Comprehension tasks reveal grammatical weakness in young children with autism

Background: Research on the grammatical abilities of children with autism has revealed both strengths and weaknesses; they demonstrate steady growth in MLU across development but frequently omit grammatical morphemes and use complex grammatical constructions such as Wh-questions much less frequently than typically developing children.  These findings have relied on production data, though, which can be unreliable in a disorder in which children are disinclined to speak.  Moreover, some have conjectured that ASD children’s avoidance of Wh-questions is pragmatic rather than grammatical.

Objectives: We investigate ASD children’s comprehension of Wh-questions, using intermodal preferential looking.

Methods:   Children are tested every four months in this ongoing longitudinal study.  At the study’s onset, the ASD children had a mean age of 33 months and had language scores comparable to 20-month-old typical children.  At Visit 4, when the children viewed the Wh-question (WHQ) video, they averaged 45.3 months of age and produced on average 34% of the words on the MacArthur CDI checklist.  The typical children at the same visit averaged 33 months and produced 68.5% of the CDI checklist. The WHQ video (Seidl et al., 2003) showed ‘hitting’ events (e.g., an apple hitting a flower), followed by test trials showing the apple and flower on separate screens.  The WHQ test audios were “What did the apple hit?/What hit the flower?”; the Name control audios were “Where’s the apple/flower?” Children’s eye movements were coded off-line.

Results:   The ASD children showed no significant looking preferences when hearing the Wh-questions, either at Visit 4 or Visit 5.  The comparison typical group demonstrated significant looking to the match at both visits.

Conclusions: Young children with autism have not yet demonstrated the ability to understand Wh-questions.  These findings support the claim that even higher-functioning autistic children have specific grammatical weaknesses that are not completely attributable to pragmatic impairments.