9 language in autism - university of california, san diego

9 Language in Autism

Matthew Wa lens ki. Helen Tager- Flusherg. and Michael 7: UElman

CONTENTS

Introriuction ......... .. .............................................................................................. 175 Langl~age itnd Communication in ASD: The Evidence ........................................ 176

Pragmatic Deficits ................... ..., .................................................................. 176 Nonverbal Communicative Gesture .......................................................... I77 Speech Acts ........................................................................ 177 Conversational Discourse .................................................................... I77 Pragmatic Function5 of Prosody .............................................................. 178 Intcrprcting Nonliteral Languacge .............................................................. I78

Gmmrnnr and Lexicon ............................ .... .................................................... 178 Grarnm;iticul Abilities ................................................................................ 179 Lexical Abi li tics ......................................... ... ............................................. 181 Neurtrirnaging Studies ................... ... .... ............... .................................... 181

Formulaic Speech ....................................................................................... 183 integrative Theories of Language i n ASD ......................................................... 183

Pragmatics and Thenry of Mind ................................................................ I83 Grammar . Lexicon . and the PDH .................................................................. 185

ASD Profile of ProceduraI System Functions .......................................... 188 ASD Profile of the Declarative Memory System ..................... .. .......... 191 The Neurobiology of Procedural and Declarative Memory Brain Structures in ASD ................. ., ..................................... 193

Summary and Conclusion ................... ..,.....,, ................................................ 193 .4cknowledgments .... ......................................................................................... 194 References .............................................................................................................. 1'31

.Autism. often rcfcrrcd to as autism spcctrum disordcr CASD). is a dcvclolpmsnt:~l Jisorder of thc brain rha t. i s strongly associated wilh dcficits in language and cotn- munication as well as a variety sf other Impairments . including abnormal social interaction and motor fi~nctian.' Roughly 20% of cl~ildren with autisln are essentially t~onverba t . u ~ i ng fewer than five words per day.' Others acquire fitnctional language to varying degrees . although the exact profile of language and communiuativc ahil- i ties appears to be somewhat heterogeneous .

176 Understanding, Autism: From Basic Neuroscience ioTreatment

In this chapter. we first summarize the cvidence regarding language abilities in ASD. (Note that we use the terns autism and ASP in the broadest sense, including diagnoses of autism and Asperges'li syndrome.) The data suggest that ASD is assaciatcd with a particular pattern of relatively spared and impaired language functions. We thcn examine in depth two explanatory theories of language in ASD. The two theories are uomplerncn tary in that they focus on different sets o f language functions. However, both theorics take the same broad approach in that they address relation% bctween lanpuagc and nonlanguage domains with a view !o exploring whether similar behavioral profiles across apparently distinct cognitive domains can be explained by cornmon neuracngni tive substrates.

According to one theory, deficits of "theory of mind" in ASD can explain prrrnq~~~nric. impairments of language and communication in zerms of social deficits and their neurocognitive underpinning^.^-^ In contrast, the groccdural deficit hypothesis (PDH) posits that granrmaficd impairments in The disorder - including syntax. morphology, and phonology - can be largely explained by neurocognitive abnar- rnalities of the procedural memory system, whereas lexical knowlcdge, which depends on the declarative memory system. remains relatively s p a r ~ d . ~ . ~ Thus, rather than investigating language deficits in isolation, we examine integrative explanatory tl-leories that attempt to account. Tar these deficits i n the broader context of brain and behavior in ASD.

LANGUAGE AND COMMUNICAT1ON IN ASD: THE EVIDENCE

Here wc first present data related to pragmatic linguistic functions and then turn to evidence pertaining to grammatical and lexical aspects OF language. Formulaic speech in ASD, which may be related to more than one of these domains, i s addrcsried az the end nf this section. Our discussion will concentrate on cross-sectional studies and. by necessity, on those children who have acquired at least same functional language.

Pragmatics concerns the practical knowledge that is necessary to use and interprez language appropriately for the social and real-world contexts in which utterances arc made. Scrcitli aspects of pragmatics crucially include know ledge of the social rules that govern speaker-hearer interactions (which aften involves interpreting a speaker's intended meaning across different social contexts)., whereas red-world aspects of pragmatics include knowledge of people and objects and haw they arc likely to interact (e.g.. knowing that girls are mare likely than boys to play with dolls). Pragmatics encompasses bath verbal and nonverbal aspects of cornrnunica- tion, including gestures. prosodic cues Ei.e., of intonation and rhythm). and facial expressions. all of which combine to enhance effective communication in face-to- face social contexts. A range of impairments consistent with pragmatic deficits are inherent 20 ASD. These impairments are widespread and are found i n both children and adults, with diagnoses ranging from classic autism to Asperger's syndrome.

Language in Autism

Nonverbal Communicative Gesture

Young ASD children present with wcll-documented problcms in nonverbal commu- 1 i iua t ion .~ -~ '9~a ta bascd on parental report indicatc significant delays in the use of early gesti~rcs.~~ Studies of nonverbal intentional gestural cornrnunicatfon i n young ASD children consistently show that pointing gesaurcs that are used to communicate requests arc generally produced and understood, whereas pointing gestures r hat are used to share interest in an abjcct or to direct attention to an event are virtually absent (even though thesc emeye at the same time as requesting gestures in typically developing uhildren).l2

Speech Acts

Speech rrcrs are utterances that serve a oomn~unicntive function. such as requests. comments-. or commands. Importantly. thcy rcquire knowledge of how languagc is ~ll;ed within a c u l t ~ r e . ' ~ Evidence suggests that ASD children arc missing speech acts that emphasize social engagement rather than speech acts that regulate otl~crs' behavior. lJ Weftherby and Prutt inglhxaamined the rangc of speech acts that were expressed by ASB children in both gestural and spokcn languqe at early stages of developmen1 in comparison to Iztnguazc-matched typically developin2 childrcn. They found that the ASD children were not significantly different From rhe controls in their use of Ianguagc for requesting objects or actions, for protests. and for sclf- regulation (c-g.. "Don't do that."). However. speech acts with social Tt~nctions. such as comments. showing off, aacknow ledging the listener, and rcquesti ng information. wcrc completely absent from ASD discome. In another study. AS13 children used declarative sentences that werc direct responses to questions but did not make other typcs of declarative statements or comments. which are thought to involve more 4gnificant social awatene~s. '~ Cornparcd to children with specific language impairment (SLI), ASD children have been found to use fewer affirming or agreeing utterances.I7 Finally. in a study comparing ASD chiIdren to Down syndrome children, rhe ASD children rarely communicated about objects that werc the fwus of their mothers' attenfion.

Conversational Discourse

Deficits in conversnt ional ability arc found throughout childhood and adolcsccncc i n ASDA1"0lder. higher-functioning ASD adolescents are likely to speak too much and in a monologue stylc during i n t e r ~ i e w s . ~ " . ~ They have problems responding adequately to questions. especially when discussing an unusual event or personal narrative," and tend to havc difficulty making clear reference in their conversa- tions to people or places." These latter findings are consistent with evidence of impaired performance on referential communication tasks.2J They also have difficulty judging the nrnount of information that necds to be included for effective communi~ation.'~

Other studies have shown thns people with autism seem less able to shift their discourse when there arc failures in communication -for example. if a lisrener has not hcard or understood an utterance. Paul and C.nhen2Vfound that althou~h adults

1 78 Undcrstancling Autism: From Rasic Neurascieslce to Treatment

with autism werc just as likcly to respond to requcsts for clarification as mentally retarded nd~tlrs matched on nonverbal IQ. their answers were less specific than those c>f the nonautistic participants. They were also less likcly to add information that might be of help to the listener. supgcsting pmblerns in judging thc relevance of ;t piece of information. This finding was recently replicated in high-functioning school- clgc children with autisn~.'~ Finally. we note that several studies have demonstrated abnormalities in autism in othcr kinds of discourse skills, especially in tclling s~ories or recounting personal narrative^.:^-^^'

Pragmatic Functions of Prosody

Pmsnd? describes the timing, rhythm. and intonation of speech. It has numemu5 pragmatic functions, including nongrarnmatical pauses and the use of stress to dircct attention to words or other elements. In a recent survcy of investigatians of prosod! i n-ASDA 11-studieG ten-f-tern)+-whiek-prqmtiettmctiCsmf-prosody could he isolated reported deficits in those fi~nctions:~'

Interpreting NonliteraE Language

Spokcn and wrirtcn discourse often includes different forms of nonliteral language. including idioms. metaphors, lies. jokes, and so forth. To understand nonliteral forms of language, one rnust infer the speaker's inzended meaning (as in jokes or metaphors. tbr example) or understand cultural-linguistic expressions (as in idioms). Rorh of these entail pra~rnatic knowledge of how language is usccl in differen[ socia1 contexts. ASD individuals, cven those whn are older and high functioning. have p a t difliculty interpreting nonliteral or figurative ~peech.~'-~"hesc difficulties include pmblerns with idiomsq3" with metaphor and and wizh the ability to explain nonliteral utterances embcdded in stories (e.g.. Iies. jokes, pretence. irony, sarcasmm, or doluhle bluff) .33.3K31' Using a more stnrclvrcd task. i t has also been found thnr children with autism have difficulty interpreting a speaker's intended meaning in a conversational can text and. unlike rnatchcd controls, they interpret utterances in 3 literal way instcad of in relation to the speaker's stated dc~ire .~"

Language dcpends upon two mental abilirie~?.~' Idiosyncratic information rnust bc memorized in some sort of mental dictionnry. which is oftcn referred to as thc mental levicorr. The lcxicon necessarily includes at1 words with arbitrary sound-meaning pairings. such as the noncompositional ("sirnple"")word ciir. Ruf language also consiszs of regularities. which can he capturcd by rules of gmmrzlrrw. The rulcs constrain how lcxical rorms and othcr basic units in language combine to make complcx representations. including phrases and sentences le.g., the rnr rotlk the frniu In 1.1-ork; syntax), complex words (e.g. , ~vi i lk -t - H I -> ~ ~ l l i e d ; morphnlogy). and the structured sound patterns of words (e.g., t + ~r c p -> 111~7; phonology). Importantly. although complex representat ions (such as the phrase 1I7e c ~ ) could be computed anew each time (tJ~e + car). they could In principle also he stored in the mental lexicon (c.g.. rltp CCII could be stared as a single unit). As we will see. the evidencc

Language in Autism I79

suggcsts that grammatical composition Is impaired in ASD. whereas lexical knowledge remains relatively spared.

Grammatical Abilities

Sylr[t.r.r underlies the rule-governed combination of words into the sequential and hierarchical stnlctures of phtascs and sentences. Impairments of sentence comprehension have been widely reportcd in ASD in both auditory and visual (i.e., reading)

However, not all subjects show these Intriguingly. one study of ASD adults reports normal accuracy but abnormally fast reacfion times in a sentence comprehension task.51

In expressive language li.e.. in speaking), studies have found thar the xpontanc- ous speech of ASD children shows reduced syntactic cornpIexity compared 20 children with Down syndrome or la developrncntal!y delayed or typically devefopin_e control ~hildren.~~-~"irnilarIy, ASD children have been found to omit required closed-class items (e-g.. definite aflicles such as the). which play important grammatical r ~ l e s . ~ ~ . ~ ~ Children with ASD have also been shown to have lower mtes of novel. nanirni tative rlttesances compared to typically developing and Dclwn syndrome children, instead relying more art "formulaic*' utterances (see following text for furthcr discussion on fornulaic s p e ~ c h ) . ~ ~ Finally, ASD chiidrcn have shown impairments on the portion of the Clinical Evaluation of Language Fundarncntals (CELFI that tests immediate scntcnce repetition and reflects (at least in part) cxpres- sive s y n t a ~ . ~ ~ + ~ ~

Morphoir~~qy. which refers to thc structure of words with respect to their mean- ingful parts, comes in two flavors. B~si~pl~fio~.rcrl t?rarpkoS~g?- (the crei~tion of new words; e,g., .role~nt~ip and tolr,qhiw.r.~ are derived from stll~t??n and ~nu,qltl has. to our knowledge. no1 been investisated in ASD. Scveral studies of ASD have examined ityYectin~ta/ ntnrpi~olr>gy. which concerns the modification of a word to fit its grammatical role (c.g., snir<q and walked arc pasf-tense inflected). It i s important to note that inflectional rnarphdogy involves hoth morphosyntax (the choice of inflecxion bnscd on the syntax of a sentence - for example. choosing present tense or past tense. depending on the syntactic context) and morphophonulogy (phonological changes to n word that reflect morphological processes, such as thc vowel change in the irregular past tense formation of d f i ~ from dig). Although marphosyntax strongly depends upon cornbinanorial mle-governed (i.e.. grammatical) processes, this i s not neccssarily the case for marphophonolsgy. Whereas mgillar rnorphopho- nological transforrnatir)ns, ns in English regular past tenses, follow rule-governed composilional patterns (~?tiEk + - ~ ' d ) . irregular morphophonology is at Eciist partly unpredictable (c-g., .s irt~-ssrng, I~r.ir~~-!~rnli,qht). and thcrefnre must rely an stored lexical knowledge,

In two studies that examined samples of spontaneous speech. ASD children omitted inflectional morphemes Ic.g.. they produccd pi(!! instead oFpl( i~l t lg) more csfren than unimpaired or Inentally retarded control s u b j c c ! ~ . ~ ~ . ~ This pattern appeared ro hold particular1 y for certain regularly inflected forms (espccia11 y -ing- suffixed forms, such as pIuyirr~. but also for regular past tenses in the Rartolucci et al, study), whereas irregular past-tenses were relatively spared in hoth studies.

1 80 Understanding Autism: From Basic Neuroscjence toTreatmcnt

Similarly. a more recent study of spontaneous speech reported impairments on third- person singular present tense (e.9,. in the production of r-unsfr~s), which is almost cnrnplctely regular,5"

A recent examination of elicited vcshal morphology (subjects werc asked to produce inflected forms in past tense picture sentence contexts) found impaired production of past tense forms in ASD {regulars and irre~utars werc not distin- g ~ i s h e d ) . ~ ~ Similarly. high rates of nrnissions and incorrect inflections in present tense and (both regular and irregular) pilst tense production (e.g., MJU.Y~ or wusf l in~ for rv~islrts; c~itclz or crrrcJ~irtg for cuught) were found in an elicitation study sf language-impaired ASD children cornparcd to AS13 children without evidcnt language d~ficits."'~ A nmre recent study of elicited past tense production in high- functioning ASD and age-matched typically developing children found normal accuracy scores for rc,nular, irregular. and navel le.g.. pIr,q-pIq~ecfl verbs. Howevcr. response times revealed important differences, between the groups: the production of regularized (stem + -ed c.g.. in crwiked, pkqqg~d, digI:qed) but not irregularized le.g., rluLq. splin~-.~plirnr) brms was ahnormally fast in AS D corn pared to ~ontrofs.~'

In summary, al l six studies of spontaneous or elicited spcech reported abnormalities in the production of inflectional morphology. All but one reported actual impairments. The remaining study. in which ASD children were abnormaIly fast at producing regularized forms. differed from the others in having the highcst-func- tinning ASD subjects (the highest EQs) and in having little or no requirement for social interaction during production. as the items were presented visually on a computer screen (for more discussion of this Iast point see Reference 62). Addition- ally, accuracy or response time differences between ASD and control suh~jects were found in all six studies for regulars. but were not observed for irregulars in three of the four studies that distinguished regular. and irregular morphological forms, including the rrne study in which regular and irregular forms were explicitly well-matched on frequency and other factors." T~LIS. ASD seems to he associated with abnormalities of inflectional mtlrpholagy. particularly for regular forms. although it is not yct clear to what extent this i s duc to problems of regular rnorphophonolagy, rnorpho- syntax, or both.

Pl?nrtolrjxy refers to the sound patterns of language. In ASD, ccrtain aspects of phonology may be relatively prescrued, whereas others are somewhat impaired. Phonology i s concerned both with individual speech sounds (i.e., phonemcs. such as the / l r / sound injitrher) and how ahey arc combined sequentially and hierarchically into syllables and words. Several studies have reported no particular ASD dcficits with individual phonemes in either receptive or expressivc language4' (but see Reference 3 1 : for n review of earlier studies see Reference 63), Impairments in ASD have more often been reported in the cornbinatic~n of sounds into complex structures, Thus, deficits have h e n found in the repetition of auditorily presented nonsense words (c.g.. harrcr~on).l&-~ though the presence of such impairments is less clear in some s t u d i e ~ . ~ ? . ~ ~ Because this task i s posited re involve both disassembling thc input into smaller units (e.g., phonemes or syllables) and then reassembling these units in production, it is expectcd to involve co~npositional processes. The evidence therefore sugzests that at leas1 compositionaI aspects of phonology may be somewhat impaired.

Language in Autism

Behavioral studies suggest that lexical knowledge remains largely notnial in ASD. First of i1I1. word-learning abilities seem to be essentially i n ta~ t .~ ' Second. pertbr- mancc has bcen found to be unimpaired an a range of receplivc lexical tnsks, such

L .1 as word-picture correspondence ("Is this a ... ? ). word-picture matching ("Pick the comct picture to match .. ."I, picture selection ("Show nle all the . . .*'I. and word definition ("What is a . ..?').44..'0"5 Third, expressive lexical abilities seem to remain largely spared in single-word production tasks. such as picture naming. synonyni and antonym generation. and reading single letters or words out l o ~ d . ~ ~ - ~ ~ - ~ ~ ~ ~ When ASD subjects are asked to name picturcs as rapidly as pclmible (so-called "'rapid automatic naming"). a mixcd protilc can bc seen, with some subjects showing normal perfnrmance, whereas others are Similarly. performance on verbal fluency tnsks. in which sub.jects are asked to name as many words as passiblc in n given period of time. seems to be generally, hut not aIways, spared. Norma1 performance has been found in both letter flucncy (e.,o.. "Name as many words as you can that begin with the letter FA") and category fluency (e.g.. "Narnc as many animals as you an,").^-'-^^.^"(^^ Other studics, in contrast. have reported deficits in both types of verbal flucncy [a~kd ' ' . (~ as well as in unconstrained ("rnisccIlaneous") verbal flucncy (e-g., "Say as many words as you can think of, any words at ~ ~ l l , " ) . ~ ~ Finally. as we havc seen earlier, the production of irrcgulclr past-tense forms (which depcnd on memorized lexical knowledge) i s gentrally. but not always. spared. Thus, the evidence suggests that although lexical knowledge itself may remain spared in ASD. ~hcl-e seem to be some deficits in retrieving or searching for this knowledge, perhaps pnrticularly under speeded conditions.

Neuroimaging Studies

There havc been few neuroirnaging studies of Icxicon or grammar in ASD. We arc aware of only two functional neuroirnaging studies of cither domain. bath of' which examined syntactic processin?. (A st~ldy of conceptual processing in ASD i s discussed helow.) In addition. one strlrctt~ral rnaynetic resonance imagine (MRI) exper- iment examined the relation bet wcen brain abnormalities and language i~npairments in ASD. In presenting these three studics. we focus specifically on structures, prcvi- ousTy implicated in language functions (i.e.. wc do not discuss visual areas. the pons etc.). In the section on the PDH. we discuss the potential significi~nce of thcsc structures. specificallly [he relation betwccn frontal. basal ganglia. and ccrchcllar ctructures and procedural memory and grammar nn the one hand and between ternpnmll tcmporo-parietal ~~~~~~~~ex and declarative memory and lexici~l memory on the other.

A functional MRI (fMRT) study of vir;ual sentence comprehension (as compared to visual fixation) found that ASD adults showed greater activation than unimpaired age- and IQ-matched controls in the left posterior supcrior temporal suIcus i betwecn the supcrior and middle temporal gyri. i.e.. between Brodrnann's areas - RA - 22 and 2 1 ). in right temporo-pnrietal/inferior-parietal cortex (BA 39). and in the para- hippocampal gyrus bilaterally: but less activation than controls in the left inferior frontal gyrus I"Rrocaqs region." i.e.. BA 44.45. and 47). rishr inferior frontal gyrus.

182 Understanding Autism: From Basic Neuroscience to Treatment

and left lateral and medial prernotor coraex and nearby frontal cortex, including the SMA/pre-S MA (SMA refers to the supplementary mator area in medial BA 6; pre- SMA lies just anterior to the SMA).-I1

Similarly, a positron emission tomography (PET) study of auditory sentence comprehension (in comparison co silence) reported greater mean regional activation (we do not discuss deactivations. which are difficult to interpret) in adult ASD subjects (who were impaired at an off-line measure of' auditory sentence comprehension) than age-matched healthy controls in the right superior remporal gyrus (RA 22,41 f42) and right inferior parietal cortex (BA 401, but less activation thnn controls in left prernotor cortex (rniddldinferior BA 6 ) as well as in the mid portion li.c., not posterior) of the left middle temporal gyms (BA 21).j7 Additionally, within left temporal cortex, the ASD subjects showed a peak activation in a posterior portion of this region. whereas the control subjects showed peak activations in mid and anterior portions. Similar to the fMRT study. there was also less activation in ASD thnn control subjects in left inferior frontal gyrus (within Bmca's region. i.e., in BA 44,45, and posterior 47), ns we11 as in the left caudate nucleus and the Ieft lentiform nucleus (both of which are part of the basal ganglia), although these differences did not reach significance. In a follow-up reanalysis of a subset of this data?R the authors reported reduced activation in the ASD subjects. compared to the controls. in left dorsolateral prefrontal cortex (BA 46) and in the right dentate nucleus of the cere- belIurn, suggesting abnormalities of frantaE-cerebellar circuitry, in particular between Icft frontal cartcx and the right cerebellum (note that the right cerebellum i s connected primarily to the left cerebrum).

These functional imaging findings are paralleled by findings from a structural MRT study that related neurnanarornical volumes to performance on the CELF (which probes rnultiplc domains of language) and a nonword repetition task.""ASD children were first divided into those who performed within the normal range on these two language tests and those who did nat. These two ASD groups were compared against two age-matched comparison p u p s , one with specific language impairment (SLI) and one composed of' typically developing control children. Both language-impaired groups (ASD with impaired languagc and SLI) had significant right hemisphere asymmetry (i.e.. right larser than left). as compared to the two cornparison groups. in classical Broca's area - that is, the pms tiangularis (BA 45) and pars opercularis (BA 44) of the inferior frontal gyms. Reported volumes from this region in each hemisphere suggest that the increased asy rnrnetry srernmcd from both decreased Ieft herni sphere vvolumc and i~lcreased right hemisphere volume in the two language-impaired p u p s . In contrast. both language-impaired groups showed Ieft hernispherc asymmetry (left larger than right) in the planum temporale (the posterior portion of the upper bank nf the tcmporal lahe). whereas the two comparison groups did not. Analogously to the Eroca*s area asymmetry, this increased leftward asymmetry in the planurn temporale seems to be explained by u combination of decreased righr hemisphere volume and increased left hemisphere volume in the two language-impaired groups.

In sum. in both of the functional neuroimaging studies of receptive syntax, ASD subjects showed greater activation in posteri w superior-temporallinferior-parietal cortex - including the posterior superior temporal gyrus and sulcus and temporo- parielalhnferior-parietal cortex - but Icss in certain left frontal regions, i.e.. left

Language in Autism 183

prernotar cortex. dorsalatera1 prefrontal cortex. and Broca's region, ns well as in left basal ganglia structures and thc right cerebellum (note that only one of the studies addressed subcortical or cerebellar structures). The szructural MRI study found that the languaee-impaired ASD (and SLS) subjects showed a vnlnrnetric decrease in left IYontal regions. especially in Broca's area, but an increase in a left temporal lobe region - an intriguingly similar pattern to the patterns of increased and decreased ;tctiuatinn found in rhc two functional neutoimaging studies. Intcrpretc?tic>ns of these findings are discussed below.

ASD specch is marked by several striking features: repetitive and stereotyped utter- anccs [e.g., overuse of mutine utterances such as ''thank you" lor 'you're weIcorneW): idiosyncratic sound-meaning associations (i.e., "metaphorical language" such as "I want to go blue'' to cxpress a desire to go outside): excessive1 y literal Ianguage (e.g.. responding *'No. it's raining watei' to the statement "It's raining cats and dog''): difficulty with pronouns le.g., a child with autism may say "Would you like a cookieL?* to request one) and with other deictic terms (i.e., terms whosc reference depends on contextual factors. such as this, ISFCII. JTPYP, tftere, etc.): and immediate or delayed echolalia (lexically. prnsodically. an8 syntactically faithful repetitior~s).~~"~."'"

These various aspects of spcech are highly characteristic of ASD and feature among the diagnostic crltcria irt DSM-IV for qualitative communicaticm impairments in the disorder.'They are observed at much higher rates in ASD than in typically developing children (for n review, see Reference 63.) or children with athcr developrncntal disorders, including SLJ7' and Down's

These aspects of au t istio spcech arc often described together aqfr,,nr~rlrric s p ~ e u l ~ . 57-7' F o ~ ~ ~ L ~ I c I s arc defined as prefabricated sequences of words 1 hat are stored and retrieved whole from memory (in any population. impaired or ~nimpaired) .~ ) Tt has bcen suggested that at least certain aspccXs of formulaic speech in ASD (e-g., echolalia and an ovcmeliance on a restricted range of formulaic forms) may reflcct pragmatic or social deficits i n language ~zse . ' ' ~ .~ ' I t has also been suggested that formulas o k r a shortc~~t (via memorization) by which ro bypass gramn~atical processing, pnrzicularly when grammatical processing is difficult.73 Thus the overreli- ance in ASD on these forms may reflect more than one underlying deficit.

INTEGRATIVE THEORIES OF LANGUAGE IN ASD

Here we prescnt two explana~ory neurocogni tive theories of language in ASD. Both thcories integrate iindings from language and nonlanguage dornnins. One focuses an pragmatics and thc ather on lexicon and grammar,

As we have secn earlier. ASD involves pervasive impairments in the pragmatic aspects of language use. The disorder is also strongly associated wizh nonverbal social d e f i ~ i t s . ~ ~ . ~ ~ The rlrenry qj'r~irld h?;nntlresis c?#' srrr!f.r~n has been proposed to

184 Unclerstanding Autism: From Basic Nei~roscience lo Trratment

integrate dcficits across these According to this hypothesis, people with ASD are fundamentally impaired at causally l inking their own and other people's behavior to mental states. Thcse deficits of theory of mind are posited to underlic impairments both of pragmatics and of nonverbal social abilitie~.~?

The evidence suggests that individuals with ASD have (henry of mind impairments. as indicated by deficilts in tasks that zap their understanding of minds and mentat states such as bclicf (e.g., false 'belief). knowledge. and c r n o t i ~ n . ~ ~ ~ " ' Cru- cially. these impairments appear to be linked not only to social deficits in ASD75 but also to pragmatic deficits..'' Several lines of evidence support an association between impairments of thcory of mind and of pragmatics.

Fir-st of all. whercas ASD is associated with in~pairrnents of pragmatics that involve viewing penplc as mental hein_es, certain other aspects of language use that do not rcquire knowledge of people's mental states tend to remain relatively spared - such as tum-taking skills. requesting hehaviar, or regulatory speech acls (sce preceding tcxt ).

A second line of cvidcnce comes fiom studies of joint attention in young children. Joint attention invoives the triadic engagement between a child. another person, and an object or event of interest. It has bccn vicwed as an early thcory of mind ability in that j t demonstrates !he child's capacity to manitor and manipulate 'the attentional focus of another person. Deficits in joint attention are evident in looking behavior, communicative pointing. and sharing inkrest wit11 another person.n-'" In~pairmentx in joint attention may cxplain why language is delayed in ASD. as joint attention ski1 Is are important developmcntal predictors of the rate of language development in the ~Iisnrdcr.~"

Third, clcficits in conversational discourse in ASD seem to reflect problems in understanding that communication is both about exchanging informalion with others and about the expression and interpretation of the speaker's intended ~ncaruinp .~~ .~~ For example, Capps et al."'fotund a positive cornlation betwcen performance on theory of mind tasks and the ability to maintain an ongoing topic of conversation among children with autism, SimiIarly. Hale and Tager-Flusherg" found an inverse relationship betwcen performance at theory of mind tasks and the frequency of non-topic related ~rttcrances. a relationship that was independent of overall language ability. Difficulties with rlnnlitcral language also seem to stem from impairments with theory of mind problems, in particular with understanding intentional aspects of communication. For example. Happe's studies have demonstrated a direct relationship between thcory ol' mind perfimnance and the ability to interpret nonli teral mcnning in pcoplc with E ~ u z ~ s ~ . ~ ' . ' . '

Finally. i t is interesting to note that functional neuroirnaging studies in ASD suggest abnormalities in thc neural structures that underlie the processing of theory of mind. Several bmin structures have been implicaled in typically dcvcloping individuals in theory of mind processing. in particular tnedial frontal cortex (RA 819. bardcring on the cirigulate gyms). Interal inferior frontal cortex (primarily Rrosa's area. namely BA 44/45), and posterior wperior temport?Iltemporo-parietal cortc~.~?-~. ' Intri~uingly, the small number of nevrnimaging studies of theory of mind in ASD suggest n tendency for decreased activation (rclaitive to controls) in both of these frontal regions and increased activation (relative to con~rols) in the temporal


This pattern of decreased activation in Broca's area and increased activation in posterior superior temporaI/tempom-parietal cortex is strikingly similar to that found in the functional neuroimaging studies of language discussed earlicr, It remains to be seen whether similar patterns are Found in the pt.ncessing of pragmatic aspects of language.

GRAMMAR, LEXICON, AND THE PDH

As we have seen in the pmceding text, ASD i s associated with impairments of comhinatorial aspects af prammas. across syntax, morphology. and phonology. According ta the procedural deficit hypothesis (PDH) of autism, thesc languagc impairments are explained by abnormalities of the brain structures that underlie the prncm'rrmi rnemolp .vysrnn, resulting in impairments of the particular language and nanlanguage functions [hat depend on thesc structures. For a discussion of the PDH with respect to a number of different disorders, including autism. SLl. attention deficit hyperactivity disorder IADHD), and dysiexia. scc Reference 6. For an in- depth examination of the hypothesis as it applies tn SLl, see Reference 7. For a related hypothesis of procedural learninp deficits in ASD. see Reference 88.

The procedural rncrnery systcm (for the sake of simplicity. we also refer to the system as the "procedural system'.) is implicated in [he learning of new. and the control of long-established, motor and co~nitive "skills." "habits." and other pracc- dures, such as typing. riding a bicycle. or skilled game playing.*''-''3 The system is particularly important for acquiring and performing skills involving s e q ~ e n c e s ~ ~ ~ ' ~ ~ and has becn shown to underlie nonlinguistic rule Iearr~ing."~-~-' Evidence also sup- scsts that the system subserves aspects of the teaming and use of grammatical rule- governed combination. across syntax. morpholo_gy. and phonology. in both receptive and expressive lang~age.~.~. '

The procedural systc~n is composed of a network of several interconnected brain structures. It depends especially on structures in the left hemisphere of the cere- h r ~ r n . ~ ~ " T r a is routed in neural circuits that encompass the frontal. hhes and the basal ganglia (a set of subcortical structures that include the caudate nucleus and the pufamcn, which together constitute the ncostrinturn). Within frontal cortex, two areas play particularly important roles: prernotor areas (espcci all y the reginn of the supplementary motor area [SMA) and pre-SMA) and Bsnca's area. Othcr brain structures also f o m part of the procedural system network, including portions of both inferior parietal cortex and the cerebellum ( inc lud ing the dentate n u ~ l e u s ) . ~ ~ ~ ~ + ~ ' ~ ~ ~ ~ ' ' Finally. [he procedural sy s tern appears to be doscly rcl ated to the so-called "dorsal" stream path~ay.""~'

Thc PDH of autism posits ahat ASD i s associated with neural abnormalities of the procedural memory system, rest~lting In impairments of the various functions that depend an this system. Thus. one should expect motor and cognitive deficits. especially those involving sequencing. I n the linguistic domain, one shorlld obscrve deficits of grammnr, across syntax. morphology and phonology. One should also find impairments of any other functions that depend on the brain structures that undcrlie the procedural system. such as aspects of temporal processing and working memory {whether or not these functions are related to procedural memory).'

186 Undcrstandinp, Aulism: Frnm Basic Netrrascience to Treatmen:

Thus the PDH is a hrf~in-based hypotlhcsis. That is, unlike many other account> of language deficits in autism ar other deveIopmenta1 disorders, the PDH i s based on thc premise that that certain brain structures are abnormal, and that the functions that dcpend en these strucFures are therefore likely to be impaired.

The nature of the observed impairments is expected to vary with thc precise nature of the neural dysfunction, potentially leading to substantial heterogeneity in the linguistic and nonlingwistic impairments in ASD. First of all. because each brain structure in the proccdurnl system plays a somewhat different functional role. the dysfunction of different structures should lead to different typcs of impairment. For exzlmplc. abnormalities of structures thought ta underlie the acquisition of procedl~rril knowledge. such as the basal ganglia. should yield different behavioral phenotype\ than abnarma! i t ies of structures ahat may subsewe the execution of procedural skills. which might be the case for certain frontal regions.[(" We expect that a portio~~ of the hetera_reneity of language (and other cognitive functions) in ASD may bc explained by variation across individuals as to which structures are affected and t(7

what degree. Just as abnormalities of different structures within thc procedural memory sys-

tem should lead to behavioral heterogeneity, so should the rlysfunction of diffcrent portions of structures. - espcciallly of those stnlctures that constitute frontallbasal canzlia and frontal/ccrebetlar circuits. These circuits are composed uf parallel and L

larscly functionally segregated "channels" (also referred to as For exani- ple. in the bmal ganglia, cach channcl receives projections at the neosttiatum - some channejs primarily at the caudate and others at the putamen - from n particular set of cortical and subcortical structures. Each chnnneI then follows the same set of internal connections within the basal ganglia and then projects outward via the thalamus to a particular cortical region (from which it also rcoeives input). primaril? in frontal cortex. A somewhat analogous architecture can be found in the circuitt-! projecting from the cerebellum via the thalamus to fron'tal cortcx. Each of the liontall basal ganglia and frontal / cerebclIar channels r~nderlies functions associated with the cortical rcgion to which it projects. For examplc. thc frontallbasal ganglia channel projecting to the primary motor cortcx subserves motor functions.

Dysfunction in a given stmctlrr~ in this circuitry (e.g., in the neostriatum) i5

unlikcly to bc limited to a single channel. Rathcr, evidence from neumdegenerative and neurodevelopmentai d isordcrs suggests a rendency for dysfunction to co-occur across multiple c h a n n c I ~ . ~ ~ ~ ~ = " ~ Howevcr. it is unlikely that exactly the same channel\ will 'be affected in all individuals with a particular disorder. Therefore abnormalities in ~hese structures in ASD should result in variability across individuals wifh respect to the comhinatiun of channels thaz are affected and the severity of their dysfunction. Ncvcriheless. evidence from other disorders suggests that within a given dil;orcIcr. certain channels should be mare likely than others to be dysf~nctional.~"~ So. although variability with regard ta which channels are affected should lead to some functional heterogeneity across subjects. an important degree of similarizy among individuals wilh the disorder is ro he expected.

Additionally. different types of deficits should be associated with the dysfunction of different pathways within these circuits, in particular with differential dysfunction of the "direct" and "indirect'" pathways wizhin the basal: ganglia. These two pathways

Languay in Autism 187

have opposing effecrs on the basal ganglia's influence on frontal cortcx. Via rr series of inhibitory and cxcitatary projections, the direct pathway ~~ltirnately disinhibits rrontal cortical activity. whereas the indirect pathway ultimately inhibits it. t~nbal- ances between the two pathways can lead to the excessive inhibition or dixinhihition of funulions that depend on the fmnral cortical regions 10 which the basal ganglia p r t l j ~ c t . l ~ ~ - " ~ This i s thought to explain thc inhibit~dlx~ppr~sscd ("hypo") and disinhi hi ted/unsuppressed ("hyper") motnr. grammar. and or her behaviors fourrd in various neurodevelopmental as well as ncurodegenerative disorders affecting the basal ganglia. including Parkinson's disease. Huntington's disease. Tourette syndrome. obsessive-compulsive disorder (OCD) and ADHD.Mu.lrfi-F."+loX For e x 1 , mplc. Huntington's disease paticnts show unsuppressihIe grammatical rule use (e-g.. in atlixation, saying wlkrnd~d and dtrg,qod) as well as unsupprcssible movements. whereas Parkinson's disease patients show thc suppression of both grammatical rule use and of movement .H('K

Given thc highly plastic nature oi'the developing brain, compensation i s likely to occur. It has bcen shown that the func~irrns of abnormal neural tissue can be taken over by similar or proximate intact ti~suc.~'" Thus, abnormalities of specific portions of' the striatrim or frontal cortex may be compensated for by other portions of these structures in thc same. or perhaps even the contralatcsal. hemisphere. Tn addition, if a function can be performed by more than one csmpura~ional mechanism. il could be taken over by a brain structure whose type of computation i s distinct from that of' the abnormal tissue. Indeed. we pasit that the dcu/u~-t tr i~v~ c.rfzerm-)ty sTrfuar can and wi l l take over certain grammatical l'i~nct ions from the abnormal procedural memory system.

The declarative memory system normally subserves the long-term Icarnf ng, representation. and use of knowfcdgc about facts (conceptual-semantic knowIedge, i.e.. "semantic memory") and personally expcricnccd events ("episodic mern- ~ r y ' ' ) . ~ [ ] . ~ ~ I The knowledge learned in this system i s at lcast partly (but not cam- plctely) explicit - that is, available ta conscious a ~ a r e n e s s . ~ " - ~ l ~ Media1 temporal structures (including the hippocampus and ncarhy regions such as the pamhippoc- ampal gyrus) consolidate new memories. which eventually depend largely on neo- cortical regions, purticularjy in the temporal l o b e ~ . ~ ~ ~ . ~ ~ . I ~ ~ - l ~ ~ Other brain structtlres rilso play a rolc in declarative memory. including portions of Brwa's region (RA 44,45, and 47) and frontal polar cortex (BA 10). which seem to underlie aspects ol' the sercction o r setriev;tl 01' dedarative and he right cerebellum. which may underlie searching for declarative knawlcdge.lw Importantly, some al' these other structures also constitute part of thc procedural memory systcm. suggesting that prt~cedural system abnormalities may be expected to resull in impair- iiients of lexical search or rctricval - though no[ necessarily of the acqalisition and c>rganization uf lexical k n ~ w l e d g e . ~ Finally. evidence sug2ests that the decl;lri~tii~e memoly system subserves the learning and use not only of fact and event knowledge but also of lcxisal knowlcdgc. Middle and inferior aspects of lthc tcmporal lobe rnnj be particularly important for storing word meanings. whereas supcrior temporal and rernporo-parietal regions tnay he more important for storing phonolopical tvc.ord forms, and possibly also for stored morphological and syntactic structures le.g.. in formulaic l'"

188 Undcrstanding Autism: From Basic Ncuroscicnce to Trea t m w .

The declarative and procedural memory systems do nat function ccsrnpletcl~ independently from each other. Rather, evidence suggests that they interact. yielding both cooperativc and cornpet itive learning and pr~cessing.~~~~'~'~ First, the tu c +

systems can camplemenr each other in acquiring the same or analogous knowledge. including knowledge of sequences. Second, animal and human studies suggest that the two systems also interact competitively. This leads to a "seesaw effe~t.''~ such that a dysfunction of one system results in an enhancement of the other. or rhar learning in one system depresses the function of the other.

We posit that in ASD the declarative memory syslern will tend to take over certain grammatical functions from the dysfunctional procedural memory system. In particular, complex structures that can be composed by the grammaticalJprrsce- dural system (e.g.. ~tw/rEX- + -ed) in normally developing individuals may simply br srorcd ns chunks (e.g., ~ d k e d ) in lexicalldeclarative memory in individuals wit11

I

ASD, Structures that are easier to memorize should be morc likely to be stored. Thus. forms that arc of higher frequency, shorter, and perhaps less complex should be particularly likely to be rnernorixed. Moreover, ASD individuals should bc able to compensate for their grammatical deficits by learning explicit rules in declarative memory. such as "add -ed to the end of the verb when the event has alread> happened." Such increases in reliance on the lexical/declarative memory system in ASD should also be reflected in measures at the level of the brain, such as activation changes (compared to controls) in lexical/declarative memory bmin structures in functional neuroirmaging studies and possibly even in changes in the neuroanatorn! of these structures ( e . ~ . . in ltheir volumes or areas). Indeed. both behavioral and brain evidcnce suggests that such declarative-memory compensation takes place; in other populations who appear to be afflicted with a grarnmaticalJpr~cedural dysfunction. including children with S t 1 and agrammatic aphasics."?

Finally. such compensation should depend on the extent to which declarative memory abilities remain intact and should vary with respect to lthe functionality ol' this system. Thus. whew declarative memory i s dysfunctional. such compensation should be less available, In the cxtreme, a Tack of such compensation in the conrext of a highly dysfunctional procedural memory system wauild be expected to result in a virtual absence of language. However, we suggest thalt declarative, memory is often (though not necessarily) largcly spared in ASD. resulting in a relative sparing of lexical knowIcdge, though the retrieval or search of this knowledge may tend to bc problematic (see preceding text).

In the fo'ollowing two sections we provide brief overviews of the status i n ASD oi' language and other functions that depend on the brain structures that underlie the procedural atad declardtive memory systems.

ASD Profife of Procedural System Functions

Grammar As we have seen, rufc-governed cornposiltional aspects of grammar seem to be largely abnormal in ASD. This pattern holds across language domains. from syntax to morphology to phonology. Tn contrast, aspects of knowledge ss processing i n these domains that do not as clearly involve composition, such as irregularly


inflected forms or individual phonemes, seem to remain relatively spared. These contrasting patterns support the prediction of the PDM that compasitional aspects of grammar. which are pclsired to depend on the procedural system, should be impaired i n ASD.

Evidcnce also suggests that ASD individuals compensate for these impairments by relying on IexicalJdeclarative memory, The widespread depcndencc on fosmu- laic speech indicates a dependence on the use of rnemorizcd complex representations. even of structures that would normally be composed in It i s important to point out that even ASD lanpage that appears to be normal may actually depend mare heavily on lexicafldeclarative memory than it does in typically developing individuals. Although such a dependence tnay be difficult to dctect in spontaneous language samples or in many standardized tests, it can he revealed by a variety of behavioral rncasures or analyses (e.g.. "frequency effcctv' cotrelations between the forms' frequency of use and production or recognition times, susgesting storage of these forms) as well as by ncuroirnaging ~echniques.' Indeed, we havc seen that neuroimnging studies of grammatical processing lind not only reduced activation and volumes in procedural systcm brain structures but also increased activation and volumes in certain declarative memory structures. which may reflect an increased dependence on these structures due to compensation and the seesaw effect.

The two recent investigations that seportcd bozh normal accuracy and faster than normal _emmmatical processing in syntax and regular morphology are highly intrigu- ing.-5i-h1 It is not yet dear what to make of these findings. One possibility i s that thcse sub-jects are highly successftil at compensating for compositiona1 deficits with the lexical/dcclarative memory system. The increased temporal lobe activation in the Jusr et al. study is consistent with this view. Alternatively, ~hese ASD subjects may possess a disinhibited /unsuppressed (hyper) profile. leading to particular!y rapid responses. whereas sub+jccts in orher studies may instead display an inhihited/suppressed (hypo) profile. leadin2 to impaired accuracy. Indeed, as we will see, both profiles seem to be found in ASD in certain nonlinguistic domains. A hyper profile is consistent with u different interpretation o f the underactivation observed in prn- cedural system brain structures during grammatical processing. namely that this reflects panicularly efficient racher than deficient processing. Interestingly. the ASD subjects were high functioning in both of the investigations that rcported fast processing, suggesting the possibility thar such fast processing might bc at least somewhat associated with high-functioning ASD.12J Whcther or not such an association turns out ta hold, i t is important to note that very few languaze studies. of ASD report response ;times (we are not aware of any apart from the two discussed here), sug~esting the possibility that such fast perfnrmance i s actually not uncommon.

Non jinguistic Procedurai System Functions Autism is strongly associated with a number o f motor and nonmotor deficits that suggest abnormalities of thc brain structures of the procedural systcm. First of all, rhc processing. as well as the acquisilion of both verbal and nonverbal nonmotar sequences. has been reported to be abnormal in ASD. particularly for complex

190 Wnderstnnding Autism: From Basic Neuroscience to Treatment

sequence^.^-^^-"^ For example. abnarmalitics have been reporrcd in the processing (production or recall) of hierarchically structured sequences, with relative sparing of linear, repetitive s e q ~ c n c e s . ~ ~ ~ . ~ ~ ~ This finding i s particularly relevam to claims of grammatical impairments because grammar (and especially complex syntax) depends an the hierarchical rather than simply linear arrangement of smaller units into larger oncs Ce.g.. the hierarchical cornhination of words to form phrases and sentences in s y n t a ~ ) . " ~ . " ~ ~ I n addition to these findings concern Eng the processing of sequences, the one published study we are aware of that examined nonmotor sequence learning in a serial reaction time task found no evidence of learning in high-f~~nctianing ASD children and adolescents. whcreas age- and IQ-matched typically developing control subjects indeed learned the sequence over the course of the task.xx

A Jary number of studies have shown the existence of' motor iri~pairrnents in ASD (for reviews, see Reference 124 and Reference 130). Indeed. thc prevalence of such findings has led to the suggestion that motor dysfunction in autism should hc considered a core feature of the d i~arder .~-" .{~~ Consistent with procedural memory abnorrnalities, deficits seem to bc much worse for performing complex sequential motor skills than simple motor acts such as finger tappingu A large number of studies havc reported impairments in the pantomime and imitation of complex actions (for a review. see Reference 129). One particularly comprehensive study found that pantomiming acrions (e.2., "Show how you would use a toothhrtlsh:') was highdy impaired, especially Ibs sequential as compared to single actions: imi- tating actions was somewhat less impaired (but again. more so for sequential than single actions); and actual object uric did not show deficits at all.T37 This is exactly the pattern seen in idearnotor aprdxia. which i s linked to left SMA. basal ganglia. and inferior parietal structures.*h Finally. i t has also been observed that individuals with ASD have dificulty teaming complex sequential motor skills such as dancing or skipping.

Other motor deficits in ASD are suggestive of abnormalities of specific brain structures of the procedural system. Impairments of balance and other motor-rclated functions linked tn the cerebellum have often becn reported in ASD.1y-'.1fA135 Still other motor impairments suggest basal ganglia abnormalities and seem to indicate the existence of both hypo and hyper deficits in the disorder. On the one hand. a rlumber of studies have reported hypokinetic (i-e.. bradykinetic, or slow) movements similar to those of Parkinson's patient^.^?^.'^" On the other hand, ASD is alw strongly associated with unsuppressed motor activity. such as motor (and vocal) tics and stcreotypies (repetitive movements or bchavior~).'.'~ which have been linked [a basal ganglia abnormalilies not only in devclopmcntal disorders such as Tourette ~yndrorne'~~.'~"bu! also in ASD it~eff'-"'~'~l (but see Reference 142). Hypcrkinetic (choreiform) movements such as isolated jerking of the extremities. which arc characteristic of Huntington's diseasc, have also been obsenrcd in ASD.[j5 Interesr- ingly, in this last study, all tcn subjects showed hyperkinetic movements. whereas none showed movement impairments typical of Parkinson's disease. Thus. there appears to be at least some separation between ASD p u p s [hat show hypo and hyper behaviors. Consistent with such a separation, Mari et al.124 found that low- functioning ASD individuals showed bradykinetic; movements similar to thosc found

Languagti in Autism 191

in Pnrkinsan's disease, whereas high-functioning individuals instcad showcd abnor- ~nally fast movements relative to normal controls. This lends credence to the hypoth- tsis ~liscussed earlier that hypo and hyper profjles may also he found in ASD in thc domain of language.

Finally. other functions assaciatcd with the brain stnlctvrex of the procedural system have also bccn found to be impaired. (Fur a discussion of thc relationship of the following functions to prc)sedural mcmury. see Refercnce 7.) Thus, investi- _cations of ASD have reported deficits of rapid tcmporal processing1J3 as well as impaimencs in the estimation of stimulus durationlU I but see Reference 88). Addi- tionaIly. working memory has been found to be impaired in ASD in some s t ~ t d i e s ~ ~ ~ . ' ~ ~ (but see Reference 62). Thcrc have also been rcports of deficits in ASD of rnotion processing, a dorsal stream T u n c t i ~ n . ~ ~ ~ . ~ ~ ~ Indeed, one studyiJf found deficits in motion processing, hut not in form processing, which depcnds on the ventral stream (which is closely relaled to declarative memoryh). Further examination in ASD of these and other functions that may depend on procedural system brain structures weems warranted.

ASD Profile of the Declarative Memory System

Lexical Memory As we havc seen. evidence suggests that word learnins and lexical knowledge remain largely normal in ASD. For example, individuals with the disorder show intact performance at receptive lexical processing tasks. Similarly. expressive lexical abilities seem ro be spared in single-word production tasks. though impairments have bccn found in both rapid naming and verbal ffucncy tasks, suggesting deficits in lexical retrieval or search, especially undcr speeded conditions. This profile of lexical abilities is consisicnt with a relatively nurrnal lexical memory. accompanied by abnorrnitIirtics to brain structures such as frontal and cerebellar regions that underlie lexical search and retrieval as well as aspects of rapid

Con cep tua l Kno wl'cdgc Conceptual knowledge appears to bc larsely spared in ASD. T h i s seems to hold for both individual word meanings and their categorical o r g a n i z a t i ~ n . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ For cxam- plc, ASD children have been fot~nd to show a normal pattcrn of prototypicality racings of members of numerous categories at both basic and superordinate levels. such as chairs. furniture, dogs. and animals.'") Additionally, scrnantic priming (e.g.. J Z I ~ W P will be responded ta faster or more accurately following docmr than following rsrhk due to semantic associations between c/ocrnr and nlr,se) has been found to be r1orrna1.I~~ suggesting intact conceptual-semantic representations in ASD. Another study reported normal interference in a Stroop task with a range of word categories. including not only color names but also concrete and abstract words.l*' Similariy. in a word-picture matching task. ASD subjects did not differ from con!rols on either concrete or abstract wards.u1

Howevcr. not all canccptunl categories seem to be sparcd in ASD. Specifically. ASD individuals have been found to show impairments at processing words that are

f 92 Understanding Autism: From Basic Neuroscicncc to Tres t men t

related to mental or emotiona1 (but see Reference 154). Importantly. thesc impairments may hc explained in terms of deficits of emotion or theory nf rnir~d.~.~~~ Indeed. ASD dcficits in the conceptual knowledge of verbs that are related to mental states ("cognition"' verbs such as knob: think) were found to be related to measures of theory of mind (but not to grammatical pcsf~rmance).~

We are aware of only one f'unctional neuroimaging study of conceptual processing in ASD.IFh This FMRl study examined conceptual-semantic processing ("Judge if a word is positive or negative:') as compared to perccptua! processing ("Judge if a ward is upper or Tower case."). In this, comparison, control subjects robustly activated regions in the left RA 45, left EA 47 [and, borderline significantly. right BA 47), and left medial frontal cortex (two clusters of activation. one on the border of BA 6 and BA 8 - i.e.. pre-SMA - and one morc anterior i n BA 8). ns well as in the right cerebellum. In contrast. ASD suhjcct.c showed only minimal frontal activation. with a small clustcr in left BA 47. They showed no activation in the cerebellum or in any other frontal region. including left BA 45. medial frontal cartex, or right BA 47. Moreover, and unlike the controls. they activated thc posterior portion of the superior temporal sulcudrniddFe temporal gyrus. This is strikingly similar to the neuraimaging pattcrns discussed earlier and reinforces the view that the temporal lobe structures of lexicalldeclarative memory may not only be spared, hut in fact might be relied an to a greater degree than in typically developing subjects. even for conceptual prwcssing.

Learning in Declc-rative Mcmory An increasing body of evidence suggests that the learning of verbal and nonverbal knowledge in declaralive memory ( i.e.. the acquisition nf know ledgc in semantic memory) i s essentially spared in ASD.c5-1JS-157.1-iX (%sks probing immediate rccal l or recognition will not be discussed herc as they may reflect processing in short- term or working memory rather ~ h a n learning in declarative memory, which can perhaps be most clear1 y ascertained with rccall or recasni tion afier a delay: we only discuss such post-delay performance here.) First of all, numerous studies have sugeested that lcarning in "rote memory" (i,c,, memorizing individual items such as telephone numbers or addresses) is a strcngrh in ASD.1J5+1s'+1wJ Second. ASD subjects show normal performance on tasks of paired-associate Ienrning (c.g., presenting a pair of wards that are studied together. then later prompting with the fin1 word of the pair and asking the subject to recall the second ~ t l r d ) . ~ - " ~ ~ ~ ~ ~ ~ Third, normal performance is also observed on cued recall measures after a delay (e.p.. after thc presentation of a list of words, subjects can be successfully cued with the initial sounds of the word, such a s f r ro cue the rccaI1 of.frltit).thl-In>

However. evidence also sirggests the existence of episodic memory impairments in ASD.125-'Jk85K11r-l Thus. memory for recent. personally experienced events is con- sistcntly reported to be impaired in ASD relative to ~ o n t r o l s . ~ ~ ~ - ~ ~ ~ The hasis of this episodic memory impairment in the cantext of spared semantic memory is not yet clear, although i t has been suggested that these impairments may reflect the particular depencience of episodic memory on frontal Note that in any case such impairments would not be expected to impact the acquisition of lexical knowledge. which should depend on semantic rather than episodic memory.

Languagc in Autism

The Neurobiology of Procedural and Declarative Memory Brain Structures in ASD

Although studies of the neurobioIogy of procedural and declarative rncrnary brain structures have produced quite a few inconsistent results, some patterns arc besinning to crnerge (for overviews of these and athcr brain structurcs in ASD. see References 169, 170. and 17 1 ). Of particular interest here are abnormalities of left frontal cortex. and cspecialIy Broca's area, which have consistently been found in s~udies that have examined this region.'"-17" Inrerestinply, neuroanatornical abnormalities of frontal-cerebellar circuit~y have also been implicated in several recent rcports. 175-v7

Moreover. postmortem studies rcliabl y report reduced numbers of Purkf nje cells i n the cerebellaim. 17'~everthcless. results from experinren ts probing the cerebeFlum with other methods. such as structural MRI. have been more variable (for a rcccnt review. see Refercncc 131 ). Studies of other brain structures of the procedural and declarative memory systems, including the hippocampus and the basal ganglia, have also produced somewhat inconsistent results. with some (hut not other) studics suggesting abnormalities (see reviews mentioned earlier and discussion in Rct'crence 142). (Notc that the hippocampus was not examined in any of the Inn- guage-relrrtcd imaging studies discussed earlier. and the basal _ganglia were only examined in one of'thcse studies.) Further investigations arc clearly needed to clarify thc inconsistencies in these (and other) structures and to examine the funolional conscqucnces for language of any observed structural abnormalities.

SUMMARY AND CONCLUSION

In this chapter we have argued that language and cnmmunicntive impairments i n autism can he bcttcr understooci in light of integrative explanatory frameworks that examine h e s e deficits in thc broader context nf brain and behavior in ASD. First, impairments in pragmatic Eanguagc abilities arc argued to be sclated to theory rmf mind impairments. Second, we show that ASD is associated with impairments of compositioni~l aspects of grammar, across linguistic domains. and argue that tllesc impairments are related to abnormalities of brain structures of the procedz~rnl memory system, including at least Broca's area. In contrast. lexical and declarative (especially semantic) memory appear to he il relative strength, and evidence suzgests that these capacities may he used to cornpensale for deficient _eramrnatical processing in ASD,

The two classes of language and cnrnmunicatian impairments (i.e.. of pragnsatics and grammar) have hecn presented as independent deficits. Mowevcr, possible rela- tions betwcen the two havc yet to be explored. Of particular interest is evidence sug~esting that frontal cortex. including Broca's area. is irnp1icated in both theory of mind and grammaticallproced~~ral functions. Intriguingly, it has been sugsested that theory of mind depends an the dorsal stream." suggesting another potential neumanatornical link between the two domains. Thus. i t may be worthwhile to explore the possibility that pragmatic and grammatical deficits in ASD result from distinct functional deficits. which nevertheless both ultimately depend on the same or related underlying brain structures.

194 Understanding Autism: From Rasic Neuroscience to Treatment

In conclusion, autism i s associated with a particular profile of impaired (pragmatic. grammatical) and spared (lexical) languase abilities. The status of these abilities in ASD can he at least paslly explained in terns of their dependence on neurocogni ti ve substrates that also subscrve non language functions. speci fical 1 y theory of mind (pragmatics), procedural memory (grammar), and declarative memory [lexiconS. Whether or haw the neurocngnitive abnonnarities underlying theory of mind and procedural memory are related to each other or to other abnormalities in ASD (c.g.. underconnectivity. 5t . '7X weak central ~oherence,~~%or impaired exec- utive remains to be exiimined. Overall. the cxplanat~ry theories presented hcre not only prnvide wide-ranging accounts of linguistic and nonlinguistic data but also generate new questions and new directions for research.

ACKNOWLEDGMENTS

Support was provided to MTU by NSF SRR-9905273, NTH ROI MH58 189. and NIH RO I HD049347, and to MTU and MW by research grants from the National Alliancc for Autism Research and the Mabel FIwy Trust. Support was provided ro HTF by U19 DC 03610. which is pal7 of the NICHQ/NIDCD funded CPEA. and US4 MH 66398. which is part of the NIH Funded STAART Centers. and a grant from the Nancy Lurie Marks Fmi Jy Foundation. Please address correspondence ti1

Michael T, Ul lman ([email protected] ).

REFERENCES

I . APA. DSM-iV. 4th ed., American Psychiatric Association. Washington, D.C.. 1994.

2. Lord. C.. Risi. S.. ant1 Pickles. A.. Tra.jectory or language development. in De~lell~p- nrerrtrrl k~~r,qrirt,qr Discrtr1cr:r. Eds.. Rice, M.L. and Warren. S.E, Lawrence Erlbaum Associates. Mahwnh. NJ. 2004. pp. 7-29.

3. Baron-Cnhcn, S.. Social and pragmatic defici~s i n autism: cognitive or al'fecdve:'. Jrrtrrrrrrl cfArtrian o~rd Dtvelr~pntc~~rol Dist~rui~,a 18. 379402, 1988.

4. Tager-Flusherg. H.. What Iicnguage revcals abuut thc understanding of minds in chi ldren with autism. in Utdl~rsrtr~j~Ir'r~g OfIwr Mitrdx: Pt~r.cp~crizw.t$r~m Alrrisn!. Eds.. Raron-Coben, S.. Tnser-Flusherg, H.. and Cohen. D.J ., Oxford University Press. Oxford. 1493. pp. 138-157.

5 . Taper-Flustrer2. H.. L;~ngua_ee and understanding minds: connections in autism. in Unrlersrcr~tdinii: Orhcs Minds, 2nd ed., Eds.. Rraron-Cohen, S.. Tager-Flusberg. El.. and Cahen. D.9.. Oxford Uniwssiry Press. New plork. 2000, pp. 124-149.

b. Ullman. M.T.. Crmtributions of memory circuits to language: the declarativc/proce- dtrral modcl. Co~:f~iric~rr 92( 1-2). 23 1-270, 2004.

7. Ullman. M.T. and Pierponr, E.I., Specific language impairment is no1 specific to I i~n~uage: the prc~ccdural deficit hypothesis, Corfr.v 4 1 (3), 399433, 2W5.

8. Mundy. I?. Sigman. M.. and Knsari, C.. A lortgitudinal srt~dy of joint attention and languagc devclnprnent in autis~ic children. Jottrnrrl of At1ti.cr11 D~t~elopt~~c~nf tr l J ~ ~ s ~ I s ~ ~ L ~ I : ~ 20. 1 15- 123, 1990.

9. Mundy. P.. Sigrnan. M.. Ungcrer. 3.. and Sherman. T.. Defining the social deficits in autism: The contribution of nonverbal cammun tcation measures, Jorrrrlnl qf Child P.rwql1o1og.v arrd Psyrlrjcttt~ 27. 657-669. 1 986.


1.7.

14.

IS.

16.

17.

18.

19.

20.

Mundy. P.. Sigman. M.. and Kasari. C.. The theory of mind and join[-attentian deficits in ziutism. in Ucrti~lursr~rrrdir~g Oiher Mir~cls: Pct:sperril~es *from Atrristlt. Eds.. Baron- Cahen. S.. Tager-Flusberg. H.. and Cohen. D.J.. Oxford University Press. Oxford. 1 943. pp. 1 8 1 -203. Chiinnan. T.. Drew A.. Baird. C.. and Raird. G., Measuring early language development in preschool children with aurisrn spcctnlm disorder using [he MacAnhur cornmuni- carive development inventory. Jourrrcrl q/'Clrifd ti~(tt~grir~,qe 3U1 21 3-236. 20413. Mundy. P.. Signran. M.. and Kasari. C.. Nonverbal comrnunica~ion. develo-pmcncal level and symptom presentation in autism. Der7e/0/?)n~~lt rrlrd P.r~.c-/top~~!lro/o~q~ 6. 389-401. 1994. Gricc. H.P.. Lopic and conversation, in Sntmm c lrrd Sennrnric..r, Eds.. Cole. F? and Morgan. J.L., Academic Press. New Ynrk. 1975. Wethcrby. A.. Onlo_eeny of communication functions in autism, Jtrlrnwl of Airris~n ( i r r d f l ~ ~ * ~ l o p ~ r ~ ~ ~ ~ ~ i ~ l Disorders I 6. 295-3 I 6, I 9x6. Wetherby. A. 3 r d Pn~tting. C., Profiles of communicative and cognitive-social abilities in autistic children, J~T I I~P IC I~ of . T ~ P ~ c * I P tr11~1 HPCJI'~)IS RL:SPCI~C~I 37, 364-377. 1984. Ball, J., A Prqmatic Analysis of Autistic Children's hnguagc with respect to Aphasic and Yonnal Language Devcltrpmcnt, Doctoral disse;eaation. Melbourne Urtiversiry. 1978. Loveland. K.. Landry, S.. Hughes. S., Mall. S., and McEvoy, R.. Speech zcts and the pragmatic deficits of autism. Jortnmr'c!fSl?e~u!a nttd Hr~arin~ R~.wlr . i r 3 1.5934~M. 1988, Rollins. P.. A Case Study of the Development of Communicative Skills for Six Autistic Children. Doctorat dissertation. Elatv~lul University. l994. Capps. I... Kehms. J.. and Si,nmnn. M.. Conversational abilities among children with autism and children with devclopmenral delays. Atitisnl 2, 325-334. 1998. Ghaziuddin. M, and Gerstein, L., Pcdantic speaking style dilfcrentintcs A~perger syndrome from high-functioning atctisrn. Jmrrni~I of Arrri.~m nrrd D e i ~ e I n ~ ~ ~ ~ ~ c ~ t t ~ d Di.7- nrd~ss 26, 585-595. 1996. Ramberg. C.. Ehlcrs. S.. Nyden. A.. Jahanssw. M.. and Citlbers, C.. Language and pragmatic functions in school-agc children on the a ~ ~ t i s m spectrum. E[tn~pelrrr Jo'orrrntrl r!f Dixotr l~t- .~ q( Cr3r~unrmicariorl 3 1 . 387'4 I 3, E 996. Adams. C.. Green, J., Cilchrist. A.. and Cox. A.. Conversational bchaviour of chi tdren with Aspergcr sy nclromc and conduct disorder. Jnlrrnal qf CJril'd P,~yho lo ,q~ nrld P s y r h i n r ~ 43. h79-690. 2002. Finc. J.. Rartolucci. G.. S7armari. P.. and Ginsberg. 63.. Cohesive discourse in pervasive developmental disorders. Jt)rwrml of Atrtism rrrrd Dcl\~~/ol~r?nrt~rrrE Dirord~/ :~ 24. 315-330. 1994. 1,oveland. K.. Tunali. R.. Kellcy. M.L,, and McEvay. R.E.. Rcfcrential communication and response adequacy in autism and Down"s syndrome. Appl i i~ l P s ~ c / ~ o ! i r ~ ~ t r i , ~ ~ f c ~ . r 10, 301-313. 1980. Surian. L.. Rarnn-Cohen, S., and Van dcr Lely. H.K.J.. Are children with autism deaf In Gaician Maxims'?. Co~ni t i re krrrr~psyhinrr?. 1 ( 1 ). 55-7 1. 1996. Paul. R. ant1 Cohcn. D.J.. Responses to contingent queries in adults with mental retardation and pervasive dcvelr~pmental disorders, A,pplictl P ~ h o l i ~ ~ ~ r ~ i . . 5. 349-357. 1984. Volden. J., Conversational repair in speakers with autism spectnlm disorder. J I I ~ P U I ~ I - i i o , ~ d Jwrnrrr/' of l~nrgrrn,q~ s u t d Crlr~ut~ra~ic~crfinn Disordprs 79, I 7 1 - I 89. 2004. CnweFand. K. and Tunali. B.. Narrative langttage in autism and the theory of mind hypothesis: a wider perspective, in I,'tldcrsrnrrrlin~ Orher. mind.^: PPI:EIFPC.!~IJPS,~~Y)~~~ Alrriswr. Eds.. Baron-Cohcn. S.. Tager-Flusherg. ?I.. and Cnhen. D.J.. Oxtbrd Univer- s i ~ y Press. Oxford. 1993.

196 Understanding Autism: From Basic Neuroscience to Treatmenl

29. Triger-Flusber?. H.. "Once upon a ribhit": stnrics namted by autistic children. Brifish Jt~rrnlal #f D~tgefr~p~rlrnral P . s ~ i r o I o g ~ 13. 45-59. 1995.

30. Tager-Flusberg, H. and Sullivan. K.. Attrib~rt ing mental states to story characlefi: a comp;~rison of narratives produccd by autistic and mentally retardcd individuals. Appjicd Psylrc~lir~griisrics 16. 24 1-256, 1 995.

31. Shsiberg, L.. Prlul. R.. McSwccney. J.L.. Klin, A,, Cohen. D.. and Volkmnr. FA.. Specch and prosody characteristics of adolescents and adults wiih high-functioning ~ ~ t t ism and as perger sy ndmmc. Jrjr rrrral of SIJQPL-~I fmr~~tu~qr a d Hl.rrr-itt,r( Rc.r~~~rrt+lt 44(51, 1097-1 1 15.2001.

32. HappC. E:. Comn~unicntive competence and theory of mind in autism: a test af re1evfinc.c theory. Coprifinn 48. I0 1 - I IS). 1993.

3.1. HappC. E. An adv;~nced test of ~hcory of mind: Understanding ol' stc~ry charr~crcrs' thoughts and feelings by able autistic. mentally handicappcd and. normal children and adulls. Jolrtnrrl c!f Arrri.v/rl nnd D ~ ~ ? ~ l o p t n ~ t ~ r ~ d D~SOITIPIJ 24, 1 29-1 54, 1994.

34. MacKay. G. and Shnw. A.. A cornparat ive study of figurative language in children with autistic spectrum disorders. Clrikd i~ln,q~rof~~ T~mhhif: rrrrd T / J P I ~ / I ~ 20. I 3-32. 2NM.

35. Minshew. N.J., Swceney. J.A.. and Fnrman. J.M.. Evidence for a primary neocanical systems abnormality in autism. Sori~ry f ir N ~ r r t n s c i r t ~ t ~ r Ab.~trnrm?.r 2 1 . 735. 1 995.

36. Kerbel. D. and Grunwell. P., A study of idiom comprehension in children with semantic-pragmaric dil'ficulties. Pan 11: Between-:roups results and discussion. I t l r t~r .

tlniiorlrrS J(?ttmnl c$Lnngirrr~e on(/ Coa~n~rrrricrrri i~~~ Di.~or.d~rs 3 3 1 ), 23-44. 1998. 37. Martin, i. and McDonald. S.. An exploration of causcs of non-literal fangatage prob-

lems in individu;ils with Aspergcr syndrornc. Jorrnlal of Auti?;nt trrld r)~rls~lr)p~?irnral Disnrb~~-,s 34. 3 1 1-328. 2004.

38. Baron-Cohen. S.. Hey! It was just a joke! Understanding proposirions and prr~posi- tionnl attitudes by normally developinp children and children wirh autism. I,r-rcrel Jolrnrirl of P.r!c-I7irrrr:1. 34. t 74- 178. 1907.

39. Jollif'fe. T. and Baron-Cohen. S.. The strztupe stories tcst: a, replication with high- functioning adults with autism or Asperger ?;yndrome. Journrrl r!f Arrrixm and D~lqcl- opwrni rcr i Dr'.~o/ul~t:~ 29. 395406. 1999.

40. Mitchell, I?. Saltmarsh. R.. and Russell. H.. Ovcrly l i tcrr~l interpretations of spccch in autism: understanding that messages arise from minds. Jor~nlrrl rd'Clrild P.ryAr~lofi. cud Ps~c;'ziltrr? 38. 685-691, 1997.

4 I. Chornsky, N.. Asj?ec.ls I?~ I I IP Tiwo~? qf'Sy~to.~, The MIT Press. Cnrnbridgc. MA. 1965. 42. Pinker. S.. T . r L~n,qitct,qe J~r.vrittct. William Morrow. Ncw York, 1 994. 43. Taper-Flusberg. H.. Sentence cornprehcnsicsn in autistic children, Applierl P.vyci~lnli~i-

glii.stic.s 2. 5-24, 1 94 I . 44. Minshew, N.J.. Gold.qtein. G,. and Siepel, D., Neurnpsychologic fttnctioning in

autism: prolile of n complcx infor~nation processing disorder. Jotrr-rtal r!frl~e IIIIPX- ncrrirrrfal N~I~I'~~~.F~I~~?~IIIR~c'cI/ Soc.i~ry 314). 303-3 1 0. I 997.

45. Minshew, N.J.. Goldstein. G.. Taylor, G.H.. and Siegel. D.J.. Ac;tdcrnic achievelnent in high functioning autistic individuals. J ~ ~ ~ ~ r t r . / t c r f qf Cli~rirul md Erper-int~~tml h ' r r r - rr)ps~rlrolojy 1 h(2). 26 1-270. 1 995.

46. Kjclgwrd. M.,M. and Tager-Flusberg, H., An i nvestiptinn of language i mpairnlenr in autism: implications for genetic suhgroups. htrgricrge onr/ Cr~,qvirivc Pmt.a.vsc~v 1 6(2). 287-308.200 1 .

47. Miiller, R.A.. Rehen. M.E.. Rothenncl. R.D., Chugani, D.G., Muzik. 0.. Mangncr. T.J.. and Chusani. KT., Brain mapping of language and auditory perception in high- functioning autistic adults: a PET study. Jmtmcrt (!f A~lrisrw arrd J)~ .~-e lnpm~nlaI [)is- nr.d~r.s 291 1 ). 19-3 I . 1999.


Minshew. N.J ,, Goldstein. 6.. and Siegel. D., Speech and lansuage in hi@-functioning aut Estic individuals. Neumpsyc/tdngy 9, 255-261. 1995. Minshew. N.J.. Goldstoin. G.. Muenz. L.R., and Ryton. J.R.. Ncuropsychological functioning in nonn~entally retarded autistic individuals. Jottrr~aJ cf Ctirlilwl atld

E~p~~rirnerrrtrl Nerrmo~-c.Erolr 14(5). 749-761, 1992. Prior. M.K. and Hsll. L.C.. Comprehension of transitive and intransitive phruses by atltistic. retarded, and normal children. Jottn~ni ~ ? f ' Ct~tnrrtltrtircrrinn ZIi.son/er,~ 12(2). 103-1 11. 1979. Just, M.. Chcrkassky. V.. Keller. T., and Minshew. N.. Cortical activation and syn- chrr)nization durins scntencc comprehcnsian in high-functioning autism: evidencc of' underconnectivity. Hrcrifl 127lPt. 8). 1x1 l - I82 I . 2004. Scarburough. H.S.. Rescnrln, L.. Trrgcr-Flusberg. H.. Fowler. A.E.. and Sudhalter. V.. The relation of utterance Sength to grammatical complexity in normal and language- disordered groups. Applied P.s?.c-11nlitr~rristic.s: P.vyc.ltoligi-ol Srlrciies of'hngrrtr,qr Pm- ce.v.ws I221 1 ). 23-45. 199 I . Eigsti, 1.-M. and Bcnnato. L.. Syntactic abilities and mcmory functions in young children with autism. under review. Tager-Flushcrg. H.. Calkins, S.. Nalin. T., Baurnbcrger. T.. Anderson. M., and Chud- wick-Dias, A., A longitudinal study of Ianguap acquisition En autistic and Down syndromc children. Jotrrrral of Attci.ml crrtd Lle~~elr~pnrerrral Disorders 201 1 1. 1-2 1 . 1990. Bi~rt~Iucci, G.. Piercc, S.J.. and Strciner, D., Cross-sectional studies nf grammatical morphemes in atit istic and mentally retarded children, Jorrnznl r!fA!cti.~m trttd flrihpi- O ~ I I I P I F I ~ I ~ Jlisoder.5 1 0( 1 ). 39-50. I 980. Howlin. F?, The acquiai~ion of grammatical morphemes in autistic children: a critiqt~e and replication of the findings of E3arrolucci. Pierce. and Streiner. 1980. Jot~n~nl r!f Au!i.vtn andl D E I * P / O ~ I I I P I I ~ ( ~ I l i ~ ~ t d ~ r . ~ 14(2], 127-1 36. 1984. Tagcr-Flusberg. H, and Calkins. S.. Does imitation facilitate rbc acquisition oF grammar"? Evidence from a study of' autistic. Down" syndrome and normal children. Jr)i~irrcrT Cl~ild ~ ~ n ~ ~ t i t r , q e 1 7(3), 59 1 -606. 1 990. Rotting, N, and Conti-Ramsden. G,. Autism, primary pragmatic difficulties. and specific language impariment: can we distinguish thcm using psycholinguistic murk- ers'!. I~e~lal iq~tn~arol Medicit re n t ~ c l Cl~i /d N ~ r ~ r n l o p ~ 45. 5 1 5-524. 2003. Condozrris. K.. Ev;~rrcic. L.. and Tager-Fl usherg. H.. Tense enor parterns i n children with azitism and children with SLI, I~rr~r~~nriotwE M P P F ~ I ~ f i r Alrrist~r Rcsi~arcnlr. Orlando, FL. 2002. Robetls. J.A.. Rice. M.L.. and Tager-Flusherg. H., Tense markins in children with ail tism , Al~plied Ps?'rk tdi~~,qaiisri 25. 4 3 9 4 . 20114. Walcnski. M.. Mostofsky. S.H., Gidley Lason. J.. and Ullrnan. M.T.. Fist Grammilt- ical Processing in autism. in 7irutfih A~rtrtic~i Meerirlg of ~ h c Co,q?litil~ N P I I F ~ C ~ P ~ ~ ~ P SocTet~. New Ynrk. 2005. Oznnoff. S. and Straycr. R.L.. Fttrthcr evidcnce of i n w t working memory in :~utism. Jntcrnrrl r?f At~fi .~trl DC~\YP~~I~IPIL'III~II I>~~.CIIY!PT.T 3 113). 257-263, 200 I, Lnrd. C. and Piul. R.. 1,angunge and communication in autism. in Hrrrarll>cw)k ol' Al~f isr~ i rarii Pcnnari~~~ Dev~tr~pt nentrrl Disr~rders. 1 497, pp. 1 95-225. Eyer-Flusberg. H.. DLI autism and specific language impairment represent overlap- ping language disorders. in O(w=!ctprrr~rrm/ Lnrr.yrrrrge Disord~r .~ . Eds.. Rice. M.1-. and Wanen. S.F.. 21304. pp. 3 1-52. Tagcr-Fl ushers, H., The conceptual basis for referential word meaning in children with iaitism, Clrik! lZIetvloj?t~r~~lf M I S ) . 1 167-1 178. 1985.

1 98 Understanding Autism: From %sic Neuroscicncc to Tredtmunl

66. Rurnsey. J.M.. Neuropsycholcrgical diveqence of high-level rlutism and severe dys- lex ia. Jortrrtcrl of Arrtis~n allil C/rilrllrenti Uisr~rder.~ 20(2). 1 55- E 68. 1 990.

67. Roucher. J.. Word Fluency in High-Functionirrg Autistic Children. JnrrntitS qfAarri.nn crrrrl Ctlil~lhood Sinhi:ophre~?i~. I K(4). 637-1145. 1 988,

68. Miiller. R.-A.. Chugani. D.C.. Rehen, M.E.. Rothcrmel, R.D.. Muxik. 0.. Chnkrrtborty. and Chugani, H.T.. Impairment of denlo-thalama-cortical pathway in autistic

men: langua~e activation data fmm positron emission lomography. Nerrmu-irrtc-p L~IIPI-.~ 243 1 ). 14, I 998.

69. Dc Fossd. L.. Hodge, S.. Makris. N.. Kennedy, D.. Giiviness. V.. McGrarh, L.. Sreclc, S., Ziegler. D., Hcrhett. M.. Frazier, J.. Tager-Flusberg, M.. and Harris. G., Language association cortex asymmetry in autism and specific lanzuagc irnpairmenl. Amlols r,rf' Nf'tiFl~lO8~ 56. 757-766. 2004,

70. Howlin, P. and Ruttcr. M . . T m h n e t t t c!f Aitfi.rrir- Chikdt~jrr. John Wiley & Sonr. Chichestes, U.K.. 1987'.

71. Rnrtak, L.. Rutter. M.. and Cox. A.. A comparative study nf infantile rtutisnl and specific dcvelopmcntal receptive language disordct. Rriris/l Jo~rrrtir/nfP.~ycinicdr?. 126. 127-145, 1975,

72. 1)obbinsan. S., Pcrkins. M,. and Baucher. J.. The interactional signitiaancc ot' formulas in autistic langua~e. C1init.d Lin,qtrisfius r r t d P11nti~iio.r 1 714-5 1, 299-307. 3003.

73. Wray, A. and Perkins. M.R.. The functions of formulaic Sanguaps: an inleprntcrl model, I ~ n g r t n ~ q ~ nt~d Cntn~~rtu~irlrrio~~ 20, 11-28, 21KH).

74. T:qcr-Flushcrg. H., A psychnlogical approach to understanding ~ h c social and l ;u~- guage impairments in autism. ilntrntorir~rrcrl R~r ip lv ~ ; ~ f P . v ~ c i ~ i u t ~ I 1. 325-334, 1999.

75. Tdgcr-Flusberg, H., Exploring the relationships between theory of mind and social- camrnunicaaivc filnctioning in children with autism, in /~~di~-idtcni D$%~rrrc.i=.~ iirl

TIzcwy r!fMitzrl.. It~r~?/fcc~rir~rr.~fi~r Trpirnl nrrd Al!;rlirrrf Dt~~~r iop~n~wt . Eds.. Repacholi. B. and Slaughter. V.. Psychology Press. Wove, UK. 2003.

76. Ramn-Cohen. S.. Leslie. A.M.. and Frith. U.. Does the atrtistic child h i l v ~ a theor! of mind'?. C,,~lrifiotr 2 1 . 3 7 4 6 . 1985.

77. Barnn-Cohcn. S..Tager-Flusberg. H.. and Cohcn. D.J.. Uitr!~rsttt,rrEirr,q Olhrr. Miracl,~: I~rrspccth~e.~ .frn~tt Attfisrtt, Oxford University Press, Ox ford, 1 993.

78. Ramn-Cohen. S. , From attention-goal psychology la belief-desire psychology: the development aTa theory of'rnind, and its dysfunction, in i!lnckr:r~crniijtl,q O ~ ~ I F I ' M i ~ i i ~ v : P~r.spet.fivc..s .fr.r)st Arrtism, Bamn-Cohen. S , . Tap-Flusberg. H,. and Cohcn. D.J.. Oxford Universiry Press. Oxfbrd. 1993. pp. 611-82.

79. Hobson. R.P.. Understanding prsons: the rolc of affect. in Utrclrr.~fandirr,q Otlrer. Minds: P ~ ~ t q ? e ~ t i v ~ . s . f i ~ ~ t ~ t Arrtisrll. Bamn-Cohen. S. . Tager-Flusberg. H.. and Cohen. D.J.. Oxrnrd University Press, Oxford. 1993. pp. 204-227.

80. Sigman. M. and Ruskin. E.. Cmtfi~~rriy nrrd Cl~orip in firp Socit~E Co~~rpetctrrc r?[ CI~ilclmtr wifh Aiiris~r~. Dnwrr S y ~ d r o ~ ? l ~ C I ~ L ! D P I - F I I E ~ ~ ~ I I I ( I E Deitr~~v, Rlnckwell Puh- lishers. Mitldcn. MA. 1999.

8 1 . Sperhcr. D. and Wilson. D.. Rcl~~ua~rr~: Cotmtr~o~ii~rr~iarl rrrrd C~~pririrrtt, Harvnrd University Prcss, Cambridge. MA. 1986.

82. Hale, C.M. and Tagcr-Flusberg, H.. Social communication in children with autism: the ~larionship hetween theory of mind in discourse development. A~rfisrrr. 912). 157-178, 2005.

83. Frith. C.D. itnd Fri'th. U.. Inlernctinp minds. - a biological basis, Si-icrrcc# 286. 1691-1 A95, 1949.

I-angi~age in Autism 199

$7.

XX.

Happi. F.. Theory of mind and thc scl f. Anno/.v ofthe NPII' YoliC. Acljc/~~rry r f S ( . i r tw .~ .~ l(H31. 1.34-144, 2003. Baron-Cohen. S.. Rirr:. H.A.. Wheelwright. S.. Bullmore. E.T,, Rrammcr. M.J,. Simmons. A.. and Williams. S.C.. Social intelligence in thc normal and autisltic brain: an WRI study. Ew-operta Joirrtrnl qt' Nt~rrroxriivtr~a 1 1 . 1891-C 898. 1999. Happd. F.. Ehlcrs. S.. Fletcher. P.. Frith. U.. Johansson. M.. Gil lbcr~. C.. Dolnn. R., Frr~ckowink. R,. and Frith. C.. '"Theow of mind"' in the brain: evidence from a PET Kiln study o f Asperger syndrome. Ne~rr~rrpar~ 8. 197-20 1 , 1996. Frith. C.. What do irnasing studies tell us about the neural basis or autism'!. Nowrris fi~rrr~rhtinn Syntpositrnf 25 1. 149- 166. 2003. Mostofsky. S.H.. Goldhery. M.C.. I,anda. R.J.. and Denckla. M.B.. Evidence for a deficit i n procoduritl learning in childrcn and nclolescents with autism: implications for cercbc t lar contri hut i nn. j o ~ i t + ~ ~ i r d I$ f h ~ I ~ F I P ~ I C I I ~ O I I ~ ~ N ~ t ~ m ~ ~ . ~ ~ c ; C l t ~ l ~ ~ , q i ~ ~ i ~ l Sni- i~f i 6(7). 752-759, 2000. Schacter. D.L. and Tulvinp. E.. Mc~rnop Sy.rr~nw 1994. The MIT Press. Cambridsc. MA, 1994. Squire, C.R. and Knowlten. B.J.. Thc medial temporal Eobc. the hippocampus. and the memory systems o f the brain. in 'The Nelt* C o ~ l t i t i ~ r Ncrrmsrir~~i-r-7. Ed.. G. 6it~nnIga, . . MA. MIT Press. Cambridge. MA. 20IlO. pp. 765-780. Willingham. D.B.. A neun~psyshological theory of malor skill learning. P . v ~ r l ~ o ~ ~ ~ , q i ~ ' ~ ~ l R ~ ~ * i e t l - 105(3). 558-584. i 99X. Grayhiel, A.M.. Building action repertoires: Memory and learning firnctions of [he basal ganplin. Grr-rent Opi~tinr~ i ! ~ N~~iml>ior'o~:y 5. 733-741. 1995. Knnwllon. B.J. and Squire. L.R.. Artificial grammar learning depends on implicit acquisilion oF but h absrmc~ and exemplar-speci f ic information. Jorirrml r f E~perE- t ? ~ i ~ t ? ? i t I P.vyrho/o#qy: I3~1rt1 i ~ l g Mmot? ( ~ ~ t d Cogrr i t i t ~ n 22( I 1. I 69- 1 8 I . I 996. Poldrack. R.A.. Prabhakarr~n. V., Seger. C.A.. and Gabricli. J.D.. Striatal activation during ncquisitiun of a coenilive ski 11. Nerrtr)y~.ry~.l~oI'~~gy 13141. 564574. 1998. Ullman, M.T.. A neuroco~nitivc perspeclive on langungc: zhc declarative/proccdi~ral model. Nrrrurr. Rp11ierr.s Nt~tr~r?sr~irv~cc 2 . 7 1 7-726. 200 F . Heilman. K.M., Watson. R.T.. and Rothi, L.G.. Disorders of skilled movements: limb npraxia. in B~hn~-iorcrl N~trralr:qy crnd N e ~ r l u ~ p s ~ . . Eds., Feinkrg. T.E. and Farah, M.J., McGraw-Hill. New York. 1997, pp. 227-235. Schluter. N.D.. Krams. M.. Rush\vorth. M.ES.. and Passingham. R.E.. Ccrchral dominance for action in thc human bmin: the selection or ac~irjns. N e ~ r m ~ ~ . ~ ~ c ~ l ~ e l ~ ) , q r ' ~ t 39I2). 105-1 13.2001. Squire. L.R. and Zoln. S.M.. Stntcture and fi~ncrion of declarative and ncmdectarrttivc memory systems. P~ore~*cii~~,qs c!J' f k ~ 1NRiirtlinJ Acrrilrtrn. nt' Si.ic.r~r~ USA 9.3. 13515-13522. 3996. Hikosaka. 0.. Sakai. K.. Nnknhara. H., Lu. X.. Miyachi, S.. Naknmura. K.. and Rand. M.K.. Neural mcchanisn~s for learning of sequential prncedures, in Tlrc. Nc~\z. Ct)mgrliri~hrp iV~t~m~-iet~,.r,s. Ed.. Gnzzanip. MS.. MI'S Press. Cambridge. MA. 2oUJ. pp. 553-571. Gaodale. M.A.. Perception and action in the human visual system. i n Tlrr iV'<>lt.

Co,s.ni~i~-c Ne~r!r>.~c-ipni.er. Ed.. Gnzznnign. M.S.. M IT Press. Can~bridgc. MrZ. 1000. pp. 365-378. Ullrnan. M .T.. Is Brocn's area part of a basal gang t ia thalamocortical circuil'!, Ct)rtr*v, in press. Middleton, EA. and Strick. Basal sanglia and cerebellar loops: motor and cognitive circuits. Rrrritr R C J $ ~ ~ ~ I I 3 T (3-31, 2.76250. 2000.

200 Undcrstr;tandi ng Auf ism: From Basic Neuroscience to Treatment

1 06.

1437.

I OR.

I l l .

112.

113.

114.

I IS.

116.

Jankovic. .I. and Tolosa. E.E.. Parkirt.~on :F Di.rcnse a!ldMm?e~nr~lr Disntd~rs, Williams and Wilkins. Ballirnore. MD. 1993. Brads haw. J . L.. I ) e ~ ~ ~ l ~ ~ p t ? t r ~ r ~ ~ l Disnrkrs t!f rh P f r f~nf~~ .~r rk f f f~d S ~ F I P ~ I . Psychology Press. Hove, East Sussex. Grenlt Britain, Z(M1. Young. A.B. and Pcnncy. J.B.. Biochemical and functional orysnization of the basal gang1 i a, in Pctrki~t.~on '.F Dis~rrse rrtlrl Mrr ~vwrcl?? ESisord~r~s. 2nd ed.. Ed s.. d i m kov ic. J. and Tolosa. E.. Williams and Wilkins, Baltimore. MD. 1993, pp. 1 - 1 1 . Albin. R.L.. Young, A.B.. and Pcnney. J.B., The functional anatomy nf basal ganglia disorders. Ti .~ t l r l t in N ~ ~ t ~ r n . r ~ i e r ~ r e . ~ 12I 10). 366-375. 1989. Micldleton. F. A. and Strick, P.L.. Basal pnnytia output and cognition: cvidence fmrn anatomical. behaviors I. and clinical studies. Hlnirz nrtd Cogr~irfnr~ 42(2). 183-200.2NHI. UlIman. M.T., Corkin. S.. Coppola. M.. Hickok. G.. Growdon. J.H.. Koroshetx. W.J.. and Pinker. S.. A neural dissociatinn within language: evidence that the mental disticlnury is part of declarative memory. and that grammatical rules are prt)cessed by the procedur~11 ssy stem. Jr~rrrrtctl qf' Cr>,g~t ifi1.p ~errm.~cirrrce 912 ), 266-276, 1 997. Mcrzenich. M.M.. Recanzone. G.. Jenkins. W.M.. Allard. T.T., and Nudo. R.J., Corrical represenfationat pl;lsticity. in N~ttmhit)k<),qy r?f N~nrnrre.r. Eds.. Rakic. P. and Singer. W.. John Wiley Rr Sons. Chiuhcster, U.K.. 1988. pp. 4 1-67. Squire. L.R., Clark. R.E.. and Raylcy. PJ.. Medial tempor;tl lohe fitnction and memory, in The CogrlirE1-P N ~ l i t s t ~ s c i ~ ~ ~ r ~ s , 3rd cd.. Ed.. Gnzzaniga, M.S.. MIT Press. Cambridge, MA. 2004. pp. 691-70R. Eichcnbaurn. H. and Cohen. N.J.. Frnm Cm~rliriarrirt~ to Cnnscinrrs Rr~lrlEet.tior1: M c n ~ c ~ p Sysrerlts r!f I ~ P Brcritl. Oxford University Press. Ncw York. 2001. Chun. M.M.. Contextual cueing of visual attention. Twnr!,~ irl Cognirive Scierrcc~s 4(5). 1 70- 1 78.2000. Hodges. J.R. and Pattcrs~n. K., Semantic memory disorders. Trertds: irt Cngnirive Sci~naes 1(2);, 68-72. 1997. Martin, A,. Ungerleider. L.G.. and Hilxby, d,V.. Category specificity and the brain: the sensorylmotor rnodcl of semantic rcprcsenrntions of' ~ b j c a s . in TIIP Cr~gnirhv N~rtmsc-i~ttr-[as. Ed.. Gazzaniga, M.S.. The MIT Press. Cambridge, MA. 2000, pp. 1023- 1036. Binder, J.R.. Frost. J.A.. Hammeke. T.A.. Rno. S.. and Cox. R.W.. Function of the left planum temporale in auditory and linglrisilic product ion. t3rnfr1 I 19. 1239-1 247. 1996, Thompson-Schi I I. S.L., DmEsposito. M.. Aguim. G.K., i d Farah. M.J.. Rolc of left inferior prcfrontal cortex in rctrieva! of sem;lntic knowledgc: a reevaluation. P r n ~ c ~ d - irrps of t h ~ Nnrionrr! Acndetny 5rient.c~ USA 94(26). 14792-1 4797. 1 997. Ruckner. R.L. rind Whceler. M.E.. The cognitive neuroscience of remembering. Norrtrcj R m i ~ n * Nc.smsc.ir.nee 219). 634-634. 200 1 . Desrnand. J.E. and Fiez. J.A.. Neuraimaging studies of the cerebellum: language, learnins. and memory, Twlrrls irr Cognititil?~ .Yricrrces 2(9), 355-362, 1998. Ul lman. M.T.. A co~nitive neuroscience perspecrive on second language acquisition: the dcclamtivefpmccdural model. in M i d atrd Cr~nr~.rr in Arlitlr Second L ~ n ~ l ~ n g c Arcjui~irion. Ed.. Sanz. C.. Georgetown University Press. Wi~shington. D.C.. 200.5, ~>p. 141-178. Poldrack. R.A. and Packard, M.G.. Competition among multiple memory systems: converging cvidence from animal and human brain studies. NcrtmpFvycitr~/n,pin 41{3). 245-25 1.2003. hckard. M.G. and Knowlton. B.J.. Learning and memory functions of the basal ganglia. Arzni~ol R P I ~ ~ P ~ I r t f N~rn7).vrir~trrp 25. 563-593, 21302.


Fletcher. P.C.. Zafiris. 0.. Frith,C.D.. Honey. R.A.E., and Corlelt. P.R.. On thc benefits nf no! trying: brain activity and connectivity reflecting the interactions of explicit nnd implicit sequence teaming. Cerehrd Cnrrcr. 1 S(7). 1002- 1 0 15. 2004. Drury. J.E. and Wllman. M. T.. The memorization of complex I'orms in aphasia: implications for recovery. Bmirl n~rd L l t t ? , y t t ( ~ $ ~ 83, 139-141. 2002. Mari. M.. CastieTlo, U.. Marks. R.. Marraf'fa. C.. and Prior, M.. The reach-to-grasp rnovcrncnt in chi ldren with LIU t ism spectru rn disorder. Plrilosophircd T'.~n,s~/t-~r'n~~s

r ? f f J ~ ~ R B ~ C I I So(*i~!\' of Lr~ncJon. Scrips R: Rir~/o,qic.al S r i r ~ t c ~ s 358( t 430). 393-4133, 2003. Bnucher. J.. "Lost in a sea of time": timing-parsing and autism. in T i r n ~ ~ r ~ r r ! M t ~ r ~ w y r : issrre.~ ~ J J Philnsoph~ utrd P . v y ~ / ~ ~ l i o p , Eds.. Perkins. M. and Howard. S.. Clarendotn Press. Oxford. 2001. pp. 1 1 1-135. Hcrrnclin. H. and Frith. U.. Psychologiw;J srudies of childhnod autism: can autistic children make sense of what they see and henr?. Foc-rr.r or1 Atrti.v?Ec' Reho~qior A( t 1. 6 1 3 . 1991. Chomsky. N.. S ~ t ~ f m r i r . Sfrrictrrre.~. The Hague, Mouton. 1 957. Chnmsky. N., The Mirtit~rnllist Pmgmnr. The MIT Press. Cambridge. MA. 1995. Smith. I.M. and Bryson, S.E., Imitation and action in autism: n critical review. Ss~ci~olo,qir~i~/ RrrlEerirr E 1 6{2), 259-273. 1994. tcnry. M.R. and Mi[l. D.A.. Moving on: autism and rnclvcmcnt dismrhancc. M~~r~f l r l R~mr-hrion 34( 1 ). 39-53. 1996. Rngcrs. S. and Pcnnington. t3.E. A theoretical approach ro the deficits in infantile autism. De~~e io l~rnc~nr m ~ d P s ~ c E ~ o p o 3. 1 37- 1 63. 1 99 l . Mijller. R.A., Pierce, K.. Amhmqc. J.B.. Allen. G.. and Caurchesne. E.. Atypiclil patrcrns of cerebral motor activation in autism: a functional magnetic rcsonance study. Biologirv~l Psyclrinrr? 49I X 1, 665476. 200 1 . Ro~gers, 3.1.. Rennetto. L.. McEvoy. R.. and Pennfngton. R.F.. Imitation and panto- rnirne in hi_eh-functioning adolescents with autism spectrum disorders. ClriTtI Dr~~lc l - optrwtlr 67 15). 2060-2073. 1996. Pierce. K. and Courchesne. E.. Evidcncc for a cercbeltar mlc in rediiced exploration and stereotyped behavior in autism. Biologiurrl' P s ~ c l ~ i n r n 4'11X). 655464 . 200 1 . Joncs, V. and Prior. M., Motor imitation abilities and neurological signs in nutictic children. J~~t i t - t r~l l o f A i ~ f i ~ m D P I J C / ~ > ~ ~ O ~ ~ C I / D i ~ r r r d ~ r , ~ 15Q 1 ). 37-46. 1985. Vilensky. J.A.. Damasin A.R., and Maurer. R.G.. Gait disttlrbances in patients with autistic behavior: a preliminary study. Arrl?i\~o.s r?f N~rrn)lo,qy 3X( 101.64h-649. 198 I . Rinynnn. J. and Jankovic, J . . Occurrence of tics in Asperger's syndromc and autistic disorder. Ji~rmwi r!f Clrifd Nertnrlo~y E 5(6), 394-400. 2000. Swcrdlow. N.R. and Young. A.R .. Ncuropathotngy in Tourcttc syndrome: an update. Ad~-ctrrc~s irl N ~ ~ r . n I o , ~ y 85. 158-1 61. 2001. Mink. J.. Neurobioloyy oi'basal pan~l ia ci rc~~i ts in Tourene syndrome: faulty inhihi- tion 111' unwanfed motor p:~ttem~~~'?. A l l ~ l c ~ t t t ~ ~ . ~ in JVPtrmlo~y 85, I 13-122. 2001. Sears. L.L.. Vest. C.. Mnhamed. S.. Bailey. J., Ranson. B.J., andl Piven. 9.. An MRT study of t hc basal gang1 ia in autism. N~rt~)~).~~~rl~nph~rrt~~ncnI~~,q~ crrtrl Rinl&~~iritJ P.r!~-hinrr? 23(4). AI3-624. 1999. Hollander. E., Anagnostou. E., Chaplin, W.. Esposito. K.. Haznedar. M.. Liciilzi. E,. Wasscrmun. S.. Soorya. L.. and Buchsbaum. M.. Striatal volume on magnetic resonance imaging and repetitive behaviors in autism. BiirCo,qirburrl Psy~.I~irirry. 2005. Hnrdan. A.Y.. Kilpatrick. M., Kcshavan, MS.. and Minshew. N.J.. Motor pcrfnrmtlnce and anatomic magnetic rcsonance imaging (MRIS o f the basal ganglia in autism. Jnrtrnal r!f'ChiId Ner~n)fr):\~ IXCS), 317-324. 2003.

202 Understanding Autism: From Basic Neuroscicncc to Treatment

Oram Cardy. J.E.. F l a g E.J.. Roberls, W.. Rrinn. J.. and Roberts. T.P.. Magnetoen- cephalography identilies rapid temporal processing deficit in autism and language impaimen[. N~uml-epot-t t (,(4), 329-332, 2005. Szclag. Em. Kowalskn. J.. Galkowski. T.. and Poppel. E.. Tcmporat processing deficit% in high-li~nctioning children with autism. Rr J iJ.~ychnJ 95(Pt. 3). 2W-282. 2M4. Bennetto. L.. Pennington. R.E. and Rogers. S.J.. Intact and impaired mcrnory func- rions in autism. Chilrl D r ~ ~ / r q ~ ~ ~ i e r i t 67. I X t6-1835. 1996. Luna. B.. Minshcw. N.J.. Garvcr. K.E.. Lazar. N.A.. Thulhom. K.R.. Eddy, W.F.. and Sweeney. J.A.. Ncocoflical system abnormalities in autism: an IMRI stt~dy of spatial working memory. N(b~rtnfr)gj 59(6). 834-840, 2002. Spencer. J.. O'Rricn, J.. Rig~s. K.. Braddick. 0.. Atkinson. J. , and Wattam-Bell. J . . Motion processins in aietism: evidence for a dorsal stream deficiency. hrc,rtrr>r.q~orr I I ( f 2). 2765-2767. 2000. Pellicano. E.. Gibson. L.. Mayhery. M.. Durkin. K.. and Raclcock, D.R.. Abnarrn;~l global procelisinp along the dorsal visu~l pathway in autism: a pnssiblc rnech:inisni ii3r weak visuospatial coherence?, N~~rrt>/>s~-rl~x)Iogitr 431 7). 1 043-1 953. 2005. Tager-Flusbcrg. H .. Basic level and superordinate Icvel categorization by auristic. ~ncn~ally retnrdcd, and normal children. Jmrtnrrrl of E.rp~ritm~t~rrrE UtiM F ~ s ~ e ~ h r n l r y ~ 4013). 450469. 1985. Eskes. G.A.. Bryson, S.E., and McCorm ick. T.A.. Gm-tprehcnsiun of concrete and abstract words in autistic children. Jotrrnrrl c!f'A~tfirru rr)?rl C'lrikdhond S~.l~i,-r>phr.r')ricl 20(1). 61-73. 1990. Habson. R .P. and Lee. A.. Emotion-related and abstract concepts in auristjc gerbple: evidcnce from the British picture vncahlulary scale, Jotrnanl r$Atrfisrn fwd DPIYJI~~?- merrtfil D i s t ~ r d e ~ ~ i 9(4). 601-62.1, 1 989. Brysnn. S.E.. Interfcrcnce effccrs in autistic childrcn: evidence for the comprehension of single stimuli, Jortrncrl rfdhr~or-rrml Ps!'chnlo,qy 921 2). 250-234. 198.7. Toichi. M. and Kamio. Y.. Verbal assaciazion for sirnplc common words in high- I'ttmctioning autism. Jtrtrrtlclf qfAlriixrln crrld Devi~loj~nrelrzrrrl Di.~orr/rrs 3 1.483490. 200 1 . Van Lancker. 13.. Cnrnelius. C.. and Needlem;in. K.. Comprehension 01' verbal tenn?; for emotions in nmmnl. autistic, and schizophrenic children. D~\velr~y~rnertrcrl Nerrrryt- sycholo:?. 7CI). 1-18. 15191. Rcversdorf. D.Q.. Anderson. J.M.. Manning. S.E., Anderson. S.L.. Nordgren. R.E.. Fclopulos, G.J.. Nadcnu. S.E., Heilmnn, K.M.. and Bauman, MI,.. Thc cl'fect nl' semantic and emotional context un written recall h r vcrbal language in high functioning adults wi I h autism spectrurn disorder. Jourrlnl of Ncrrrr3logy. Neri t~~stt~~et: \: lrnd P .c~ i - l r i r r~r~~ 65(5) , 87344737. t 998. Harris. G.J.. Chahris. C.F. .. Clark. J.. Urban. T.. Ahamn. I . . Steele. S.. McGrath. L.. Condouris. K.. and Taper-Flusher?. M.. B sai n act ivatinn during sernanlic processing in autism spccrrlrm disorders via functional ~nagnctic resonance irnng ing, Btniir o ld

G r g j ~ i ~ i o r l . in prcss. Minshew. N.J. and Gulds~ein. G.. The patrcrn of intact and impaired memory f~~nction.; in autism. jourt~nl r.$ Clrirlif P.~yc l rn to~~~ nrtd Psyliirrrry 42. 1 095- 1 I0 I . 200 1 . Ren Shalom. D.. Mcmory in autism: review and synthesis. Ct>r.rc,r 339(4-5). 1 129-1 t3X. 2003. VanMeter. L.. Fein. D.. Morris. R.. Waterhouse. L., and Allcn, D.. Delay versus deviance in autistic social behavior. huntrrE cfRrrrisat nrrrd I l ~ ~ r f n ~ ) ~ n r r ~ r c l l r')Esnr-clr.~:v

27(5). 557-569, 1997. Minshew. N,J, and Goldstein, G.. Is autism an amnesic disorder? Evidence from the CnIifornia verbal learning tcst, hrer r~r~p .~~cJ~n/o~y 7. 209-2 16. t 993.


161. Rnucher. J. and Warrington, E.K.. Mcrnory deficits in early infantile autism: srrme similarities to the amnesic syndrome. 1lrifi.d Joltmnf r?f P.~yc/rrn!ogy 671 1 ). 73-87, 1976.

t 62. Bnwlcr. D.M.. Matthews, N.J.. and F;mliner. J.M.. Aspcrger's syndrome and memory: similarity w autism but not amnesia, Netrr-i>psyrh{jIo,~in 35, 65-70. 1997.

163. Tager-Flu5berg. H..Semantic processing in the freerccall of autistic children: furtlrer evidence for a cognitive deficit. Brirish Joltnlai vf P.v,rr:l~oI~~,yy 9. 41 7430. 199 1 .

164. Toishi. M. and Kami0.Y.. 1-ong-term memory in high-f'unc~ioning aurism: conlrovcrsy on cpisodic memory in nurism rcconsidcmd. Jotrl-met r!f Auri.wn ~rrrd De\*~lrq?t~rcnrrr/ Di.~~rlr<f~r,~ 33(2), 151-161. 2007.

165. Bouol-rer. J . . Mcmory for recent cvent in autistic children. Jmrrr~crl qf Attrirn~ r r r ~ d Cl~iltl/~lood .Yr.lri:oy~hrt~rticr t I ( ? ) . 293-30 I. [ O X 1 .

166. Klien. S.B.. Chsn. R.L.. and Loflus. J.. lrtdependence of episodic and semantic sclf- knowlcdse: rhe ewe from autism. SorirrE Co,qttiriorl 17(4). 41 3-436, 1999.

367. Millwarri. C.. Powell. 5.. Mcsser. D.. and Jordan. R.. Rccall for self and other in autism: chi ldrcn's memory for cvents experienced by themselves and their peers. JOIII.IICI/ qf 'A i i r i~~ t t ittld DP~'C*IOI?~IIPIIICE Diso~der.~ 330( 1 5 . 13-28. 2000.

I . RowJer. 13.M.. Gardiner, J.M.. Gricc. S.. and Sanvnlaincn. P., Memory illusions: false recall and rccngnitfon in adults with Aspergcr's syndrome. Jocfrnnl of Rhrro~nci!

Psyi~ol(~,p!. 1 (69(4 3. h63472.21XH). 169. Rumscy. J. itnd Ernsl. M.. Funcrionnl Ncuraimaping nf autistic disorders, Mrpt~rnf

Rr~frrrd(rtic~rr crr~d I I e ~ * r ~ l n p m ~ t ~ i ~ ~ i Di.~nhiiiric.s R~spctrrh Rt>~*ie~\:c 6(3). 17 1 - 1 79. 2800. 170. Hatrn~nn, M.L. and Kernper. T.L.. The Nellr-ohic~lqgy r!fAlrfi.~r,r. The Johns Hopkins

University Press. Rnltimorc. MD. 2005. p. 404. 7 I. Bramhilla. P.. Hardan, A,. di Mcrni, S.W.. Perez, J., Soases. J.C.. and Rmale. E. Rrain

anatomy and development in aurism: review af srructuri~l MRI studies. Bri~i~l Re.v~wr11 Rt t I /~ t i l r 6 1 (6 ) . 557-569. 2003.

172, Abel!. F.. Krams, M.. Ashburner, J.. Passinsham, R.. Friston. K., Frackowiak. R.. Happe. E. Frirh. C.. and Frirh. U.. The neumanatomy of autism: a voxcl-based whole brain arialysis d stnrcturnl scans. N~rrtnrepor/ t O(X). 1647- I65 1. 1999.

173. Herbert. M.R.. Harris. G.J.. Adsien. K.T.. Ziegler. D.A., Maktis. N.. Kennedy. D.N.. Lanpe, N.T.. Chabris. C.F., Rakardjiev. A.. Hodgson. J..Takeokrl, M..Tagcr-Flusberg. H. . and Caviness. V.S.. Abnormal asymmetry in languape association cortex in autism. Anltnls o f h r ~ i t m k ~ q ~ 52(5) , SRR-596. 2002.

173. Hcrhcrt. M.R., Ziegler. D.A.. Dcuzsch. C.K., 0' Brien. L.M.. Kennedy. D.N.. Filipek. P.A.. Bakardjiev. A.1.. Hod~son. .I.. Takeoka. M.. Makris. N.. and Cavincss. V.S.. Jr.. Brain nsymmctricq in autism ;md developmental language disorder: a nested wholc- brain analysis. Rroiir 1281R. 1). 2 13-226, 2005.

175. Cnurchesnc. E. and Pierce. K.. Brain overgrowth in autism during a critical time i n development: implications for fnlntal pyramidal neuron and interneuron dcvcloprnent and connectivity. i t ~ r ~ r ~ ~ ~ ~ t i o r r ( r f JO~W~IU I of 138~~~Io~~r1rc~rr t id N P I ~ ~ ~ . T ( ' ~ P E I C P 2 3 (2-3 3. 153- 1 TO. 2005.

176. Carper. R. and Courchcsnc. E.. Localized enlarsement of the frontal cortex in early autism. Riofr;qic.rrl Psvchintr~ 5712). 12G-133. 2CX)S.

177. C;~spcr. K. and Courchesne. E.. Inverse correlation hetween frontal lobe and cerebellum sizes in children with autism, Btnirr 123(43. 836-844. 2000l.

178. Relmonze. M.K.. Allen, G.. Beckcl-Mitchcner. A., Boulanger. L.M.. Carper. R.A.. and Webb. S.J,. Autism and abnormal development of brain connectiviry. Jr)rr,-r~al of Nc.rrrnsr.ielrc~ 24(42). 9228-923 1. 2004.

179. FrEth. U. and Happd, E.Autism: beyond "theory ol'mind:'Ci~,qnirii~n 5C). 1 15-1 32. 1994. 180. Hi!l.E.L..Execu~ivcdysfrrnutioninautism.T~~11~~~i~~C~~~~1iti~~S~ir~~~'e.~X(I).2&-12.20(b4.

9 language in autism - university of california, san diego

Documents