Invited symposium
Chairs: Philip R. Zelazo and Jake Burack
Discussant: Marian D. Sigman
Autism is a developmental disorder that is evident, indeed diagnosed,within the first three years of life. The vast majority of research onautism has focused on its presumed biological origins in general, andgenetic basis, in particular, and with older children who are beyond theages of emergence. A large segment of this work was conducted eitheroutside the realm of Psychology by Psychiatry, Neurology and Genetics and,when studied by psychologists, by clinical psychologists who have lessexpertise in the area of development, particularly during the first threeyears. Attempts to bridge the gap between clinical and developmentalpsychology, have been initiated with the subdiscipline of DevelopmentalPsychopathology. We propose to intensify this effort and to focus on theearly development of autism using data from both normal and deviantdevelopment during the first three years of life. As a general corrective,we bring an openness to the role of experience both in the course of normaldevelopment and in the etiology of autism. In this symposium, we will address the thorny problems of definingappropriate controls for studying development and psychopathology andbridging different disciplines and perspectives. Results from an earlyautism screening project, particularly results for pretend play, jointattention, imitation and social communicative behaviors will be presented.We will review studies of attachment with autistic children and theirimplication for research on joint attention and theory of mind. Adevelopmental cognitive-behavioral account of early autism includingetiology, assessment of information processing ability and treatment will beproposed. Our discussant will integrate these findings with results fromexisting extensive longitudinal studies.
Details of individual items:
paper
Developmental methodologies and paradigms are routinely applied to the studyof autism. The most common methodological strategy based in developmentaltheory is the use of developmental level to match between persons with autism andthose who are studied as comparison subjects. Typically, the paradigms are adaptations of those used in studies of typically developing children. Boththese types of developmentally-inspired strategies contribute to enhanced empirical initiatives, but neither ensures that the research in this area isconsistent with developmental guidelines or that the findings are meaningfulto understanding the development of persons with autism. In this paper, I willargue that current notions of developmental approaches to studying autism represent important initiatives but do not adequately integrate therequirements of the two disparate disciplines.I will briefly outline primary developmental innovations in theory,methodology, and the interpretation of findings in the study of persons with autism. Specifically, I will address two sets of issues that arise from the general notion of developmental level. One set is relevant to the choice of personsthat comprise the comparison group(s) and the other to the various implicationsof the subjects' level(s) of functioning. These issues will be addressedwithin the context of: 1) common but avoidable misrepresentations of developmental issues that lead to erroneous conclusions about the implications of thefindings and 2) more severe challenges to comprehensively charting the development ofpersons with autism. As the integration of the study of generaldevelopmental processes with that of differences between a specific atypical populationand other groups of persons are largely incompatible, these challenges need tobe addressed in a systematic and collaborative manner. Empirical work shouldbe fine-tuned within the context of developmental theory and methodology, and interpretation. This will lead to scientifically compelling work and to increasingly heuristic and collaborative approaches to the study of personswith autism.
paper
Tony Charman, Institute of Child Health, University College London, 30Guilford Street, London WC1N 1EH. e-mail: t.charman@ich.ucl.ac.ukA screening study that prospectively identified a small group of childrenwith autism and pervasive developmental disorder (PDD) in infancy(Baron-Cohen et al. 1996) has provided a unique opportunity for us to studyinfant social communicative behaviours in individuals with autism closer tothe time at which they emerge than has been possible previously. Severalimportant findings have emerged. These will be briefly reviewed. Theimplications for our understanding of typical and atypical infantsocial-communicative development will be discussed. First, impairments in pretend play and joint attention behaviours enabledus to identify approximately 40% of the children with autism from thepopulation who was screened. Second, in line with previous longitudinalstudies by Marian Sigman, Peter Mundy and Wendy Stone certain aspects ofjoint attention, play and imitation behaviours have been shown to predictlater language development. Third, in a small sample of typically developinginfants included in the study, both general and specific longitudinalassociations were found between these infant 'precursor' socialcommunicative abilities and the later development of language and theory ofmind abilities. Fourth, the associations between the abnormal social andcommunicative behaviours and the repetitive behaviours and restrictedinterests that also characterize autism were examined. There was someevidence that the repetitive abnormalities may appear later than the socialand communication abnormalities.The findings will be discussed in relation to emerging theoretical accountsof social and non-social cognitive development in infancy and to accountsthat attempt to explain autism at a psychological level. While the findingsare of theoretical interest they may also have important practicalimplications for the early diagnosis and treatment of autism.
paper
Following Bowlby (1969/82, 1973, 1980) attachment is conceptualized as the'affectional bond or tie that an infant forms between himself and his motherfigure' (Ainsworth et al., 1978). This special bond is associated withconcurrent and later social and cognitive abilities in normally developingchildren. Bowlby suggested that the attachment system is activated wheninfants are tired or distressed, when there are threats in the environment,and when the attachment figure moves away or is absent. Ainsworth andcolleagues (1978) designed the Strange Situation procedure to assess theattachment status or working model of children between the ages of 12 - 18months toward their caregivers. Patterns of secure and insecure attachmentsmay be identified based on this procedure in which infants go through aseries of separations from, and reunions with their caregivers. The studiesconducted with children with autism, employing various modifications of theStrange Situation procedure, suggest that similar to other groups ofchildren about 50% of the children with autism show secure attachment towardtheir caregivers. The findings that many children with autism reveal secureattachment behaviors is surprising, especially given the earlier impairmentsin joint attention behaviors and the later impairments in social behaviorand cognition including theory of mind abilities. These findings will bereviewed in relation to attachment theory in general and more specificallyto the core deficits involved in autism as well as to intervention programs.Philip R. Zelazo, Ph.D.Keywords: Autism; Developmental; TheoreticalEARLY AUTISM: A DEVELOPMENTAL, COGNITIVE-BEHAVIORAL PERSPECTIVEPhilip R. Zelazo, Department of Psychology, 2300 Tupper, Montreal, QuebecH3H 1P3, Canada.e-mail : plecpsy@mch.mcgill.caAutism is classified in the DSM-IV (1994) as a Developmental Disorder. Thedefining behaviors include delayed expressive language and play, aberrantsocial interactions and, in 75% of the cases, mental retardation. Moreover,these characteristics must be present before three years of age.Considerable basic and clinical research exists with older children but workwith children under three is only beginning (e.g., Baron-Cohen, et al, 1995;Lovaas , 1987; Sigman et al, 1999; Zelazo, 1997a, 1997b). Even moresurprising, dominant theoretical accounts either fail to address the earlymanifestations of autism by relying on data from older children who displayadvanced manifestations of the disorder or accord little weight to the roleof experience in its development , e.g., Rutter, 1996).I will present a theoretical account derived from research on normalpsychological development during the first three years, point out the needfor unconfounded measures of intelligence for children with autism and PDD,propose how development of the defining behaviors may go awry in autism, andprovide supportive evidence for this view. Specifically, I suggest thatnon-compliance to task demands (Carr & Newson, 1985; Kearsley, Ungerer &Zelazo, 1984) is a core behavior in autism with cascading negative effectson affective, linguistic, social and cognitive development. We showed thatnon-compliance to task demands was greater to moderate and high difficultytasks for a sample of children with autism and PDD than for developmentallymatched controls (Rogers, Zelazo & Mendelson, 1998).Non-compliant and resistant behaviors, be they tantrums, crying, eyeavoidance, or a host of other behaviors allow children to escape from taskdemands and can arrest the development of expressive abilities such aslanguage, play and social interaction. Evidence will be presented to showthat treatment to foster compliance, first with actions and later withexpressive language ( Zelazo et al; 1984) early in life, not only can reducedelays on tests of intelligence, but facilitate symbolic play, talking andappropriate social interactions. The role of experience in the etiology ofautism will be discussed.
paper
Autism is classified in the DSM-IV (1994) as a Developmental Disorder. The defining behaviors include delayed expressive language and play, aberrant social interactions and, in 75% of the cases, mental retardation. Moreover, these characteristics must be present before three years of age. Considerable basic and clinical research exists with older children but work with children under three is only beginning (e.g., Baron-Cohen, et al, 1995; Lovaas , 1987; Sigman et al, 1999; Zelazo, 1997a, 1997b). Even more surprising, dominant theoretical accounts either fail to address the early manifestations of autism by relying on data from older children who display advanced manifestations of the disorder or accord little weight to the role of experience in its development , e.g., Rutter, 1996).I will present a theoretical account derived from research on normal psychological development during the first three years, point out the need for unconfounded measures of intelligence for children with autism and PDD, propose how development of the defining behaviors may go awry in autism, and provide supportive evidence for this view. Specifically, I suggest that non-compliance to task demands (Carr & Newson, 1985; Kearsley, Ungerer & Zelazo, 1984) is a core behavior in autism with cascading negative effects on affective, linguistic, social and cognitive development. We showed that non-compliance to task demands was greater to moderate and high difficulty tasks for a sample of children with autism and PDD than for developmentally matched controls (Rogers, Zelazo & Mendelson, 1998).Non-compliant and resistant behaviors, be they tantrums, crying, eye avoidance, or a host of other behaviors allow children to escape from task demands and can arrest the development of expressive abilities such as language, play and social interaction. Evidence will be presented to show that treatment to foster compliance, first with actions and later with expressive language ( Zelazo et al; 1984) early in life, not only can reduce delays on tests of intelligence, but facilitate symbolic play, talking and appropriate social interactions. The role of experience in the etiology of autism will be discussed.