 500 ms to 2.25 s). It follows that the longer presentation duration used in the present study cannot by itself explain the absence of sequential processing disorder in our sample of dyslexic children. Deficits in processing information at short ISIs have been viewed as consistent with the sluggish attentional shifting theory of dyslexia (Hari & Renvall, 2001). Dyslexics’ difficulty under sequential presentation would then result from sluggishness in disengaging attention from the first stimulus to the second, thus leading to poor encoding of the second stimulus. Obviously, the present findings suggest that our dyslexic participants did not suffer from longer attentional dwell time. On the other hand, when deficits were evidenced for longer SOAs (Ben-Yehudah & Ahissar, 2004; Ram-Tsur et al., 2006), the sequential pro-
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
cessing disorder was viewed as resulting from a decay of the perceptual memory trace (Magnussen & Greenlee, 1999). The absence of sequential processing disorder in the current study does not invalidate this hypothesis since the deficits when observed were typically reported for longer than 200 ms SOAs. Another explanation of the discrepancy between the current findings and previous ones might deal with differences in the type of experimental tasks used to assess sequential processing skills in dyslexic readers. Indeed, it was hypothesised that dyslexic readers might be more specifically impaired in visual paradigms requiring sequential comparisons (Amitay et al., 2002; Ben-Yehudah & Ahissar, 2004). In line with this task effect hypothesis, those studies using comparison tasks typically showed that sequential presentation yielded better performance among most of the controls than the simultaneous condition. On the contrary in the current study, CA controls performed better in the simultaneous than in the sequential report task. It is noteworthy however that, in their studies, Ram-Tsur et al. (2006) and Ben-Yehudah and Ahissar (2004) used low-contrast non linguistic stimuli. The use of high contrast alphabetic stimuli in the current study might have triggered some visual attentional processes more specifically involved in letter-string processing and reading (as postulated by Whitney & Cornelissen, 2005). Second and as expected, the dyslexic children with no phonological problems were found to exhibit a simultaneous processing disorder. This finding is in agreement with previous research indicating that a VA span disorder (i.e., a visual simultaneous processing disorder) typically dissociates from phonological problems in developmental dyslexia (Valdois et al., 2003; Bosse et al., 2007). However, the current results extend previous findings in showing that this letter-report disorder is specific to simultaneous processing. Indeed, the two tasks of letter report we used here were designed to mostly differ on the simultaneous/sequential dimensions. Both tasks require not only the integrity of the visual processes involved in letter identification but also preservation of the high-level processes involved in the activation of letter names in long-term memory and their maintenance in verbal short-term memory. The good performance of nonphonological dyslexic children in the sequential report task suggests that these levels of processing were preserved. In particular, the current findings suggest that the poor performance of dyslexic children in simultaneous processing cannot be attributed to verbal short-term memory problems. This conclusion is in line with previous findings showing that deficits in global report (simultaneous processing) typically dissociate from verbal short-term memory problems in developmental dyslexia. Indeed in the case studies we previously reported, we showed that dyslexic children with phonological problems and poor verbal STM performed the global report task as well as control children of the same chronological age (Valdois et al., 2003). In contrast, children with poor performance in glo-
985
bal report had preserved verbal short term memory (Dubois et al., submitted for publication; Valdois et al., 2003). In another study conducted on a large sample of typically developing children (N = 417) in 1st, 3rd and 5th grade, Bosse and Valdois (submitted for publication) further found that visual attention span abilities predicted reading performance independently of the child phoneme awareness and verbal short term memory. This finding again suggests that performance in the simultaneous condition of global report is not primarily affected by phonological or verbal STM abilities. This conclusion is supported by other findings showing that performance in similar global report tasks was not affected by concurrent activity in verbal short-term memory (Pelli, Burns, Farell, & MoorePage, 2006; Scarborough, 1972). It is further noteworthy that if verbal short term memory was strongly involved in the two tasks of simultaneous and sequential letter report then one would predict performance to be sensitive to the task verbal memory load. Whereas all 5 letters are simultaneously displayed for 200 ms in the simultaneous condition of global report, each individual letter is presented for 200 ms in the sequential condition so that letter names have to be maintained in verbal short term memory for 1000 ms (200 ms 5 letters) before being reported. Accordingly, memory load is far higher in the sequential than simultaneous condition of letter report. However, dyslexic children performance does not differ from that of the controls in sequential processing suggesting that their performance was not affected by differences in verbal short term memory. As previously stressed by Shovman and Ahissar (2006), the present findings suggest that our dyslexic participants did not suffer from a general visual processing deficit or a letter identification problem (see also Prado et al., 2007). This conclusion is further supported by the findings of Bosse et al. (2007) that dyslexic children did not differ from normally developing children in single letter processing rate. Our dyslexic participants, in particular those with no phonological problems, thus demonstrate a letter-string processing disorder despite preserved letter identification skills. In line with these findings, Pelli et al. (2006) have suggested that single letter identification skills and letterstring processing might correspond to two independent visual subsystems. In their study, they trained non dyslexic participants to learn foreign alphabets and showed that their ability to recognise previously unfamiliar letters increased very quickly with experience. A few hundred trials with the new alphabet were enough for the novice participants to become as efficient as fluent readers for which the alphabet was familiar. However, despite their expertise in single letter processing, the trained participants could only report a few of the foreign characters (1 or 2) when displayed in random strings. In fact, they reported as a few letters as the participants seeing them for the first time and far less than fluent readers. These results suggest that efficiency for letter identification has no direct impact on the ability to process letters in strings. Letter-string pro-
986
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
cessing skills would then be acquired independently based on specific visual or visual attentional mechanisms (see Whitney & Cornelissen, 2005, for a similar account). Unexpectedly however, the phonological dyslexic group also demonstrated poorer simultaneous visual processing skills than the controls even if performing significantly above the other group of dyslexic children. Although unexpected, this result is compatible with previous findings showing that at least some dyslexic individuals exhibit both a phonological and a VA span disorder (Bosse et al., 2007). A post-hoc analysis was thus conducted to determine the proportion of children who exhibited a simultaneous processing disorder (score < 1.65 SD from the norm) in each of the two dyslexic groups, with or without phonological problems. The analysis revealed that 38% of the dyslexic children with a phonological disorder further exhibited a simultaneous visual processing disorder against 85% in the group with no phonological problems. Another important finding of the current study is the specific and significant relationship we found between reading performance and simultaneous processing skills. However, this relationship might be considered as trivial if the simultaneous letter-report task was viewed as just a particular type of reading task. According to dual route models (Coltheart, Curtis, Atkins, & Haller, 1993; Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001), unfamiliar word reading relies on a number of cognitive components including visual analysis, grapheme–phoneme conversion rules, phonemic blending and verbal short term memory. We already acknowledged that the simultaneous letter report task requires some of the visual processing skills involved in reading. The task however involves no reading-based chunking mechanism which might have helped recalling the identity of letters in the string. Indeed, the 5-consonant letters never corresponded to a lexical word skeleton and two subsequent consonants never formed legal graphemes in French. Moreover and in contrast to unfamiliar word reading, grapheme–phoneme conversion rules do not apply, since letter names, not letter sounds, have to be reported. As a consequence, phoneme blending is not involved since the output sequence is a sequence of letter names, not of letter sounds as in reading. Verbal short term memory is involved in the task for letter names to be maintained in short term memory until their pronunciation but this component was assumed to be preserved in our dyslexic participants since involved in sequential processing as well. Performance on the simultaneous report task thus primarily reflects the efficiency of the visual attentional processes involved in reading. It is further noteworthy that the 5-consonant strings used in this task did not keep to the standard reading format, since the distance between two adjacent letters was increased. The adopted spacing (of 0.57°) was large enough to reduce lateral masking effects (Pelli, Palomares, & Majaj, 2004) but did not exceed the threshold above which parallel processing no longer applies (Dehaene, Cohen, Sigman, & Vinckier, 2005; Cohen, Dehaene, Vinckier, Jobert, & Montavont, 2008). It thus
appears that the task more specifically taps those of the visual mechanisms of reading which are involved in the extraction of letter information during parallel processing. The specific relationship we found between simultaneous processing skills and reading performance can be straightforwardly explained within the multitrace memory model (MTM) of reading (Ans et al., 1998). The model indeed attributes a key role to a component of the reading system, the visual attentional window, which determines the amount of orthographic information simultaneously processed while reading. Large visual attentional windows extending over the whole word length are involved when reading in global mode whereas smaller windows extending over sub-lexical units typically apply when reading in analytic mode. The model thus predicts that differences in the amount of orthographic units the system can simultaneously process will affect reading performance (Valdois et al., 2004). In particular, a simultaneous processing disorder characterised by a reduction of the visual attentional window (i.e., the visual attention span in humans) would prevent reading in global mode thus primarily affecting performance on regular and irregular words. However, because the whole letters constituting relevant orthographic units (morphemes, syllables or graphemes) have to be simultaneously processed when reading analytically, a reduction of the VA span can also affect pseudo-word reading. In line with these predictions, it has been shown that deficits of the VA span contribute to the poor reading performance of dyslexic children for both irregular words and pseudo-words (Bosse et al., 2007) and that performance on both types of items is predicted by VA span abilities in large samples of typically developing children (Bosse & Valdois, submitted for publication). The present results thus provide further support to the MTM model in showing that the relationship between reading and letter report performance is specific to simultaneous processing. References Amitay, S., Ben-Yehudah, G., Banai, K., & Ahissar, M. (2002). Disabled readers suffer from visual and auditory impairments but not from a specific magnocellular deficit. Brain, 125, 2272–2285. Ans, B., Carbonnel, S., & Valdois, S. (1998). A connectionist multipletrace memory model for polysyllabic word reading. Psychological Review, 105(4), 678–723. Bednarek, D. B., Saldana, D., Quintero-Gallego, E., Garcia, I., Grabowska, A., & Gomez, C. M. (2004). Attentional deficit in dyslexia: A general or specific impairment? Neuroreport, 15(11), 1787–1790. Ben-Yehudah, G., & Ahissar, M. (2004). Sequential spatial frequency discrimination is consistently impaired among adult dyslexics. Vision Research, 44(10), 1047–1063. Ben-Yehudah, G., Sackett, E., Malchi-Ginzberg, L., & Ahissar, M. (2001). Impaired temporal contrast sensitivity in dyslexics is specific to retainand-compare paradigms. Brain, 124(7), 1381–1395. Bishop, D. V. M., & Snowling, M. J. (2004). Developmental dyslexia and specific language impairment: Same or different? Psychological Bulletin, 130(6), 858–886. Blachman, B. A. (2000). Phonological awareness. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.). Handbook of reading research (Vol. III, pp. 483–502). Mahwah, NJ: Lawrence Erlbaum.
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988 Boets, B., Ghesquie`re, P., Wieringen, A. V., & Wouters, J. (2007). Speech perception in preschoolers at family risk for dyslexia: Relations with low-level auditory processing and phonological ability. Brain and Language, 101, 19–30. Boets, B., Wouters, J., Van Wieringen, A., & Ghesquiere, P. (2006). Auditory temporal information processing in preschool children at family risk for dyslexia: Relations with phonological abilities and developing literacy skills. Brain and Language, 97(1), 64–79. Borsting, E., Ridder, W. H., Dudeck, K., Kelley, C., Matsui, L., & Motoyama, J. (1996). The presence of a magnocellular defect depends on the type of dyslexia. Vision Research, 36, 1047–1053. Bosse, M. L., Tainturier, M. J., & Valdois, S. (2007). Developmental dyslexia: The visual attention span deficit hypothesis. Cognition, 104, 198–230. Bosse, M. L., & Valdois, S. (submitted for publication). Influence of the visual attention span on child reading performance: A cross-sectional study. Breznitz, Z., & Meyler, A. (2003). Speed of lower-level auditory and visual processing as a basic factor in dyslexia: Electrophysiological evidence. Brain & Language, 85, 166–184. Castles, A., & Coltheart, M. (2004). Is there a causal link from phonological awareness to success in learning to read? Cognition, 91, 77–111. Cestnick, L. (2001). Cross modality temporal processing deficits in developmental phonological dyslexics. Brain and Cognition(46), 319–325. Cohen, L., Dehaene, S., Vinckier, F., Jobert, A., & Montavont, A. (2008). Reading normal and degraded words: Contribution of the dorsal and ventral visual pathways. Neuroimage, 40(1), 353–366. Coltheart, M., Curtis, B., Atkins, P., & Haller, M. (1993). Models of reading aloud: Dual-route and parallel-distributed processing approaches. Psychological Review, 100(4), 589–608. Coltheart, M., Rastle, K., Perry, C., Langdon, R., & Ziegler, J. (2001). DRC: A dual route cascaded model of visual word recognition and reading aloud. Psychological Review, 108, 204–256. Conlon, E., Sanders, M., & Zapart, S. (2004). Temporal processing in poor adult readers. Neuropsychologia, 42(2), 142–157. Dehaene, S., Cohen, L., Sigman, M., & Vinckier, F. (2005). The neural code for written words: A proposal. Trends in Cognitive Sciences, 9(7), 335–341. De Martino, S., Espesser, R., Rey, V., & Habib, M. (2001). The ‘‘temporal processing deficit” hypothesis in dyslexia: New experimental evidence. Brain and Cognition, 46(1–2), 104–108. Dubois, M., Kyllingsbaek, S., Prado, C., Peiffer, E., Lassus, D., & Valdois, S. (submitted for publication). Fractionating the multi-element processing deficit in developmental dyslexia: Evidence from two case studies. Ehri, L. C., Nunes, S. R., Willows, D. M., Schuster, B. V., YaghoubZadeh, Z., & Shanahan, T. (2001). Phonemic awareness instruction helps children learn to read: Evidence from the National Reading Panel’s Meta Analysis. Reading Research Quarterly, 36(3), 250–287. Facoetti, A., Lorusso, M. L., Catteaneo, C., Galli, R., & Molteni, M. (2005). Visual and auditory attentional capture are both sluggish in children with developmental dyslexia. Acta Neurobiologiae Experimentalis, 65, 61–72. Frith, U. (1997). Brain, mind and behaviour in dyslexia. In C. Hulme et al. (Eds.), Dyslexia: Biology, cognition and intervention (pp. 1–19). London: Whurr Publishers. Griffiths, Y. M., & Snowling, M. J. (2002). Predictors of exception word and nonword reading in dyslexic children: The severity hypothesis. Journal of Educational Psychology, 94(1), 34–43. Hari, R., & Renvall, H. (2001). Impaired processing of rapid stimulus sequences in dyslexia. Trends in Cognitive Sciences, 5(12), 525–532. Hari, R., Valta, M., & Uutela, K. (1999). Prolonged attentional dwell time in dyslexic adults. Neuroscience Letters, 271(3), 202–204. Hawelka, S., Huber, C., & Wimmer, H. (2006). Impaired visual processing of letter and digit strings in adult dyslexic readers. Vision Research, 46(5), 718–723.
987
Hawelka, S., & Wimmer, H. (2005). Impaired visual processing of multielement arrays is associated with increased number of eye movements in dyslexic reading. Vision Research, 45(7), 855–863. Helenius, P., Tarkiainen, A., Cornelissen, P., Hansen, P. C., & Salmelin, R. (1999). Dissociation of normal feature analysis and deficient processing of letter-strings in dyslexic adults. Cerebral Cortex, 9(5), 476–483. Hinshelwood, J. (1917). Congenital word blindness. London: H.K. Lewis & Co.. Jacquier-Roux, M., Valdois, S., & Zorman, M. (2002). ODEDYS: Un outil de de´pistage des dyslexies. Grenoble: Laboratoire Cogni-sciences, IUFM de GRENOBLE. Laasonen, M., Service, E., & Virsu, V. (2001). Temporal order and processing acuity of visual, auditory, and tactile perception in developmentally dyslexic young adults. Cognitive, Affective & Behavioral Neuroscience, 1, 394–410. Lefavrais, P. (1967). Test de l’Alouette: Manuel. Paris: Les e´ditions du Centre de Psychologie Applique´e. Livingstone, M. S., Rosen, G. D., Drislane, F. W., & Galaburda, A. M. (1991). Physiological and anatomical evidence for a magnocellular defect in developmental dyslexia. Proceedings of the National Academy of Sciences of the United States of America, 88, 7943–7947. Magnussen, S., & Greenlee, M. W. (1999). The psychophysics of perceptual memory. Psychological Research, 62, 81–92. McAnally, K. I., Castles, A., & Stuart, G. W. (2000). Visual and auditory processing impairments in subtypes of developmental dyslexia: A discussion. Journal of Developmental and Physical Disabilities, 12(2), 145–156. McAnally, K. I., & Stein, J. F. (1996). Auditory temporal coding in dyslexia. Proceedings of the National Academy of Sciences of the United States of America, 263, 961–965. Morgan, W. P. (1896). A case of congenital word blindness. British Medical Journal, 2, 1378–1379. Mousty, P., Leybaert, J., Alegria, J., Content, A., & Morais, J. (1994). BELEC: Une batterie d’e´valuation du langage e´crit et de ses troubles. In J. Gre´goire & B. Pie´rart (Eds.). Evaluer les troubles de la lecture: les nouveaux mode`les the´oriques et leurs implications diagnostiques (pp. 127–146). Bruxelles: De Boeck. Pammer, K., Lavis, R., Cooper, C., Hansen, P. C., & Cornelissen, P. L. (2005). Symbol-string sensitivity and adult performance in lexical decision. Brain and Language, 94(3), 278–296. Pammer, K., Lavis, R., Hansen, P., & Cornelissen, P. L. (2004). Symbolstring sensitivity and children’s reading. Brain and Language, 89(3), 601–610. Pelli, D. G., Burns, C. W., Farell, B., & Moore-Page, D. C. (2006). Feature detection and letter identification. Vision Research, 46, 4646–4674. Pelli, D. G., Palomares, M., & Majaj, N. J. (2004). Crowding is unlike ordinary masking: Distinguishing feature integration from detection. Journal of Vision, 4(12), 1136–1169. Prado, C., Dubois, M., & Valdois, S. (2007). The eye movements of dyslexic children during reading and visual search: Impact of the visual span. Vision Research, 47(19), 2521–2530. Ram-Tsur, R., Faust, M., & Zivotofsky, A. Z. (2006). Sequential processing deficits of reading disabled persons is independent of inter-stimulus interval. Vision Research, 46(22), 3949–3960. Ramus, F., Rosen, S., Dakin, S. C., Day, B. L., Castellote, J. M., White, S., et al. (2003). Theories of developmental dyslexia: Insights from a multiple case study of dyslexic adults. Brain, 126(4), 841–865. Ramus, F., & Szenkovits, G. (2008). What phonological deficit? The Quarterly Journal of Experimental Psychology, 61, 129–141. Raven, J. C., Court, J. H., & Raven, J. (1998). Progressive matrices standard (PM38). EAP, Paris. Renvall, A., & Hari, R. (2002). Auditory cortical responses to speech-like stimuli in dyslexic adults. Journal of Cognitive Neuroscience, 14, 757–768. Scarborough, D. L. (1972). Memory for brief visual displays of symbols. Cognitive Psychology, 3, 408–429.
988
D. Lassus-Sangosse et al. / Vision Research 48 (2008) 979–988
Shovman, M. M., & Ahissar, M. (2006). Isolating the impact of visual perception on dyslexics’ reading ability. Vision Research, 46, 3514–3525. Skottun, B. C. (2000). The magnocellular deficit theory of dyslexia: Evidence from contrast sensitivity. Vision Research, 40, 111–127. Slaghuis, W. L., & Ryan, J. F. (1999). Spatio-temporal contrast sensitivity, coherent motion, and visible persistence in developmental dyslexia. Vision Research, 39, 651–668. Slaghuis, W. L., & Ryan, J. F. (2006). Directional motion contrast sensitivity in developmental dyslexia. Vision Research, 46, 3291–3303. Snowling, M. (2001). From language to reading and dyslexia. Dyslexia, 7(1), 37–46. Spinelli, D., Angelelli, P., De Luca, M., Di Pace, E., Judica, A., & Zoccolotti, P. (1997). Developmental surface dyslexia is not associated with deficits in the transient visual system. Neuroreport, 27(8), 1807–1812. Stein, J. (2003). Evaluation of an exercise based treatment for children with reading difficulties. Dyslexia, 9, 124–126. Stein, J., & Walsh, V. (1997). To see but not to read; the magnocellular theory of dyslexia. Trends in Neurosciences, 20(4), 147–152. Talcott, J. B., Witton, C., McClean, M., Hansen, P. C., Rees, A., Green, G. G. R., et al. (1999). Can sensitivity to auditory frequency modulation predict children’s phonological and reading skills? Neuroreport, 10, 2045–2050. Tallal, P. (1980). Auditory temporal perception, phonics, and reading disabilities in children. Brain and Language, 9(2), 182–198.
Valdois, S., Bosse, M. L., Ans, B., Carbonnel, S., Zorman, M., David, D., et al. (2003). Phonological and visual processing deficits can dissociate in developmental dyslexia: Evidence from two case studies. Reading and Writing: An Interdisciplinary Journal, 16, 541–572. Valdois, S., Bosse, M. L., & Tainturier, M. J. (2004). The cognitive deficits responsible for developmental dyslexia: Review of evidence for a selective visual attentional disorder. Dyslexia, 10, 1–25. Vellutino, F. R., Fletcher, J. M., Snowling, M. J., & Scanlon, D. M. (2004). Specific reading disability (dyslexia): What have we learned in the past four decades? Journal of Child Psychology and Psychiatry, 45, 2–40. Visser, T. A. W., Boden, C., & Giaschi, D. E. (2004). Children with dyslexia: Evidence for visual attention deficits in perception of rapid sequences of objects. Vision Research, 44, 2521–2535. Whitney, C., & Cornelissen, P. (2005). Letter-position encoding and dyslexia. Journal of Research in Reading, 28, 274–301. Witton, C., Talcott, J., Hansen, P., Richardson, A., Griffiths, T., Rees, A., et al. (1998). Sensitivity to dynamic auditory and visual stimuli predicts non-word reading ability in both dyslexic and normal readers. Current Biology, 8, 791–797. Yap, R., & Van der Leij, A. (1993). Rate of elementary symbol processing in dyslexics. In S. F. Wright & R. Groner (Eds.), Facets of dyslexia and its remediation (pp. 337–347). Amsterdam: Elsevier Science. Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia, and skilled reading across languages: A psycholinguistic grain size theory. Psychological Bulletin, 131(1), 3–29.
