Communication Autism

EARLY COMMUNICATION DEVELOPMENT AND

INTERVENTION FOR CHILDREN WITH AUTISM

Rebecca Landa1,2*1Center for Autism and Related Disorders, Kennedy Krieger Institute, Baltimore, Maryland

2Department of Psychiatry, Johns Hopkins School of Medicine, Baltimore, Maryland

Autism is a neurodevelopmental disorder defined by impair-ments in social and communication development, accompanied bystereotyped patterns of behavior and interest. The focus of this paperis on the early development of communication in autism, and earlyintervention for impairments in communication associated with thisdisorder. An overview of components of communication is provided.Communication characteristics that are diagnostic of autism are sum-marized, with consideration of the overlap between social and com-munication impairment, particularly for children with autism func-tioning at the prelinguistic level. Early communication developmentand predictors of communication functioning in autism are exam-ined, based on a review of prospective and retrospective studies. Thefocus of the discussion then turns to intervention. Consideration isgiven to the rationale for beginning intervention as early in life aspossible for children with autism. Implications of motor, imitation,and play deficits for communication-based intervention are exam-ined. Finally, issues related to the design and delivery of interventionfor young children with autism are presented, along with a review ofthe major early intervention approaches for autism. ' 2007 Wiley-Liss, Inc.MRDD Research Reviews 2007;13:16–25.

Key Words: autism; communication; early development; early intervention

Autism is a neurodevelopmental disorder defined byimpairments in social and communication develop-ment, accompanied by stereotyped patterns of behavior

and interest. Along with pervasive developmental disorder-nototherwise specified (PDD-NOS) and Asperger syndrome, au-tism is categorized as a pervasive developmental disorderwithin the DSM-IV [American Psychiatric Association (APA),1994]. In clinical practice, the term autism spectrum disorder(ASD) is often used to collectively refer to autism, PDD-NOS, and Asperger syndrome. In this paper, the term autismwill be used, since much of the research is based on childrenwho met criteria for autism. When the term ASD is used, itinclusively refers to autism and PDD-NOS. Asperger syn-drome is not discussed within this review.

DEFINITION OF COMMUNICATIONThe focus of this paper is on the early development of

communication in autism and early intervention for impair-ments in communication associated with this disorder. Com-munication is a broad concept, encompassing linguistic, para-linguistic, and pragmatic aspects of functioning. The linguistic

domain includes phonological, morphological, syntactic, andsemantic rule systems.

Phonological rules establish how speech sounds (pho-nemes) are combined to form words and how a particularspeech sound is to be pronounced given the context of thespeech sounds before and after it. Morphological rulesinvolve signaling grammatical information at the word level,as when words are inflected with past tense markers such as‘‘-ed’’. Syntax is a rule system that guides how words arecombined into sequences and hierarchical structures ofphrases and sentences. The semantic system involves themental ‘‘dictionary’’ of words and their meanings, how tocombine words to form meaningful relationships such as pos-session (‘‘my shoe’’), abstract language processing (includingliteral and nonliteral meaning), and formation of a gist froma text or discourse.

Paralinguistic communication includes proxemics (e.g.,use of space in communication as in distance between speakerand listener), facial expression (e.g., rolling the eyes to indi-cate that a comment was intended as sarcasm, or smiling ascriticism is given to convey tenderness and sincerity so thatthe listener knows that the comment was made out of con-cern rather than merely to criticize), intonation (e.g., using arising intonational contour at the end of a declarative syntac-tic construction to signal that it is to be interpreted as a ques-tion rather than as an assertion), and gesture. Pragmaticsinvolves discourse management (e.g., topic initiation andmaintenance), communicative intentions (variety expressedand understood, as well as variety of forms used to communi-cate intentions), and presupposition (making judgments aboutthe type and style of information presentation depending oncharacteristics of the context, ranging from listener-specificcharacteristics to setting).

Grant sponsor: National Institute of Mental Health; Grant numbers: MH59630,154MH066417.*Correspondence to: Rebecca Landa, 3901 Greenspring Avenue, Baltimore, MD21211. E-mail: [email protected] 29 November 2006; Accepted 11 December 2006Published online in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/mrdd.20134

MENTAL RETARDATION AND DEVELOPMENTAL DISABILITIESRESEARCH REVIEWS 13: 16 – 25 (2007)

' 2007Wiley -Liss, Inc.

DIAGNOSTIC CRITERIA FORAUTISM INVOLVINGCOMMUNICATION

The DSM-IV lists four criteriafor communication impairment in au-tism: (1) delay in, or total lack of, thedevelopment of spoken language; (2)marked impairment in the ability toinitiate or sustain conversation in chil-dren with speech; (3) stereotyped andrepetitive use of language; and (4) lackof varied, spontaneous make-believeplay or social imitative play appropriateto developmental level. Criteria forcommunication impairment, whichencompass the linguistic and discourseaspects of communication, are sepa-rated from the criteria for ‘‘socialimpairment’’. However, some of thecriteria specified for social impairmentare intimately involved in the prag-matic aspects of communication. Forexample, the criterion for social im-pairment involving nonverbal behav-iors would encompass gaze modula-tion, facial expressions, body gestures,and social regulatory gestures. As de-scribed earlier, such behaviors are inti-mately involved in cueing others aboutthe speakers’ communicative intention(i.e., how to interpret what is meantby the words or sentences being spo-ken), cueing others about the speaker’spreparation to terminate a conversa-tional turn so that the partner will notspeak prematurely (which would haveinterrupted the speaker), clarifying oremphasizing a point that is being madelinguistically, and so forth. These samebehaviors are used by listeners to com-municate interest in what the speakerhas to say, to signal confusion or dis-taste about what is being said, to createa synchrony in the communicativeexchange, and so forth. Another socialcriterion, lack of spontaneous sharingof enjoyment, interests, or achieve-ments, melds with communicationbecause such sharing represents a classof social communicative intentions thatare considered by speech-languagepathologists to fall within pragmatics.The social criterion listed in theDSM-IV as ‘‘lack of social or emotionalreciprocity’’ captures an aspect of com-munication that signals attention, in-terest, and engagement between con-versational or communicative partners.In very young children, especiallythose in the prelinguistic stage of de-velopment, the DSM-IV criteria forsocial impairment are particularly con-ceptually linked to the communicationimpairment of autism. During the pre-linguistic stage of development, chil-

dren rely on nonverbal behavior, suchas gaze, facial expression, and bodylanguage (including gesture), to com-municate their needs, wants, and socialintentions. They use gaze to indicatethat they are directing their behaviorto someone or to ‘‘point’’ to the objectof their attention/intention. They usetheir facial expression to indicateintention (e.g., affirmation, protest,request, greeting), urgency, and toinvite the engagement of others. Ges-tures are used to signal communicativeintent and content, to take a commu-nicative turn, and to maintain the‘‘topic’’ of the communicative exchangewith others through, for example,matching the behavior of the partner.

EARLY PREDICTORS OFLANGUAGE DEVELOPMENTIN AUTISM

Early social and communicationdevelopment are intimately intertwined.For example, by 9–10 months of age,infants understand that others’ directionof gaze and pointing gestures signalsomething important, and they shifttheir attention to the object being ‘‘ref-erenced’’ by these behaviors in others,thereby establishing a state of jointattention with another. This ability isimportant for learning that a wordrefers to a particular object. In typicaldevelopment, infants are heavily influ-enced by others’ joint attention cues,and are more likely to associate a newword to an object if the speaker islooking at that object than if his/herattention is not directed to that object[Baldwin, 1991; Baldwin and Moses,2001; Woodward, 2003]. Children withautism are impaired in their ability touse others’ gaze cues in word learningtasks [Baron-Cohen et al., 1997]. In-deed, young children’s joint attentionabilities are useful in predicting a laterdiagnosis of autism. Early joint attentionabilities also are predictive of later lan-guage functioning in typical develop-ment [Tomasello and Todd, 1983], au-tism [Mundy, 1995; Mundy and Gomes,1998; Sigman and Ruskin, 1999; Char-man et al., 2003; Dawson et al., 2004]and in 14-month-old siblings of chil-dren with autism [Sullivan et al., inpress; Note: the relationship betweenautism or milder developmental disrup-tions in siblings of children with autismand nonfamilial autism is not yetknown]. In addition, rate of nonverbalcommunication in 2-year-olds with au-tism is a significant predictor of com-munication and social functioning at

age 7 years [Charman et al., 2005]. Ex-pressive language at age 4 years is alsopredicted by imitation abilities measuredat 2 years of age in children with autism[Stone and Yoder, 2001].

EARLY COMMUNICATIONDEVELOPMENT IN AUTISM

Signs of social and communica-tion disruption may be present in chil-dren with autism as early as the firstyear of life, even before spoken lan-guage is expected to emerge in typicallydeveloping children. Such disruptionmay be seen in the desynchronizationof vocal patterns with the caregiver,early sharing of affective expression[Trevarthen and Daniel, 2005; Yirmiyaet al., 2006], delayed onset of babbling[Iverson and Wozniak, in press], as wellas in use of gestures and responsivenessto the communicative bids of others[Baranek, 1999].

In the second and third years oflife, communication development in au-tism is generally characterized by reducedfrequency and diversity of communica-tive forms, including complex babbling,gestures, consonants in syllables, words,and word combinations [Wetherby et al.,2004; Goldberg et al., 2005; Landa et al.,in press; Mitchell et al., 2006; Wetherbyet al., in press]. Gestures tend to be iso-lated acts, less often integrated withvocalization than in typically developingprelinguistic children [Wetherby et al.,1998]. Initiation of social communicativeacts (e.g., showing, initiating joint atten-tion), which requires integrated attentionto social and nonsocial aspects of context,is impaired relative to requesting (a non-social use of communication) in two-and three-year old children with autism[Wetherby and Prutting, 1984; Lovelandand Landry, 1986; Sigman et al., 1986;Baron-Cohen, 1989; Mundy et al., 1990;McEvoy et al., 1993; Stone et al., 1997;Wetherby et al., 1998]. Thus, childrenwith autism have very restricted meansby which to indicate their needs anddesires to others; this is likely to dramati-cally reduce their effectiveness as com-municators when compared to their agepeers. Their range of options for contin-gently extending the communication ofothers is very limited. They are likely toexhibit decreased flexibility in adaptingand responding to the dynamic, fluidcontext of communication, where com-municative topics and nature of engage-ment shift frequently. The resultingattenuation of ‘‘on topic’’ engagementwith others may compromise the natureand frequency of linguistic and social

MRDD Research Reviews DOI 10.1002/mrdd � COMMUNICATION DEVELOPMENT & INTERVENTION IN AUTISM � LANDA 17

input that children with autism elicitfrom others.

In addition, young children withautism less often initiate communicationbids to regulate the behavior of othersin order to achieve a desired object oraction [Charman et al., 1997; Wetherbyet al., 2004; Landa et al., in press;Wetherby et al., in press]. Likewise, andperhaps even more diagnostically rele-vant for autism, there is a reduced fre-quency of initiation of and response tojoint attention bids to share experiencesand objects of attention [Landa et al., inpress; Lord, 1995; Wetherby et al.,2004; Sullivan et al., in press; Wetherbyet al., in press]. This characteristic alsodifferentiates autism from other devel-opmental disorders from 2 to 5 years ofage [Mundy et al., 1990; Lord, 1995;Charman et al., 1997; Wetherby et al.,1998; Dawson et al., 2004], and is con-sidered to be a core deficit in autism[Sigman et al., 2004].

Some have suggested that youngchildren with autism have particular dif-ficulty in play and developing symbolsinto language, and that these two sys-tems, though distinct, are closely linkedin development [Riguet et al., 1981;Sigman and Ungerer, 1984; Stahmer,1995; Libby et al., 1998]. In the DSM-IV, the impairment in symbolic devel-opment is represented as a diagnosticcriterion for autism within the broaddomain of communication impairment(encompassing primarily linguistic as-pects of communication, but also aspectsof pragmatics and symbol developmentas observed in play). The symbolic im-pairment in autism may be related tothe early receptive and expressive lan-guage delays seen in most children withautism. Over time, however, single-word vocabulary development becomesa relative strength for most childrenwith autism [Lord and Paul, 1997].Children with autism tend to scorehigher on tests of single word vocabu-lary than on tests of complex language[Paul, 1987]. Symbolic skills often doemerge to some degree in autism,although they may be most apparent inhighly structured contexts [Curcio andPischeria, 1978; Ungerer and Sigman,1981; McDonough et al., 1997; Libbyet al., 1998]. Symbolic communicationdifferentiates children with autism fromthose with other developmental disor-ders from 2 to 5 years of age [Mundyet al., 1990; Lord, 1995; Charmanet al., 1997; Wetherby et al., 1998;Dawson et al., 2004], and thus, is con-sidered to be a core deficit in autism[Sigman et al., 2004].

Trajectory of CommunicationDevelopment

Defining the developmental tra-jectory of communication skills in au-tism will yield insights into diagnosti-cally relevant developmental disruptions,providing information pertinent to thedevelopment of early autism interven-tions. Our group prospectively exam-ined receptive and expressive languagedevelopment from 6 to 24 months ofage [Landa and Garrett-Mayer, 2006],and development of communicativeintention and use of gaze and affect incommunication, and variety of commu-nicative forms (gesture, consonants,words, word combinations), from 14 to24 months of age in infants at highgenetic risk for autism (siblings of chil-dren with autism) who received‘‘outcome’’ diagnostic classifications at30 or 36 months of age [Landa et al., inpress; Sullivan et al., in press]. A pro-gressive slowing in rate of receptive andexpressive language development wasnoted between 6 and 24 months in thegroup of children having outcome diag-noses of ASD, distinguishing them fromlanguage delayed and typically develop-ing groups [Landa and Garrett-Mayer,2006]. At the 14-month assessment,about half of the children who had anoutcome classification of ASD receivedtheir first ASD diagnosis. For these chil-dren, no significant gain was observedbetween 14 and 24 months in any as-pect of communication studied [Landaet al., in press]. For the children whoseASD diagnosis was not identified untilafter 14 months of age, a mix of declin-ing and plateauing development insocial aspects of communication wasobserved between 14 and 24 months ofage. This group exhibited minimal gainin linguistic aspects of communicationin this same timeframe.

On the basis of retrospective stud-ies, an atypical developmental trajectoryinvolving regression has been reportedby parents, characterized by diminishingsocial and/or communication skills inthe second and third years of life, affect-ing up to 50% of children with autism[Lotter, 1966; Kurita, 1985; Hoshinoet al., 1987; Tuchman and Rapin, 1997;Davidovitch et al., 2000; Goldberget al., 2003; Luyster et al., 2005; Ozon-off et al., 2005]. Of the children retro-spectively reported to have exhibited re-gressive patterns of early development,20–40% lost language skills [Kurita,1985; Rutter and Lord, 1987]. Lordet al. [2004b] concluded from theirstudy of children with autism, nonaut-ism developmental delay, and typical de-

velopment that word loss appears to beunique to, but not universal in, autism,suggesting that word loss is a ‘‘red flag’’for autism.

A recent study examined the re-gression issue in autism through the an-alysis of first birthday videotapes of tod-dlers with and without parental report ofregression, including a group of typicallydeveloping toddlers [Werner and Daw-son, 2005]. The nonregressive autismgroup differed from the regressive andtypically developing groups in that theyexhibited fewer instances of joint atten-tion and communication. By 24 months,however, abnormality had becomeobservable in the children with reportedregression; they exhibited fewer social orcommunication behaviors than typicallydeveloping children [Werner and Daw-son, 2005]. Prospective research on chil-dren at high risk for autism will shedmore light on the nature of regression inautism, further characterizing this phe-nomenon through direct, longitudinalassessment of children, avoiding thecomplication of recall bias and limita-tions in behavior sampling associatedwith the use of home videotapes.

Most children with autism acquireat least minimal spoken language ability.Of the approximately 80% of childrenwith autism who produce more thanfive spoken words [Lord et al., 2004a],about two thirds are estimated to havelinguistic deficits affecting receptive andexpressive language domains [Allen andRapin, 1980, 1992]. The remainingthird have pragmatic impairments with-out substantial linguistic impairment[Allen and Rapin, 1980, 1992].

IMPLICATIONS FORINTERVENTION

The literature indicates that au-tism can be detected early, and thus,intervention may begin at a young age.Early intervention for the communica-tion impairment in autism is important,since social communication deficits inautism are a major stressor for parents[Bristol and Schopler, 1984], and sincegains in communication skills are relatedto prevention and reduction of malad-aptive behaviors [Carr and Durand,1985; Reichle and Wacker, 1993]. Fur-thermore, degree of language impair-ment impacts clinicians’ impressionsabout comprehensiveness of impair-ment, as reflected in diagnostic classifi-cation as ‘‘autism’’ or ‘‘PDD-NOS’’[Lord et al., 2004a]. Linguistic out-comes are affected by social ability[Bono et al., 2004]. Since social impair-ment is an enduring feature of autism

18 MRDD Research Reviews DOI 10.1002/mrdd � COMMUNICATION DEVELOPMENT & INTERVENTION IN AUTISM � LANDA

[Mundy et al., 1990; Sigman and Rus-kin, 1999], it should be addressed asearly as possible. Through early inter-vention, communication and social de-velopment may be improved [e.g.,Rogers et al., 1986; Smith et al., 2000;Landa and Holman, 2005; Kasari et al.,2006], hopefully leading to better out-comes for the child and family.

Parents and professionals face sev-eral major challenges as they encounterthe task of intervention programming foryoung children with autism. For exam-ple, under what circumstances should achild be enrolled in early intervention?What should the instructional content(targeted goals) of the intervention be?What instructional delivery methodsshould be used to teach the targetedskills? These issues are briefly addressedlater.

WHY EARLY INTERVENTIONFOR AUTISM?

The urgency of early interventionfelt by many parents, clinicians, andresearchers is linked to recent discoveriesin the neurosciences regarding experi-ence-dependent neuroplasticity. This lit-erature demonstrates that experience,especially within social interactions [Kuhlet al., 2003], is related to cortical speciali-zation [Johnson and Munakata, 2005].Such cortical specialization involves fine-tuning of perceptual systems and increas-ing intra- and inter-regional integrationor connectivity. The resulting increas-ingly complex brain circuitry supportsmore complex behaviors and integrationof behaviors across developmental sys-tems. This process appears to be enhancedby exposure to diverse and complex input[Lewis, 2004; Quinn, 2006], yieldingcoherent and flexible patterns of behav-ior. With increasing experience, childrendevelop expertise, which leads to increasedlevels of flexibility and generalization ofknowledge. The result is more contextu-ally appropriate behavior in increasinglynovel conditions [Bloom et al., 1976;Ross, 1982; Rutter and Durkin, 1987;Eckerman and Didow, 1989]. Althoughthese principles have not been systemati-cally examined in intervention studiesinvolving children with autism, they haveimplications for designing interventionswhere diversity and complexity of stimuliwithin learning contexts are systemati-cally engineered, and tailored to thechild’s emerging and established abilities.These principles of typical developmentalso highlight the importance of provid-ing interactive experiences for youngchildren with autism, where skills may befrequently practiced in familiar and

novel, but meaningful, contexts to stimu-late flexible and generalized use of skills.

In autism, a growing body of lit-erature indicates that intervention isassociated with improvements in speech,language, and social development. In-tensity of intervention, defined by num-ber of hours, received by 2-year-oldswith autism was a significant predictorof language outcome at age 4 [Stoneand Yoder, 2001]. Better communica-tion skills translate into better prognosis,a reduction in maladaptive behaviors[Reichle et al., 1991], and new learningopportunities that yield additional accessto information about and throughothers [Yoder and Warren, 1999].

DECISIONS ABOUT WHEN TOENROLL IN INTERVENTION

The signs of autism emerge, oftengradually [Landa and Garrett-Mayer,2006; Landa et al., in press], betweenthe first year of life and the third birth-day. At present, however, autism andPDD-NOS are diagnosed, on average,between 3 and 4 years of age [Mandellet al., 2005]. Autism will be detected atyounger ages because of recent effortsby the Centers for Disease Control andPrevention [2006], Autism Speaks, theAmerican Academy of Pediatrics[2006], and others, to disseminate wide-spread information to the public (Learnthe Signs, Act Early, CDC, 2006), andthe publication of new guidelines fromthe American Academy of Pediatrics[2006] for developmental surveillanceand screening. At present, however,caregivers are usually the first to notedisrupted development in children laterdiagnosed with autism. Parents usuallyexpress concern to their pediatricians ataround 18 months of age, often becausetheir child is a late talker [Rogers andDiLalla, 1990; Wimpory et al., 2000].Unfortunately, unless multiple mile-stones are delayed or a single delay isstriking, professionals often respond toparents’ concerns with a watch-and-seestance. Such a stance may be acceptablein some situations, but certain ‘‘redflags’’ signal the need for a more thor-ough developmental screening, or evenassessment.

Information about ‘‘red flags’’ ofdevelopmental disruption comes fromprospective, longitudinal studies of youngchildren with autism. Wetherby et al.[2004] identified nine red flags that differ-entiated autism from typical or delayeddevelopment at a mean age of 21 months.These red flags included: lack of appro-priate gaze; lack of warm, joyful expres-sions with gaze; lack of sharing enjoy-

ment or interest; lack of response toname; lack of coordination of gaze, facialexpression, gesture, and sound; lack ofshowing; unusual prosody; repetitivemovements of the body; and repetitivemovements with objects. Occasionally, achild may be identified as being at risk forautism based on these red flags, but overtime, these behaviors become less promi-nent or undetectable [Fein et al., 2005].Early intervention providers, togetherwith caregivers, should closely monitorthe developmental trajectory of childrenwho show signs of being at high risk foran autism diagnosis, and implementappropriate developmental stimulation(through parent training and/or directservice delivery to the child) or interven-tion programming at levels of intensitydeemed necessary.

INTERVENTION CONTENTCONSIDERATIONS:NONLINGUISTIC FACTORS

Multiple developmental systemsare impaired in autism, including aspectsof perceptual, motor, cognitive, social,and cross-modal processing systems.Since language development represents atransactive process involving nonlinguis-tic aspects of development, such as imi-tation [Tomasello and Farrar, 1986], de-velopment within one domain isexpected to affect development in otherdomains [Thelen, 2000]. Several aspectsof development have particularly strongconcurrent and predictive relationshipsto language development. These includemotor, imitation, and play development.Below, these are briefly discussed. Directintervention targeting two of these (imi-tation and play) has been documentedto have enhancing effects on communi-cation development [Play: Rogers andLewis, 1989; Stahmer, 1995; Whalenand Schreibman, 2003; Kasari et al.,2006; Imitation: Leaf and McEachin,1999; Ingersoll and Schreibman, 2006].

MotorDisruption in motor development

has been detected in the first year oflife in infant siblings of children withautism [Iverson and Wozniak, in press],in 9–12-month-olds later diagnosedwith autism [Baranek, 1999], and inolder children with autism [e.g., Jansie-wicz et al., 2006], but is not diagnosticof autism. Motor disruptions mayinclude, for example, delayed onset ofearly motor milestones, abnormalitiesin motor tonus, postural instability, ex-cessive mouthing, and poor movementmodulation [Vernazza-Martin et al.,


2005]. These motor differences mayimpede the timing and grading ofmovements involved in contingentcommunication and imitation, result-ing in desynchronized interactionsand, possibly, missed opportunities forengagement and language learning. Insome children, disruption in aspects ofmotor functioning (e.g., involvingplanning) may even contribute to thefailure to acquire spoken language[Rogers et al., 2006].

From early in life, disruption ofthe motor system could have implica-tions for communication development.For example, motor and vocal systemsare tightly coupled in infancy. This cou-pling, or synchronicity, has been shownto be abnormal in infants at high geneticrisk for autism [Iverson and Wozniak, inpress]. Specifically, these infants exhibitedreduced integration of rhythmic limbmovement with vocalization. This dis-ruption in the ‘‘coupling’’ between man-ual and oral/vocal systems early in devel-opment could have implications for thecoordination of vocalization and gestureduring intentional communication acts[Iverson and Thelen, 1999]. In typicaldevelopment, infants’ frequent experi-ence with rhythmic arm movements,along with the stability of this motor pat-tern prior to the emergence of reduplica-tive babbling, may facilitate vocal activity,eventually leading to babbling character-ized by rhythmically organized conso-nant-vowel sequences [Iverson and The-len, 1999]. Infants with autism may notbe able to benefit from such natural facil-itation if their motor development is dis-rupted and if they have difficulty inte-grating behaviors across developmentaldomains.

Intervention for communicationimpairment in autism should take intoconsideration the developing motor sys-tem. Factors such as the child’s posturalcontrol and the positioning require-ments of communication activities, thecomplexity of motor demands of inter-vention activities, and the complexityand familiarity of movements requiredfor vocal, gestural, action schema inplay, or imitative sequences should beconsidered. In early intervention, vocal-ization may be enhanced when com-munication-based activities occur withintasks having predictable limb move-ment, simple rhythmic patterns (e.g.,song-gesture games in which the child’sarms are gently moved in a rhythmicpattern that keeps time with the pro-duction of simple, repetitive vocal pat-terns), ample response time, and securephysical positioning.

ImitationImitation is intimately related to

communication learning [Rogers, 1999].In addition, it provides a vehicle for com-municative reciprocity. For example, animitation of another’s behavior serves toacknowledge their act, confirming atten-tion and responsivity in a reciprocal,meaningfully contingent way. The imita-tion ‘‘invites’’ a response from the otherperson, and thus initiates an interactiveexchange. A ‘‘teachable moment’’ hasarisen. In children with autism, such nat-urally occurring teachable moments arealtogether too sporadic. Despite thesocially disengaged echoic behavior seenin many children with autism, childrenwith autism rarely exhibit spontaneous,meaningful, and socially engaged imita-tion of others’ actions on objects, vocal-izations, and body movements [Dawsonet al., 1998; Smith and Bryson, 1998;Bennetto, 1999; Hobson and Lee, 1999;Aldridge et al., 2000].

Imitation is impaired by 20months of age in autism [Charman et al.,1997]. Impaired ability to imitate othersat such a young age has been shown tobe related to language functioning laterin the preschool years [Stone and Yoder,2001]. Likewise, imitation ability inyoung children with autism is related toother aspects of development (e.g., jointattention and play) that are closelylinked to language development [Love-land and Landry, 1986; Mundy et al.,1990; Baron-Cohen and Swettenham,1997; Sigman and Ruskin, 1999; Char-man et al., 2003].

Targeting imitation in interven-tion for autism is a longstanding prac-tice. Since imitation is a powerful toolfor communication and contingentinteractions with others, the movementpatterns that children are taught to imi-tate should incorporate the movementpatterns needed for targeted communi-cative gestures and for actions onobjects within meaningful play sequen-ces. As skills improve, mapping languageonto the production of the imitatedmovement (e.g., ‘‘bye bye,’’ ‘‘come,’’‘‘big,’’ ‘‘push’’) will strengthen the linkbetween motor and language modalitiesfor communicative purposes.

PlayPlay is the platform for much of

the social engagement that young chil-dren have with others [Doctoroff,1996]. It is also a context for establish-ing social coordination with otherswithin which many communicationexchanges are made [Ross and Lollis,1989]. In autism, play development is

disrupted. This is characterized byreduced object exploration [Pierce andCourchesne, 2001], reduced diversity ofplay acts on objects [Baranek et al.,2000; Stone et al., 1990; Wetherbyet al., in press], fewer novel play acts[Charman and Baron-Cohen, 1997],and delay in symbolic play development[Ungerer and Sigman, 1981]. Play hasbeen found to be concurrently associ-ated with expressive language function-ing as early as 20 months of age[Ungerer and Sigman, 1984; Charmanet al., 1997].

Systematically building the ca-pacity for representational play mayfacilitate the development of representa-tional thought, which is linked to sym-bolic language development. At around18 months of age, children developstrong preferences for combinatorial,constructive, and symbolic play activ-ities. Unlike their age peers, young chil-dren with autism exhibit strong prefer-ences for simple cause-effect activities[Losche, 1990]. To exhibit more sophis-ticated and symbolic play, children withautism often require highly structuredcontexts, within which therapists mayalso stimulate the coordination of playwith symbolic language development.The intervention context providesenriched, scaffolded opportunities forchildren with autism to practice increas-ingly symbolic and decontextualizedplay and language skills, as well as theintegration of these skills.

Intervention programming forplay can be strategically orchestrated todirectly enhance the language learningprocess. This may be accomplishedthrough carefully selecting the objectsto be used for targeting concurrent orfuture linguistic concepts and combina-tions of such concepts into semanticrelations. For example, Samuelson andSmith [2005] have shown that toddlerswith typical development learn andgeneralize novel words best when theinitial referent object has perceptuallysimple features. To the extent that thisprinciple may be generalized to childrenwith autism, the process of building cat-egories that are represented by words(animate or inanimate) may be en-hanced if the exemplar objects are atfirst perceptually simple, and the fea-tures that link the category memberscan be made salient. As concepts repre-sented by nouns are taught to childrenwith autism, sufficient experience witha variety of exemplars will be needed[Quinn, 2006]. Objects that afford a va-riety of actions have relevance to thechild’s daily experience, and lend them-


selves to being combined in thematicsequences of play will be useful in thetherapeutic context. Likewise, objectsshould be considered that foster‘‘teachable moments’’ because they mo-tivate the child’s continued engagement(e.g., toys that easily generate a sensoryexperience, as by shaking a clear tubewith plastic balls, versus a novel switchactivated toy, where the action affordedby the switch may be novel and is dis-tally related to the effect that it acti-vates).

HOW TO DESIGN ANDDELIVER INTERVENTION FORYOUNG CHILDREN WITHAUTISM?

Families of children with autismare faced with a serious dilemma: howmany hours of intervention are neededand what intervention method(s) shouldbe used in their child’s intervention pro-gram? Although the literature addressingthe first question is sparse, the literaturesupports intensive intervention for chil-dren with autism [Lovaas, 1987; Shein-kopf and Siegel, 1998; Harris and Han-dleman, 2000; Smith et al., 2000; Elde-vik et al., 2006]. Number of hours ofspeech-language therapy received betweenages 2 and 4 years is related to the de-velopment of spoken language in chil-dren with ASDs [Stone and Yoder,2001]. Intervention intensity is a some-what elusive concept, since the qualityof the intervention, degree to whichthe child’s attention and engagement aresecured and sustained during therapeu-tic activities, number and nature ofresponse opportunities and other relatedfactors are likely to contribute to the‘‘intensity’’ (or dosage) of intervention,and to intervention response. The waythat these intervention ingredients inter-act with hours of intervention per weekand characteristics of children with au-tism or their parents (e.g., parents’ buy-in to the intervention, or their respon-sivity to their child [Yoder and Warren,2001]) has not been addressed in theempirical literature.

Since autism is a heterogeneousdisorder affecting multiple systems, andsince children with autism have differ-ent needs at different points in their de-velopment, it is unlikely that a singleintervention method will be optimallysufficient for all children with autism[Beglinger and Smith, 2005; Schererand Schreibman, 2005]. Studies are be-ginning to examine the types of childcharacteristics that may be related tolow versus high levels of improvementafter exposure to a particular interven-

tion approach. The most commonlystudied predictor is pretreatment intelli-gence quotient (IQ) [Bibby et al., 2002;Eikeseth et al., 2002; Goldstein, 2002].Younger age at onset of intervention[Bibby et al., 2002; Goldstein, 2002;Harris and Handleman, 2000) and rapidlearning, particularly involving imitationand receptive language, during the first3–4 months of intervention [Newsomand Rincover, 1989; Weiss, 1999] havebeen reported as predictors of positiveoutcome. Recently, pretreatment socialbehavior has been identified as a predic-tor of outcome. For example, Beglingerand Smith [2005] examined the rela-tionship between social subtype of au-tism and IQ change in early intensivebehavioral intervention, where childrenreceived 30–40 hr of 1:1 interventioninvolving discrete trial teaching, usuallywithin the home. They found that chil-dren with autism who were categorizedas ‘aloof ’ according to Wing and Gould’s(1979) subtyping guidelines showed thesmallest gain in IQ (7.6 points) whencompared to IQ gains of 19.25, 27.17,and 36.75 points for children in the au-tism subtype categories of Passive,Active-but-odd, and Typically develop-ing based on parent report on the WingSubgroups Questionnaire [Castelloe andDawson, 1993]. In a similar vein, Sal-lows and Graupner [2005] reported thatoutcome for children enrolled in an in-tensive behavioral intervention (ABA)was predicted by pretreatment imitation,language, and social responsiveness.

Studies have not yet examined theimpact of changing intervention ap-proach for children showing no or slowresponse to a certain instructional method.Thus, there are no prescriptive formulasto help providers select which interven-tion method(s) to use for children withparticular behavioral profiles. At Ken-nedy Krieger’s Center for Autism andRelated Disorders, early intervention isguided by ‘‘feature match’’ design, wherethe instructional features are selected toaddress individual children’s patterns ofrelative strength and impairment. Evi-dence to support this approach has beenreported [Rogers et al., 1986; Rogers,1996; Stahmer and Ingersoll, 2004;Landa and Holman, 2005]. A featurematch design cannot be equated withwhat the literature refers to as an‘‘eclectic’’ approach [e.g., Eldevik et al.,2006], where a variety of educationalsettings and instructional approaches arerepresented in a child’s interventionrepertoire, but without evidence thattherapists have mastered the interven-tion techniques.

Below, major intervention ap-proaches to autism are summarized. Theyare conceptually grouped into: traditionalapplied behavior analysis (ABA), contem-porary ABA, transactional, and comple-mentary approaches. To greater or lesserdegrees, these approaches achieve theguidelines presented by the NationalResearch Council [2001] on EducatingChildren with Autism.

TRADITIONAL ABAINTERVENTIONS

Traditional ABA, often referred toas Discrete Trial Teaching or Early In-tensive Behavioral Intervention, is anestablished and extensively studied inter-vention method [Smith, 2001; Gold-stein, 2002]. It is based on principles ofoperant conditioning [Newsom, 1998]where skills are dissected into discreteintervention targets based on task analy-sis and the child’s task performance.Intervention targets are addressed throughmassed trials of antecedent–behavior–consequence chains, initiated by the adult,using adult-selected materials and tasks,and presented in massed trials to promotesuccess. The therapist maintains tight con-trol over antecedent stimuli, prompt hier-archy, and reinforcers, which are usuallynot specifically related to the content ofthe child’s behavior. Teaching occurswithin a nondistracting environment, dis-embedding behavior from meaningfulactivities during initial skill acquisition.After initial skill acquisition, the emphasisis on systematically generalizing skills toactivities typical of the child’s daily life.Curriculum manuals provide step-by-stepguidelines for teaching component skills,usually within the domains of language,nonverbal cognitive, and preacademic skills[Partington and Sundberg, 1998; Lovaas,2003]. Studies of the effects of traditionalbehavior analytic intervention deliveredfor a period of 1–4 years, with 30–40 hrper week of 1:1 intervention, report anaverage IQ gain of 20 points for pre-schoolers with autism [Lovaas, 1987; Sal-lows and Graupner, 2005], with greaterimprovement being associated with greaterintensity (e.g., 40 hr per week versus 10hr per week) (e.g., McEachin et al., 1993;Smith et al., 2000; Howard et al., 2005;Cohen et al., 2006]. The number of chil-dren attaining age-appropriate IQ (usuallydefined as 85) by the end of the studyranges from 0 to about 50%. Short-termdiscrete trial teaching has been successfulin teaching 4- to 6-year-olds with autismto combine gesture with simple verbalresponses [Buffington et al., 1998]. Moreresearch is needed to assess the impact ofthis intervention approach on social func-


tioning, particularly in core deficits of au-tism such as joint attention and affectiverelatedness.

Contemporary ABA Approaches tointervention modify the execution ofABA principles such that intervention iscentered around the child’s interests andactivities. This variation in techniquewas largely a response to children’s diffi-culties in generalizing skills from thehighly structured, adult-led therapyactivities that were disembedded frommeaningful contexts in the traditionalABA approach. Contemporary applica-tions of ABA are seen in approachesknown as Incidental Teaching [McGeeet al., 1999], Natural Language Para-digm [Koegel et al., 1987], PivotalResponse Training [Koegel and Koegel,1995; Schreibman and Pierce, 1993],and the Milieu Teaching approach[Warren and Bambara, 1989]. Theyemphasize increasing children’s ‘‘moti-vation’’ to communicate, and are basedon the literature on development, prag-matics, and ABA. Contemporary ABAapproaches employ strategies aimed atfacilitating spontaneous language andcommunication development and focus-ing on the child’s role as an active com-munication partner (motivation), usingnatural rewards, embedding teachingactivities within natural environments(including group contexts), identifyingtopics for communicative exchange thatinvolve child-preferred and child-se-lected activities, where interactions aremore natural and fluidly structured toencourage child-initiated communica-tion. These approaches are also designedto easily capitalize on spontaneity, butstrategically infuse the teaching contextwith opportunities to increase frequencyand duration of responses and engage-ment. Empirical support for these pro-grams with children with autism is alsoavailable [Koegel et al., 1987; Warrenet al., 1994; Pierce and Schreibman,1995; Fey et al., 2006], with evidencethat generalization of some communica-tion skills is increased when teachingtrials are embedded throughout the dayin meaningful communicative contexts[e.g., Neef et al., 1984]. Furthermore,preliminary evidence has been reportedby Stahmer and Ingersoll [2004] thatearly intervention based on contempo-rary ABA [specifically, McGee’s Waldenprogram; McGee et al., 1999] may ben-efit early communication developmentin young children with autism whenother intervention strategies are incor-porated [e.g., Picture Exchange Com-munication System; Bondy and Frost,2003] within a classroom setting where

typically developing peers are included.Stahmer and Ingersoll’s (2004) interven-tion study did not employ an experi-mental design, so more research isneeded to evaluate the benefit of such aprogram for communication, cognitive,adaptive, and social functioning ofyoung children with autism.

Another approach based on prin-ciples of ABA is the Picture ExchangeCommunication System [PECS; Bondyand Frost, 2003], providing a visuallybased alternative/augmentative commu-nication system for children with lim-ited speech. Children are taught toexchange a single picture for a desireditem and eventually to construct pic-ture-based sentences using planned gen-eralization behavioral principles. Evi-dence is emerging that PECs is a usefulsystem for developing aspects of com-munication development for some chil-dren with autism [Ganz and Simpson,2004; Rogers et al., 2006].

Transactional, developmental, andsocial-pragmatic approaches view languagelearning as a co-created process sharedby child and other. Perhaps the mostdistinguishing features of the transac-tional/developmental approaches pertainto the emphasis on reciprocal, affective,self-regulatory, relationship-building,and discovery processes. Through inter-actions with others, such as joint actionroutines, shared experiences and mean-ings are developed. The interventioncontext is the natural environment,with instruction embedded within day-to-day natural routines. An emphasis, asin other approaches, is on building fromconcrete to abstract concepts, with par-ticular focus on integration of these lan-guage- and socially based concepts intoa variety of meaningful experiences,thus promoting generalization. Theanticipated result is the development ofa well-rounded communication systemwhere a variety of communicative in-tentions are initiated using a variety oflinguistic and nonlinguistic forms acrossa variety of contexts. Transactional/de-velopmental intervention methods areoften characterized by multi-modalintegration of sensory stimulating, motorplanning, and visual input augmenta-tion, as well as family involvement.Approaches falling into this categoryinclude Floortime [Greenspan andWeider, 2003], Social Communication,Emotional Regulation and TransactionalSupport (SCERTS; Prizant et al., 2006],the Denver Model [Rogers et al., 1986],and the Miller Method [Miller and Eller-Miller, 2000]. There have been no exper-imental studies that support the efficacy of

these methods, but quasi-experimentalevidence indicates that a well-executedand defined developmentally based ap-proach, such as the Denver model, is asso-ciated with developmental gains for chil-dren with autism [Rogers et al., 1986].

Complementary approaches to thosedescribed earlier may include environ-mental engineering, augmentative com-munication systems, sensory-motor in-terventions, and more. One of the mostwidely used approaches for engineeringthe learning environment is the Treat-ment and Education of Autistic andrelated Communication-handicappedChildren (TEACCH) program [Scho-pler and Mesibov, 1984; Ozonoff andCathcart, 1998]. The features of thisapproach include physical organizationof the teaching environment, schedulesof activity that increase organizationand predictability, individual work sta-tions to promote independent goal-directed activity, and learning task orga-nization using visually guided cues tosuccessful task completion. Anotherapproach provides guidelines and strat-egies for facilitating children’s ability toautomatically process complex sensoryinformation, improve motor coordina-tion, reduce over-or under-reactivity,and improve emotional adjustment aswell as social functioning. These ap-proaches, referred to as Sensory Integra-tion [Ayres and Mailloux, 1981] or Sen-sory Registration, provide systematicand individualized ‘‘doses’’ and types ofsensory experience, coordinating sensa-tion with motor planning, using a vari-ety of equipment and sensory-enhanc-ing materials tailored to a child’s inter-ests. These principles are incorporatedinto some of the Developmental ap-proaches (such as Floortime, SCERTS,and the Miller Method), but their effi-cacy has not been examined in autism.

SUMMARY AND CONCLUSIONSThe communication impairment

in autism is universal, appears early, andaffects multiple aspects of development.Fortunately, autism diagnostic indicatorsfor young children are being revealedby research, so earlier identification ofautism is becoming a reality. Communi-cation intervention for children withautism will envelop many aspects of de-velopment, including social engage-ment, social reciprocity, joint attention,imitation, play, vocal-manual coordina-tion, language, flexible communicativecontingencies, and social communica-tive abilities. A variety of instructionalapproaches are available for use in tar-geting communication and related goals,


some of which are evidence-based.Now, more than ever before, clinicianshave resources from which to draw inthe design and delivery of individualizedcommunication interventions for chil-dren with autism. n

ACKNOWLEDGMENTSAppreciation is expressed to Alli-

son O’Neill for her administrative assis-tance in the preparation of this manu-script.

REFERENCESAldridge MA, Stone KR, SweeneyMH, et al. 2000.

Preverbal children with autism understandintentions of others. Dev Science 3:294–301.

Allen D, Rapin I. 1980. Language disorders inpreschool children: Predictors of outcome.A preliminary report. Brain Dev 2:73–80.

Allen D, Rapin I. 1992. Autistic children are alsodysphasic. In: Naruse H, Ornitz E, editors.Neurobiology of infantile autism. Amster-dam: Excerpta Medica. p 73–80.

American Academy of Pediatrics. 2006. Identify-ing infants and young children with devel-opmental disorders in the medical home: Analgorithm for developmental surveillance andscreening. Pediatrics 118:405–420.

American Psychiatric Association. 1994. Diagnos-tic and statistical manual of mental disorders,4th ed. Washington, DC: American Psychi-atric Association.

Ayres AJ, Mailloux Z. 1981. Influence of sensoryintegration procedures on language develop-ment. Am J Occup Ther 35:383–390.

Baldwin DA, Moses LJ. 2001. Links betweensocial understanding and early word learn-ing: Challenges to current accounts. SocDev 10:309–329.

Baldwin DA. 1991. Infants’ contribution to theachievement of joint reference. Child Dev62:875–890.

Baranek GT, Reinhartsen DB, Wannamaker SW.2000. Play: Engaging young children withautism. In: Heubner R, editor. Autism andrelated disorders: A sensorimotor approachto management. Gaithersburg, MD: Aspen.

Baranek GT. 1999. Autism during infancy: A ret-rospective video analysis of sensory-motorand social behaviors at 9–12 months of age.J Autism Dev Disord 29:213–24.

Baron-Cohen S, Baldwin DA, Crowson M. 1997.Do children with autism use the speaker’sdirection of gaze strategy to crack the codeof language? Child Dev 68:48–57.

Baron-Cohen S, Swettenham J. 1997.Theory ofmind in autism: Its relationship to executivefunction and central coherence. In: CohenDJ, Volkmar FR, editors. Handbook of au-tism and pervasive developmental disorders,2nd ed. New York: Wiley. p 880–893.

Baron-Cohen S. 1989. Perceptual role taking andprotodeclarative pointing in autism. Br J DevPsychol 7:113–127.

Beglinger L, Smith T. 2005. Concurrent validityof social subtype and IQ after early intensivebehavioral intervention in children with au-tism: A preliminary investigation. J AutismDev Disord 35:295–303.

Bennetto L. 1999. A componential approach toimitation and movement deficits in autism.Diss Abstr Int B 60(2):0819.

Bibby P, Eikeseth S, Martin NT, et al. 2002. Pro-gress and outcomes for children with autismreceiving parent-managed intensive interven-tions. Res Dev Disabil 23:81–104.

Bloom L, Rocissano L, Hood L. 1976. Adult-child discourse: Developmental interactionbetween information processing and linguis-tic knowledge. Cognit Psychol 8:521–552.

Bondy A, Frost A. 2003.Communication strat-egies for visual learners. In: Lovaas OI, edi-tor. Teaching individuals with developmentaldelays: Basic intervention techniques. Austin,TX: Pro-Ed, p 291–304.

Bono M, Daley T, Sigman M. 2004. Relationsamong joint attention, amount of interven-tion, and language gain in autism. J AutismDev Disord 34:495–505.

Bristol MM, Schopler E. 1984. A developmentalperspective on stress and coping in families ofautistic children. In: Blacher J, editor. Severelyhandicapped young children and their families.Orlando, FL: Academic Press. p 91–141.

Buffington DM, Krantz PJ, McClannahan LE,et al. 1998. Procedures for teaching appro-priate gestural communication skills to chil-dren with autism. J Autism Dev Disord 28:535–545.

Carr EG, Durand VM. 1985. Reducing behaviorproblems through functional communicationtraining. J Appl Behav Anal 18:111–126.

Castelloe P, Dawson G. 1993. Subclassification ofchildren with autism and pervasive develop-mental disorder: A questionnaire based onWing’s subgrouping scheme. J Autism DevDisord 23:569–583.

Centers for Disease Control and Prevention (n.d.).Learn the Signs, Act Early. Available from www.cdc.gov/ncbddd/autism/actearly/. Accessed July19, 2006.

Charman T, Baron-Cohen S, Swettenham J, et al.2003. Predicting language outcome ininfants with autism and pervasive develop-mental disorder. Int J Lang Commun Disord38:265–285.

Charman T, Baron-Cohen S. 1997. Brief report:Prompted pretend play in autism. J AutismDev Disord 27:325–332.

Charman T, Swettenham J, Baron-Cohen S, et al.1997. Infants with autism: An investigationof empathy, pretend play, joint attention, andimitation. Dev Psychol 33:781–789.

Charman T, Taylor E, Drew A, et al. 2005. Out-come at 7 years of children diagnosed withautism at age 2: Predictive validity of assess-ments conducted at 2 and 3 years of age andpattern of symptom change over time. JChild Psychol Psychiatry 46:500–513.

Cohen H, Amerine-Dickens M, Smith T. 2006.Early intensive behavioral treatment: Repli-cation of the UCLA model in a communitysetting. J Dev Behav Pediatr 27:S145–S155.

Curcio F, Pischeria EA. 1978. Pantomimic repre-sentation in psychotic children. J AutismChild Schizophr 8:181–189.

Davidovitch M, Glick L, Holtzman G, et al.2000. Developmental regression in autism:Maternal perception. J Autism Dev Disord30:113–119.

Dawson G, Meltzoff A, Osterling J, Rinaldi J. 1998.Neuropsychological correlates of early symp-toms of autism. Child Devel 69:1276–1285.

Dawson G, Toth K, Abbott R, et al. 2004. Earlysocial attention impairments in autism: Socialorienting, joint attention and attention to dis-tress. Dev Psychopathol 40:271–283.

Doctoroff S. 1996. Parents’ support of infants’object play: A review of the literature. InfantToddler Interv 6:153–166.

Eckerman CO, Didow SM. 1989. Toddlers’ socialcoordinations: Changing responses to anoth-er’s invitation to play. Dev Psychol 25:794–804.

Eikeseth S, Smith T, Jahr E, et al. 2002. Intensivebehavioral treatment at school for 4- to 7-year old children with autism: A one-yearcomparison controlled study. Behav Modif26:49–68.

Eldevik S, Eikeseth S, Jahr E, et al. 2006. Effectsof low-intensity behavioral treatment forchildren with autism and mental retardation.J Autism Dev Disord 36:211–224.

Fein D, Dixon P, Paul J, et al. 2005. Brief report:Pervasive developmental disorder can evolveinto attention deficit/hyperactivity disorder(ADHD): Case illustrations. J Autism DevDisord 35:525–534.

Fey ME, Warren SF, Brady N, et al. 2006. Earlyeffects of responsivity education/prelinguisticmilieu teaching for children with develop-mental delays and their parents. J SpeechLang Hear Res 49:526–547.

Ganz J, Simpson R. 2004. Effects on communica-tive requesting and speech development ofthe picture exchange communication systemin children with characteristics of autism.J Autism Dev Disord 34:395–409.

Goldberg WA, Jarvis KL, Osann K, et al. 2005.Brief report: Early social communicationbehaviors in the younger siblings of childrenwith autism. J Autism Dev Disord 35:657–664.

Goldberg WA, Osann K, Filipek PA, et al. 2003.Language and other regression: Assessmentand timing. J Autism Dev Disord 33:607–616.

Goldstein H. 2002. Communication interventionfor children with autism: A review of treat-ment efficacy. J Autism Dev Disord 32:373–396.

Greenspan SI, Wieder S. 2003. Engaging autism:The floortime approach to helping childrenrelate, communicate, and think. Cambridge,MA: De Capo.

Harris SL, Handleman JS. 2000. Age and IQ atintake as predictors of placement for youngchildren with autism: A four- to six-yearfollow-up. J Autism Dev Disord 30:137–142.

Hobson RP, Lee A. 1999. Imitation and identifi-cation in autism. J Child Psychol Psychiatry40:649–660.

Hoshino Y, Kaneko M, Yashima Y, et al. 1987.Clinical features of autistic children with set-back course in their infancy. Jpn J PsychiatryNeurol 41:237–245.

Howard JS, Sparkman CR, Cohen HG, et al.2005. A comparison of intensive behavioranalytic and eclectic treatments for youngchildren with autism. Res Dev Disabil 26:359–383.

Ingersoll B, Schreibman L. 2006. Teaching recip-rocal imitation skills to young children withautism using a naturalistic behavioral ap-proach: Effects on language, pretend play,and joint attention. J Autism Dev Disord 36:487–505.

Iverson J, Thelen, E. 1999. Hand, mouth, andbrain: The dynamic emergence of speechand gesture. J Conscious Stud 6:19–40.

Iverson J, Wozniak RH. Variation in vocal-motordevelopment in infant siblings of childrenwith autism. J Autism Dev Disord (in press).

Jansiewicz EM, Goldberg MC, Newschaffer CJ,et al. 2006. Motor signs distinguish childrenwith high functioning autism and Asperger’sSyndrome from controls. J Autism Dev Dis-ord 36:613–621.


Johnson MH, Munakata Y. 2005. Processes ofchange in brain and cognitive development.Trends Cogn Sci 9:152–158.

Kasari C, Freeman S, Paparella T. 2006. Jointattention and symbolic play in young chil-dren with autism: A randomized controlledintervention study. J Child Psychol Psychia-try 47:611–620.

Koegel RL, Koegel LK. 1995. Teaching childrenwith autism. Baltimore: Brookes.

Koegel RL, O’Dell MC, Koegel LK. 1987. Anatural language teaching paradigm for non-verbal autistic children. J Autism Dev Disord17:187–200.

Kuhl PK, Tsao FM, Liu HM. 2003. Foreign-languageexperience in infancy: Effects of short-term ex-posure and social interaction on phonetic learn-ing. Proc Natl Acad Sci USA 100:9096–9101.

Kurita H. 1985. Infantile autism with speech lossbefore the age of thirty months. J Am AcadChild Adolesc Psychiatry 24:191–196.

Landa R, Garrett-Mayer E. 2006. Developmentin infants with autism spectrum disorders: Aprospective study. J Child Psychol Psychiatry47:629–638.

Landa R, Holman K, Garrett-Mayer E. Earlysocial and communication development asso-ciated with early and later onset of autism.Arch Gen Psychiatry (in press).

Landa R, Holman K. 2005. The effects of target-ing interpersonal synchrony on social andcommunication development in toddles withautism. Presented at the Annual CPEA/STAART meeting, Washington, DC, 2005.

LeafR,McEachin J, editors. 1999. Awork in progress:Behavioral management strategies and a curric-ulum for intensive behavioral treatment of au-tism. NewYork: Different Roads to Learning.

Lewis MH. 2004. Environmental complexity andcentral nervous system (CNS) developmentand function. Ment Retard Dev Disabil ResRev 10:91–95.

Libby S, Powell S, Messer D, et al. 1998. Sponta-neous play in children with autism: A reap-praisal. J Autism Dev Disord 28:487–497.

Lord C. 1995. Follow up of two year olds referredfor possible autism. J Child Psychol Psychia-try 36:1365–1382.

Lord C, Paul R. 1997. Language and communica-tion in autism. In: Cohen DJ, Volkmar FR,editors. Handbook of autism and pervasivedevelopmental disorders, 2nd ed. New York:Wiley. pp 195–225.

Lord C, Risi S, Pickles A. 2004a. Trajectory oflanguage development in autistic spectrumdisorders. In: Rice M, Warren S, editors.Developmental language disorders. New Jer-sey: Erlbaum. p 7–29.

Lord C, Shulman C, Dilavore P. 2004b. Regressionand word loss in autistic spectrum disorders.J Child Psychol Psychiatry 45:936–955.

Losche G. 1990. Sensorimotor and action devel-opment in autistic children from infancy toearly childhood. J Child Psychol Psychiatry31:749–761.

Lotter V. 1966. Epidemiology of autistic condi-tions in young children. Soc Psychiatry 1:124–137.

Lovaas OI. 1987. Behavioral treatment and nor-mal educational and intellectual functioningin young autistic children. J Consult ClinPsychol 55:3–9.

Lovaas OI. 2003. Teaching individuals with devel-opmental delays: Basic intervention techni-ques. Austin, TX: Pro-Ed.

Loveland K, Landry SH. 1986. Joint attention andlanguage in autism and developmental languagedelay. J Autism Dev Disord 16:335–349.

Luyster R, Richler J, Risi S, et al. 2005. Earlyregression in social communication in autismspectrum disorders: A CPEA Study. DevNeuropsychol 27:311–36.

Mandell DS, Novak MM, Zubritsky CD. 2005.Factors associated with age of diagnosisamong children with autism spectrum disor-ders. Pediatrics 116:1480–1486.

McDonough L, Stahmer A, Schreibman L, et al.1997. Deficits, delays, and distractions: Anevaluation of symbolic play and memory inchildren with autism. Dev Psychopathol 9:17–41.

McEachin JJ, Smith T, Lovaas O. 1993. Long-term outcome for children with autism whoreceived early intensive behavioral treatment.Am J Ment Retard 97:359–372.

McEvoy RE, Rogers SJ, Pennington BF. 1993.Executive function and social communica-tion deficits in young autistic children. JChild Psychol Psychiatry 34:563–578.

McGee GG, Morrier MJ, Daly T. 1999. An inci-dental teaching approach to early interven-tion for toddlers with autism. J Assoc PersSev Handicaps 24:133–146.

Miller A, Eller-Miller E. 2000. The Miller method:A cognitive-developmental systems approachfor children with body organization, socialand communication issues. In: Greenspan S,Weider S, editors. ICDL clinical practicesguidelines: Revising the standards of practicefor infants, toddlers and children with devel-opmental challenges. Bethesda, MD: ICDLPress. p 489–516.

Mitchell S, Brian J, Zwaigenbaum L, et al. 2006.Early language and communication develop-ment of infants later diagnosed with autismspectrum disorder. J Dev Behav Pediatr 27:S69–S78.

Mundy P, Gomes A. 1998. Individual differencesin joint attention skill development in thesecond year. Infant Behav Dev 21:469–482.

Mundy P, Sigman M, Kasari C. 1990. A longitu-dinal study of joint attention and languagedevelopment in autistic children. J AutismDev Disord 20:115–128.

Mundy P. 1995. Joint attention and social-emo-tional approach behavior in children withautism. Dev Psychopathol 7:63–82.

National Research Council. 2001. Educating chil-dren with autism. In: Lord C, McGee JP,editors. Washington, DC: Committee onEducational Interventions for Children withAutism, Division of Behavioral and SocialSciences and Education, National AcademyPress.

Neef NA, Walters J, Egel AL. 1984. Establishinggenerative yes/no responses in developmen-tally disabled children. J Appl Behav Anal17:453–460.

Newsom C, Rincover A. 1998. Autism. In: MashEJ, Barkley RA, editors. Treatment of child-hood disorders. New York: Guilford. p 286–346.

Ozonoff S, Cathcart K. 1998. Effectiveness of ahome program intervention for young chil-dren with autism. J Autism Dev Disord28:25–32.

Ozonoff S, Williams B, Landa R. 2005. Parentalreport of the early development of childrenwith regressive autism: The delays-plus-regression phenotype. Autism 9:461–486.

Partington JW, Sundberg, ML. 1998. The assess-ment of basic language and learning skills:An assessment, curriculum guide, and track-ing system for children with autism or otherdevelopmental disabilities. Danville, CA:Behavior Analysts Inc.

Paul R. 1987. A comparison of language charac-teristics of mentally retarded adults withFragile X Syndrome and those with nonspe-cific MR and autism. J Autism Dev Disord17:457–469.

Pierce K, Courchesne E. 2001. Evidence for acerebellar role in reduced exploration andstereotyped behavior in autism. Soc BiolPsychiatry 49:655–664.

Pierce K, Schreibman L. 1995. Increasing com-plex social behaviors in children with au-tism: Effects of peer-implemented pivotalresponse training. J Appl Behav Anal 28:285–295.

Prizant BM, Wetherby AM, Rubin E, et al. 2006.The SCERTS MODEL: A comprehensiveeducational approach for children with au-tism spectrum disorders. Baltimore: Brookes.

Quinn PC. 2006.On the emergence of perceptualorganization and categorization in younginfants: Roles for perceptual process andknowledge access. In: Balter L, Tamis-LeMonda C, editors. Child psychology: ahandbook of contemporary issues, 2nd ed.Philadelphia: Psychology Press. p 109–131.

Reichle J, Wacker DP. 1993. Communicativealternatives to challenging behavior: Inte-grating functional assessment and interven-tion strategies, Vol. 3. Baltimore: Brookes.

Reichle J, York J, Sigafoos J. 1991. Implementingaugmentative and alternative communica-tion: Strategies for learners with severe dis-abilities. Baltimore: Brookes.

Riguet CB, Taylor ND, Benaroya S, et al. 1981.Symbolic play in autistic, Down’s, and nor-mal children of equivalent mental age. J Au-tism Dev Disord 11:439–448.

Rogers SJ, DiLalla, D. 1990. Age of symptomonset in young children with pervasive de-velopmental disorders. J Am Acad ChildAdolesc Psychiatry 29:863–872.

Rogers SJ, Hayden D, Hepburn S, et al. 2006.Teaching young nonverbal children with au-tism useful speech: A pilot study of the Den-ver model and PROMPT interventions. JAutism Dev Disord.

Rogers SJ, Herbison J, Lewis H, et al. 1986. Anapproach for enhancing the symbolic, com-municative, and interpersonal functioning ofyoung children with autism and severe emo-tional disorders. J Div Early Childhood 10:135–148.

Rogers SJ, Lewis H. 1989. An effective day treat-ment model for young children with perva-sive developmental disorders. J Am AcadChild Adolesc Psychiatry 28:207–217.

Rogers SJ. 1996. Brief report: Early interventionin autism. J Autism Dev Disord 26:243–246.

Rogers SJ. 1999.An examination of the imitationdeficit in autism. In: Nadel J, ButterworthG, editors. Imitation in infancy. Cambridge:Cambridge University Press. p 255–283.

Ross HS, Lollis SP. 1989. A social relations analy-sis of toddler peer relationships. Child Dev60:1082–1091.

Ross HS. 1982. Toddler peer relations: Differen-tiation of games and conflicts. Can J BehavSci 14:364–379.

Rutter DR, Durkin K. 1987. Turn-taking inmother-infant interaction: An examination ofvocalizations and gaze. Dev Psychol 23:54–61.

Rutter M, Lord C. 1987. Language disorders asso-ciated with psychiatric disturbance. In: YaleW, Rutter M, eds. Language developmentand disorders. Philadelphia, PA: Lippincott.p 206–233.

Sallows GO, Graupner TD. 2005. Intensive be-havioral treatment for children with autism:


Four-year outcome and predictors. Am JMent Retard 6:417–438.

Samuelson LK, Smith LB. 2005. They call it likethey see it: Spontaneous naming and atten-tion to shape. Dev Sci 8:182–198.

Scherer MR, Schreibman L. 2005. Individual be-havioral profiles and predictors of treatmenteffectiveness for children with autism. JConsult Clin Psychol 73:525–538.

Schopler E, Mesibov G, editors. The effects of au-tism on the family. NewYork: Plenum, 1984.

Schreibman L, Pierce K. 1993. Achieving greatergeneralization of treatment effects in childrenwith autism: Pivotal response training andself-management. Clin Psychol 46:184–191.

Shienkopf SJ, Siegel B. 1998. Home based behav-ioral treatment of young children with au-tism. J Autism Dev Disord 28:15–23.

Sigman M, Dijamco A, Gratier M, et al. 2004. Earlydetection of core deficits in autism. MentRetard Dev Disabil Res Rev 10:221–233.

Sigman M, Dijamco A, Gratier M, Rozga A.2004. Early detection of core deficits in au-tism. Ment Retard Dev Disabil Res Rev 10:221–233.

Sigman M, Mundy P, Sherman T, et al. 1986. Socialinteractions of autistic, mentally retarded, andnormal children and their caregivers. J ChildPsychol Psychiatry 27:647–655.

Sigman M, Ruskin E. 1999. Continuity andchange in the social competence of childrenwith autism, Down syndrome, and develop-mental delays. Monogr Soc Res Child Dev64:1–114.

Sigman M, Ungerer J. 1984. Cognitive and lan-guage skills in autistic, mentally retarded, andnormal children. Dev Psychol 20:293–302.

Smith IM, Bryson SE. 1998. Gesture imitation inautism I: Nonsymbolic postures and sequen-ces. Cogn Neuropsychol 15:747–770.

Smith T, Groen AD, Wynn JW. 2000. Random-ized trial of intensive early intervention forchildren with pervasive developmental disor-der. Am J Ment Retard 105:269–285.

Smith T. 2001. Discrete trial training in the treat-ment of autism. Focus Autism Other DevDisabl 16:86–92.

Stahmer A, Ingersoll B. 2004. Inclusive program-ming for toddlers with autistic spectrum dis-orders: Outcomes from the children’s toddlerschool. J Pos Behav Interv 6:67–82.

Stahmer AC. 1995. Teaching symbolic play skills tochildren with autism using pivotal responsetraining. J Autism Dev Disord 25:123–141.

Stone W, Lemanek K, Fishel P, et al. 1990. Playand imitation skills in the diagnosis of autismin young children. Pediatrics 86:267–272.

Stone W, Ousley O, Littleford C. 1997. Motorimitation in young children with autism:What’s the object? J Abnorm Child Psychol25:475–485.

Stone W, Yoder P. 2001. Predicting spoken lan-guage level in children with autism spectrumdisorders. Autism 5:341–361.

Sullivan M, Finelli J, Marvin A, et al. Responseto joint attention as an early marker for au-tism spectrum disorder: A prospective study.J Autism Dev Disord (in press).

Thelen E. 2000. Grounded in the world: Devel-opmental origins of the embodied mind.Infancy 1:3–28.

Tomasello M, Farrar MJ. 1986. Joint attention andearly language. Child Dev 57:1454–1463.

Tomasello M, Todd J. 1983. Joint attention andlexical acquisition style. First Lang 4:197–212.

Trevarthen C, Daniel S. 2005. Disorganizedrhythm and synchrony: Early signs of autismand Rett syndrome. Brain Dev 27:S25–S34.

Tuchman R, Rapin I. 1997. Regression in perva-sive developmental disorders: Seizures andepileptiform electroencephalogram corre-lates. Pediatrics 99:560–566.

Ungerer JA, Sigman M. 1981. Symbolic play andlanguage comprehension in autistic children.J Am Acad Child Psychiatry 20:318–337.

Ungerer JA, Sigman M. 1984. The relation of playand sensorimotor behavior to language in thesecond year. Child Dev 55:1448–1455.

Vernazza-Martin S, Martin N, Vernazza A, et al.2005. Goal directed locomotion and balancecontrol in autistic children. J Autism DevDisord 35:91–102.

Warren SF, Bambara LM. 1989. An experimentalanalysis of milieu language intervention:Teaching the action-object form. J SpeechHear Disord 54:448–461.

Warren SF, Gazdag GE, Bambara LM, et al. 1994.Changes in the generativity and use ofsemantic relationships concurrent with mi-lieu language intervention. J Speech HearRes 37:924–934.

Weiss MJ. 1999. Differential rates of skill acquisi-tion and outcomes of early intensive behav-ioral intervention for autism. Behav Interv14:3–22.

Werner E, Dawson G. 2005. Validation of thephenomenon of autistic regression usinghome videotapes. Arch Gen Psychiatry 62:889–895.

Wetherby A, Prizant B, Hutchinson T. 1998.Communicative, social/affective, and sym-bolic profiles of young children with autismand pervasive developmental disorders. Am JSpeech Lang Pathol 7:79–91.

Wetherby A, Prutting C. 1984. Profiles of com-municative and cognitive-social abilities inautistic children. J Speech Hear Res 27:364–377.

Wetherby A, Watt N, Morgan L, et al. Socialcommunication profiles of children with au-tism spectrum disorders late in the secondyear of life. J Autism Dev Disord (in press).

Wetherby A, Woods J, Allen L, et al. 2004. Earlyindicators of autism spectrum disorders inthe second year of life. J Autism Dev Disord34:473–493.

Whalen CM, Schreibman L. 2003. Joint attentiontraining for children with autism usingbehaviour modification procedures. J ChildPsychol Psychiatry 44:456–468.

Wimpory DC, Hobson RP, Williams JMG, et al.2000. Are infants with autism socially engaged?A recent study of retrospective parental reports.J Autism Dev Disord 30:525–536.

Woodward AL. 2003. Infants’ developing under-standing of the link between looker andobject. Dev Sci 6:297–311.

Yirmiya N, Gamliel I, Pilowsky T, et al. 2006.The development of siblings of childrenwith autism at 4 and 14 months: Socialengagement, communication, and cognition.J Child Psychol Psychiatry 47:511–523.

Yoder P, Warren S. 1999. Maternal responsivitymediates the relationship between prelinguis-tic intentional communication and later lan-guage. J Early Interv 22:126–136.

Yoder P, Warren S. 2001. Relative treatmenteffects of two prelinguistic communicationinterventions on language development intoddlers with developmental delays vary bymaternal characteristics. J Speech Lang HearRes 44:224–237.


Communication Autism

Documents

Transcript of Communication Autism