Grammatical Difficulties in Autism as Revealed by a Sentence-Picture Matching Task

Background: Although deficits of language and communication are diagnostic of autism, not all aspects of language appear to equally affected in the disorder.

Objectives: To examine syntactic processing in autism with a sentence-picture matching task – the Test of Active and Passive Sentences – that is sensitive to grammatical deficits in children with Specific Language Impairment (TAPS; van der Lely, 1996).

Methods: Participants listen to a sentence and choose the picture (one of four) that is most consistent with the meaning of the sentence. The sentences are either active (“The boy cuts the elephant”), passive (“The elephant is cut by the boy”), a shortened passive sentence (“The elephant is being cut”), or an ambiguous shortened passive sentence (“The elephant is cut”, which is ambiguous between  a verbal interpretation, “The elephant is being cut” and a grammatically simpler adjectival interpretation, “The elephant is in a state of having-been-cut”). We tested native-English-speaking high-functioning children with autism (n=9) and typically-developing control children (n=12).

Results:   Preliminary results indicate that for the three unambiguous sentence types the children with autism performed worse than the control children (79% vs. 92%). For the ambiguous sentences, the controls were split roughly evenly between the verbal interpretation and the adjectival interpretation, while the children with autism appeared to show a stronger preference for the adjectival interpretation.

Conclusions: Both the reduction in accuracy on unambiguous sentences and the preference for a simpler grammatical  structure are consistent with prior claims of syntactic and other grammatical abnormalities in autism. Additional implications will be discussed.