Working Memory Functioning in Children with Learning Disorders and Speciﬁc Language Impairment

Home , Dyscalculia

Top Lang Disorders Vol. 33, No. 4, pp. 298–312 Copyright c 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins Working Memory Functioning in Children With Learning Disorders and Speciﬁc Language Impairment

Kirsten Schuchardt, Ann-Katrin Bockmann, Galina Bornemann, and Claudia Maehler

Purpose: On the basis of Baddeley’s working memory model (1986), we examined working memory functioning in children with learning disorders with and without specific language impairment (SLI). We pursued the question whether children with learning disorders exhibit similar working memory deficits as children with additional SLI. Method: In separate analyses, we compared the following groups of children: (1) 30 children with dyslexia (DYS) and 16 children with DYS receiving special language education and (2) 19 children with combined disorder of scholastic skills (CDSS) and 18 children with CDSS receiving special language education. A control group of 30 typically developing children was included in each comparison. All of the children receiving special language education met criteria for SLI. To assess the 3 subcomponents of working memory (phonological loop, visual–spatial sketchpad, central executive), the children worked individually on an extensive test battery. Results: We found deficits in the phonological loop and central executive functioning for children with dyslexia (and CDSS) as well as for children with additional SLI. Deficits in phonological functioning were broader and more profound for children with SLI. Deficits in visual–spatial sketchpad could only be found for children with CDSS without SLI. Con- clusions: Children with isolated learning disorder and children with additional SLI demonstrate similarities and differences in working memory functioning. These findings support our hypothesis that underlying working memory deficits for the different disorders partly overlap but also are distinct and partly distinguish between certain disorders. Key words: combined disorder of scholastic skills, dyslexia, learning disorder, specific language impairment, working memory

EARNING DISORDERS REPRESENT one impairments in acquiring the cultural tech- L of the most frequent causes for school fail- niques of reading, spelling, and calculating. ure. Children with specific learning disorders Dyslexia describes specific deficits in read- such as dyslexia and dyscalculia show general ing acquisition (often combined with spelling disorder), and dyscalculia is characterized by deficits in arithmetic skills. The exact definitions and diagnostic criteria of learning disorders differ widely, but Author Affiliations: Department of Diagnostic and some sources are used commonly across the Educational Psychology (Drs Schuchardt, Bockmann, and Maehler and Ms Bornemann), world. International Classification of Dis- University of Hildesheim, Hildesheim, Germany. eases, Tenth Revision (ICD-10, 2011) by the The authors have indicated that they have no financial World Health Organization defines interna- and no nonfinancial relationships to disclose. tionally accepted diagnostic criteria. Here, Corresponding Author: Kirsten Schuchardt, PhD, learning disorders are described as poor per- Department of Diagnostic and Educational Psychol- formance in reading, spelling, and calculat- ogy, University of Hildesheim, Marienburger Platz 22, 31141, Hildesheim, Germany (schuchar@uni- ing, respectively, that must be significantly hildesheim.de). lower than expected with regard to age, in- DOI: 10.1097/01.TLD.0000437943.41140.36 telligence, and schooling. 298

It is not unusual that learning disor- WORKING MEMORY ders occur together with specific language impairment (SLI), which is characterized by As possible causal factors underlying SLI serious qualitative and quantitative deficits and learning disorders, researchers have iden- in productive and/or receptive language pro- tified deficits in working memory and diverse ficiency. The ICD-10 (Code F80) defines aspects of phonological information process- “specific developmental disorders of speech ing, such as phonological awareness (Catts and language” as “disorders in which nor- et al., 2005; Eisenmajer, et al., 2005; Nithart mal patterns of language acquisition are dis- et al., 2009). Although various models of turbed from the early stages of develop- working memory have been developed, the ment.” It specifies further that the conditions model by Baddeley (1986) has proved a par- are not directly attributable to neurological ticularly useful theoretical tool in numerous or speech mechanism abnormalities, sensory studies in this area. The model distinguishes impairments, mental retardation, or environ- between different components of working mental factors. Specific developmental dis- memory, with the modality-free central ex- orders of speech and language are often ecutive acting as a kind of supervisory system followed by associated problems, such as that serves to control and regulate the cog- difficulties in reading and spelling. nitive processes occurring in its two limited- Approximately 25%–75% of all children capacity slave systems, the phonological loop with language impairment develop read- and the visual–spatial sketchpad. Further ing difficulties (Catts, Adlof, Hogan, & Ellis functions of the central executive that have Weismer, 2005; McArthur, Hogben, Edwards, since been identified by Baddeley (1996) in- Heath & Mengler, 2000; Tomblin, Zhang, clude coordinating the slave systems, focusing Buckwalter, & Catts, 2000). On the contrary, and switching attention, and retrieving repre- according to Catts et al. (2005), every fifth sentations from long-term memory. child with dyslexia shows a history of dif- In contrast, Baddeley’s (1986, 1996) two ficulties in language acquisition. If language slave systems perform modality-specific op- disorders persist until school age, the rate of erations. Verbal and auditory information children demonstrating specific reading dis- is stored temporarily and processed in the ability climbs to 50% (McArthur et al., 2000). phonological loop. Whereas verbal or audi- Children with SLI also can experience diffi- tory information enters the phonological store culties. These difficulties can emerge prior directly, visual information has to be trans- to formal schooling and persist during the lated into phonological code before it can school-age years (Cowan, Donlan, Newton, do so. Two components of the phonologi- & Lloyd, 2005; Donlan, Cowan, Newton, & cal loop are distinguished: the phonological Lloyd, 2007; Eisenmajer, Ross, & Pratt, 2005; store and the subvocal rehearsal process. The Fazio, 1994, 1996, 1999). visual–spatial sketchpad is concerned with re- Given this high comorbidity rate, recent remembering and processing visual and spatial search has started to investigate the common- information; it comprises a visual cache for alities and shared causal factors of learning static visual information and an inner scribe disorders and SLI. Within recent years, one for dynamic spatial information (Logie, 1995; question has been addressed in particular: Are Pickering, Gathercole, Hall, & Lloyd, 2001). SLI and learning disabilities distinct disorders Later, Baddeley (2000) added a fourth com- with different causal factors or are they var- ponent to the working memory model, the ious manifestations of the same underlying episodic buffer, for linking long-term mem- cognitive factors (Baird, Slonims, Simonoff, & ory and integrating information from all of the Dworzynski, 2011; Catts et al., 2005; de Bree, other systems into a unified experience. To Wijnen, & Gerrits, 2010)? date, however, research on working memory

has mostly focused on the three original sub- Deficits in working memory functioning components, probably because it turned out have been investigated extensively also for to be difficult to create valid tasks measuring children with SLI. Children with SLI display the functioning of the episodic buffer. severe deficits in phonological loop functioning (Archibald & Gathercole, 2006b, 2007; WORKING MEMORY AND LEARNING Marton & Schwartz, 2003). Results concern- DISORDERS ing visual working memory are inconsistent. Although some studies did not find any impair- There is considerable evidence that chil- ment of visual spatial sketchpad in children dren with dyslexia have deficits in phono- with SLI (e.g., Archibald & Gathercole, 2006a; logical processing and storage (Schuchardt, Riccio, Cash, & Cohen, 2007), others have re- Maehler, & Hasselhorn, 2008; Vellutino, ported significantly lower scores for children Fletcher, Snowling, & Scanlon, 2004). Chil- with SLI on tasks assessing the visual spatial dren have been found to exhibit a reduced sketchpad than those for typically develop- memory span for acoustically presented ing children (e.g., Hick, Botting, & Conti- words, numbers, and nonwords. Numerous Ramsden, 2005; Hoffman & Gillam, 2004). studies on dyslexia and accompanying deficits In addition to these phonological and visual– in complex abilities such as text comprehen- spatial difficulties exhibited by children with sion also detect deficits in central-executive SLI, deficits in central–executive processing working memory functioning (Landerl, Bevan are evident (Archibald & Gathercole, 2006b; & Butterworth, 2004; Schuchardt et al., 2008; Marton & Schwartz, 2003; Montgomery & Siegel & Ryan, 1989). In contrast, hardly any Evans, 2009). These deficits in central execu- reliable correlations between visual–spatial tive are not confined to phonological material, sketchpad functioning and dyslexia could they also occur with visual–spatial material. be found (O’Shaughnessy & Swanson, 1998; This lends support to the hypothesis that chil- Schuchardt et al., 2008). dren with SLI demonstrate a broader impair- Researchers studying working memory ment of central-executive functioning, which functioning in children with dyscalculia have is not restricted to phonological information reported deficits in the visual–spatial sketch- processing. pad (Schuchardt et al., 2008), whereas mixed In summary, it can be stated that specific results have been reported for the phonolog- patterns in working memory functioning have ical loop and central executive. Some stud- been detected both for learning disorders and ies show them not to be impaired (McLean & SLI and that some similarities in deficit pro- Hitch, 1999; Schuchardt et al., 2008), whereas files are evident. Nevertheless, only a few others do report deficits in the phonological studies have directly compared working mem- loop or central executive (McLean & Hitch, ory profiles in children with both specific 1999; Swanson & Sachse-Lee, 2001). Children and comorbid learning deficits although such with a combined disorder of scholastic skills comparisons are crucial to understanding the (CDSS; a term used in Germany on the basis of potential common or distinct cognitive im- the ICD-10 to describe learning difficulties in pairments associated with different learning reading and writing as well as mathematics) profiles. Catts et al. (2005) examined chil- have been found to have severe deficits in all dren with dyslexia, SLI, and combined SLI and three working memory components (Maehler dyslexia. Only children with reading difficul- & Schuchardt, 2009; Schuchardt et al., 2008). ties (with or without SLI) performed poorly Some of this research (Maehler & Schuchardt, on phonological awareness and phonologi- 2009) has shown similar performance pat- cal working memory tasks, whereas the SLI terns between children with intelligence and control groups did not. Catts et al. con- below average and children with CDSS, who, cluded that SLI and dyslexia are distinct dis- by definition, have average intelligence. orders with overlapping difficulties in basic

cognitive functioning. However, Catts et al. special language education (SLE) (DYS; n = relied on a preschool SLI diagnosis and did not 30); (2) children with dyslexia and referred examine whether the children with SLI still for SLE (SLI + DYS; n = 16); (3) children met the criteria for SLI at school age. There- with CDSS but not referred for SLE (CDSS; fore, it is an open question whether school- n = 19); (4) children with CDSS and referred aged children with a current diagnosis of SLI, for SLE (SLI + CDSS; n = 18); and (5) a con- dyslexia, or both SLI and dyslexia will show trol group of typically developing children overlapping or different patterns of working matched for chronological age (C; n = 30). memory functioning. The two groups of children with learning dis- We also could find no studies on chil- orders (DYS and CDSS) were recruited from dren with SLI and comorbid severe learning the counseling center for children with learn- disorder in reading, spelling, and arithmetic ing disabilities, which is part of a university skills (CDSS). We, therefore, wanted to inves- programinacityofGermany.Allofthemat- tigate whether children with SLI and comor- tended regular primary schools; due to learn- bid CDSS exhibit working memory impair- ing problems, they voluntarily attended the ments that are similar in profiles and as severe diagnostic procedures described later and re- as those observed for children with CDSS but ceived the diagnosis DYS or CDSS. The two without SLI. groups of children referred for SLE (SLI + The purpose of this research was to exam- DYS and SLI + CDSS) attended a school for ine whether children with persisting language SLE. Their SLI-diagnoses were present before impairment and children with learning disor- they participated in the study. Being referred ders in either reading only or across scholastic for SLE and attending a special school re- domains exhibit the same deficits in working veals severe problems of language develop- memory. We hypothesized that deficits in the ment in the two SLI groups. The typically de- same working memory components would be veloping children in the control group were suggestive of a common underlying cognitive second to fourth graders from a public ele- impairment whereas unique deficits would mentary school. Only native German-speaking correspond to different underlying impair- children were included in the study sample. ments. In our study, children with learning No information on socioeconomic status of disorders and children with combined learn- the families was available. ing disorders and SLI underwent an extensive working memory test battery with tasks on all subcomponents with the exception of the Diagnostic classification episodic buffer. We then compared the pat- The two groups of children with SLI (SLI + terns of the following groups of children: (1) DYS and SLI + CDSS, who all attended a children with isolated dyslexia and children school for SLE) were tested for language de- with comorbid SLI and dyslexia and (2) chil- velopment and intelligence before they were dren with CDSS (i.e., dyslexia and dyscalculia) referred to special education and before they and children with SLI and CDSS, and a control took part in the study. Because of time re- group of typically developing children. strictions, we accepted their intelligence testing but we confirmed their language deficits. The two groups with learning disorders (DYS METHODS and CDSS) had not been tested before and received their diagnosis through our testing. We did not examine language performance Participants in these groups, because there was no indi- Five groups of children (second- to fourth- cation for these tests in the context of the grade students) participated in the study: counseling center (no language problems re- (1) children with dyslexia but not referred for ported by parents or teachers). Of course, one

might argue that language impairment in chil- on grade level) and the nonword reading dren with learning disorders might be more subtest “Wortunähnliche Pseudoworter.”¨ subtle and harder to identify but yet, observ- Based on the test battery described previ- able through their lower school achievement. ously, the operational criteria for the learn- Unfortunately, we cannot rule out this pos- ing disorder subgroups in this study were as sibility because, due to the routine proce- follows: (a) IQ ≥ 80; (b) below-average read- dure within the counseling center and time ing, spelling, and/or arithmetic scores (T < restrictions, we did not apply any direct mea- 40 [i.e., T scores: mean of 50 and SD of 10] sures for language development to these chil- or percentile < 16); and (c) a critical dis- dren. However, given the severe language crepancy of at least 1.2 SDs between IQ and impairment of the children receiving SLE, we overall performance on the standardized tests dare to rely on a significant difference in lan- of school achievement (DYS groups: discrep- guage development between the learning dis- ancy between tests of reading and spelling order and SLI groups. and intelligence; CDSS groups: discrepancy All children were screened with standard- between tests of reading, spelling, and arith- ized tests for intellectual ability, spelling, metic compared with intelligence). The cri- reading, and arithmetic. Different German in- teria for the control group of typically de- telligence tests had been used for the children veloping children were normal intelligence with SLI before (CFT-1, Cattell, Weiss, & (IQ > 85) and performance at average in Osterland, 1980; IDS, Grob, Meyer, & all standardized tests of school achievement Hagemann-von Arx, 2009; K-ABC, Melchers (T ≥40 in reading, spelling, arithmetic). & Preuss, 2001; SON-R, Tellegen, Winkel, & Table 1 summarizes the five groups’ de- Laros, 2003), in the counseling center the scriptive statistics. On average, the CDSS and K-ABC (Melchers & Preuss, 2001) was used. SLI + CDSS groups performed significantly For our study, we relied on the measures lower on the mathematics abilities test than for nonverbal intelligence in the different did the C, DYS, and SLI + DYS groups. At the intelligence tests that captured a similar same time, the DYS, SLI + DYS, CDSS, and concept of logical reasoning. Mathematical SLI + CDSS groups scored significantly lower skills were assessed using standardized on spelling and reading tests than the C group German mathematical achievement tests for did. Inspection of gender distribution patterns second, third, and fourth graders (DEMAT across learning disorder groups showed that 2+, Krajewski, Liehm, & Schneider, 2004; more (60%) children with CDSS were female, DEMAT 3+, Roick, Golitz,¨ & Hasselhorn, whereas more children with DYS (60%), SLI + 2004; DEMAT 4, Golitz,¨ Roick, & Hasselhorn, DYS (81%), and SLI + CDSS (61%) were male. 2006). These multicomponent tests include Analysis of variance revealed that the five ex- computation problems, word problems, and perimental groups did not differ significantly geometry problems. Spelling abilities were in terms of age, F (4, 107) = 1.17, η2 = .04, assessed by the Weingartener spelling tests p = .329. The groups differed significantly in for second and third graders (WRT 2+, Birkel, terms of intelligence, F (4, 107) = 8.11, η2 = 1994a; WRT 3+, Birkel, 1994b) and the West- .23, p = .001. We, therefore, included gen- ermann spelling test for fourth graders (WRT eral intelligence as covariate in all subsequent 4/5, Rathenow, 1980). In both of these stan- analyses. dardized German achievement tests, children All of the children receiving SLE completed insert dictated words into given sentences. additional language measures. For receptive Reading speed abilities were classified on and expressive vocabulary, we carried out the basis of scores on two subtests of the the computerized German vocabulary and Salzburg reading test (SLT; Landerl, Wimmer, word finding test for 6- to 10-year-old chil- & Moser, 1997): the word reading subtest dren (WWT 6–10; Gluck,¨ 2007). For gram- “Textlesen” (short or long version, depending mar, we used the subtests of plural formation

Table 1. Means (SDs) for descriptive characteristics of subgroups: sex, age, IQ, DEMAT mathematic T scores, WRT spelling T scores, and SLT word and nonword reading T scores

DYS SLI + DYS CDSS SLI + CDSS (n = 30) (n = 16) (n = 19) (n = 18) C (n = 30)

Sex (male/female) 18/12 13/3 8/11 11/7 15/15 Age (months) 108.57 111.56 104.00 108.89 108.77 IQ 99.67 (7.31) 93.87 (9.65) 94.06 (6.93) 89.83 (7.67) 100.53 (6.41) Mathematic 48.00 (8.19) 50.25 (8.60) 31.00 (5.77) 27.72 (5.74) 50.34 (5.51) Spelling 33.13 (4.83) 28.19 (3.83) 32.11 (6.30) 26.06 (3.62) 50.73 (6.42) Word reading 32.62 (7.02) 27.81 (7.96) 31.53 (8.32) 28.17 (8.03) 52.17 (8.42) Nonword reading 33.48 (8.06) 29.81 (10.37) 33.35 (11.42) 30.56 (11.34) 49.04 (7.51)

Note.C= normally performing control children matched for chronological age; CDSS = children with combined disorder of scholastic skills; DYS = children with dyslexia; SLI + CDSS = children with speciﬁc language impairment and combined disorder of scholastic skills; SLI + DYS = children with speciﬁc language impairment and dyslexia.

(Plural-Singular bildung) and imitation of (T <40) for vocabulary, grammar, and lan- grammatical structures (Imitation gramma- guage comprehension. tischer Strukturformen); and for language comprehension, we used the subtest understanding of grammatical structures (Verste- Working memory assessment hen grammatischer Strukturformen). These Working memory was assessed by a bat- measures are all part of the Heidelberger tery of 16 tasks: 7 phonological tasks (mem- Sprachentwicklungstest (H-S-E-T; Grimm & ory spans for digits, one-syllable and three- Scholer,¨ 1991). Table 2 illustrates our ﬁnd- syllable words, one-syllable and three-syllable ings that both the SLI + DYS and SLI + CDSS nonwords, and images; nonword repetition), groups performed below average on language 5 visual–spatial tasks (memory span for loca- tasks (exception: receptive vocabulary of the tions, matrix span, corsi-block), and 4 cen- SLI + DYS group). tral executive tasks (double span, backward All children in the SLE subgroups met spans for one-syllable words and digits, count- the diagnostic criteria for SLI: (a) IQ ≥ ing span). A detailed description of all tasks 80 and (b) below-average language scores follows below.

Table 2. Means (SDs) for language characteristics of SLI subgroups: expressive and receptive vocabulary T scores, imitation of grammatical structures and plural-forming T scores, and language comprehension T scores

Vocabulary Grammar

Imitation Of Grammatical Language Expressive Receptive Structures Plural Forming Comprehension

SLI + DYS 30.36 (17.27) 46.33 (19.10) 18.13 (4.39) 31.40 (8.22) 35.13 (6.75) SLI + CDSS 21.69 (18.52) 35.29 (10.96) 17.00 (0.00) 29.00 (8.36) 31.06 (4.34)

Note.SLI+ CDSS = children with speciﬁc language impairment and combined disorder of scholastic skills; SLI + DYS = children with speciﬁc language impairment and dyslexia.

Phonological loop spatial component of visual–spatial memory. The digit span task is one of the con- Nine red blocks were nailed in random po- × ventional measures used to assess phono- sitions on a gray board (23 27.5 cm). The logical short-term capacity. A series of one experimenter tapped a sequence of blocks at to nine digits was presented acoustically therateofonepersecond.Thechildthen at a rate of one digit per second, start- attempted to reproduce the sequence of taps ing with two and continuing up to a max- in the correct order. We used two variations imum of eight digits. Participants had to of the Corsi-block task: simple sequences in- repeat the digits immediately in the pre- volving short distances between blocks with- sented order. The one-syllable and three- out path crossings, and complex sequences syllable word span tasks and the one- involving long distances between blocks with syllable and three-syllable nonword span path crossings. tasks were presented in the same manner as A matrix span task was incorporated in the the digit span measure. In the one-syllable battery as well to measure the static compo- and three-syllable word span tasks, familiar nent of the visual–spatial sketchpad. This task German nouns (e.g., Stern = star, Fisch = assesses memory for random visual–spatial ﬁsh, Erdbeere = strawberry, Briefkasten = patterns of increasing complexity. Patterns of × letterbox) were used; the one-syllable and white and black boxes in a 4 4 matrix were three-syllable nonword span tasks are word- presented on the computer, beginning with like nonwords (e.g., fen, sim, bestrugeln, two black boxes and continuing up to a max- reseubelt). imum of eight black boxes. Immediately after In the images span task, participants were presentation, children were asked to repro- presented a series of pictures of easily rec- duce the pattern in a blank matrix. Two vari- ognizable objects (e.g., sun, umbrella, door, ations of this task were implemented as well: car) on a computer screen and were asked a simple matrix span with the black boxes to recall them in the order of presentation. arranged in simple patterns and a complex This is considered a phonological loop task matrix span with the black boxes located at in which the phonological information is pre- some distance from one another. sented visually instead of acoustically because Central executive the pictures had to be named internally in order to recall them and report them by name. Measures of the central executive are those The German nonword repetition task was that require remembering and processing at developed by Hasselhorn and Korner¨ (1997). the same time. The same items were used for Children had to repeat 24 word-like nonwords the backward digit and word span tasks as of 2, 3, or 4 syllables immediately after their for the forward spans, the only difference be- presentation. Nonwords of different lengths ing that participants were required to recall were presented acoustically in random order. the sequences of items in reverse order. In ad- The number of correctly repeated nonwords dition, a double span task was implemented was taken as the score for this task. This task to assess the children’s ability to coordinate is not a span task, as there was an immediate the functioning of the phonological loop repetition after each word. and the visual–spatial sketchpad. The same pictures as in the images span task were presented but this time in different locations on Visual–spatial sketchpad a3× 3 matrix. Children had to recall the In the location span task, children were pictures simultaneously by verbally recoding shown a series of green dots at different loca- the semantic content (phonological demand) tions on a 3 × 3 matrix and asked to recall and their location (visual–spatial demand) in these locations in the correct order. Corsi- the order of presentation. Thus, this task is block tasks were used to assess the dynamic properly viewed as a central executive task

because of its coordinative requirements. It is sessions with children with SLI and children not a dual task because there is only one reac- with learning disabilities. The DEMAT and tion required: remember the correct pictures WRT measures were carried out with control in the correct order and location. group children in classroom learning groups. The complex counting span task, a mea- All other tests were conducted individually sure of storage and processing efﬁciency, was within a period of 3 weeks. Except for the based on a task designed by Case, Kurland, corsi-block task, all working memory tasks and Goldberg (1982). A series of yellow cir- were administered by computer. The order cles (target items) and squares (distractor of presentation of the working memory tasks items) was presented in a random, computer- was the same for all children (images span, lo- generated pattern. Children were instructed cation span, double span, one-syllable word to count the number of circles. Subsequently, span, three-syllable word span, corsi-block, another map was presented and children nonword repetition, backward word span, again had to count the number of circles. backward digit span, counting span, digit Finally, the experimenter asked the child span, matrix span, one-syllable nonword span, to recall the number of circles counted on and three-syllable nonword span) and was each map. The number of maps presented carried out in one session. per sequence was steadily increased up to a maximum of eight. RESULTS

Stop criterion The first question of our study targeted We used the same stop criterion for all span working memory functioning of children with tasks. The length of the sequences presented DYS with and without SLI. To answer this was increased gradually, beginning with a question, we compared children in the DYS, minimum of two, and increasing to a maxi- SLI + DYS, and control groups for each work- mum of eight items. There were four trials at ing memory subsystem separately. each sequence length. If an error was made, Table 3 presents means and standard devia- the child was given a second attempt at an tions for all working memory measures by the item of the same length. If a child succeeded three groups. The scores of the seven tasks on two successive trials of the same length, assessing the phonological loop functioning the task continued with the next span length. were entered into a multivariate analysis of If a child failed on two successive trials of the variance (MANCOVA). The multivariate main same length, he or she was not presented with effect proved to be significant, F (14, 136) = any further sequences of the same length, but 7.53, η2 = .437, p < .001. The univariate with a sequence of one item shorter. The de- tests also showed significant differences be- pendent measure for all span tasks was the tween groups for all phonological tasks (im- longest sequence of items repeated in correct ages span, F (2, 73) = 12.51, η2 = .255, p < order. Children were credited an extra one- .001; digit span, F (2, 73) = 20.26, η2 = .357, fourth point (0.25) if they repeated a further p < .001; one-syllable word span, F (2, 73) = sequence of the same length correctly (e.g., 14.00, η2 = .277, p < .001; three-syllable word a score of 5.25 was awarded if two of four span, F (2, 73) = 12.85, η2 = .260, p < .001; 5-item sequences were recalled correctly, 5.5 one-syllable nonword span, F (2, 73) = 21.43, if three of four sequences, and 5.75 if all four η2 = .370, p < .001; three-syllable nonword sequences were recalled correctly). span, F (2, 73) = 12.53, η2 = .255, p < .001; nonword repetition, F (2, 73) = 82.44, η2 = Procedure .693, p < .001). We administered standardized tests for In the same way, the scores of the five tasks spelling, reading, arithmetic, intelligence, and assessing the visual–spatial sketchpad were working memory individually in two separate entered into a second MANCOVA. In this case,

Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 306 TOPICS IN LANGUAGE DISORDERS/OCTOBER–DECEMBER 2013 + p .001 < Test, DYS/SLI DYS Post Hoc p DYS/C .001.001.001.001 .001 .002 .036 .689 .001 .003 .001 .001 .001 .001 .020 + < < < < < < < < Test, Post Hoc SLI children with speciﬁc language impairment = DYS + p .001 .001 .001 < < < Hoc Test, DYS/C Post children with dyslexia; SLI = DYS C + DYS SLI s) for working memory measures of subgroups and post hoc tests SD Means ( normally performing control children matched for chronological age; DYS = Images span 4.22 (0.77) 3.41(0.61) 4.60 (0.84) .151 Digit spanOne-syllable word span 4.28 (0.74) 3.50 (0.60) 4.63 (0.64) 4.62 (0.66) 3.88 (0.83) 5.19 (0.61) .149 .005 Three-syllable word spanOne-syllable nonword span 3.64 3.47 (0.47) (0.89) 3.35 2.92 (0.45) (0.60) 3.86 4.27 (0.43) (0.50) .154 Three-syllable nonword spanNonword repetition 1.33 (1.22) 1.06 (1.10) 17.47 (3.13) 2.42 (0.74) 9.56 (3.86) 20.93 (1.78) Location spanCorsi-block simpleCorsi-block complexMatrix span simpleMatrix span complexBackward digit spanBackward 5.68 word (1.34) span 4.80 (0.85) 5.27Double (1.00) spanCounting span 6.72 6.08 (1.11) 4.45 (1.05) 5.17 (1.63) (0.91) 5.50 (1.31) 6.61 (1.62) 4.23 3.45 (1.97) 5.77 (0.66) (1.36) 4.98 (1.08) 3.58 4.98 (0.56) (0.80) 3.31 6.53 (0.64) (1.29) 4.72 (1.52) 3.23 (0.54) 3.78 (0.76) .952 3.81 3.90 .742 (0.89) (0.68) .480 3.82 (0.63) 3.33 .836 (0.55) 3.08 .809 (0.46) 4.04 .027 (0.72) 4.45 (0.97) .250 .731 .803 .215 .976 .600 .334 .013 .015 .005 .575 .429 .749 .962 .895 .005 .781 .150 .099 .C Phonological loop Visual–spatial sketchpad Central executive and dyslexia. Table 3. Note

the multivariate group effect was not signifi- The scores of the tasks assessing the phono- cant, F (10, 140) < 1, η2 = .060 (univariate logical loop were entered into a MANCOVA. tests: location span, F (2, 73) = 1.56, η2 = The multivariate main effect proved to be .041; corsi-block simple task, F (2, 73) < 1, significant, F (14, 116) = 11.84, η2 = .588, η2 = .014; corsi-block complex, F (2, 73) < 1, p < .001. The univariate tests also showed η2 = .022; matrix span simple, F (2, 73) < 1, significant differences between groups for all η2 = .004; matrix span complex, F (2, 73) < 1, phonological tasks (images span, F (2, 63) = η2 = .013). 20.08, η2 = .389, p < .001; digit span, F (2, Third, the scores of the four tasks assess- 63) = 41.85, η2 = .571, p < .001; one-syllable ing the central executive were entered into word span, F (2, 63) = 30.69, η2 = .493, a MANCOVA: here, the multivariate group p < .001; three-syllable word span, F (2, 63) effect, F (8, 142) = 3.69, η2 = .172, p = = 28.26, η2 = .473, p < .001; one-syllable .001, proved to be significant. Univariate nonword span, F (2, 63) = 26.63, η2 = .458, tests showed significant differences between p < .001; three-syllable nonword span, F (2, groups on all central executive memory tasks 63) = 27.70, η2 = .468, p < .001; nonword (digit backward span, F (2, 73) = 5.37, η2 = repetition, F (2, 63) = 90.73, η2 = .472, p < .128, p = .007; word backward span, F (2, .001). 73) = 5.24, η2 = .126, p = .007; double span, Second, the scores of the five tasks assess- F (2, 73) = 5.38, η2 = .128, p = .007; counting ing the visual–spatial sketchpad were en- span, F (2, 73) = 13.71, η2 = .273, p < .001). tered into a second MANCOVA: the multivari- Post hoc tests (Tukey, Table 3) for further ate group effect, F (10, 122) = 2.65, η2 = analysis of group differences revealed that the .179, p = .006, and all univariate tests (loca- DYS and control groups differed in phonologi- tion span, F (2, 64) = 3.06, η2 = .087, p = cal loop and central executive functioning for .047; corsi-block simple task, F (2, 64) = 3.27, most tasks (forward and backward digit span, η2 = .093, p = .045; corsi-block complex, one- and three-syllable nonword span, non- F (2, 64) = 7.93, η2 = .199, p = .001; matrix word repetition, counting span). The SLI + span simple, F (2, 64) = 6.52, η2 = .169, p = DYS group exhibited deficits in all phono- .003; matrix span complex, F (2, 64) = 6.71, logical and central executive tasks compared η2 = .137, p = .002) proved to be significant. with group C. Here, we presume a broad In the same way, the scores of the four deficit concerning all tasks. These deficits be- tasks assessing the central executive were came even more evident in tasks assessing entered into a MANCOVA. Here, the multi- phonological loop and in aspects of central variate group effect proved to be significant, executive compared with children with iso- F (8, 124) = 6.10, η2 = .282, p < .001. Uni- lated DYS (images span, digit span, one- and variate tests showed significant differences three-syllable word span, one-syllable non- between groups on all central executive mem- word span, nonword repetition, counting ory tasks (digit backward span, F (2, 64) = span). Just as we expected, no differences ap- 8.48, η2 = .209, p = .001; word backward peared between the groups on tasks assessing span, F (2, 64) = 14.13, η2 = .306, p < .001; the visual–spatial sketchpad. double span, F (2, 64) = 13.90, η2 = .303, In a second step, we compared working p < .001; counting span, F (2, 64) = 28.94, memory performance of children who met η2 = .475, p < .001). criteria for learning disorder in reading, writ- Post hoc tests (Tukey), illustrated in ing, and arithmetic only (CDSS) with those Table 4, demonstrate that both clinical groups who had an additional diagnosis of SLI (SLI + (CDSS and SLI + CDSS) displayed phono- CDSS) with controls (C). Table 4 presents logical working memory deficits compared mean scores on all working memory tasks by with controls (all phonological tasks). Deficits subgroup. were more pronounced for children with

Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 308 TOPICS IN LANGUAGE DISORDERS/OCTOBER–DECEMBER 2013 p + .001 SLI children with < Test, Post Hoc = CDSS/CDSS CDSS + p + .001.001 .340 .009 .001.001.001.001 .034 .074 .329 .024 .001 .001 .825 .001.001 .589 .262 SLI < < < < < < < < < < Test, CDSS/C Post Hoc p .001 .001 .001 .001 .001 .001 .001 .001 Test < < < < < < < < CDSS/C Post Hoc children with combined disorder of scholastic skills; SLI = CDSS C + CDSS SLI s) for working memory measures of subgroups and post hoc tests SD Means ( normally performing control children matched for chronological age; CDSS = Images span 3.68 (0.56) 3.34 (0.54) 4.60 (0.84) Digit span 4.21 (0.50) 3.65 (0.60) 5.19 (0.61) One-syllable word span 3.43 (0.60) 3.22 (0.44) 4.62 (0.66) Three-syllable word span 3.24 (0.47) 2.92 (0.38) 3.86 (0.43) One-syllable nonword spanThree-syllable nonword span 3.21 (0.98) 1.22 (0.83) 2.88 (0.55) 0.36 (0.83) 4.27 (0.50) 2.42 (0.74) Nonword repetition 18.16 (4.57) 7.35 (3.93) 20.93 (1.78) .018 Location spanCorsi-block simpleCorsi-block complexMatrix span simpleMatrix span complexBackward digit spanBackward word span 4.95 (1.13) 3.80 (1.23) 4.36 (0.78) 5.26 (1.16) 3.22 (1.03) 5.81 (1.01) 4.97 (0.39) 4.90 (1.18) 3.26 (0.47) 6.08 (1.04) 4.01 3.20 (1.73) (0.32) 5.77 (1.36) 4.98 (1.08) 4.98 (0.80) 3.07 (1.01) 3.10 6.53 (0.47) (1.29) 4.77 (1.52) .049 .001 3.90 .048 (0.68) 3.82 (0.63) .002 .003 .012 .996 .999 .960 .433 .083 .001 .048 .005 .155 .100 .611 .705 Double spanCounting span 3.34 (0.55) 3.22 (0.73) 3.14 (0.54) 2.82 (0.34) 4.04 (0.72) 4.45 (0.97) .001 .C Phonological loop Visual–spatial sketchpad Central executive speciﬁc language impairment and combined disorder of scholastic skills. Table 4. Note

SLI + CDSS in comparison with children CDSS. Incidentally, we found that the three- with CDSS only (digit span, one-syllable word syllable nonword span did not differentiate span, three-syllable nonword span, nonword the groups because of significant floor effects. repetition). On the contrary, we detected the highest In contrast, only children with CDSS (with- difference in performance on nonword rep- out SLI) showed deficits associated with etition tasks for measuring the phonological visual–spatial sketchpad compared with con- store. trols (all visual–spatial tasks), whereas chil- The finding that children with additional dren with SLI + CDSS did not. With regard known SLI scored even lower on phonolog- to central executive functioning, both disabil- ical loop tasks than children with DYS and ity groups differed from the control group but CDSS who had not been referred for SLE is not from each other (CDSS and SLI + CDSS: partly in line with results of other studies deficits in all central executive tasks). (Archibald & Gathercole, 2006c; Archibald & Gathercole, 2007; Baird et al., 2011; de Bree DISCUSSION et al., 2010). Nonword repetition is viewed as a rather pure indicator for the phonolog- Researchers currently regard deficits in ical store, because the task does not imply working memory functioning as one major retrieval from long-term memory. Thus, chil- characteristic for language impairment and dren with the additional diagnosis of SLI seem learning disorders (Archibald & Gathercole to be more likely to exhibit a particularly pro- 2006a, 2006b; Schuchardt et al., 2008). Our nounced storage impairment. study investigated the question of whether Extending these findings, Hasselhorn and children with learning disorders and language Werner (2000) conducted a study in which impairment present with the same working they varied acoustic presentation as well as memory deficits. Comparing working mem- syllable length in nonwords. They highlighted ory profiles of children with learning disor- half of the words with white noise to lead to ders (DYS or CDSS) only with profiles for acoustic distortion. In their extended model children with additional language impairment of the phonological loop, Hasselhorn, Grube (SLI) was intended to help us to discover and Mähler (2000) regarded the acoustic dis- whether patterns of working memory deficits tortion effect as a marker for quality of the are disorder-specific. In summary, our results phonological store. Children with SLI and suggested that learning disorders (in reading, language-matched controls performed signifi- spelling, and calculating) and learning disor- cantly worse with increasing syllable length ders combined with SLI share some deficits on acoustically distorted tasks. The perfor- in underlying working memory and are also mance gap between the groups was reduced associated with distinct patterns of working in a condition in which children had to memory difficulties. repeat distorted three- or four-syllable non- Deficits with regard to the phonological words. Hasselhorn and Werner (2000) con- loop and central executive were found for cluded that children with SLI are disturbed learning disorders both with and without SLI; less by white noise (with increasing sylla- however, the phonological impairment was ble length) than language-matched controls, more severe and broader in children who because children with SLI benefit less from met criteria for SLI. In fact, all clinical groups accurate acoustic presentation. The authors displayed significant deficits in tasks involv- concluded that working memory impairment ing the phonological loop. Children with in children with SLI, therefore, is due to re- combined language and learning impairment duced processing quality in the phonologi- (SLI + DYS and SLI + CDSS) performed even cal store, especially when the soundscapes worse, however, on the majority of phonolog- exceed a certain number of information ical tasks than children with isolated DYS or units.

In addition to the phonological working Findings of Donlan et al. (2007) and Fazio memory deficits, all four clinical groups dis- (1999) point to the same line of argument. played deficits in central executive function- Thus, we would expect mathematical prob- ing. Therefore, deficits in central executive lems in children with SLI especially on tasks functioning appear to be characteristic for that severely depend on language comprehen- children with DYS and CDSS (Schuchardt sion but not on tasks that can be solved by et al., 2008) as well as for children with SLI intact visual–spatial perception and memory. (Archibald & Gathercole, 2006b; Marton & In summary, we would conclude that ana- Schwartz, 2003; Montgomery & Evans, 2009). lyzing working memory functioning in-depth Nevertheless, it should be considered that could help us to differentiate between vari- most of the central executive tasks used also ous disorders according to underlying cogni- involve the participation of the phonological tive deficits. Certainly, the assessed patterns loop to some extent because phonological in- of deficits in different clinical groups do not formation has to be processed. Therefore, we allow for distinct diagnoses of specific dis- cannot rule out the possibility that the phono- orders. Furthermore, other disorders besides logical deficit is the predominant impairment the ones included in this study are associ- in children with learning disorders with or ated with working memory problems. But at without additional SLI. least we could become more precise in rec- When we look at the visual–spatial sketch- ommending intervention measures that rely pad, we find different patterns of results. In on different strengths and difficulties of the line with earlier studies, we could observe children. that children with CDSS exhibit deficits in According to the results of this study, this domain (Maehler & Schuchardt, 2009; phonological and central executive problems Schuchardt et al., 2008). Interestingly, we must be expected in children with learn- could not provide evidence for disadvan- ing disorders with or without SLI. Therefore, tages in visual-spatial working memory for teaching should take these deficits into ac- the children with CDSS + SLI (see Nithart count and try to reduce the task demands et al., 2009). As only children with combined with regard to the amount of information to learning disorder (CDSS) showed deficits in be processed or integrated. And children with the visual–spatial working memory, it may SLI should be supported to profit from their be that visual–spatial malfunctions influence apparently intact visual–spatial memory, for arithmetic problems of these children. In con- example, by providing training on relevant trast, we assume that children with addi- memory strategies or visual supports to lan- tional language problems (CDSS + SLI) could guage intervention activities. Future research exhibit arithmetic problems because of lan- is needed to evaluate working memory train- guage problems, as their visual–spatial working programs in order to explore the possibil- ing memory turned out to be unimpaired. ity to overcome working memory deficits.

REFERENCES

Archibald, L. M., & Gathercole, S. E. (2006a). Visuospa- Archibald, L. M., & Gathercole, S. E. (2006c). Nonword tial immediate memory in language impairment. Jour- repetition in specific language impairment. Psycho- nal of Speech, Language, and Hearing Research, 49, nomic Bulletin & Review, 14, 919–924. 265–277. Archibald, L. M., & Gathercole, S. E. (2007). Nonword Archibald, L. M., & Gathercole, S. E. (2006b). Short- repetition in specific language impairment: More than term and working memory in specific language a phonological short-term memory deficit. Psycho- impairment. International Journal of Lan- nomic Bulletin & Review, 14, 919–924. guage & Communication Disorders, 41, 675– Baddeley, A. D. (1986). Working memory.Oxford:Uni- 693. versity Press.

Baddeley, A. D. (1996). Exploring the central executive. Journal of Speech, Language, and Hearing Research, Quarterly Journal of Experimental Psychology, 49, 42, 420–431. 5–28. Gluck,¨ C. W. (2007). Wortschatz- und Wortfindungstest Baddeley, A. D. (2000). The episodic buffer: A new com- fur¨ 6- bis 10-Jahrige¨ (WWT 6–10).Munchen,¨ Ger- ponent of working memory? Trends in Cognitive Sci- many: Elsevier, Urban & Fischer. ences, 4, 417–423. Golitz,¨ D., Roick, T., & Hasselhorn, M. (2006). Deutscher Baird, G., Slonims, V., Simonoff, E., & Dworzynski, K. Mathematiktest fur¨ vierte Klassen (DEMAT 4) (2011). Impairment in non-word repetition: A marker [German mathematics test for fourth grades (DE- for language impairment or reading impairment? De- MAT 4).Gottingen,¨ Germany: Hogrefe. velopmental Medicine and Child Neurology, 53, Grimm, H., & Scholer,¨ H. (1991). Heidelberger 711–716. Sprachentwicklungstest (H-S-E-T). 2. verbesserte Aufl. Birkel, P. (1994a). Weingartener Grundwortschatz Gottingen,¨ Germany: Hogrefe. Rechtschreib-Test fur¨ zweite und dritte Klassen Grob, A., Meyer, C. S., & Hagemann-von Arx, P. (WRT 2+) [Weingarten basic vocabulary spelling (2009). Intelligence and Development Scale (IDS). test for second and third grades (WRT 2+)]. Intelligenz- und Entwicklungsskalen fur¨ Kinder von Gottingen,¨ Germany: Hogrefe. 5–10 Jahren. Bern, Switzerland: Huber. Birkel, P. (1994b). Weingartener Grundwortschatz Hasselhorn, M., Grube, D., & Mähler, C. (2000). Rechtschreib-Test fur¨ dritte und vierte Klassen (WRT Theoretisches Rahmenmodell fur¨ ein Diagnostikum 3+) [Weingarten basic vocabulary spelling test for zur differentiellen Funktionsanalyse des phonologis- third and fourth grades (WRT 3+)].Gottingen,¨ chen Arbeitsgedächtnisses. In M. Hasselhorn, W. Germany: Hogrefe. Schneider, & H. Marx (Hrsg.), Diagnostik von Case, R. D., Kurland, M., & Goldberg, J. (1982). Opera- Lese-Rechtschreibschwierigkeiten (Reihe “Tests und tional efficiency and the growth of short-term memory Trends”, Neue Folge, Band 1, S. 167–181). Gottingen,¨ span. Journal of Experimental Child Psychology, 33, Germany: Hogrefe Publishing. 386–404. Hasselhorn, M., & Korner,¨ K. (1997). Nachsprechen Cattell, R. B., Weiss, R., & Osterland, J. (1980). Grundin- von Kunstwortern:¨ Zum Zusammenhang zwischen telligenztest CFT1.Gottingen,¨ Germany: Hogrefe. Arbeitsgedächtnis und syntaktischen Sprachleistun- Catts, H., Adolf, S., Hogan, T., & Ellis Weismer, S. (2005). gen bei Sechs- und Achtjährigen [Repeating non- Are specific language impairments and dyslexia dis- words: The relationship between working memory tinct disorders? Journal of Speech, Language, and and syntactic competence in six- and eight-year- Hearing Research, 48, 1378–1396. olds]. Zeitschrift fur¨ Entwicklungspsychologie und Cowan, R., Donlan, C., Newton, E. J., & Lloyd, D. (2005). Padagogische¨ Psychologie, 29, 212–224. Number skills and knowledge in children with spe- Hasselhorn, M., & Werner, I. (2000). Zur Bedeu- cific language impairment. Journal of Educational tung des phonologischen Arbeitsgedächtnisses fur¨ Psychology, 97, 732–744. die Sprachentwicklung. In H. Grimm (Ed.), Enzyk- de Bree, E., Wijnen, F., & Gerrits, E. (2010). Non-word rep- lopadie¨ der Psychologie (Part C, Series III, Vol. etition and literacy in Dutch children at-risk of dyslexia 3: Sprachentwicklung, pp. 363–378). Gottingen,¨ and children with SLI: Results of the follow-up study. Germany: Hogrefe. Dyslexia, 16, 36–44. Hick, R. F., Botting, N., & Conti-Ramsden, G. (2005). Donlan, C., Cowan, R., Newton, E. J., & Lloyd, D. (2007). Short-term memory and vocabulary development in The role of language in mathematical development: children with Down syndrome and children with spe- Evidence from children with specific language impair- cific language impairment. Developmental Medicine ment. Cognition, 103, 23–33. & Child Neurology, 47, 532–538. Eisenmajer, N., Ross, N., & Pratt, C. (2005). Specificity Hoffman, I. M., & Gillam, R. B. (2004). Verbal and spa- and characteristics of learning disabilities. Journal of tial information processing in children with specific Child Psychology and Psychiatry and Allied Health language impairment. Journal of Speech, Language, Professions, 46, 1108–1115. and Hearing Research, 47, 114–125. Fazio, B. B. (1994). The counting abilities of children Krajewski, K., Liehm, S., & Schneider, W. (2004). with specific language impairment: A comparison of Deutscher Mathematiktest fur¨ zweite Klassen (DE- oral and gestural tasks. Journal of Speech, Language, MAT 2+) [German mathematics test for second and Hearing Research, 37, 358–368. grades (DEMAT 2+)].Gottingen,¨ Germany: Hogrefe. Fazio, B. B. (1996). Mathematical abilities of children Landerl, K., Bevan, A., & Butterworth, B. (2004). De- with specific language impairment: A 2-year follow- velopmental dyscalculia and basic numerical capaci- up. Journal of Speech, Language, and Hearing Re- ties: A study of 8–9-year-old students. Cognition, 93, search, 39, 839–849. 99–125. Fazio, B. B. (1999). Arithmetic calculation, short-term Landerl, K., Wimmer, H., & Moser, E. (1997). Salzburger memory, and language performance in children with Lese- und Rechtschreibtest (SLT) [Salzburg reading specific language impairment: A 5-year follow-up. and spelling test (SLT)]. Bern, Switzerland: Huber.

Logie, R. H. (1995). Visual-spatial working memory. Further evidence for the fractionation of visuo-spatial Hove, United Kingdom: Erlbaum. memory. The Quarterly Journal of Experimental Psy- Maehler, C., & Schuchardt, K. (2009) Working memory chology, 54, 397–420. functioning in children with learning disabilities: Does Rathenow, P. (1980). Westermann rechtschreibtest 4/5 intelligence make a difference? Journal of Intellectual (WRT 4/5). Braunschweig, Germany: Westermann. Disabilities Research, 53, 3–10. Riccio, C. A., Cash, D. L., & Cohen, M. J. (2007). Learn- Marton, K., & Schwartz, R. G. (2003). Working mem- ing and memory performance of children with spe- ory capacity and language processes in children with cific language impairment (SLI). Applied Neuropsy- specific language impairment. Journal of Speech, Lan- chology, 14, 255–261. guage, and Hearing Research, 46, 1138–1153. Roick, T., Golitz,¨ D., & Hasselhorn, M. (2004). Deutscher McArthur, G. M., Hogben, J. H., Edward, V. T., Heath, Mathematiktest fur¨ dritte Klassen (DEMAT 3+) S. M., & Mengler, E. D. (2000). On the “specifics” of [German mathematics test for third grades (DEMAT specific reading disability and specific language im- 3+)].Gottingen,¨ Germany: Hogrefe. pairment. Journal of Child Psychological Psychiatry Schuchardt, K., Mähler, C., & Hasselhorn, M. (2008). and Allied Health Professions, 41, 869–874. Working memory deficits in children with different McLean, J. F., & Hitch, G. J. (1999). Working memory im- learning disabilities. Journal of Learning Disabilities, pairments in children with specific arithmetic learning 41, 514–523. difficulties. Journal of Experimental Child Psychol- Siegel, L. S., & Ryan, E. B. (1989). The development of ogy, 74, 240–260. working memory in normally achieving and subtypes Melchers, P., & Preuss, U. (2001). Assessment battery of learning disabled children. Child Development, 60, for children (K-ABC), German version (5th ed.). 973–980. Gottingen,¨ Germany: Hogrefe. Swanson, H. L., & Sachse-Lee, C. (2001). A subgroup Montgomery, J. W., & Evans, J. L. (2009). Complex sen- analysis of working memory in children with reading tence comprehension and working memory in chil- disabilities: Domain-general or domain-specific defi- dren with specific language impairment. Journal of ciency? Journal of Learning Disabilities, 34, 249– Speech, Language, and Hearing Research, 52, 269– 263. 288. Tellegen, P. J., Winkel, M., & Laros, J. A. (2003). Non- 1 Nithart, C., Demont, E., Majerus, S., Leybaert, J., Poncelet, verbal intelligence-test (SON-R 5 /2-17), German ver- M., & Metz-Lutz, M. N. (2009). Reading disabilities sion.Gottingen,¨ Germany: Hogrefe. in SLI and dyslexia result from distinct phonological Tomblin, J. B., Zhang, X., Buckwalter, P., & Catts, H. impairments. Developmental Neuropsychology, 34, (2000). The association of reading disability, behav- 296–311. ioral disorders, and specific language impairment in O’Shaughnessy, T., & Swanson, H. L. (1998). Do im- second grade children. Journal of Child Psychology mediate memory deficits in students with learn- and Psychiatry and Allied Health Professions, 41, ing disabilities in reading reflect a developmental 473–482. lag or deficit? A selective meta-analysis of the lit- Vellutino, F. R., Fletcher, J. M., Snowling, M. J., & Scanlon, erature. Learning Disability Quarterly, 21, 123– D. M. (2004). Specific reading disability (dyslexia): 148. What have we learned in the past four decades. Jour- Pickering, S. J., Gathercole, S. E., Hall, M., & Lloyd, S. A. nal of Child Psychology and Psychiatry and Allied (2001). Development of memory for pattern and path: Health Professions, 45, 2–40.