Impaired Detection of Temporal Synchrony for Social and Nonsocial Events in Children with Autism Spectrum Disorders

Background: Intersensory processing develops across the first year of life and is a fundamental building block for typical social and communicative functioning (Bahrick & Lickliter, 2002). However, children with autism spectrum disorders (ASD) appear to show altered intersensory processing with impaired temporal synchrony perception (Bebko et al., 2006), impaired use of visual speech in noise (Smith & Bennetto, 2007), and a wider temporal window for audiovisual integration (Foss-Feig et al., under review). Furthermore, they show social attention deficits, including impaired disengagement and decreased attention to social as compared with nonsocial events (Bahrick et al., 2009; Newell et al., 2007). Given that social events provide an extraordinary amount of intersensory redundancy (synchrony, rhythm, tempo, and intensity changes invariant across face, voice, and gesture), a general intersensory processing disturbance could lead to impairments in social orienting in ASD. 

Objectives: The objective of the current research was to assess intersensory processing of audiovisual temporal synchrony for social and nonsocial events in young children with ASD, typical development (TD), and developmental delay (DD) using the intersensory measure from our Behavioral Attention Assessment Protocol (BAAP; Newell et al., 2007). We predicted that compared to TD and DD children, children with ASD would show intersensory processing impairments, particularly for social events. 

Methods: Children with ASD (N = 7; 3.98 yrs), DD (N = 7, 4.18 yrs), and TD (N = 10, 3.18 yrs) (matched on Mullen composite scores) were presented with trials consisting of a 3 s central stimulus followed by two side by side peripheral events (10 s). One peripheral event was synchronous with the soundtrack and the other was out of synchrony. Three event types were presented in blocks of 20 trials depicting social neutral (woman speaking with neutral affect), social positive (woman using infant directed speech with positive affect, exaggerating intersensory redundancy), and nonsocial events (objects impacting a surface in an erratic temporal pattern). The proportion of total looking time (PTLT) spent fixating the sound synchronous event was calculated for each event type.

Results: An ANOVA revealed a significant main effect of group, with TD children showing greater PTLTs than children with ASD (p = .01). In addition, TD and DD children showed significant matching of films and soundtracks (PTLTs greater than chance, .50) for both types of social events (ps < .05) and TD children also matched nonsocial events (p = .04). In contrast, children with ASD showed no evidence of audiovisual matching for any events (ps > .50).

Conclusions: Consistent with predictions, our findings indicate that children with ASD show impaired intersensory processing skills. They showed no evidence of detecting face-voice synchrony in social events or object-sound synchrony in nonsocial events. In contrast, typically developing children detected audiovisual synchrony in all event types, and children with other developmental delays detected face-voice synchrony in social events. These findings are consistent with prior findings of impaired audiovisual temporal synchrony detection in children with ASD and are compatible with a general intersensory processing disturbance underlying the development of social orienting impairments in ASD.