Narrative Ability in Children with Autism Tested through the Karmiloff-Smith (1985) Stories

Background: Difficulties in personal and fictional narratives are an important aspect of autism spectrum disorders, yet there is little evidence of the precise nature of the problems with narrative in autism. In particular, it is not clear whether difficulties stem from not understanding the emotional or mental state content, in developing story schema or with the technical devices that give stories coherence and cohesion. Objectives: To compare the performance of children with autism on a narrative task to a comparison group of children with developmental delay and the ‘norms' established by Karmiloff-Smith... Methods: A replication of the Karmiloff-Smith (1985) study of narrative ability, using five wordless ‘story' books, .was carried out with 15 children with autism and a comparison group of 15 children with developmental delay, matched by verbal ability. The child's description of a picture from each of the ‘stories' presented as a single picture, was compared with the description of the same pictures when embedded in a wordless ‘story' book. Results: Overall, the children with autism did not perform significantly less well in terms of correct use of anaphoric reference, establishing a ‘thematic subject', using linking devices across pictures and using devices to engage the listener. However, there were more bizarre responses and a significant difference in relation to two stories; in one this stemmed from a poorly drawn event leading to misinterpretation by the children with autism. In the other it related to the failure of the children with autism to give a causal interpretation to an emotional event. Conclusions: The results agree with other studies showing the key problem with narrative in autism as the failure to make causal statements linking the emotions of characters to events.