Are Specific Language Impairment and Dyslexia Distinct Disorders

7/29/2019 Are Specific Language Impairment and Dyslexia Distinct Disorders

1/20

University of Nebraska - Lincoln

DigitalCommons@University of Nebraska - Lincoln

Special Education and Communication DisordersFaculty Publications

Department of Special Education andCommunication Disorders

1-1-2005

Are Specic Language Impairment and DyslexiaDistinct Disorders?

Hugh W. CasUniversity of Kansas

Suzanne M. AdlofUniversity of Kansas

Tiany HoganUniversity of Nebraska - Lincoln, [email protected]

Susan Ellis WeismerUniversity of WisconsinMadison

Follow this and additional works at: hp://digitalcommons.unl.edu/specedfacpub

Part of the Special Education and Teaching Commons

is Article is brought to you for free and open access by the Department of Special Education and Communication Disorders at

DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Special Education and Communication Disorders Faculty

Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.

Cas, Hugh W.; Adlof, Suzanne M.; Hogan, Tiany; and Weismer, Susan Ellis, "Are Specic Language Impairment and DyslexiaDistinct Disorders?" (2005). Special Education and Communication Disorders Faculty Publications. Paper 13.hp://digitalcommons.unl.edu/specedfacpub/13
http://digitalcommons.unl.edu/?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/spec_ed?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/spec_ed?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/801?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub/13?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub/13?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/801?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/spec_ed?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/spec_ed?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/specedfacpub?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://digitalcommons.unl.edu/?utm_source=digitalcommons.unl.edu%2Fspecedfacpub%2F13&utm_medium=PDF&utm_campaign=PDFCoverPages


2/20

In recent years, there has been considerable interestin the relationship between developmental disordersof oral and written language (Bishop & Snowling, 2004;

Catts & Kamhi, 2005). The most widely investigateddevelopmental written language disorder is dyslexia,which is characterized by a signicant decit in printedword recognition in the face of adequate instruction andgeneral cognitive abilities (Lyon, Shaywitz, & Shaywitz,2003). Research has shown that a phonological process-ing decit underlies word-reading difculties in manychildren with dyslexia (Fletcher et al., 1994; Gillon,2004). In the case of oral language, the most frequentlystudied developmental disorder is specic language im-

pairment (SLI). Children with SLI exhibit decits in se-mantics, syntax, and discourse in the presence of normalnonverbal cognitive abilities (Leonard, 1998; Tager-Flus-

berg & Cooper, 1999).At rst glance, it would seem that SLI and dyslexia

are two distinct developmental language disorders; SLIprimarily represented by difculties in semantics, syn-tax, and discourse, and dyslexia characterized by prob-lems in phonological processing and word reading.However, recent ndings suggest there may be a closerassociation between these developmental language dis-orders. Children with dyslexia have been shown tohave early decits in semantics and syntax (Gallagher,

Published inJournal of Speech, Language, and Hearing Research 48 (December 2005), pp. 13781396; doi: 10.1044/1092-4388(2005/096) Copy-

right 2005 American Speech-Language-Hearing Association. Used by permission.

Submitted October 26, 2004; accepted March 13, 2005.

Are Specifc Language Impairmentand Dyslexia Distinct Disorders?

Hugh W. Catts, Suzanne M. Adlof, and Tiffany P. HoganUniversity of Kansas, Lawrence

Susan Ellis WeismerUniversity of WisconsinMadison

Corresponding author Hugh W. Catts, Department of Speech-Language-Hearing, 1000 Sunnyside Avenue, University of Kansas,

Lawrence, KS 66045; email [email protected]

Abstract

Purpose: The purpose of this study was to determine whether specic language impairment (SLI) and dyslexia are distinct

developmental disorders.

Method: Study 1 investigated the overlap between SLI identied in kindergarten and dyslexia identied in 2nd, 4th, or 8th

grades in a representative sample of 527 children. Study 2 examined phonological processing in a subsample of par-

ticipants, including 21 children with dyslexia only, 43 children with SLI only, 18 children with SLI and dyslexia, and 165

children with typical language/reading development. Measures of phonological awareness and nonword repetition

were considered.

Results: Study 1 showed limited but statistically signicant overlap between SLI and dyslexia. Study 2 found that children

with dyslexia or a combination of dyslexia and SLI performed signicantly less well on measures of phonological pro -

cessing than did children with SLI only and those with typical development. Children with SLI only showed only mild

decits in phonological processing compared with typical children.

Conclusions:These results support the view that SLI and dyslexia are distinct but potentially comorbid developmental lan-

guage disorders. A decit in phonological processing is closely associated with dyslexia but not with SLI when it oc -

curs in the absence of dyslexia.

Keywords: specic language impairment, dyslexia, phonological processing, phonological awareness, nonword repetition

1378


3/20

Are Spe ci fi c LAnguAge impAirment An d dySLexiA d i S t i nct d i S o rde rS? 1379

Frith, & Snowling, 2000; P. Lyytinen, Poikkeus, Laakso,Eklund, & Lyytinen, 2001; Scarborough, 1990, 1991;Snowling, Gallagher, & Frith, 2003), and children withSLI have often been noted to have phonological process-ing decits and subsequent problems in word recogni-tion (Catts, 1993; Snowling, Bishop, & Stothard, 2000).These ndings have led some to conclude that dyslexia

and SLI represent variants of the same developmentallanguage disorder (Kamhi & Catts, 1986; Tallal, Allard,Miller, & Curtiss, 1997). However, in a recent review ofbehavioral, neurological, and genetic evidence, Bishopand Snowling (2004) concluded that SLI and dyslexiaare best treated as two different but overlapping devel-opmental disorders. In this article, we present the resultsfrom a longitudinal study that provide further evidencefor a distinction between SLI and dyslexia.

Dyslexia

According to the International Dyslexia Association(IDA), dyslexia is a specic learning disability charac-terized by difculties with accurate and/or uent wordrecognition and spelling (Lyon et al., 2003). The IDAdenition further proposes that these difculties typi-cally result from a decit in the phonological componentof language and are unexpected in relation to age andother cognitive and academic abilities. The phonologicaldifculty most often associated with dyslexia is a decitin phonological awareness, ones sensitivity to, or explicitawareness of, the sound structure of language (Stanov-ich, 1988). It is generally argued that problems in pho-

nological awareness make it difcult for children withdyslexia to learn how to apply the alphabetic principleto decode and spell printed words (Gillon, 2004). Nu-merous studies have documented a decit in phonologi-cal awareness in children with dyslexia or in children atrisk for this disorder (Bradley & Bryant, 1983; Gallagheret al., 2000; Fletcher et al., 1994; H. Lyytinen et al., 2001).

The phonological processing problems associatedwith dyslexia also extend to areas other than phonolog-ical awareness. Specically, children with dyslexia oftendemonstrate problems in phonological memory (Brady,Shankweiler, & Mann, 1983; Vellutino & Scanlon, 1982).Among the phonological memory tasks with which chil-

dren with dyslexia have difculty is the nonword repe-tition task, in which participants must store and repeat aphonological sequence that could be a word in the lan-guage but is not. Research has shown that children withdyslexia consistently perform less well than control par-ticipants on nonword repetition tasks (Brady, Poggie,& Rapala, 1989; Catts, 1986; Hulme & Snowling, 1992;Kamhi & Catts, 1986; Snowling, 1981; van Daal & vander Leij, 1999; van der Bob & van der Pijl, 1997). Stud-ies have also demonstrated that heritability for dyslexia

is higher when the disorder is combined with a decitin nonword repetition (Bishop, 2001; Bishop, Adams,& Norbury, 2004; Raskind, Hsu, Berninger, Thomson,& Wijsman, 2000). Finally, research suggests a link be-tween decits in phonological memory and phonolog-ical awareness in that both decits may result from aninefciency in the formation of phonological representa-

tions (Elbro, 1996; Metsala & Walley, 1998).Other research indicates that the language problems

in dyslexia may go beyond those in phonological pro-cessing. Studies show that children with dyslexia mayalso have problems in semantics, syntax, and discourse(Catts, Fey, Tomblin, & Zhang, 1999; McArthur, Hog-ben, Edwards, Health, & Mengler, 2000; Plaza, Cohen, &Chevrie-Muller, 2001). For ease of reference, these prob-lems are referred to in this article as oral language difcul-ties and do not include a phonological processing def-icit. Some of these oral language difculties could bethe result of reading problems themselves. Poor readersdo not read as much as good readers do, and as a re-sult may not have the same language learning opportu-nities as do good readers. However, a growing numberof studies demonstrate that oral language difculties arepresent in children at risk for dyslexia prior to school en-try (Gallagher et al., 2000; P. Lyytinen et al., 2001; Scar-borough, 1990, 1991). For example, Scarborough (1990,1991) followed 20 children with a family risk of dys-lexia from 30 months through second grade. The at-riskchildren who later developed dyslexia showed syntac-tic decits in terms of reduced mean length of utteranceand restricted use of syntactic structures during the pre-school years. Whereas these oral language difculties

were present, they were typically not severe enough forchildren to have been identied as having SLI (Scarbor-ough & Dobrich, 1990). This has also been the case forother studies that have documented oral language prob-lems in children with a family risk for dyslexia (e.g., Gal-lagher et al., 2000).

SLI

Specic language impairment represents a disorderin the development of oral language (Leonard, 1998). Itis specic in that children with SLI have nonverbal IQ

scores within normal limits and no hearing or socioemo-tional decits. The oral language problems observedin SLI include problems in semantics, syntax, and dis-course (Paul, 2001). Particular attention has been givento decits in morpho-syntax (Leonard, 1998). For exam-ple, children with SLI have been shown to have prob-lems in the acquisition of tense marking, and this decithas been posited by some as a psycholinguistic or clini-cal marker of SLI (Conti-Ramsden, Botting, & Faragher,2001; Bedore & Leonard, 1998; Rice & Wexler, 1996).


4/20

1380 cAttS et AL . in Journal of Speech, lan gu a ge, an d hearing re S e a rch 48 (2005)

Children with SLI have also been reported to haveproblems in phonological processing. These include def-icits in phonological awareness (Briscoe, Bishop, &Nor-bury, 2001; Catts, 1993; Joffe, 1998; Nathan, Stackhouse,Goulandris, & Snowling, 2004; Snowling et al., 2000)and phonological memory (Bishop, North, & Donlan,1996; Briscoe et al., 2001; Dollaghan & Campbell, 1998;Edwards & Lahey, 1998; Ellis Weismer et al., 2000; Gath-ercole & Baddeley, 1990; Kamhi & Catts, 1986). In fact,considerable attention has been paid to a link betweenSLI and decits in phonological memory. Specically,

Gathercole and Baddeley (1990) observed that childrenwith SLI performed poorly on measures of phonologicalmemory, especially nonword repetition. On the basis oftheir results, they proposed that SLI involves a specicdecit in the phonological loop component of workingmemory, which causes difculties in semantic and syn-tactic development. Furthermore, Bishop et al. (1996)proposed that difculty in nonword repetition may bea good phenotypic marker for SLI (also see Conti-Rams-den et al., 2001).

Given the problems that children with SLI appear tohave in phonological processing, it would be expectedthat these children would also have difculties in word

reading. Indeed, studies have shown that children withSLI often have problems in learning to recognize printedwords (Bishop & Adams, 1990, Catts, 1993; Catts, Fey,Tomblin, & Zhang, 2002; McArthur et al., 2000; Snowl-ing et al., 2000; Tallal, Allard, & Curtiss, 1988). For ex-ample, Tallal et al. (1988) found that approximately67% of children with SLI at 4 years of age showed lowachievement in word recognition at age 8. Silva, Wil-liams, and McGee (1987) also reported evidence of lowword reading achievement in children with SLI, but at a

lower prevalence rate (approximately 35%). In addition,McArthur et al. (2000) found in a series of three studiesthat approximately 50% of school-age children with SLIconcurrently had a specic reading disability charac-teristic of dyslexia. Snowling et al. (2000) also reportedhigh rates of dyslexia in children with SLI.

Relationship Between Dyslexia and SLI

Given the documented overlap between SLI and dys-lexia, what is the best way to characterize the relation-

ship between these disorders? Three possible models ofthis relationship are depicted in Figure 1. According toModel 1, dyslexia and SLI are different manifestationsof the same underlying cognitive decit (Kamhi & Catts,1986; Tallal et al., 1997). In this model, a phonologicalprocessing decit is responsible for both disorders. Thedifferent manifestations (SLI vs. dyslexia), however, re-sult from variations in the severity of the phonologicalprocessing decit. If the decit is severe, children willshow problems in word reading as well as difculties inoral language (i.e., SLI). If, on the other hand, the def-icit is less severe, children will demonstrate problemsin word reading and show limited or no problems in

oral language (i.e., dyslexia). If Model 1 is correct, thereshould be a great deal of overlap between SLI and dys-lexia. Children with SLI and those with dyslexia shouldhave problems on tasks involving phonological process-ing and word reading; however, these problems shouldbe more severe in children with SLI.

Model 2 indicates that dyslexia and SLI are partiallysimilar but distinct disorders. A model such as this wasproposed by Bishop and Snowling (2004) in a recent re-view of the literature. According to Model 2, both disor-

Figure 1. Models of the relationship between specic language impairment (SLI) and dyslexia.


5/20


ders are similar in that they are characterized by a pho-nological processing decit that underlies word-readingproblems. Unlike Model 1, the severity of the phonolog-ical decit is equal, on average, in dyslexia and SLI. Thedisorders, however, are different in that SLI involvesan additional cognitive decit or decits, which oper-ates independently of the phonological processing def-

icit and causes problems in the development of oral lan-guage. If this model is accurate, it would be expectedthat there would be considerable overlap between dys-lexia and SLI in that both disorders would have similarproblems in phonological processing and word reading.However, the disorders would be distinct in that chil-dren with SLI would have difculties in oral language,and those with dyslexia would show normal or at least

low normal development in this area.Model 3 depicts a third possible relationship between

dyslexia and SLI. According to this model, dyslexia andSLI are distinct developmental disorders with differ-ent cognitive decits and behavioral manifestations. Asshown in this model, a phonological processing decitis the core decit in dyslexia and is responsible for theword reading problems of children with this condition.Children with SLI, on the other hand, have a differentdecit(s) at the core of their disability that causes prob-lems in the development of oral language. Unlike Model2, in which the overlap results from both disordersshowing a decit in phonological processing, the over-lap in Model 3 is due to comorbidity (Caron & Rutter,1991). That is, although the disorders are distinct, theyare related and sometimes occur together in the same in-dividual. If this view is correct, it would be expected that

greater-than-chance overlap should be found betweenSLI and dyslexia. However, numerous cases should beobserved of children with SLI who do not have wordreading problems (and a phonological processing de-cit) and children with dyslexia who do not have a his-tory of oral language difculties.

In this article we report the results of two studies thatsought to determine which of the above models best char-acterizes the relationship between dyslexia and SLI.1 InStudy 1, we used a large longitudinal database to studythe overlap between these developmental disorders. Thisdatabase included measurements of oral language (andIQ) in kindergarten, second, fourth, and eighth grades

and assessments of word recognition in second, fourth,and eighth grades. In our analyses, we examined thepercentage of children with SLI in kindergarten whohad dyslexia in second, fourth, and eighth grades. Con-versely, we also determined the percentage of childrenidentied as having dyslexia in second, fourth, or eighthgrades who showed SLI in kindergarten.

In both of the above cases, SLI was identied duringkindergarten. The decision to identify SLI at this pointwas based on several factors. First, because SLI is char-acterized by problems in the development of oral lan-guage, it has traditionally been diagnosed during thepreschool years (Leonard, 1998; Rice & Wexler, 1996).Second, it is preschool problems in oral language that

have often been argued to be an early manifestation ofdyslexia (Scarborough, 2005; Snowling et al., 2003; Tal-lal et al., 1997). Third, and perhaps most important, byidentifying SLI in kindergarten, prior to formal readinginstruction, we reduce the possibility that the oral lan-guage impairments associated with SLI are the result ofdyslexia rather than an early manifestation of the dis-order. As noted above, children with dyslexia read lessoften and thus are not as able to take advantage of thelanguage learning opportunities that accompany read-ing experience (Stanovich, 1986). This may in turn leadto the development of language problems during theschool years (Share & Silva, 1987). Thus, by identifyingoral language impairments in kindergarten, one can re-duce the impact of poor reading on this diagnosis.

Study 1: Overlap Between SLI and Dyslexia

Method

Participants

Children with SLI and children with dyslexia wereselected from a population-based sample of children

participating in a longitudinal study of language andreading development. The specic criteria used to selectparticipants with SLI and those with dyslexia are de-scribed at the end of the Method section. In this section,the participant sample from which these children weredrawn is described. This sample included 527 school-age children. These children originally participated inan epidemiologic study of language impairments in kin-dergarten children (Tomblin et al., 1997). The epidemi-ologic investigation used a stratied cluster sample of7,218 children. This sample was stratied by residen-tial setting (i.e., rural, urban, suburban) and cluster-sampled by school building. The sample was 33% rural,37% urban, 30% suburban; 51% male, 49% female; and83% White, 12.7% African American, and 4% other. Allavailable kindergarten children in selected schools werescreened for language impairments using a test of 40items taken from the Test of Language Development2: Primary (TOLD2:P; Newcomer & Hammill, 1988).These items had been shown to have high sensitivity for

1 This study provides evidence concerning the relationship between dyslexia and SLI in English-speaking children. Although these disordersare thought to be constitutional in origin, the nature of the spoken language and/or orthography could inuence the specic relationship be-tween them.


6/20


the identication of SLI (see Tomblin, Records, & Zhang,1996). Children who failed the screening, and a randomsample who passed, were given a diagnostic test batteryof language abilities and other measures. Data from thisassessment were used to estimate the prevalence of lan-guage impairments in kindergarten children (Tomblinet al., 1997).

On completion of the epidemiologic study, a subsam-ple of children was solicited to participate in a follow-uplongitudinal investigation conducted by the Child Lan-guage Research Center (Tomblin, Zhang, Weiss, Catts,& Ellis Weismer, 2004). Because the primary purposeof the center is the study of language impairments, allchildren who displayed these impairments on the kin-dergarten diagnostic battery were asked to participate.Of the 642 children who met this criterion, permissionto participate was received for 328. In addition to thesechildren, a random sample of the children without im-pairments was recruited. Permission to participate wasobtained for 276 nonimpaired children, yielding a totalsample of 604 children. These children, segregated bydiagnostic category, did not differ signicantly in termsof demographic characteristics or language and cogni-tive abilities from those children who were not asked ordid not choose to participate. All children were mono-lingual English speakers and had no history of sensorydecits or neurological disorders. In addition, no childhad been diagnosed with autism or mental retardationin the epidemiologic study.

All the above 604 children completed the kindergar-ten and second-grade test batteries. Thirty-four chil-dren were lost to attrition by fourth grade and another

43 were lost by eighth grade. The latter 77 children didnot differ signicantly in language or nonverbal cogni-tive abilities from the remaining 527 children; however,the children who remained in the study throughout theproject had signicantly higher reading achievement insecond grade than those who dropped out. This differ-ence in reading achievement could have inuenced theestimate of the prevalence of dyslexia in participantswith SLI; however, analyses showed no evidence of suchinuence. Children with SLI from the sample of 604 (N= 123) had rates of dyslexia in second grade (the onlygrade in which rates were available for both groups) al-most identical to those of the subset of children with SLI

who remained in the study through eighth grade (N=106). Therefore, to better allow for comparisons acrossgrades, children with SLI (and/or dyslexia) were drawnfrom the 527 children who completed testing througheighth grade.

Materials

Language. In kindergarten, language abilities wereassessed by ve subtests of the TOLD2:P (Newcomer

& Hammill, 1988) and a narrative story task (Culatta,Page, & Ellis, 1983). Local norms were used to convertraw scores to z scores. These norms were based on datafrom 1,502 children who received the kindergarten testbattery in the epidemiologic study. The z scores fromthe TOLD2:P Picture Identication and Oral Vocabu-lary subtests were combined to form a vocabulary com-

posite score. The z scores from the TOLD2:P Gram-matic Understanding, Grammatic Completion, andSentence Imitation subtests were used to form a gram-mar composite score, whereas z scores from the com-prehension and recall portions of the narrative taskwere used as a narrative composite score. To derive areceptive language composite score, z scores from thePicture Identication, Grammatic Understanding, andnarrative comprehension tasks were combined. To ob-tain an expressive language composite score, z scoresfrom the Oral Vocabulary, Grammatic Completion,Sentence Imitation, and narrative recall tasks wereused. An overall language composite score was alsocalculated using the expressive and receptive languagecomposite scores.

Intelligence. The criteria we used to identify SLI anddyslexia required estimates of nonverbal and Full ScaleIQ. As part of the diagnostic battery in kindergarten,children were administered the Block Design and Pic-ture Completion subtests of the Wechsler Preschooland Primary Scale of IntelligenceRevised (Wechsler,1989). These subtests were combined to form a compos-ite measure of nonverbal IQ (Bishop & Adams, 1990;LoBello, 1991). Nonverbal IQ was assessed again in sec-ond and eighth grades. In second grade, the full Perfor-

mance scale of the Wechsler Intelligence Scale for Chil-drenIII (Wechsler, 1991) was administered. In eighthgrade, the Block Design and Picture Completion sub-tests from the Wechsler Intelligence Scale for Chil-drenIII were given.

Full Scale IQ was also estimated in second, fourth,and eighth grades. At each of these grades, the PeabodyPicture Vocabulary TestRevised (PPVTR; Dunn &Dunn, 1981) served as an index of verbal intelligence.Scores on the PPVTR were combined with those ontests of nonverbal IQ to form a composite z score to es-timate Full Scale IQ at each grade. Because no measureof nonverbal IQ was available in fourth grade, we com-

bined childrens scores on the second grade measure ofnonverbal IQ with that on the fourth grade PPVTR tocreate an estimate of Full Scale IQ for fourth grade.

Word recognition. The Word Identication and WordAttack subtests of the Woodcock Reading MasteryTestsRevised (Woodcock, 1987) were administered insecond, fourth, and eighth grades. The Word Identica-tion subtest measured participants ability to accuratelypronounce printed English words ranging from highto low frequency of occurrence. The Word Attack sub-


7/20


test assessed participants ability to read pronounceablenonwords varying in complexity. To form a compositescore for word recognition, the standard scores for thesesubtests were converted to z scores and combined toform a composite z score.

Criteria for SLI

The criteria we used for SLI were used in the originalepidemiologic study (Tomblin et al., 1996). These criteriawere developed to be consistent with research ndingsin child language disorders and to have high sensitivityand specicity when compared to clinical judgments ofSLI. The criteria are also similar to those used by manyothers to identify the disorder (Paul, 2001; Silva, 1980).Our approach is based on a model of language that in-cludes three domains of language (vocabulary, gram-mar, and narration) and two modalities (receptive andexpressive). A composite score is calculated for each do-main and modality of language. Children are identiedas having a language impairment if their performanceon at least two of ve language composite z scores fallbelow 1.25 SD (approximately the 10th percentile basedon local norms). This criterion is approximately equal tohaving an overall language composite z score of below 1.14 SD (Tomblin et al., 1996). Furthermore, children areconsidered to have a specic language impairment(SLI) if they also demonstrate normal or above-normalnonverbal IQ (>1 SD) and normal sensory and socio-emotional development (Stark & Tallal, 1981).

Data from the kindergarten diagnostic battery wereused to identify children with SLI. When the above crite-

ria were applied to these data, 106 of the 527 children inthe sample were identied as having SLI. These childrenhad a mean language composite standard score (basedon local norms) of 76.9 (SD = 5.4) and a mean nonverbalIQ standard score of 99.4 (SD = 8.6) in kindergarten.

Criteria for Dyslexia

We used multiple sets of criteria for dyslexia to cap-ture the variability in the way the disorder has been de-ned. Our most liberal denition of dyslexia requiredlow achievement in word recognition ability alone (Sie-gel, 1989). This was referred to as the low-achievement

denition. We operationalized low achievement as per-formance of at least 1 SD below the mean on the com-posite measure of word recognition. This cutoff value isconsistent with that frequently used by other research-ers in the study of reading problems in young children(McArthur et al., 2000; Meyer, Wood, Hart, & Felton,1998; Snowling et al., 2003) and represents a compro-mise criterion level compared with that found in moreliberal denitions (25th percentile; Fletcher et al., 1994;Stanovich & Siegel, 1994) or in more conservative deni-

tions of reading disabilities (1.5 SD; Badian, McAnulty,Duffy, & Als, 1990). It is also comparable to the severitylevel of the overall language composite score reectedin our criteria for SLI.

Whereas dyslexia has occasionally been dened onthe basis of low achievement alone, most traditionaldenitions require that low achievement occur in the

presence of normal intelligence (Vellutino, Scanlon,& Lyon, 2000; Wimmer, Mayringer, & Landerl, 2000)or that a signicant discrepancy exist between read-ing level and intelligence (Frankenberger & Fronzaglio,1991; Rutter & Yule, 1975; B. A. Shaywitz, Fletcher, Hol-ahan, & Shaywitz, 1993). Therefore, we used severaldenitions that referenced intelligence. First, in the IQ-cutoff denition, children were considered to have dys-lexia if they had low achievement in word reading (< 1 SD) and scored above a cutoff value (1 SD) in theirmeasured intelligence. Separate analyses were under-taken using either estimates of Full Scale IQ or nonver-bal IQ as the index for intelligence. Whereas Full ScaleIQ is most often used in dening dyslexia (Pennington,Gilger, Olson, & DeFries, 1992; S. E. Shaywitz, Shay-witz, Fletcher, & Escobar, 1990), a few researchers haveused nonverbal IQ in studies of the reading outcomesof children with SLI (e.g., Bishop & Adams, 1990). Thelatter approach, although less common, reduces the roleof verbal intelligence in identifying dyslexia and there-fore might be expected to lead to more children with ahistory of SLI being identied as having dyslexia thanif Full Scale IQ is used.

Second, we used an IQachievement discrepancy deni-tion. The IQ-cutoff denition assures that children with

dyslexia have normal intelligence but does not alwaysresult in a signicant discrepancy between reading abil-ity and intelligence. To address this issue, it is commonto use an IQachievement discrepancy approach, espe-cially one that controls for the correlation between read-ing and intelligence (Snowling et al., 2000). In this ap-proach, children are identied as having dyslexia if theirachievement level is signicantly below that predictedby their intelligence. In operationalizing this approach,we used regression equations based on data from theentire sample. Estimates of Full Scale IQ and nonverbalIQ were each used to predict word recognition scores.Participants were identied as having dyslexia if their

actual word recognition score was more than 1 SD be-low their predicted word recognition score. Finally, wealso calculated prevalence rates for dyslexia using crite-ria that required that children not only show the abovediscrepancy but also have low achievement in word rec-ognition. Such an approach has been suggested in or-der to eliminate children from the category of dyslexiawho have normal word recognition, but at a level signif-icantly below that predicted by their intelligence (Dyk-man & Ackerman, 1992).


8/20


Results

In the rst set of analyses, we examined the preva-lence of dyslexia in second, fourth, and eighth gradesamong children with SLI in kindergarten. The percent-ages of children with SLI in kindergarten who met thevarious criteria for dyslexia at each grade are shown inTable 1. These results indicated that approximately onethird of the children with SLI had low achievement inword recognition in second, fourth, and eighth grades;however, only about 19% to 21% of the children met the

low achievement plus Full Scale IQ-cutoff criteria fordyslexia. As expected, slightly higher prevalence rates(25%26%) were found when nonverbal IQ rather thanFull Scale IQ was used as the IQ-cutoff criterion. Theprevalence rates and the difference between estimatedFull Scale IQ- and nonverbal IQ-based criteria were es-sentially the same when the regression-based IQ-dis-crepancy criteria were used. In addition, similar resultswere observed when the requirement of low achieve-ment was added to the regression-based IQ-discrepancycriteria. The latter nding indicates that there were veryfew children with SLI who had reading achievementsignicantly below that predicted by IQ but still in the

normal range.Given the relatively low rate of dyslexia among chil-

dren with SLI, it is important to ask if this rate is higherthan the base rate of the disorder in the general popu-lation. Our calculations showed that the base rate ofdyslexia (using the Full Scale IQ-discrepancy and low

achievement criterion in fourth grade) in our sampleof 527 children was 8.6%. A two-sample binomial testdemonstrated that the observed prevalence of dyslexiaamong children with SLI (17%) was signicantly higherthan this base rate (z = 3.1, p = .002). Also, when similarcriteria involving nonverbal IQ are used, the observedrate of dyslexia in children with SLI (24.5%) was signif-

icantly higher than the base rate of this condition in ourpopulation (9.7%; z = 4.1, p < .001). Results were simi-lar when we compared rates based on dyslexia in sec-ond and eighth grades.

In a second set of analyses, we examined the relation-ship between SLI and dyslexia from the opposite per-spective; that is, we determined the percentage of chil-dren with dyslexia in second, fourth, and eighth gradeswho met the criteria for SLI in kindergarten. For thisanalysis, we used the regression-based IQ-discrepancyplus low achievement criteria. Estimates of Full Scaleand nonverbal IQ were used in separate calculations.Using criteria involving Full Scale IQ, we identiedfrom our sample of 527 participants 72 children withdyslexia in second grade, 74 in fourth grade, and 68 ineighth grade. Using nonverbal IQ, we identied 85 chil-dren with dyslexia in second grade, 89 in fourth grade,and 75 in eighth grade. For each method, there was con-siderable overlap in those children identied with dys-lexia across grades. Approximately 70% to 75% of thechildren identied as having dyslexia at a given gradealso met the criteria for dyslexia in at least one of theother grades.

To calculate the percentage of children with dyslexiawho had SLI in kindergarten, we used weighted scores.

Such a procedure was necessary to reduce the bias thatis introduced by the fact that the sample from which weidentied children with dyslexia (N= 527) had a higherpercentage of children with SLI in kindergarten thanwould be found in the general population. This biascould lead to an overestimation of the prevalence of SLIin children with dyslexia. To reduce this bias, we deter-mined how likely it was that a child in our sample of527 children with his or her gender, language, and non-verbal prole would have participated in the represen-tative sample seen in the epidemiologic study. Then,each childs scores were weighted accordingly. In otherwords, although our sample contained more children

with language impairments than would be found in arepresentative sample, the scores of these children weregiven proportionally less weighting to assure the repre-sentativeness of the results.2

Our analyses showed that a relatively small per-centage of children identied with dyslexia in second,

Table 1. Percentages of children with specic language im-pairment in kindergarten (N = 106) who met various crite-

ria for dyslexia.

Criteria 2nd grade 4th grade 8th grade

Low achievement 33.0 31.1 35.8

Low achievement

+ IQ cutoffFull Scale IQ 18.9 19.8 20.8

Nonverbal IQ 26.4 25.5 26.4

IQ discrepancy

Full Scale IQ 17.9 17.0 18.8


IQ discrepancy

+ low achievement

Full Scale IQ 17.9 17.0 17.9


An estimate of Full Scale IQ was used that included the Peabody Pic-

ture Vocabulary TestRevised as a measure of Verbal IQ.

2 For example, the epidemiologic study estimated that boys with SLI (no nonverbal decits) compose 3.9% of the general population. In our sam-ple, however, these children composed 12.1%. To assure that the children from this group did not contribute disproportionately to our results,we adjusted their scores by weighting them by a constant that was equal to the expected prevalence of these children (3.9%) divided by their ac-tual prevalence in our sample (12.1%; constant = .322).


9/20


fourth, or eighth grades met the criteria for SLI in kin-dergarten. The data in Table 2 show that 14.8% to 16.5%of the children with dyslexia based on estimated FullScale IQ discrepancy (and low achievement) had SLI inkindergarten. A slightly higher, but still low, rate (19%)was observed when dyslexia was based on nonverbalIQ-discrepancy and low achievement criteria. We againexamined whether these prevalence rates were signif-

icantly higher than would be expected, given the baserate of SLI in our sample. A series of two-sample bino-mial tests showed that the observed rates of SLI in chil-dren with dyslexia based on estimated Full Scale IQdiscrepancy plus low achievement were signicantlyhigher than the base rate of the disorder in second andeighth grades (zs = 2.0 and 2.2, p < .05). The differencebetween the observed rate and base rate at fourth gradeapproached but did not reach statistical signicance (z= 1.9, p = .057). Signicant differences were found be-tween the observed rates and base rates at all threegrades when the nonverbal IQ-discrepancy plus lowachievement criteria for dyslexia were used (zs = 2.9

3.1, p < .005).

Discussion

These results demonstrate a somewhat limited butstatistically signicant overlap between dyslexia andSLI. About one third of children with SLI in kinder-garten met the most liberal criteria for dyslexia in latergrades. If more conservative (and more widely used)criteria involving reference to IQ were used, fewer chil-dren with SLI could be identied as having dyslexia.These data showed that 17% to 29% of children with

SLI in kindergarten met IQ-referenced denitions ofdyslexia in the school grades. A slightly higher rate ofdyslexia was found when nonverbal IQ was used as abenchmark than when estimated Full Scale IQ was used.Again, this difference was expected because childrenwith SLI generally have lower verbal than nonverbalIQs and thus should show less of an IQachievementdiscrepancy when an estimate of verbal IQ is includedin the IQ benchmark.

The prevalence rates of dyslexia in children with SLIthat we observed are lower than those found in many

other studies (e.g., McArthur et al., 2000; Snowling et al.,2000; Tallal et al., 1988). Various differences between ourstudy and those of others could account for this discrep-ancy. One primary difference concerns the way partici-pants were recruited. We used a quasi-random approachto select children from a representative population basedsample. Most other studies in this area have used conve-

nience sampling techniques to select participants largelyfrom clinical populations (e.g., McArthur et al., 2000;Snowling et al., 2000). Whereas the latter procedures arecommon in clinical research, they often result in the re-cruitment of participants with more severe disordersand concomitant conditions than participants who areobtained through population-based sampling (Berkson,1946). Thus, in the case of studies of SLI, this procedurecould lead to the inclusion of children with more severelanguage impairments and a higher incidence of dys-lexia than in the present investigation. One other investi-gation has used a population-based sampling procedurelike ours and reported data on the reading outcomes ofchildren with SLI. In this study, Silva et al. (1987) identi-ed children with SLI (at or below the 5th percentile ontests of language) from a population of approximately1,000 three-year-olds. When these children were seen atages 7, 9, and 11 years, 44.1%, 30.4%, and 30.6%, respec-tively, were found to show low achievement in wordrecognition. No data were provided concerning the pro-portion of the children that met IQ-referenced criteriafor dyslexia. Nevertheless, the rates of low achievementthat they report are comparable to those observed in thepresent study.

There is at least one other important difference be-

tween our study and some other investigations. In thepresent study, we examined the incidence of dyslexiaduring the elementary and middle school grades in chil-dren identied as having SLI in kindergarten. In thestudies reported by McArthur et al. (2000), SLI and dys-lexia were identied concurrently during the early ele-mentary school grades. As such, the language problemsobserved in these studies could have been inuenced inpart by poor reading achievement, which in turn couldhave led to a higher overlap of the disorders. We choseto identify SLI prior to reading instruction to limit theimpact that a reading disability could have on the devel-opment of language problems.

Besides examining the prevalence of dyslexia in chil-dren with SLI, we also looked retrospectively at theprevalence of SLI in children identied as having dys-lexia. Our results indicated that only approximately15 20% of children identied with dyslexia (in second,fourth, or eighth grades) met the criteria for SLI in kin-dergarten. Such a prevalence rate is lower than that re-ported by some investigators. Specically, McArthur etal. (2000) found in a series of four studies that an aver-age of 55% of children with dyslexia also had signicantoral language impairments (met criteria for SLI similar

Table 2. Percentage of children with dyslexia in second,fourth, and eighth grades (based on IQ discrepancy and lowachievement criteria) who had specic language impairmentin kindergarten.

Discrepancy 2nd Grade 4th Grade 8th Grade

Full Scale IQ 15.4 14.8 16.5


An estimate of Full Scale IQ was used that included the Peabody Pic-

ture Vocabulary TestRevised as a measure of verbal IQ.


10/20


to ours). Again, this higher rate is likely inuenced bythe way participants were recruited. Children with dys-lexia in the studies reported by McArthur et al. (2000)were selected by convenience sampling from clinicalpopulations. Such a procedure could have led to partic-ipants with more severe reading problems and a higherrate of SLI.

McArthur et al.s (2000) studies also used concur-rent identication of dyslexia and SLI in school-age chil-dren. As noted above, such a design could result in ahigher degree of overlap between SLI and dyslexia thanwas found in our study. This conclusion is supportedby other studies that have used a design like ours, inwhich language problems have been observed duringpreschool prior to the emergence of reading disabilities(Gallagher et al., 2000; P. Lyytinen et al., 2001; Scarbor-ough, 1990, 1991; Snowling et al., 2003). These studiesidentied children who were at high risk for dyslexia onthe basis of a family history of reading disabilities. Re-sults showed that at-risk children who later developeddyslexia often had oral language problems during thepreschool years. These problems, however, tended to berather mild and sometimes disappeared by school entry(Scarborough, 1990; Gallagher et al., 2000). Seldom werelanguage problems severe enough for the children to bediagnosed as having SLI. For example, Gallagher et al.(2000) reported that only 9 of 63 (14%) at-risk childrenperformed at least 1 SD below the mean in languageabilities (no information was provided concerning non-verbal IQ). Whereas some of these at-risk children didnot develop dyslexia in the school years, the propor-tion that had SLI is still quite low and, in fact, no greater

than would be expected in the general population giventhe criteria they used.Finally, a word of caution is warranted in terms of

the implications of Study 1 for clinical/educationalpractice. Our ndings of a limited overlap betweenSLI and dyslexia should not diminish the impor-tance of oral language decits in reading disabilities.This limited overlap was observed between two spe-cic and rather narrowly dened clinical categoriesin children selected from a population-based sam-ple. Children with SLI who are referred for servicesin the schools or in clinics are likely to have a greaterincidence of dyslexia than we observed. In addition,

many children with language impairments that co-occur with nonverbal cognitive decits or are not se -vere enough to meet our criteria of SLI go on to haveword reading problems like those seen in dyslexia.Many others experience signicant problems in read-ing comprehension (Catts et al., 2002). As such, orallanguage decits should remain an important earlyindicator of risk for reading disabilities and shouldbe addressed with appropriate clinical/educationalintervention.

Study 2: Phonological Processing in SLI andDyslexia

The results from Study 1 showed a statistically sig-nicant overlap between SLI and dyslexia. However,this overlap was rather limited. Only a small percent-age of children with SLI in kindergarten met the crite-ria for dyslexia in the school grades and, conversely,only a small percentage of children with dyslexia in theschool grades met the criteria for SLI in kindergarten.Given that the overlap between SLI and dyslexia is lim-ited, we are left with the question of how children withthese disorders could be characterized by the same de-cits in phonological processing. Recall that research hasoften shown that children with SLI and those with dys-lexia have decits in phonological awareness and pho-nological memory (Catts, 1993; Fletcher et al., 1994; Ka-mhi& Catts, 1986; Snowling, 1981). One possibility forthis puzzling set of ndings may be that studies of pho-

nological processing have often included heterogeneoussamples involving a mix of children, some with bothSLI and dyslexia and some with SLI only. Such stud-ies could show differences between the target popula-tion and typically developing children when in fact aphonological processing decit is primarily characteris-tic of one disorder and not the other. The disorder mostlikely to be associated with a phonological processingdecit is dyslexia. Recall that such a decit is thoughtto be the proximal cause of word reading problems indyslexia (Lyon et al., 2003). Children with SLI in the ab-sence of dyslexia may not have problems in phonologi-cal processing; however, because of the partial overlap

(and borderline cases of overlap) of SLI and dyslexia, itis likely that when a group of children with SLI are se-lected and compared to a group of typically developingchildren, signicant differences might be found in pho-nological processing. In Study 2, we examined this is-sue by investigating phonological processing in childrenidentied with SLI only, dyslexia only, both SLI anddyslexia, and neither of the disorders.

Method

Participants

The participants in this study were a subsample ofthose identied with SLI and/or dyslexia in Study 1.Four groups were selected. One subgroup (SLI only)consisted of all children with SLI in kindergarten whohad normal reading achievement in fourth grade (wordrecognition composite score above the 40th percentile; N= 43). A second subgroup (SLI/dyslexia) was composedof all participants who had SLI in kindergarten and whoalso met the regression-based Full Scale IQ-discrepancyand low achievement criteria (N= 18). A third subgroup


11/20


(dyslexia only) consisted of all children with dyslexiain fourth grade (same criteria as above) who had nor-mal language in kindergarten (i.e., did not meet the cri-teria for SLI or a nonspecic language impairment; N= 21). A nal subgroup (normal) included all childrenwho had normal language in kindergarten (same crite-ria as above) and normal reading achievement in fourth

grade (i.e., same criterion as above; N = 165). Fourthgrade reading achievement was used for participant se-lection because it represented the intermediate point inour reading achievement data. The criteria for SLI andnormal language status were again based on kindergar-ten language scores for the same reasons discussed inStudy 1.

The language and word recognition scores of each ofthe subgroups are displayed in Table 3. The kindergar-ten language and fourth grade word recognition com-posite scores are shown to highlight group differencesand similarities, some of which were imposed by sub-group selection criteria, while others were not. Analysesof variance (ANOVAs) indicated subgroup differencesin language, F(3, 243) = 102.7, p < .01, and word recog-nition scores, F(3, 243) = 243.1, p < .01. Tukey honestlysignicant difference tests for unequal Ns demonstratedthat the SLI-only and the SLI/dyslexia subgroups hadsignicantly lower language composite scores than thedyslexia-only (p < .01, ds = 0.81 and 0.82, respectively)and normal subgroups (p < .01, d = 1.73). Tukey testsalso demonstrated that the dyslexia-only and SLI/dys-lexia subgroups had signicantly lower word recogni-tion composite scores than the SLI-only (p < .01, ds = 1.94and 2.41, respectively) and normal subgroups (p < .01, ds

= 2.17 and 2.64, respectively). Both of these sets of differ-ences, of course, are expected on the basis of subgroupselection criterion. Other similarities and differences ingroup comparisons were not predetermined by partici-pant selection criteria. Group comparisons showed thatthe SLI-only and SLI/dyslexia subgroups did not dif-fer signicantly in their language composite scores (p >.05, d = 0.01); however, the dyslexia only subgroup didhave a signicantly lower language score than the nor-mal subgroup (p < .001, d = 0.92). In the case of wordrecognition, the SLI-only and normal control groups didnot differ signicantly (p > .05, d = 0.23), but a signi-

cant difference was observed between the SLI/dyslexiaand the dyslexia subgroups (p < .05, d = 0.47).

Materials

The same measures of language, intelligence, andword recognition that were used to identify children

with SLI and dyslexia in Study 1 were used to select par-ticipants in this study. In addition, measures of phono-logical awareness and phonological memory were ad-ministered to the participants.

Phonological awareness. A syllable/phoneme deletiontask was given to participants in kindergarten and sec-ond and fourth grades. This task required children to re-peat a real word produced via live voice by a trained ex-aminer. The examiner then instructed the participant tosay the word again but to delete a designated syllable orphoneme. The kindergarten version included 21 itemsthat required the deletion of the initial syllable or pho-neme (Catts, Fey, Zhang, & Tomblin, 2001). In secondand fourth grades, 9 additional items were added thatrequired the deletion of a nal consonant or member ofa nal consonant cluster. The score was the total num-ber of items produced correctly.

In eighth grade, a more complex phoneme deletiontask, adapted from Gayan and Olson (2003), was ad-ministered to participants. It required participants to re-peat 46 nonwords individually and then delete a pho-neme to derive a real word. The phoneme to be deletedwas a singleton consonant or a consonant in a two- orthree consonant cluster. Nonwords were presented viaheadphones and a high-quality audio recorder, and the

participants responses were recorded. The score wasthe number of items correct or partially correct (partialcredit was given for responses that were incorrect butphonetically similar). The scores from both phonologi-cal awareness tasks were converted to standard scoresbased on the weighted means and standard deviationsof the entire sample.

Phonological memory. A nonword-repetition task,which was administered in second and eighth grades,served as a measure phonological memory. This taskwas developed by Dollaghan and Campbell (1998) andconsisted of 16 nonwords ranging from one to four syl-

Table 3. Language and word recognition proles of Study 2 subgroups.

SLI only Dyslexia only SLI and dyslexia Normal

(n = 43) (n = 21) (n = 18) (n = 165)

M SD M SD M SD M SD

Language (K) 77.0 5.6a 90.4 8.1b 76.9 5.9a 106.5 13.2c

Word recognition (4th grade) 105.7 6.1a 75.1 6.4b 67.7 12.1c 109.3 7.9a

Means in the same row that do not share subscripts differ at p < .05 in Tukey honestly signicant difference test for unequal Ns. SLI = specic

language impairment; K = kindergarten.


12/20


lables in length (four words at each length). Each of thenonwords was composed of early developing phonemesand contained syllables that did not correspond to Eng-lish lexical items. The latter constraint was imposed toreduce the effects that differences in vocabulary knowl-edge might have on performance on this task (see Dol-laghan & Campbell, 1998). The nonword-repetition taskwas administered to children via headphones and ahigh-quality audio recorder, and participants responseswere recorded. These responses were scored in terms

of the percentage of consonants produced correctly.Scores were converted to standard scores based on theweighted mean and standard deviation of the availablesample at second (N= 604) and eighth grades (N= 527).

Results

The subgroups performances on measures of phono-logical awareness are displayed in Figure 2. UnivariateANOVA procedures were used to examine group dif-ferences. Because tests (or items) used to measure pho-nological awareness varied at some grades, grade level

was not evaluated as a repeated measure. The results in-dicated that there was a signicant group difference ateach grade, Fs(3, 243) = 32.482.4, p < .01. In kindergar-ten, Tukey honestly signicant difference tests for un-equal Ns showed that only the normal subgroup per-formed signicantly different from the other subgroups(p < .001, ds = 1.031.29). In the other grades, both thenormal and the SLI-only subgroups scored signicantlybetter than the dyslexia-only and SLI/dyslexia sub-groups (p < .001, ds = 1.082.09). The normal and SLI-only subgroups differed signicantly from each other

in second grade (p < .05, d = 0.48) but not in the fourthand eighth grades (p > .05, ds = 0.130.19). The dyslexia-only and SLI/dyslexia subgroups did not perform sig-nicantly different from each other on the phonolog-ical awareness tasks at any grade tested (p > .05, ds =0.110.44).

Data for the nonword-repetition task are shown inFigure 3. A 4 (group) 2 (grade) mixed-model ANOVAwas used to examine group differences at each grade.This analysis revealed a signicant main effect of group,

F(3, 242) = 31.2, p < .001, and grade, F(1, 242) = 57.0, p .05. Follow-up tests of group differ-ences (collapsed across grades) indicated the dyslexia-only and SLI/dyslexia subgroups did not differ signif-icantly from each other, F(1, 242) = 1.0, p > .05, but eachdid differ signicantly from the normal subgroup, Fs(1,242) = 45.4 and 55.3, p < .001. Results further showedthat the SLI-only subgroup performed signicantly bet-ter than the dyslexia-only subgroup, F(1, 242) = 13.9, p .05. Follow-up analyses (p .05, ds = 0.050.65) were ob-served between the other subgroups. To rule out theinuence of IQ in comparisons involving the normalsubgroup, we conducted ANCOVAs using fourth gradeestimated Full Scale IQ as a covariate. The results ofthese comparisons were the same as those when no co-variate was used, with one exception. The normal andSLI-only subgroups did not differ in nonword repetitionin this ANCOVA; however, when a less restrictive mea-

sure of nonverbal IQ (either at second or eighth grade)was used as a covariate, these groups differed signi-cantly, as they had in the original analysis.

Another set of post hoc analyses involved compari-sons between the SLI-only and SLI/dyslexia subgroups.A primary nding in this study was that these sub-groups differed in phonological processing. Given thesignicance of this nding, it is important to rule outother subgroup differences that may have inuencedthis result. As noted above, the SLI-only and SLI/dys-lexia subgroups did not differ in IQ. Also, recall that

Figure 3. Nonword repetition performance of subgroups in the second and eighth grades.


14/20


these subgroups did not differ signicantly in terms ofthe severity of their language impairment in kindergar-ten. Whereas severity of language impairment was freeto vary in these groups, they had almost identical meanlanguage composite scores. Further post hoc analysesshowed that these subgroups did not differ signicantlyon any of the language subtests that were used to form

the kindergarten language composite score (p > .05, ds=0.060.49). Kindergarten language data were also avail-able on an experimental measure of grammatical tensemarking (see Rice, Tomblin, Hoffman, Richman, & Mar-quis, 2004) for approximately 60% of the participants inthese subgroups. Analysis of these data indicated thatthe SLI-only and SLI/dyslexia subgroups performedcomparably in this aspect of language (p > .05, d = 0.41).Additional post hoc analyses indicated that these sub-groups did not differ signicantly on language compos-ite scores in second grade (p > .05, d = 0.33) or fourthgrade (p > .05, d = 0.40). However, in eighth grade theSLI/dyslexia group had a signicantly lower languagecomposite score than the SLI-only subgroup (p < .05, d= 0.72). This latter difference could represent a differ-ence in constitutional language abilities that was not ap-parent until a later grade. Alternatively, this differencecould be the result of subgroup variation in readingachievement and experience.

Whereas the SLI-only and SLI/dyslexia subgroupsdid not generally differ in severity of language impair-ment, further post hoc analyses did suggest that theremay have been differences in intervention history. Par-ents of participants with both SLI and dyslexia more of-ten reported that these children had received clinical

services in kindergarten and/or primary grades thanhad parents of children with SLI only, 2(1, N = 61) =55.3, p < .001. This result is not surprising given otherresearch showing that in clinical samples (i.e., those re-ceiving intervention) there is a high overlap between SLIand dyslexia. Last, although we could not rule out dif-ferences in environmental inuences among subgroups,we found no signicant differences in mothers educa-tion between the SLI-only and SLI/dyslexia subgroups(p > .05, d = 0.33).

Discussion

In this study, we predicted that a phonological pro-cessing decit would be more closely associated withdyslexia than SLI. Our results were consistent with thisprediction. Children with dyslexia only and those witha combination of dyslexia and SLI (i.e., the SLI/dyslexiasubgroup) performed poorly on measures of phonologi-cal awareness and nonword repetition across the grades.Children with SLI only, on the other hand, did not showsignicant decits on measures of phonological process-ing. This subgroup, however, had lower scores than the

normal subgroup on all measures of phonological pro-cessing. Although these differences were not statisticallysignicant in all cases, they may indicate that childrenwith SLI only, on average, have a mild decit in phono-logical processing.

These various ndings are consistent with a largebody of research that indicates that a decit in phono-

logical processing is central to dyslexia (e.g., Fletcheret al., 1994). They are also in line with the most recentIDA denition of dyslexia, which proposes that a decitin phonological processing lies at the core of the wordrecognition problems in the disorder (Lyon et al., 2003).Our results, however, appear to be in contrast to thoselinking SLI with a decit in phonological processing.This is particularly true for the ndings concerning non-word repetition. Recall that many studies have reportedthat children with SLI have decits in nonword repeti-tion (Dollaghan & Campbell, 1998; Gathercole & Bad-deley, 1990; Kamhi & Catts, 1986). Furthermore, prob-lems in nonword repetition have been argued to be apotential psycholinguistic marker of SLI (Bishop et al.,1996; Conti-Ramsden et al., 2001). Our ndings, how-ever, indicate only a weak association, at most, betweenSLI and problems in nonword repetition.

The discrepancy between our ndings and those ofothers concerning a link between SLI and a decit inphonological processing can be explained largely onthe basis of the comorbidity between SLI and dyslexia.In Study 1, we found that the overlap between SLI anddyslexia was greater than expected given the base ratesof the two disorders. This overlap indicates that a por-tion of children with SLI will also have dyslexia. Fur-

thermore, if this comorbidity involves an overlap of def-icits in abilities that are continuously distributed, wemight also expect that children with SLI who do notmeet the criteria for dyslexia to still be lower, on aver-age, in word reading and phonological processing thanchildren with normal language. Thus, it seems quitepossible that previous studies of SLI and nonword rep-etition have involved samples of children with SLI thatincluded enough children who also had dyslexia or bor-derline dyslexic-like problems such that SLI groups, asa whole, would score signicantly below that of controlgroups on nonword repetition. Indeed, in our longitudi-nal sample, which had rather limited overlap between

SLI and dyslexia compared with other studies, the posthoc analysis indicated that when all children with SLIin kindergarten were combined (including those withSLI only, SLI and dyslexia, and those on the borderlineof dyslexia; N = 106), they performed signicantly be-low that of typically developing children in nonwordrepetition. Also, Ellis Weismer et al. (2000) showed thatchildren from our same longitudinal sample who wereidentied as having SLI in second grade scored signif-icantly less well on the nonword-repetition task thandid typically developing children. Thus, in our sample


15/20


and in others, comorbidity with dyslexia may account inpart for why children with SLI, as a group, show poorperformance in nonword repetition. However, furtherpost hoc analyses indicate that such comorbidity maynot completely explain these results. These analysesshowed that when we compared all children with SLI inkindergarten (N=106) to all children without language

impairment (N = 256) and covaried out differences inword reading, the groups still differed signicantly innonword repetition. This nding suggests that at leasta portion of the low nonword repetition performance ofchildren with SLI results from factors other than comor-bidity with dyslexia.

Additional results from our longitudinal databaseprovide further converging evidence related to Study 2.Tomblin et al. (2004) reported that a factor analysis ofthe language scores of our sample at age 7 showed thatperformance on phonological awareness and nonwordrepetition loaded on a different factor than performanceon semantic and syntactic tasks. This suggests that somechildren may have problems in phonological process-ing and not in semantics and syntax (i.e., dyslexia only),and others may show the reverse pattern (i.e., SLI only).These ndings are also consistent with the results of re-cent genetics studies. Bishop and colleagues (Adams &Bishop, 2002; Bishop, 2001, 2005), in a twin study of SLI,found high heritability for grammatical morphologyand nonword repetition; however, heritability of each ofthese skills was independent of the other. Furthermore,Bishop and her colleagues reported a greater genetic as-sociation (i.e., bivariate heritability) between nonwordrepetition and dyslexia than between grammatical mor-

phology and dyslexia (Bishop, 2001; Bishop et al., 2004).This latter nding converges well with our results dem-onstrating a link between decits in nonword repetitionand dyslexia.

Whereas our results appear to be consistent with theabove related ndings, two issues need further consid-eration. One issue concerns the age at which we iden-tied children with SLI. Many studies that have exam-ined the relationship between SLI and phonologicalprocessing have selected participants on the basis oflanguage performance during the postkindergartenschool years (e.g., Dollaghan & Campbell, 1998; Conti-Ramsden et al., 2001). We selected participants on the

basis of a kindergarten language assessment (for rea-sons outlined in the beginning of this article); however,to be consistent with previous studies, we reanalyzedour results using school-age diagnostic criteria. In theseanalyses, the participants were reclassied into sub-groups using criteria based on second-grade languagestatus and second-grade word reading scores. We alsoregrouped participants using criteria based on fourth-grade language and fourth-grade word reading scores.The results in both cases were essentially the same.Children with dyslexia only and those with SLI/dys-

lexia had signicant decits in phonological process-ing, whereas those with SLI only had mild problems atmost. Thus, it does not appear that the grade at whicha language impairment is identied inuences the na-ture of the relationship between SLI and phonologicalprocessing.

A second issue concerns the direction of causality be-

tween problems in phonological processing and dys-lexia. We have argued that our results support the viewthat a decit in phonological processing underlies theword reading problems in dyslexia. However, it is pos-sible that at least a portion of the differences in phono-logical processing observed between participants withdyslexia (i.e., those in the dyslexia-only and SLI/ dys-lexia subgroups) and those without (normal and SLI-only subgroup) was a consequence of poor word read-ing. Indeed, studies have shown that word readingability itself can inuence performance in phonologi-cal processing, especially phonological awareness (Ho-gan, Catts, & Little, 2005; McGuinness, McGuinness, &Donohue, 1995). Our results showing that the SLI-onlysubgroup seemed to improve across grades in pho-nological awareness, whereas the dyslexia-only sub-group declined slightly across grades, could possiblybe a reection of the inuence of reading on phonolog-ical awareness. Alternatively, this result might indicatethat phonological awareness decits are more specic tochildren with dyslexia than those with SLI only and, assuch, are more stable over time.

General Discussion

In the beginning of this article, we offered three alter-native models concerning the relationship between SLIand dyslexia. Model 1 characterizes SLI and dyslexiaas variants of the same developmental language dis-order but differing in the severity of the disorder (e.g.,Tallal et al., 1997). Model 2 proposes that SLI and dys-lexia share a comparable decit in phonological process-ing and word reading problems but differ in terms ofthe presence/absence of oral language decits (Bishop&Snowling, 2004). Model 3 argues that SLI and dyslexiaare distinct but comorbid disorders. The results from thepresent investigation are more in line with Model 3.

In Study 1, we examined the overlap between SLIand dyslexia. If either Model 1 or 2 is accurate, weshould have found considerable overlap between SLIand dyslexia. Both of these proposals contend thatchildren with SLI have problems in phonological pro-cessing and subsequent difculties in word reading.Thus, most children with SLI should also be identiedas having dyslexia. This was not the case. Our resultsshowed a statistically signicant, but limited, overlapbetween SLI and dyslexia. Most children with SLI inkindergarten did not have dyslexia during the school


16/20


years. This result is more consistent with Model 3. Ac-cording to this model, most affected children will haveeither SLI or dyslexia. A small percentage of children,however, can have both disorders as a result of comor-bidity. Model 3 is further supported by Study 2. Thisstudy showed that whereas dyslexia was associatedwith signicant decits in phonological processing, SLI

alone was generally not.

Relationship Between SLI and Dyslexia

Taken together, the ndings from the present inves-tigation support the view that SLI and dyslexia are dis-tinct developmental disorders. According to this view,dyslexia is a developmental language disorder that ischaracterized by problems in phonological processingand word reading decits. SLI, on the other hand, is adisorder involving problems in oral language, includ-ing decits in semantics, syntax, and/or discourse pro-

cessing. It is unclear from this investigation what fac-tors may underlie SLI. The disorder may result from aspecic morpho-syntactic decit (Rice & Wexler, 1996)and/or from some other perceptual/cognitive impair-ment (Miller, Kail, Leonard, & Tomblin, 2001; Mont-gomery, 2000; Tallal, 2003). A problem in phonologicalprocessing, however, does not appear to be a major fac-tor in SLI when it occurs in isolation from dyslexia.

Whereas dyslexia and SLI may best be viewed asdistinct disorders, they appear to be comorbid in somechildren. Our results indicated that about twice asmany children had both disorders than would be pre-dicted given the base rate of either disorder. In clini-

cal populations, we would expect even more overlapto occur. Children from the latter populations gener-ally have more severe and widespread disorders andthus should more often meet the criteria of both dis-orders. Indeed, the studies we reviewed that sampledfrom clinical populations found a high level of overlapbetween SLI and dyslexia (e.g., McArthur et al., 2000;Tallal et al., 1997). Because the decits that underlieSLI and dyslexia are likely to involve continuously dis-tributed abilities (Dollaghan, 2004; S. E. Shaywitz, Es-cobar, Shaywitz, Fletcher, & Makuch, 1992), the comor-bidity of the disorders should spread its effects to theborderline of each disorder. As a result, children with

SLI alone may show low normal performance in pho-nological processing and word reading, and childrenwith dyslexia alone may have low normal oral lan-guage abilities. However, despite the additional over-lap on the borderline of each disorder, there should bemany children who meet the criteria for one disorderbut are well within normal limits in abilities related tothe other disorder.

The fact that SLI and dyslexia are distinct disordersis supported further by a growing body of research on

poor comprehenders, that is, children who demonstratea decit in reading comprehension despite normal ornear-normal word recognition ability. It is estimatedthat perhaps as many as 5% to 10% of school-age chil-dren show this reading problem (Catts, Adlof, & EllisWeismer, in press; Nation, 2005). Recent research in-dicates that these children have a wide range of de-

cits in oral language (Nation, Clarke, Marshall, & Du-rand, 2004; Nation & Snowling, 1997; Oakhill & Yuill,1996; Stothard & Hulme, 1995). These decits, how-ever, are conned to nonphonological aspects of lan-guage and do not include problems in phonologicalawareness and phonological memory. Thus, thesechildren appear to demonstrate the characteristics ofchildren with SLI alone and are quite distinct fromthose with dyslexia. Indeed, studies have documentedthat nearly 50% of poor comprehenders have a his-tory of oral language problems that are severe enough(and generally discrepant enough from nonverbal IQ)to meet the criteria of SLI (Catts et al., in press; Nationet al., 2004).

The concept of a poor comprehender is also centralto one of the alternative models concerning the rela-tionship between SLI and dyslexia. Specically, Bishopand Snowling (2004) proposed that SLI and dyslexiatypically share decits in phonological processing andword reading but differ in that SLI is also characterizedby signicant oral language problems and dyslexia isnot (i.e., Model 2). They acknowledged, however, thatsome children may have signicant decits in oral lan-guage abilities but have normal phonological process-ing abilities. They referred to the latter children aspoor

comprehenders rather than children with SLI only, as wedo. Thus, the primary difference between their pro-posal and the one we favor is the choice of terminol-ogy. However, we believe our proposal is more con-sistent with traditional practice and current researchndings. The term SLI has traditionally been used todescribe children with oral language decits regardlessof the presence or absence of phonological processingdecits (Leonard, 1998). It has also been used to char -acterize childrens oral language development duringthe preschool years and has not been dependent onreading problems. Our results suggest that at least ina population-based sample there will be many children

who meet the criteria for SLI prior to school entrancebut who do not have a phonological processing de-cit. It would seem more appropriate to refer to thesechildren as having SLI and acknowledge that this con-dition can exist by itself in some children as well as becomorbid with dyslexia in others. In such a model, theterm poor comprehenderwould be used to refer to chil-dren with a history of SLI (as well as those without)who have specic problems in reading comprehensionduring the school years.


17/20


Acknowledgments

This study was supported by a grant from the National Insti-tute on Deafness and Other Communication Disorders (1-P50-DC02726-04). The completion of this study was aided consid-erably by a valuable research team comprising the following:Bruce Tomblin, Xuyang Zhang, Marc Fey, Paula Buckwalter,Marlea OBrien, Connie Ferguson, Jodi Schwartz, Juanita Li-mas, and Amy Kundel. Appreciation is also extended to HollyStorkel and David Slegers for their assistance.

References

Adams, C. V., & Bishop, D. V. M. (2002, July). Genetic inu-ences on verb morphology decits in 6-year-old twins.Poster presented at the joint conference of the Interna-tional Association for the Study of Child Language andthe Symposium for Research on Children with LanguageDisorders, Madison, WI.

Badian, N., McAnulty, G., Duffy, G., & Als, H. S. (1990). Pre-

diction of dyslexia in kindergarten boys. Annals of Dys-lexia, 41, 221243.

Bedore, L. M., & Leonard, L. B. (1998). Specic language im-pairment and grammatical morphology: A discrimi-nant function analysis. Journal of Speech, Language, andHearing Research, 41, 11851192.

Berkson, J. (1946). Limitations of the application of fourfoldtable analysis to hospital data. Biometrics, 2, 4753.

Bishop, D. V. M. (2001). Genetic inuences on language im-pairment and literacy problems in children: Same or dif-ferent? Journal of Child Psychology and Psychiatry, 42,189198.

Bishop, D. V. M. (2005). DeFriesFulker analysis of twin datawith skewed distributions: Cautions and recommenda-tions from a study of childrens use of verb inections.Behaviour Genetics, 35, 479490.

Bishop, D. V. M., & Adams, C. (1990). A prospective study ofthe relationship between specic language impairment,phonological disorders, and reading retardation. Journalof Child Psychology and Psychiatry, 31, 10271050.

Bishop, D. V. M., Adams, C. V., & Norbury, C. F. (2004). Us-ing nonword repetition to distinguish genetic and envi-ronmental inuences on early literacy development: Astudy of 6-year-old twins. American Journal of MedicalGenetics: Neuropsychiatric Genetics, 129, 9496.

Bishop, D. V. M., North, T., & Donlan, C. (1996). Nonwordrepetition as a behavioral marker for inherited languageimpairment: Evidence from a twin study. Journal ofChild Psychology and Psychiatry, 37, 391403.

Bishop, D. V. M., & Snowling, M. J. (2004). Developmentaldyslexia and specic language impairment: Same or dif-ferent? Psychological Bulletin, 130, 858886. Bradley, L.,& Bryant, P. E. (1983). Categorising sounds and learningto readA causal connection. Nature, 301, 419521.

Brady, S., Poggie, E., & Rapala, M. M. (1989). Speech repeti-tion abilities in children who differ in reading skill. Lan-guage and Speech, 32, 109122.

Brady, S., Shankweiler, D., & Mann, V. (1983). Speech per-ception and memory coding in relation to reading ability.Journal of Experimental Child Psychology, 35, 345367.

Briscoe, J., Bishop, D. V. M., & Norbury, C. F. (2001). Phono-logical processing, language, and literacy: A comparisonof children with mild-to-moderate sensorineural hearingloss and those with specic language impairment. Jour-

nal of Child Psychology and Psychiatry, 42, 329340.Caron, C., & Rutter, M. (1991). Comorbidity in child psycho-

pathology: Concepts, issues and research strategies. Jour-nal of Child Psychology and Psychiatry, 32, 10631080.

Catts, H. W. (1986). Speech production/phonological decitsin reading-disordered children. Journal of Learning Dis-abilities, 19, 504508.

Catts, H. W. (1993). The relationship between speech lan-guage impairments and reading disabilities. Journal ofSpeech and Hearing Research, 36, 948958.

Catts, H. W., Adlof, S. M., & Ellis Weismer, S. (in press). Lan-guage decits in poor comprehenders: A case for the sim-ple view of reading. Journal of Speech, Language, and

Hearing Research.Catts, H. W., Fey, M. E., Tomblin, J. B., & Zhang, X. (1999).

Language basis of reading and reading disabilities. Scien-tic Studies of Reading, 3, 331361.

Catts, H. W., Fey, M. E., Tomblin, J. B., & Zhang, X. (2002).A longitudinal investigation of reading outcomes in chil-dren with language impairments. Journal of Speech, Lan-guage, and Hearing Research, 45, 11421157.

Catts, H. W., Fey, M. E., Zhang, X., & Tomblin, J. B. (2001).Estimating the risk of future reading difculties in kin-dergarten children: A research-based model and its clini-cal implications. Language, Speech, and Hearing Servicesin Schools, 32, 3850.

Catts, H. W., & Kamhi, A. G. (2005). Language and readingdisabilities (2nd ed.). Needham Heights, MA: Allyn &Bacon.

Conti-Ramsden, G., Botting, N., & Faragher, B. (2001). Psy-cholinguistic markers for specic language impairment(SLI). Journal of Child Psychology and Psychiatry andAllied Disciplines, 42, 741748.

Culatta, B., Page, J., & Ellis, J. (1983). Story retelling as acommunicative performance screening tool. Language,Speech, and Hearing Services in Schools, 14, 6674.

Dollaghan, C. A. (2004). Taxometric analyses of specic lan-guage impairment in 3- and 4-year-old children. Journalof Speech, Language, and Hearing Research, 47, 464475.

Dollaghan, C. A., & Campbell, T. F. (1998). Nonword repeti-tion and child language impairment. Journal of Speech,Language, and Hearing Research, 41, 11361146.

Dunn, L. M., & Dunn, L. M. (1981). Peabody Picture Vocab-ulary TestRevised. Circle Pines, MN: American Guid-ance Service.

Dykman, R. A., & Ackerman, P. T. (1992). Diagnosing dys-lexia: IQ regression plus cut points. Journal of LearningDisabilities, 25, 574576.

Edwards, J. D., & Lahey, M. (1998). Nonword repetitions ofchildren with specic language impairment: Exploration


18/20


of some explanations for their inaccuracies. Applied Psy-cholinguistics, 19, 279309.

Elbro, C. (1996). Early linguistic abilities and reading devel-opment: A review and a hypothesis. Reading and Writ-ing: An Interdisciplinary Journal, 8, 453485.

Ellis Weismer, S., Tomblin, J. B., Zhang, X., Buckwalter, P.,Chynoweth, J. G., & Jones, M. (2000). Nonword repeti-

tion performance in school-age children with and with-out language impairment. Journal of Speech, Language,and Hearing Research, 43, 865878.

Fletcher, J. M., Shaywitz, S. E., Shankweiler, D. P., Katz, L.,Liberman, I. Y., Stuebing, K. K., et al. (1994). Cognitiveproles of reading disability: Comparisons of discrep-ancy and low achievement denitions. Journal of Educa-tional Psychology, 86, 623.

Frankenberger, W., & Fronzaglio, K. (1991). A review ofstates criteria and procedures for identifying childrenwith learning disabilities. Journal of Learning Disabili-ties, 24, 495500.

Gallagher, A., Frith, U., & Snowling, M. J. (2000). Precursors

of literacy delay among children at genetic risk of dys-lexia. Journal of Child Psychology and Psychiatry andAllied Disciplines, 41, 202213.

Gathercole, S. E., & Baddeley, A. D. (1990). Phonologicalmemory decits in language disordered children: Is therea causal connection? Journal of Memory and Language,29, 336360.

Gayan, J., & Olson, R. K. (2003). Genetic and environmentalinuences on individual differences in printed word rec-ognition. Journal of Experimental Child Psychology, 84,97123.

Gillon, G. T. (2004). Phonological awareness: From researchto practice. New York: Guilford Press.

Hogan, T. P., Catts, H. W., & Little, T. (2005). The relation-ship between phonological awareness and reading: Im-plications for the assessment of phonological awareness.Language, Speech, and Hearing Services in Schools, 36,285293.

Hulme, C., & Snowling, M. J. (1992). Decits in output pho-nology: An explanation of reading failure? CognitiveNeuropsychology, 9, 4772.

Joffe, V. L. (1998). Rhyming and related skills in childrenwith specic language impairment. Cahiers de Psychol-ogie Cognitive/Current Psychology of Cognition, 17,479512.

Kamhi, A. G., & Catts, H. W. (1986). Toward an understand-

ing of developmental language and reading disorders.Journal of Speech and Hearing Disorders, 51, 337347.

Leonard, L. B. (1998). Children with specic language im-pairment. Cambridge, MA: MIT Press.

LoBello, S. G. (1991). Intelligence testing with the WISCR.New York: Wiley.

Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2003). A de -nition of dyslexia. Annals of Dyslexia, 53, 114. Lyytinen,H., Ahonen, T., Eklund, K., Guttorm, T. K., Laakso, M. L.,Leinonen, S., et al. (2001). Developmental pathways ofchildren with and without familial risk for dyslexia dur-

ing the rst years of life. Developmental Neuropsychol-ogy, 20, 535554.

Lyytinen, P., Poikkeus, A. M., Laakso, M. L., Eklund, K., &Lyytinen, H. (2001). Language development and sym-bolic play in children with and without familial risk fordyslexia. Journal of Speech, Language, and Hearing Re-search, 44, 873885.

McArthur, G. M., Hogben, J. H., Edwards, V. T., Heath, S.M., & Mengler, E. D. (2000). On the specics of spe-cic reading disability and specic language impairment.Journal of Child Psychology and Psychiatry and AlliedDisciplines, 41, 869874.

McGuinness, D., McGuinness, C., & Donohue, J. (1995). Pho-nological training and the alphabet principle: Evidencefor reciprocal causality. Reading Research Quarterly, 30,830852.

Metsala, J. L., & Walley, A. C. (1998). Spoken vocabularygrowth and the segmental restructuring of lexical repre-sentations: Precursors to phonemic awareness and earlyreading ability. In L. C. Ehri & J. L. Metsala (Eds.), Word

recognition in beginning literacy (pp. 89120). Mahwah,NJ: Erlbaum.

Meyer, M. S., Wood, F. B., Hart, L. A., & Felton, R. H. (1998).Selective predictive value of rapid automatized nam-ing in poor readers. Journal of Learning Disabilities, 31,106117.

Miller, C., Kail, R., Leonard, L., & Tomblin, J. B. (2001). Speedof processing in children with specic language impair-ment. Journal of Speech, Language, and Hearing Re-search, 44, 416433.

Montgomery, J. W. (2000). Verbal working memory and sen-tence comprehension in children with specic languageimpairments. Journal of Speech, Language, and Hearing

Research, 43, 293308.Nathan, L., Stackhouse, J., Goulandris, N., & Snowling, M.

J. (2004). The development of early literacy skills amongchildren with speech difculties: A test of the Criti-cal Age Hypothesis. Journal of Speech, Language, andHearing Research, 47, 377391.

Natio

Are Specific Language Impairment and Dyslexia Distinct Disorders

Documents

Transcript of Are Specific Language Impairment and Dyslexia Distinct Disorders