Negotiation in naturally-occurring conflictual interactions: A contribution to the differential assessment of high-functioning autistic children

Autistic children with moderate mental retardation and relatively preserved language development encounter specific difficulties relating to diagnostic and evaluation a) they are often diagnosed later than more severely affected children; b) even when autism is suspected or recognized by trained clinicians, the social environment of these children may have difficulties in identifying the nature of their communicative failures. Testing these children's language and social knowledge in experimental settings does not reveal clearly where the problems are situated. Indeed, current tests of language and social understanding abilities do not capture specifically the pragmatic dimension of autistic children's communicative behavior. The aim of this study was to conduct a precise evaluation of these children's use of verbal and nonverbal behaviors in socially-meaningful ways, when engaged in ecologically-valid interactions. We studied particular kinds of interactional events –Oppositional Episodes - because they are a) clearly identifiable b) occur naturally in everyday interactions c) are particularly demanding from the viewpoint of socio-pragmatic adaptation d) can be analyzed both qualitatively and quantitatively and e) are suitable to highlight both typical and atypical interactional functioning. Participants were 6 autistic children (diagnosed by experienced clinicians with DSMI-IV and ADI-R), 6 typically developing children matched on verbal age (VA levels: 3-4 and 6-7 years) and 3 additional 2 year-olds providing the "floor" level functioning on the behaviors studied. All the children were videorecorded at home during meals and play contexts. All naturally occurring oppositional episodes were systematically coded according to well defined criteria (inter-rater reliability: .85), including justifications and manners of insistence or acceptance. Verbally less advanced autistic children (VA: 3-4yrs) produce less justifications then verbal age matched controls (the rate is similar to that of 2yr olds) but the persuasive effectiveness of their justifications is similar. Verbally more advanced autistic children (VA: 6-7yrs) produce the same proportion of justifications then typically-developing controls but their justifications are less effective in persuading their partners who tend to counter-argument what are considered unclear reasons. Furthermore, these children do not seem to be immediately convinced by the justifications offered by their partners, who justify as much as controls their oppositional moves.Conclusions: In this approach, a relatively small sample of mother-child naturally occurring interaction is able to show areas of pragmatic functioning and dysfunctioning and to establish, for each dyad, a differential assessment of their communicative interactions in everyday life. Specific difficulties seem to reside in the partners' interpretation of each other's communicative intentions (implicit theory of mind)