Exploring Intermodal Perception In Children with Asperger Syndrome Using a Preferential Looking Task

Background: Intermodal perception (IMP) links information from more than one sensory modality so that multisensory information is perceived as a unitary event. In the amodal processing of speech stimuli, although the auditory and visual modalities are separate and discrete, their information is automatically extracted as one congruent event. Current research is inconsistent with regards to the IMP abilities of children with autism, although there is emerging research that indicates that there may be a linguistic-specific impairment in IMP in autism. Research examining these abilities in children with Asperger Syndrome (AS) is limited. Based on the emerging evidence of linguistic-specific IMP difficulties in children with autism, an examination of the intermodal perception of speech stimuli in individuals with AS would be beneficial due to the differences in language abilities between AS and autism.

Objectives: To provide insight into the intermodal perception abilities of children with AS, who diagnostically will have no language delay, and how these abilities may differ from children with autism, typical development and intellectual disabilities using a preferential looking task.

Methods: An unanalyzed AS group consisting of 18 children (M = 12.61 years, SD = 2.92) will be compared to three previously examined groups (autism, intellectually disabled [ID], typically developing [TD]) (Weiss, 2007, unpublished dissertation). Three criteria differentiated the AS group from the autism group: no significant language delay, limited or no problems in current language abilities, and no cognitive impairments. The authors presented the participants with non-speech, natural vowel (/i/, /a/), and fundamental tones that represented the vowels (3000 Hz for /i/, 300 Hz for /a/) conditions. The two videos for each condition were shown side by side on a split screen and the auditory sound matched only one of the two screens. Looking behaviours were coded using The Noldus Observer software.

Results: The duration of participants’ eye gaze at each the screens is coded for each trial to determine the percentage of looking time at the matching and non-matching screens. Data analysis is ongoing and will include comparing looking times for the AS group and the three comparison groups. Preliminary results based on Weiss (2007) indicate that there were no differences between children with autism, children with ID, and TD children in their preferential looking during the natural vowel condition. However, unlike their TD and ID counterparts, children with autism did not display preferential looking during the vowel tone condition after being primed by the natural-vowel conditions.

Conclusions: If the AS group shows looking patterns different from the autism group it would have two implications: First, it could suggest that the difficulties in the intermodal processing of linguistic stimuli found in individuals with autism may be linked to language impairments or delays, which are not present in AS samples. Second, future research could further examine the possible utility of a speech-based preferential looking task in early identification of autism-specific markers.