Evaluating Intermodal Processing in Autism and Asperger Syndrome Using the McGurk Effect

Background:
Intermodal or intersensory perception involves the coordination of information from various senses, leading to a unified perception of an event. The McGurk effect is an audiovisual task that demonstrates how we integrate what we see and what we hear during speech. The illusion involves a video of a person saying one syllable auditorily (e.g. /ba/), and saying a different syllable visually (e.g. /ga/). The listener usually integrates the two resulting in experiencing a percept that is an entirely different sound (e.g. /da/).
Children with autism have increased difficulty in intermodal speech perception.  Weiss, Bebko & Schroeder (in preparation) found that children with autism were significantly less likely to show the McGurk effect relative to both typically developing children and a cognitively impaired control group.  The authors concluded that children with autism may have difficulty with intermodal processing of speech information.  The current study aims to extend this research by exploring these processes in individuals with Asperger Syndrome (AS).
Objectives:
This research will help to determine if impairments in intermodal processing of speech are related to ASDs in general, or if they are related specifically to cognitive level or the developmental language delays and difficulties in current language that characterize autism.
Methods:
The samples ranged in age from 6-16 years and consisted of children with Autism (n=15), AS (n=15), Down Syndrome (DS; n=15), and a typically developing comparison group matched on chronological age (TD; n=19).    
Results:
Significant group differences were found, F(3, 60) = 6.59, p = .001, η² = .25.  Post-hoc contrasts revealed that the Autism group show significantly less intermodal speech perception compared to TD children, children with DS and children with AS.  Children with AS show equal levels of intermodal speech perception compared to TD and DS controls.  These differences remained after controlling for performance on the unimodal tasks.  Speech-reading is positively correlated with McGurk performance and listening performance is negatively correlated with McGurk performance across groups.
Conclusions:
It is the first study to directly investigate intermodal speech processing abilities in individuals with Asperger Syndrome as a distinct sample from Autism. This is also the first study to examine the McGurk effect in children with Down Syndrome. The results of the present study indicate that there are basic cognitive perceptual differences between children with Autism and those with AS. The difficulties for Autism do not appear to be shared by children with intellectual disabilities, including those with Down syndrome, or typically developing children. Moreover the audiovisual integration skills involved in processing speech appear to be intact in Asperger Syndrome.  These results suggest that children with Autism may have unique intermodal speech perception difficulties linked to their representations of speech sounds.