Pragmatic language impairment in relation to autism and SLI

Pragmatic language impairment:

A correlate of SLl, a distinct

subgroup, or part of the

autistic continuum?

Dorothy V.M. Bishop

Inilr

lOO BISHOP

pr:sseRts: r /ith,.language, difficultiqq rnay ,hav-e :trobl ms.,that:: ext@,beyond thetraditionatr'bound*riei'af'sl-tr, and resemble,those ieen,in: auti , Although addi_tionaLBrollemscan arise aq a

:eonseqtre*e'e of oral,languagg,*iffi;lties,,we shouldL'eware.sf,aut atie&a ing tharpiagmatic Aifnel:l{iei ,iecondar], prob-lens;'Assessmentl'of nonverb,al.cotnrnmieation ean:prilvide,v*lua'ble jn{orna.tionin'malcingdiagnostie,distinctic*s.: :,,

INTRODUCTION

Accounts ol specific language impairment (SLfemphasise the difficulties that affected childrenhave with mastering the medium of commun-ication: Grammar, vocabulary and, frequently,phonology are learned with difficulty, so the childhas to struggle to express what he or she wants tosay. Comprehension is often also impaired suchthat the child may focus on a few content wordsand deduce the meaning from these, but makeerrors in understanding spoken language if thisrequires complex vocabulary or syntax. It istypically assumed that the affected child has anormal desire to communicate and is developingunremarkably in other respects. As Miller (1991)put it:

"Children with language disorders evidencestrengths in conversation skills. They arepurposeful and responsive; however com-munication is limited by their mastery olgrammatical form." (p. 6)

Howeveq not all cases of SLI fit this description.Occasionally one sees children whose problemsare not confined to language form: content anduse ollanguage are also abnormai. pragmatic dif-ficulties are communicative problems that lrave todo with the appropriate use of language in a givencontext. Typically what one sees is a child usingutterances that are syntactically well-formed andcomplex, but which don't appear appropriate inthe conversational context in which they occur(see Bishop, 1997, for a fuller account). In addi-tion, such children may display social and behavi-oural deficits that resemble those seen in autistic

disorder. Strictly speaking, then, they don't seemto fit the category of specific langLiage impairment.Most experts will agree that such children e_ris:(although I shall argue that their numbers ma1, beunderestimated), but they disagree as to hou. torespond to them.

A common reaction is to regard tl.re pragmaricdifficulties and nonverbal impairments in socialor imaginative behaviours that are seen in SLI assecondary to the oral language difficulties. Forinstance, Brinton and Fujiki (1993) point our that:"Because language skills piay a critical role insocial interaction, it seems likely that children withlanguage difficulties would be at particular risklor social failure" (p. i95). These authors arguethat the traditional diagnosis of SLI by exclusion.with its emphasis on the specifici4; of the impair-ments in language structure, is misleading andunrealistic. lmpairments in language do not occurin a vacuum, but rather have impacts on man).aspects of development. According to this view-point, associated impairments are to be expected.

A contrasting view points to the centrai placethat pragmatic deficits play in autistic disorder.and proposes that children with serious prag_matic impairments should be regarded as casesof autism, or at least be categorised as having apervasive developmental disorder, rather than aspecific developmental disorder. Of course. bothviews might be valid in different cases. The prob-lem is that, unless we are able to specily just howand when a language disorder can lead to second-ary problems, we confront major diagnostic diffi-culties every time we see a child who presentswith a clinical picture intermediate betweenautistic disorder and SLL Furthermore, as Craig(1993) pointed out, until we establish whethersocial interactionai difficulties in SLI represent aprimary deficrt in social knowledge or a secondary

6. PRAGMATIC LANGUAGE IMPAIRMENT 101

consequence ofpoor language skills, our attemptsat intervention will be inadequate.

In this chapter, first I review studies that lookat social and pragmatic impairments in childrenwrth SLI, considering how frequent these are, andhow far they can be regarded as secondary con-sequences of structural language limitations. I thenmove on to cases of pragmatic language impair-ment that are more difficult to explain in this way,

because the child's use of language is dispropor-tionately poor in relation to structural languageskills. For such children, a critical question iswhether a diagnosis ofautism or autistic spectrumdisorder would be appropriate. My conclusion isthat there are many children who fall between thediagnostic options of SLI or autism: their develop-mental difficulties are not restricted to structuralaspects of language but, on the other hand, theydo not have the full range of pervasive impairrments that would warrant a diagnosis of autism.

SOCIAL AND PRAGMATIC DEFICITS ASSECONDARY CONSEQUENCES OF SLI

Several studies of SLI have documented probiemswith peer relationships and social interaction. Forinstance, Paul, Spangle Looney and Dahm (1991)found that late-talking 2- and 3-year-olds wereimpaired on the Socialization Scale of the Vine-land Adaptive Behavior Schedule, even if itemsinvolving langtrage were excluded from considera-tion. Nonverbal items included playing socialgames, imitating complex motor routines in play,using household objects in play, and smilingappropriately. Around one third ofthe late talkershad receptive language deficits at age 3 years, and

all but one of these children was also impairedon the Socialization Scale. Such a study tells us

that problems exist, but does not help us sort outcause and effect. Do the language problems leadto poor peer relations, or does the child have amore basic deficit in social cognition or emotionalrelations?

Empirical evidence to show that a child'slanguage level affects peer relationships wasprovided in a series of studies by Rice and her

colleagues. Rice, Sell and Hadley (1991) assessed

social interactions in a preschool playgroup forfour groups of children: those with SLI, thosewith speech disorders, those with English as asecond language (ESL), and a control group ofnormally-developing children. They found closesimilarities between the SLI and ESL groups: bothwere less likely to initiate interactions, and whenthey did so, they were more likely to communic-ate with an adult rather than another child. ln a

related study, Hadley and Rice (1991) found thatchildren with language or speech impairmentswere half as likely to be addressed by their peers

as children with age-appropriate language. Whenaddressed by other children, they were less likelyto respond. It is sometimes assumed that childrenare largely insensitive to the characteristics oftheir playmates before the age of 7 or 8 years.

This study contradicts that view, and shows thateven by 3 or 4 years ofage, children are aware ofthe communicative level of others, and prefer tointeract with peers who have age-appropriatelanguage skills. These authors concluded thatthere is a negative interactive spiral, whereby a

child becomes increasingly unwilling to engagein interactions with peers after experiencing lackof success.

Gertner, Rice and Hadley (1994) went on tomeasure children's peer relations more directly,using a sociometric method in which childrenwere asked to nominate which of their classmatesthey would like to play with, and which theywould prefer not to play with. Children with age-appropriate language were by far the most pop-ular group, and level ofreceptive language was a

strong predictor of children's popularity. These

authors note that children with English as a sec-

ond language, like children with SLI, receive fewerpositive and more negative nominations from otherchildren. They argued, therefore, that the socialdifficulties of the SLI group are a direct con-sequence of their limited language skills, ratherthan reflecting some constitutional limitation olsocial cognition that co-occurs with the languageimpairment. More recently, Redmond and Rice(1998) showed that teachers rated children withSLI as having more social problems than controlswhereas parents did not, and concluded that this

102 BISHOP

was further evidence that social impairmentswere situationally dependent, and not an intrinsiccharacteristic of the child. Viewed in this 1ight,abnormal use of language by some children withSLI could be a secondary consequence of lackof social experience, which arises as a result ofunsuccessful attempts at communication in theearly years of life.

PRAGMATIC DIFFICULTIES THAT ARE HARDTO EXPLAIN AS SECONDARYCONSEQUENCES

There is, however, reason to believe that this isnot the whole story. I sha1l locus on three lines ofevidence that favour the view that, in at least somechildren, there are pragmatic and social difficultiesthat cannot be explained away as direct or indir-ect effects ol poor mastery of structural aspectsof language.

1. Children with disproportionate pragmaticdifficulties: "semantic-pragmatic disorder"

Perhaps the greatest difficulty for any theory thatattempts to explain away pragmatic problems inSLI as consequences of oral language limitationsis the heterogeneity of SLI. Some children whopresent with SLI appear to have particular diffi-culties with the pragmatic aspects of communica-tion. Their deficits are similar in kind, thoughtypically milder in degree, than those describedin high-functioning autism. However, there is nopositive correlation between severity of expressiveor receptive language limitations and extent ofpragmatic difficulties. Problems in the appropriateuse of language can co-occur with relatively goodmastery of language form.

In the mid 1980s, two descriptive taxonomiesof developmental language disorders were pub-lished one (Rapin & Alien, 1983) in the USA,and the other (Bishop & Rosenbloom, 1987) inthe UK. Both described a subtype of languageimpairment in which structural aspects of lan-guage (phonology and grammar) were relativelyintact, but use of language was abnormal. Rapin

: Rfrpin1s,1Sg6 charaeterisatian of ,scmantic ,, ' ,"' :

. -f::g=1*t"ry1| -.

--..-...- .,i Verlrosily, , ,,;

r Comprehenrion defieits for conneeted speocl

i 'frlord finding deficits ,

I Atypical word choices

; Phoaology and syntax unimpaired: Inadequate eonversational,skills,i Speaki*g'aloud,'to'ao ono in'particular , . :

Poor mainlenance of topici', Answering besides the peint of e questiou ,

and Allen coined the term "semantic pragmaticdeficit syndrome" to describe this subtype. Bishopand Rosenbloom described a similar clinical pic-ture and, following Rapin and Allen, termed this"semantic-pragmatic disorder". The most recentclinical account by Rapin (1996) lists the clinicalcharacteristics shown in Table 6.1. Although theUK and US accounts have close similarities,there is an important difference in how they usediagnostic labels. Rapin's classification is notdesigned simply to subtype SLI, but rather tocharacterise communicative problems arising fromany cause. Thus the term "semantic pragmaticdeficit disorder" may be applied to children withknown organic etiologies, such as those withhydrocephalus or Williams syndrome, and to thosewith autistic disorder, as well as to a minority ofthose with SLl. In the UK, "semantic pragmaticdisorder" has been regarded as a subtype of SLl,and typically incorporates the implicit notion thatnonverbal IQ is normal, there is no known organicetiology, and the child does not meet diagnosticcriteria lor autism.

Conti-Ramsden, Crutchley and Botting (.1991)assessed a random sample of all 7-year-o1d chil-dren attending language units (special classes forchildren with SLI) in England, using both teacherreport and standardised tests. On cluster analysis,they identified a subgroup of children (cluster 6)whose language profi1e closely matched Rapin andAllen's "semantic pragmatic deficit syndrome".However, the difficulties of these children only

6, PRAGMATIC LANGUAGE IMPAIRMENT 103

became evident when teacher ratings were takeninto account. If reliance were placed solely onstandardised test scores, children in this subgrouplooked relatively unimpaired, despite the fact thatthe teachers regarded their communicative diffi-culties as severe. This study emphasises on the

one hand that pragmatic difficulties can be seen inthe context ofadequate structural language skiils,and on the other that there is a lack of assessments

that pinpoint pragmatic difficulties.

2. Microanalysis of conversational glitches

Another approach to understanding pragmaticimpairments is to look in detail at the com-municative behaviours that lead to a child beingregarded as a case of "semantic pragmatic dis-order". This approach was adopted by Bishop and

Adams (1989), who undertook a classificationof children's conversational utterances that hadbeen judged "inappropriate". Although some cases

could be accounted for by difficulties in sentence

formulation, semantic selection, or comprehensionof the partner's talk, others were less easy to ex-

plain this way. For instance, some children wouldprovide far too little or far too much informationto the conversational partner, or would use stere-

otyped utterances, sometimes with abnormalprosody. More recently, Bishop et al. (2000) have

extended this approach, focusing specifically onthe extent to which children's conversationalresponses mesh with an adult's prior solicitation.An attempt was made to subdivide responses intothose judged to be adequate (meeting expecta-tions), inadequate (not meeting expectations, butplausibly accounted for by limitations of languageexpression or comprehension), and pragmaticallyinappropriate (not meeting expectations, but notreadily explained by limitations of language ex-

pression or comprehension). After training, ratersachieved reasonable agreement in independentlycoding children's responses in this way. and the

method received some validation from the findingthat younger normally-developing children madea relatively high rate of inadequate responses,

but relatively few pragmatically inappropriate re-sponses. Children with SLI showed wide individualvariation in their performance, but on average

made a higher rate of both pragmatically inappro-priate and inadequate responses compared toage-matched controls. This study, then, represented

a direct attempt to distinguish pragmatic prob-lems that could be secondary to structural lan-guage limitations from those that could not, andconcluded that for some language-impaired chil-dren, the latter type of response was relativelycommon.

3. lmpairments of nonverbalcommunication

Of particular interest is nonverbal communica-tion. It is sometimes assumed that children withSLI will compensate for their oral languagedeficits by using nonverbal communication. Thisis exactly what is seen in children with severe

hearing impairments. Even if not exposed to anatural sign language, most hearing-impaired chil-dren will make healy use of gesture and facialexpression to communicate, and will, if given theopporhrnity to interact with other hearing-impairedchildren, develop a sign language de novo (Goldin-Meadow & Mylander, 1998). Use of nonverbalcommunication in children with SLI is muchless well documented though an early study byBartak, Rutter and Cox (1975) reported that justover 40oh of a sample with receptive languageimpairments did not use gesture. Several clinicalaccounts have remarked on difficulties experiencedby children with SLI in the use of gesture andfaciai expression, and difficulties in interpretingnonverbal cues emitted by others (see Goldman,1987, for a brief review). In the recent studydescribed above, Bishop et al. (2000) analysedconversations between language-impaired chil-dren and unfamiliar adults, and compared these

with adult-child conversations with normallydeveloping children. The language-impaired chil-dren had been selected to include some thoughtto have disproportionate pragmatic problems,

and others with more typical SLI. The surprisingfinding was that both these groups made less use

of nonverbal responses (e.g. head nodding) thannormally developing children. This could not be

regarded as communicative immaturity, becauseyounger language-matched control children made

104 BISHOP

Boxplot showing

distribution of nonverbal

responses (absolute

number per conversation)

for four groups of children:

Language matched

controls (LA con), age

matched controls (CA

con), typical SLI (SLl-T)

and pragmatic language

impaired (PLl). The top

and bottom o{ the shaded

bar show the 25th and

75th centiles respectively,

with the central line

indicating the median,

The error bars show 1Oth

and 90th centiles, with the

range being represented

by circular points. See

Bishop et al. (2000)

lor further data.

,6 30-oc)

co?oc-'z

1i)

LA con CA con SLI-T

the highest use of nonverbal responses (see

Fig. 6.1).Overall then, there are several lines of evid-

ence to support the view that some children havepragmatic difficuities that are disproportionate andnot readily explained as secondary consequences

of SLI. How, then, should we classify such chil-dren? One view is that they belong more natur-ally within the autistic spectrum rather than withSLI. This view needs to be given serious con-sideration, given that pragmatic difficulties are a

hallmark of autistic disorder.

PRAGMATIC PROBLEMS ASSYMPTOMATIC OF AUTISM

The relationship between autism and develop-mental language disorder was a topic that wasmuch discussed in the 1960s and 1970s. At thattime, several experts had suggested that autismwas, fundamentally, a form of language disorder,and that the abnormal nonverbal behaviours thatare part of the autistic syndrome might simply be

secondary consequences of poor communicationskills. However, this view was gradually aban-doned as research evidence mounted showing thatchildren with autism had distinctive communis-ative, social and behavioural difficulties not seenin other conditions that affected language develop-ment. As our understanding of autistic disorderincreased a consensus developed that autism anddevelopmental language disorder are qualitativelydistinct, a position which is reflected in contem-porary diagnostic systems.

The Diagnostic and Statistical Manual (DSM)of the American Psychiatric Association is a

case in point. In DSM-IV (American PsychiatricAssociation, 1994), autistic disorder is identifiedas a pervasive developmental disorder (PDD),which means that it has two salient character-istics. First, the disorder affects several areas offunctioning. In autistic disorder, there is a triadof impairments in language, social relationshipsand behavioural and imaginative repertoire. Sec-

ond the impairments in these domains go bey-ond simple immaturities, with children showingbehaviours that would be abnormal at any age.

In contrast, specific language impairment (SLI)

o

o

6, PRAGMATIC LANGUAGE IMPAIRMENT t05

....no.1ing19*o.o'.uy.i+.*..=:-.:..:iil:l::i:.]:i::..o.5...i: no phrase speech by 30 monttr.,,r,dimiaalsL*d'or attno*roa],,.bgb.blg,..r:::::,::::::ii:tr,,:::::::::::,,,,.,,.','l

t"i,.t-u.:..t .1..t..,* tLt.i65'it'aiiiiial*r*J*.*ai iaiij.*ia*iei.:',. -"'..tt=:-'..t+it-:,,,.-,,::, :',':::i:1:45'

ever thoueht deaf 84 65

delects oiarticulation 53

ir,ril'.....1:r".:,i... .,..l.i:tl..i:i"':l1 t.'.:.'.:l.:'..,i',i, i,;.i:ti.i':,tlt''

, *Joruria i."eq ' 1i;...i':1i...1i.t; .t'tti..'..ii,'..1g$,,.

' srereotyped utterances leverj or 9

metaphorical language lever) 37 0

inappropriate remarks 32 0

no spontaneous chatno ,pontun*us char' fails to respond to questions 6J 22

n.n., used eesrle

is a paradigmatic example of a specific develop-mental disorder. This is diagnosed when thereis impairment in a single domain of functioning(in this case language), and the development inthis domain is delayed rather than qualitativelyabnormal.

The distinction between specific and pervas-

ive developmental disorders was supported by a

seminal early study by Bartak et al. (1975). These

researchers explicitly compared children withautistic disorder and those with receptive develop-mental language disorder on both quantitative and

qualitative measures of communication and be-

haviour. The initial goal was to identify childrenwith severe comprehension problems in the con-

text of normal nonverbal ability. Boys aged 5-10years, with nonverbal IQ of 70 or above, wererecruited from specialist centres for children with

autism and/or language disorders. Once suitablechildren had been found, information from case

records was used to subdivide them into thosewith autistic disorder and those with receptivelanguage disorder. A small proportion of childrendid not fall clearly into either category and werekept separate. Table 6.2 shows some of the findingsof the study. In terms of communication therewere some points of similarity between childrenwith autism and those with language disorder, butalso clear qualitative differences. Two features inparticular stand out: first, almost all the childrenwith autism were repofied as not using gesture,

whereas this was true for only half of the language-

impaired children. Second, in the children withlanguage impairment, the communication difficult-ies could be described in terms of immaturityand/or lack of communicative skills. In those with

106 BISHOP

n

II

autism, there were more qualitative oddities inthe use of language that are not normal at anyage. For instance, children would use words andphrases with a private, individualised meaningbased on their own personal experiences, suchthat the listener who had not shared the child'sexperience would have difficulty making sense

of the utterance. Kanner (1946), who used theterm "metaphorical language" to describe suchinstances, gave an example ol a child who re-ferred to himself as "Blum" whenever his parentsquestioned whether he was telling the truth. Thisapparently irrelevant utterance assumed meaningonce the parents realised that the child had beenreading a large advertising sign which stated"B1um tells the truth". In the study by Bartaket al., cryptic utterances such as these, whichbecome comprehensible only when related to thechild's personal experiences, as weil as stereotlpeduse of language, were much more common inchildren with autism as compared to those withlanguage impairments. The results may be sum-marised by saying that whereas both groups hadmajor difficulties in mastering language form(syntax and semantics), plus limitations in vocabu-lary comprehension and understanding of com-plex sentences, the autistic group had addrtionalimpairments in the appropriate use of language,i.e. pragmatics.

The study by Bartak et al. (1975) was import-ant in establishing that the profile of impairments inautism cannot simply be reduced to consequencesof poor language comprehension. Over the pasttwo decades, the conceptualism of autistic dis-order has changed with greater emphasis beingplaced on impairments of joint attention, socialinteraction, and social cognition. These are allareas where children with autism can be shown tobe deficient, even when compared with controlgroups matched on mental age and/or languagelevel. Furthermore, it is now recognised thatpragmatic difficulties are a hallmark of autisticcommunication, and that other language difficult-ies are a more variable correlate (Frith, 1989).

Bartak et al. (1977') carried out a series of dis-criminant function analyses on data from the samesample. Discriminant function analysis is a statis-ticai method that assesses how accuratelv a set of

measures can be used to assign an individual to a

specific group. First, the two groups (in this case

those with autism or SLI) are compared on a set

of variabies, and each measure is assigned a

weighting which reflects how well it discriminatesbetween the groups. The weighted measures arethen summed for each child; those scoring abrove

a cutoff are allocated to category A, and thosebelow the cutoff to category B. If there is littleoverlap between the groups on the measured char-acteristics, then there should be close agreementbetween this categorical assignment and the ori-ginal allocation ofcases to groups. That is exactlywhat was found by Bartak et al. (1977). Very fewchildren had scores that were at all ambiguous.Overall then, the study by Bartak et al. (1975)was seen as supporting the distinction betweenautistic disorder and SLI, with pragmatic difficult-ies being a characteristic difficulty of the formergroup. but not lhe latter.

Does this mean, then, that all children withautistic-like pragmatic dilficulties shouid be re-garded as autistic? There are two lines of evidencethat caution against leaping to that conclusion.First, the studies concerned with dilferential dia-gnosis of autism and SLI, while noting the differ-ent clinical pictures of these disorders, aiso re-veal cases that are hard to categorise as one orthe other. Second, studies ofchildren who are iden-tified as having semantic-pragmatic disorder re-veal that only a subset of them appear to havesignilicant autistic features in nonlinguistic do-mains.

STUDIES CONTRASTING AUTISTICDISORDER AND SLI: EVIDENCE FORINTERMEDIATE CASES

Bartak et al. (1975) noted that while the datashowed a clear differentiation between autism anddevelopmental language impairment, five "mixed"cases were important in demonstrating an areaof overlap containing children who show somefeatures of both conditions. On the discriminantfunction analysis, about one fifth of the childrencould not be unequivocally classified usually

6, PRAGMATIC LANGUAGE IMPAIRMENT 107

because data from the child's clinical history tolda different story than the contemporary informa-tion. Some children had moved from an autisticprofile to a language-impaired one, whereas otherschanged in the opposite direction.

Cantwell, Baker, Rutter and Mawhood (1989)carried out a follow-up in middie childhood of29 of these children. These investigators foundthat the earlier, ciear differentiation between thegroups became somewhat blurred at the laterage. Although many aspects of communicationand behaviour tended to improve in the language-impaired group, there was a subset of children forwhom peer relations deteriorated notably. Cantwellet al. commented: "the finding of marked diffi-culties in friendships in some children whoselanguage was improving raises questions regard-ing the traditional view that the socioemotionalproblems are just secondary features that havedeveloped as a response to. . . having a languagehandicap" (p.29). Furthermore, at follow-up, 28%of the language-impaired group showed ritualisticbehaviour, 3 6% had stereotyped mannerisms, 2 I %o

used metaphorical language, and 31% producedstereotyped utterances. All of these are behavi-ours that are not typically regarded as part oftheclinical picture of SLI but are seen in autism.Mawhood (1995) studied the same sample inadulthood and found that several cases who hadbeen regarded as having a receptive languagedisorder continued to show evidence of socialimpairment and restricted interests. Overall, then,both the original study by Bartak et al., andsubsequent follow-ups of this sample confirmed,that, even when the most detailed and careful dia-gnostic process is applied there are children who,are difficult to categorise unambiguously as casesof autistic disorder or SLI.

In recent years, the notion ofa sharp boundarybetween autistic disorder and language impair-ment has come under increasing pressure. In a

study of preschool children with mental handicapand very limited language skiils, Lord and Pickles(1996) contrasted social behaviour, nonverbalcommunication, and repetitive behaviours inautistic vs nonautistic cases. They concluded that,although none of the nonautistic group had everbeen thought to have a pervasive developmental

disorder, they nevertheless had difficulties insocial behaviour that were similar to those of theautistic children. They commented on the artifici-ality of absolute diagnostic distinctions withina population of cognitively delayed language-impaired young children. The existence of inter-mediate cases has been emphasised further instudies of parents and siblings of people withautism, which find that although core autism israre in these relatives, milder difficulties, ofteninvolving only one or two of the elements of theautistic triad are common (Bolton et al., 1994).This has led to the notion of a "lesser variant" ofautism being employed in genetic studies.

AUTISTIC FEATURES IN CHILDREN WITHSEMANTIC- PRAGMATIC DISORDER

Although there has been much debate as towhether "semantic-pragmatic disorder" is partof the autistic continuum (e.g. Lister Brook &Bowler, 1992; Boucher, 1998, and associatedcommentaries on Boucher's paper), there has beenlittle hard evidence on which to base an argu-ment. A recent study using neuropsychologicaltests and measures of social cognition showedclose overlap in test profile between a group with"semantic pragmatic disorder" and children withhigh-functioning autism, both of whom contrastedwith children with a more typical form of SLI(Shields, Varley, Broks, & Simpson, 1996a, b).However, other studies, including those reviewedabove, show that many children with pragmaticdifficulties fall short of meeting diagnostic criteriafor autistic disorder. So we are left with the ques-tion of how to categorise language-impairedchildren who have pragmatic difficulties whichsometimes, but not always, occur in association .

with mild abnormalities of social interaction orrestricted interests.

Part of the problem confronting researcherswho wish to address this issue is the lack ofclear criteria for defining "semantic-pragmatic dis-order", coupled with lack of suitable diagnosticinstruments for objectively documenting pragmaticdifficulties. With most forms of SLI it is possible

-108 BTSHOP

rr

et

to devise operatioital dejrnitions" sclecting chil_dren. for instance. whose stanc-iarclisecl scores onlanguage tests fall belorv some specifiec,l lcvei.However, the clinical leatures shown in Table6.1 are not easy to assess by availabie tests. Asnoted above. Conti-Ramsden et al. (1997) tbunclthat language tests did not reveal a distinctivepattern of deficits in children .,l,ho r,r,ere thoughtto have semantic-pragr-nattc disorcier: it was onlyrvhen tcacher impression ',vas taken into accountthat the dilfrculties of these childrcn could becharacterised.

One w.ay ol respr:nding to this situation is tomake teachcr ratin-qs more objective. This approachrvas adopted by Bishop (1998) in developing rhechildren's comnrLrnication checklist (CCIC). Thisstudy had tu'o related goals. On the one hancl- itaimed to develop a trore objective way of assess_ing childrent pragmatic impairments using ratingsby teachers or other prolessionals r.r,ho kneu, thechild uell. The rationale for adopting this approachwas that many praglnatic deficits are rare inoccurrence. and/or dilficLrlt to elicit in a clinicalassessntent because they are contextually depencl_ent. A person u.ho interacts rvith the child reslr_iarly might. therefor-e" be in a better position'toevaluate these aspects of behaviour than a profes-sional rvho sees the child fbr a sin-eie assessment.A second goal ol this study \\,as to consider hou,strongly pragmatic impairments uere associatedu"ith impairments in other dornains, especiallyproblems rvith social interaction and restrictedinterests ol the kind that are character.istic ofautism. In this study, a subset of chilclren liornthe suru'ey by Conti-Ramsden et al. (1997) wereassessed on a checklist that included scalesassessing aspects ol strr_rctural language, pragmattcskills, social interaction. and restricteci interests.Ttems rvere retained in the checklist ilthey shou,edacceptable ler,'els ol agreement betu'een independ_ent raters. Diagnostic inlonnation ftom schoolrecords was contpared u'ith checklist results. Chil_dren nhose records indicateci a definite or pos-sible diagnosis of "semantic pragmatic disorcler.'did as expectecl obtain lower scores on the prag-matic scales. None of these children hacl a dia_gnosis olautism, but a subset rvas thought to havepossible or definite autistic features or Asperger

syndronte. The latter group shorved somc impaii._n'rellts on the social ancl interests scales. Horvever.another subset ol the semantic pragmatic grouphad ner.er been gir,en any autistic spectr-um dia_gnosis. Thesc cliildren did not drfler fi.om a .,typical

SLi" group in terms of their scores on the socialand intelests scales. This study, then, supporteclthe vieri' that, \\.hile pragmatic in.rpairments andother autistic features tencl to co-occllr. there arechildren with si-enificant pragmatic difficLrltieswho do not have any marked difficulties in thcdomains olpeer relations or interests. The check_list gives only a cursory exanrination of thesearcas, and in fllture u'ork it is planned to do amuch nore detailcd ar.ralysis of autistic f-eatr-rresin language-inrpaired children with pragmaticproblems. using instruments clevelopecl lor thediagnosis of autism. For the present, though, ourdata are consistent u'ith those of other str,rdiesin suggesting that there is n'ic1e variation in thespecific clinical profiles shown by children rvithpragmatic difficulties, and it would be pr.ematureto conclude that all of them have a mild fonl olar-rtistic disorder.

d

n

iti

,f

T

fiI/f,

!lj

IS THERE A SEPARATE SYNDROME OFSEMANTIC- PRAGMATIC DISORDER?

I have argued that one sees children u,ith prag_matic difllculties that arc not ji-rst secondary tostructural language problems, yet who do notappear to meet criteria lor autism either. So whereshould tliey be classified'J One solution rvould beto propose "semantic pragmatic disorder" as aseparate diagnostic entity, distinguished from bothSLI and autistic disorder. I shall argue that this isnot a satislactory solution, because there is littleevidence that the features described in Table 6.1fornr a colrerent syndrome. Rather. it seems asthough pragmatic dilficulties are a variable cor_relate ol SLI: they can be flor-urd in childrenwho meet the clinical descriptions of semantic_pragmatic disolder and rvho use fluent. complexianguage, but they can also be found in otherchildreri u,ho have more typical structural languageproblems. They may be associated with semantic

#,il

icl

&n

6. PRAGMATIC LANGUAGE IMPAIRMENT 1Og

language use

e^€phonologicg * E**synactrc € e*-^*deficit €e€ *€

€*e€ e* language structure

Notional distribution of

impairments in language

structure and language

use, according to a modelthat allows for oppositeprofiles in those withphonologic-syntactic def icitand those with semantic-pragmatic deficit. Good

skills are shown as +,and deficits as -.

+

€€€ € semantic-

* * Pragmatic=s *edeflcit

difficulties, but can also be found in children whodo not appear to have either word-finding prob-lems or unusual vocabulary.

What is the evidence to support these asser-tions? One source of data comes from studiesI conducted in the 1980s, using experimentalmethods to assess aspects of pragmatic func-tioning in language-impaired children. The goalwas to contrast children who had the clinicalfeatures of semantic pragmatic disorder with otherchildren with more typical SLI. To this end, Ideveloped various tasks that were designed topinpoint cognitive processes thought to underliepragmatic problems, anticipating that these wouldclearly differentiate typical SLI from ',semantic-pragmatic disorder". However, assessments thatwere devised to be sensitive to deficits in casesof semantic-pragmatic disorder typically revealedunexpected deficits in some children with moretypical SLL This proved to be the case withtests of referential communication skill (Bishop& Adams, 1991) and story comprehension (Bishop& Adams, 1992). In the study already describedby Bishop et al. (2000), which used derailedanalysis of more naturalistic conversational data,the same picture emerged. There was a subset ofchildren who were prone to make pragmaticallyinappropriate responses, and who were alsoabnormally restricted in their use of nonverbalresponding. However, these children are not neces-sarily the ones identified by teachers as havingcharacteristics of semantic-pragmatic disorder.

Some of them had structural language limitationsas well.

Clinical classifications have stressed the con-trast between semantic-pragmatic disorder andmore typical SLI, which typically correspondsto what Rapin (1996) would term phonologic-syntactic deficit disorder. On this basis, oneanticipates a double dissociation between the twokinds of impairment, as shown in Fig. 6.2. How-ever, in a sfudy using an earlier version of a teacherchecklist, Bishop (2000) found results moreconsistent with the scenario shown in Fig. 6.3.Rather than the predicted inverse relationshipbetween impairments of language form andpragmatic difficulties shown in Fig. 6.2,there wasa small positive correlation between the twodomains. Although there were many childrenwho did resemble the profile in Table 6.1, withrelatively good mastery of syntax and phonologyand poor pragmatics, and others who showed theopposite picture, there were yet other cases whohad a "double deficit", with poor ratings on lan-guage form and language use. Taken together,such results begin to question the notion ofsemantic-pragmatic .disorder as a separate syn-drome of language impairment. Pragmatic diffi-culties are a common, though by no meansuniversal, correlate of language impairment. Afurther result from this study was that there wasno positive correlation between semantic and prag-matic difficulties, semantic impairments beingcommon in all language-impaired children.

110 BtsHoP

€- l+€-phonologic-@*swntactic s-€

deficit €€

€€

e€€Alternative model

of distribution of

impairments, showing

additional subgroup

with impairment in both

language structure and

language use.

Model depicting

dissociable impairments

in language structure,

social use of language,

and interests, and howparticular profiles map

onto existing clinical

categories.

e€6€

€€

*€e€€

€e

€€

e€

€

language structure

+

semantic-pragmaticdeficit

languagestructure

Bishop (1998, 2000) suggested that the notionof a distinct syndrome of semantic pragmatic dis-order may have come about just because pragmaticproblems are much more obvious and hard toexplain away when they occur in the context ofgood formal language skills. Quite simply, chil-dren who speak clearly and in complex sentencesbut who use strange language draw attention tothemselves, because on the one hand we canunderstand what they are saying, and on the otherhand we don't expect odd utterances from achild with apparently mature language skills. Datafrom both conversational analysis and from teacherchecklists, however, suggest that there is a widerange of pragmatic competence in children withmore limited mastery of language form. Howeveq

we automatically tend to make allowance forpragmatic difficulties in such a child assumingthat this is a secondary consequence of limitedpowers of expression and comprehension.

Findings such as these do raise questionsabout a term such as "semantic pragmatic deficitsyndrome". The label "pragmatic language im-pairment" seems preferable. It does not imply thatsemantic and pragmatic problems will necessarilyco-occur, and it is easier to accommodate to amore dimensional view of language impairment,which treats pragmatics as one domain in whichcommunication may be impaired.

The set diagram shown in Fig. 6.4 illustratesthis more dimensional view of communication dis-orders. The three sets have some correspondence

-s*€-

social use

6. PRAGMATIC LANGUAGE IMPAIRMENT 111

t with the traditional triad of autistic impairments,but are shown as frequently being dissociated.Different children wiil have different combinationsof symptoms, varying in severity. Autistic disorderis diagnosed when a child has major impairmentsin all three domains. Asperger's disorder, definedin DSM-IV as a pervasive developmental dis-order in which structural language skills developnormally, is shown as involving a combined deficitin the social use of language and restricted inter-ests. Children who have pragmatic problems in thecontext of relatively good language form wouldin this diagram, fit in the area in the top nght ofthe figure, where only the "social/pragmatic', circleis involved. Howeveq as the diagram emphasises,there are no clear boundaries between this dis-order and autistic disorder, on the one hand andSLI on the other. Furthermore, there are childrenwho have a mixed picture ol problems with lan-guage structure and pragmatics.

PRAGMATIC DIFFICULTIES AND PDDNOS

A dimensional view may be useful in capturingthe fulI range of clinical phenomena, but it cancreate problems in practice. Quite simply, dia-gnostic labels are important in ensuring thatchildren obtain access to suitable education andintervention. A11 too often, the ciinical reality isthat many children have complex constellationsof developmental difficulties, but provision tendsto be directed towards the textbook categories.Thus one may be confronted with a stark choicebetween an educational placement for childrenwith autism, which is geared mainly to the needsof low-functioning children with difficult-to-manage behaviour, or one for children with SLI,where the emphasis is on conventional speechlanguage therapy.

One diagnosis thaf is increasingly beingadopted in such complex cases is pervasive

Developmental Disorder Not Otherwise Specifie<J,or PDDNOS (see, e.g. O'Hare, euew, & Aitkin,1998). This category was introduced in DSM-IVto be used for cases where there is subthresholdsymptomatology, or where there is significant

impairment in only one or two of the domains ofthe autistic triad. However, there are problems withthis label. PDDNOS was originally intended asa default diagnosis to be used in rare instanceswhen a child just fell short of diagnostic criteriafor autism. It is vaguely described with no cleardefining criteria, and potentialiy incorporates ahuge range ofclinical profiles. It is too nonspecifica diagnosis to be useful for identilying the kindsof services a child requires (though the labelitself may serve a function in ensuring that thechild obtains access to some kind of services).It may be that, as PDDNOS becomes an increas-ingly popular diagnosis, better services will bedeveloped for high-functioning children who donot have major behaviour problems but who dohave complex difficulties. Some speech-languagetherapists have developed considerable expertisein working with such cases (e.g. Brinton & Fujiki,1989; Gallagher, 1991), but many professionalsstiil are seriously challenged by their complexdeficits. Although there is some work concernedwith facilitating nonverbal communication in chil-dren (e.g. Nowicki & Duke, 1992), this has beendeveloped independently by clinical psychologistsand there has been little application to language-impaired children.

My own view is that it would be premature toconclude that children with pLI either do or donot properly belong in the autistic spectrum inour eurrent state of knowledge. Future researchon etiology or underlying cognitive processes maygive clearer answers. Meanwhile, an importantmessage from the research to date is that we needto look carefully at pragmatics when assessing achild with language impairment, and not auto-matically assume that poor use of communicationis a secondary symptom. To date we have beenseriously limited by the lack of reliable and prac-tical assessment tools. Methods such as the check-list described by Bishop (1998) may provide astart in alerting practitioners to cases where thechild has a wider range of communicative diffi-culties. In addition, video analysis of nonverbalcommunication looks like a promising method foridentifying the child whose communication diffi-culties extend beyond oral language. I suspect thatultimately we may need to abandon the sharp

112 BISHOP

ffiilfilltliil

flu:_

rl

f;qi

distinction betr,r'een autism and SLI. and mor,e toa more quantitative approach, u.hich depicts PLIas literally intermediate betrveen autistic disorderanci SLl, rather than belon_eing with one or theotl.rer. One challenge fbr the future is to devisemore appropriate pror.ision for children with suchcomplex problems.

REFEREhIOES

American Psychiatric Association. (.1994). Diagrutstic:cold .\tatistical ntutuul of' mentctl clistnrlers, llheclition. (DSM IIt1. Washington. DC: AmelicanPsychiatric Association.

Bartak. L.. Rutter, M.. & Cox, A. (1975). A comparat-ive stud,v of infantile autism and specific clevelop-mental receptive language disorder: I. The chilclren.Brilish .lottrnctl o.l' Psrchian'.t," I 26, 127 145.

Bartak. I-., Rutter. M.. & Cox. A. (1977). A com-parative stud-v of infantile autism and specificclevclopmental rcceptive langr-rage disorders. IILDiscliminant function analvsis. Journal of Autisntantl Childhood Schi:ophrenia.7. 383 396.

Bishop. D.VM. (1997). Uncoutntott t.rnderstanding.Det,elopntent und clisorcler.s of langtruge comprehen-sion in childreil. Hove, UK: Psychology Prcss.

Bishop. D.VM. (1998). D,^r,elopment of the children'scommunication chccklist (CCC): A rncthod fbrasscssing elualitatile aspccts of conmunicativeir.npairment in childlen. Journa| of Chiltl P,stclto!og.t.

und P.st.'chianl. J9. 879 892.Bishop. D.VM. (2000). What s so specral about Asperger

syndromc? The need fbr furthcr explolatton of thcbolder'lands of autism. In,{. Klin. F.R. \rolkmar. &S.S. Sparror'r' (Eds). Ly;erger "s1:pllyouru (pp. 25:1

277). Nov Yori<: Gr-rilfbrd Pless.

Bishop. D.\{N1., & ,A.dams. C. (1989). Conr.,ersarioniLl

characteristics ol children r.r'ith sernantic pragntarrcdisolder. II. What leatures lead to a judgen-rent ofinappropliac.v'.) B ri t i s' h J o m n u l o f D i.s o rtl er "s o /' C.- o n tntrulicLtrion. 21" 21I 263.

Bishop, D.\{M.. & Aclams, C. (1991). What clo reftr'-entral commrLnication tasks measure? A stucl1' ofchiIdren rvith specific iangr-Lage impairment. App|iedPsvt'hr.tlingttisllc,s, 11. 1 99 215.

Bishop, D.\1M.. & Adams. C. (1992). Conrprehensionproblems in children with specilic language impair-rnent: Literal and inf'erential meaning. Jotnurl o.;f

Specc:h untl Itearing Reseat't-lt. -t-t. 119 129.

Bishop. D.VM.. Chan, J.. Hartley.. J.. Adams. C.. &\\'eir" F-. (2000). t'onr,ersational responsir,eness inspccific langr-rage irnpairment: Eviclcnce of dis-proportionate pl"agmatic dtfficulties in a sr.rbset ofclrldren. Developtneri drul Ps.t'rhopathology, 12,177 199.

Bishop. D.\llvl., & Rosenbloorn. L. (i987). Classifica-tion of childhood language disorders. hr \&'. Yule &N{. Rutter (Ed,s')" Language clet,elopmenr und dis-orders; Clinics in Developntentul Medir:ine fdoubieissue]. London: MacKeith Press.

Bolton, P, MacDonald- H.. Pickles, A., Rios. P. Goode,S., Cror','son, M., Bailey. A., & Rutter, M. (1994).A case-control family history study of auttsm.Jotu'nul oi Child Psycholagt, uncl Psychiutr.y, 35,877 900.

Boucher, J. ( I 998). SPD as a distinct diagnostic entity:Logical considerations and directions for futureresealch. Inlerncttional,Iournal oJ' Longuage andL'ontntttnic:atictn Disorclers" JJ, 71 81.

Brinton. 8.. & Fuiiki. M. (1989). Conyerscrtionolttluntrgenent u,i.th language-impuiretl children:P ru g n1 ct t i c a s.\ e.\ s meil t ctntl i n t e rv enlion. Rockville.MD: Asperr Publishers.

Brinton, 8., & Fujiki. M. (1993). Language. socialslrills, and socioemottonal behavior. Language,Speech, and Hetu'ing Sert,ices in Schools, 24, 1941 98.

Cantu'el1. D.. Baher. L., Rutter. M., & Manhood. L.( 1 989). Infantile autism and developmental r'eceptive

dysphasia: A comparative fbllor,v-up into micldlcchildhood. Jout'nul of Atttisnt crntl DevelopmentulDisortlers. 19, 19 31.

Conii-Rarrsclen. G.. Crutchley. A., & Botting. N.( 1 997). The extcnt to ri,hich psychornctric tests dif-lerentiate subgroups of children with SLl. ,loLtntu/o/ S1teeth, Lctnluage. und Hecu.ing Reseurch, 10.765 777.

Clraig, il K (1993). Social skills of chiich'en u,ithspecific language impairment: Peer lelationships.Lctrtguttge, Speech, und I{eoring Sen.'ice-s in St'lt.ool.s.

21.2A6 215.Frith. U. (1989). A ner.r. look at langua-qe and com-

rnunication in arLtisr-n. B ri ti s h Journ o / of' I)i..t ct rde rso.f Contuuurit'ution. )1.123 150.

GaJlaglrer. T (Ed.). (.1991l. Prugntdtic.\ of lunguoge;(llinicttl prct('ti(e rs.sae.r. Szin Dicgo: Singr-rlar Pub-lishing Group.

Gertner, B.L.. Rice. Nlt.L.. & Hadlel'. PA. (199,+).

Influence of corrlnunicatire cornpetenee on feerpreferences in a pleschool classloom. Journul of'Speech untl Itettrin,g Reseut'c.h, 3i. 913 923.

{p

Ln

6. PRAGMATICLANGUAGE IMPAIRMENT 113

it.ii

;0

;$l

Goldin-Meadow, S., & Mylander, C. (1998). Spontane-ous sign systems created by deaf children in twocultures. Nature, 39 I, 219281.

Goldman, L.G. (1987). Social implications of languagedisorders. Reading, Writing and Lenrning Disabilit-ies, 3,119-130.

Hadley, PA., & Rice, M.L. (199i). Conversationalresponsiveness of speech and language impairedpreschoolers. . Journal of Speech and HearingResearch, 34, 1308-1317 .

Kanner, L. (1946). Irrelevant and metaphorical languagein early infantile autism. American Journal of Psy-chiatry,103,242 246.

Lister Brook, S., & Bowler, D. (1992). Autism byanother name? Semantic and pragmatic impairmentsin children. Journal of Autism and DevelopmentalDisorders, 22, 6l-82.

Lord, C., & Pickles, A. (1996). Language level andnonverbal social-communicative behaviors in autisticand language-deiayed children . Journal of the Amer-ican Academy of Child and Adolescent Psychiatry,35,1542 1550.

Mawhood L. (1995). Autism and developmental lan-guage disorder: Implications from a follow-up inearly adult 1ife. Unpublished Ph.D. thesis, Univer-sity of London, UK.

Miller, J. (1991). Research on language disorders inchildren: A progress report. In J. Mil1er (Ed.).

Research on child language disorders (pp. 3-22),Austin, TX: Pro-Ed.

Nowicki, S., & Duke, M.P. (1992). Helping the childwho doesn't ft in. Atlanta'. Peachtree Publishers.

O'Hare, A.E., Quew, R., & Aitken, K. (1998). The iden-tification of autism in children referred to a tertiaryspeech and language clinic and the implications forservice delivery. Autism, 2,171 180.

Paul, R., Spangle Looney, S., & Dahm, P.S. (1991).Communication and socialization skills at a1es 2and 3 in "late-ta1king" young children. Journal ofSpeech and Hearing Research, -t4, 858-865.

Rapin, I. (1996). Developmental language disorders:A clinical update. Journal ofChild Psychology andPsychiatry, 37, 643-655.

Rapin, L, & Allen, D. (1983). Developmental lan-guage disorders: Nosologic considerations. In U.

Kirk (Ed.), Neuropsychology of language, reading,andspelling (pp. 155-18a). New York: AcademicPress.

Redmond" S.M., & Rice, M.L. (1998). The socioe-motional behaviors of children with SLI: Socialadaptation or social deviance? Journal of Speech,Language, and Hearing Research,4l,688 700.

Rice, M.L., Sell, M.A., & Hadley, P.A. (1991). Socialinteractions of speech and language impaired chil-dren. Journal ofSpeech and Hearing Research,34,1299 1307.

Shields, J., Varley, R., Broks, P., & Simpson, A. (1996a).Hemispheric function in developmental languagedisorders and high-level autism. DevelopmentalMedicine and Child Neurology, 38, 473-486.

Shields, J., Varley, R., Broks, P., & Simpson, A. (1996b).Social cognition in developmental language disordersand high-level auttsm. Developmental Medicine andChild Neurology, 38,487 495.

l

$

Fllt

Speech and languageimpairments in children:

Causes, characteristics, interventionand outcome

edited by

Dorothy V.M. BishopUniversity of Oxford, UK

Laurence B. LeonardPurdue University, lndiana, USA

!el: fl {: