Social Orienting Impairment in Autism: Relations among Look Length, Disengagement, and Symptom Severity

Background: Children with autism spectrum disorders (ASD) show impairments in social orienting, with slower times to orient and less looking to social versus nonsocial events (Dawson et al., 2005). Our prior study (Newell et al., 2007) revealed social orienting impairments in ASD, with reduced looking to social events, particularly affectively positive infant directed speech, but comparable looking to nonsocial events relative to TD children. Affectively positive infant directed speech exaggerates intersensory redundancy (rhythm, tempo, and intensity changes common to audible and visible stimulation).  Intersensory redundancy is highly salient and underlies the typical development of social orienting (Bahrick, 2008).

Objectives: To more precisely characterize social orienting impairments in autism and their relation to symptom severity, we explored in finer detail the nature of attention to social versus nonsocial events in ASD and TD children from our prior study. We expected social orienting impairments to be demonstrated in children with ASD by more disengagement and shorter looks to social events, particularly affectively positive social events and that greater social orienting impairments would be related to greater symptom severity among children with ASD.

Methods: Eleven children with ASD (M=3.6 yrs) and 11 TD children (M=2.1 yrs) matched on functional age (using the ABAS) were tested in the Behavioral Attention Assessment Protocol (Newell et al., 2007). In this procedure, children receive trials of a central stimulus (3s) followed by two side-by-side peripheral events (10s) along with the soundtrack to one event. Blocks of trials depicted nonsocial, social neutral, and social positive events. The proportion of available looking time, number of disengagements per minute, and average length of look to the peripheral events were calculated.

Results: ANOVAs indicated that children with ASD showed more disengagements per minute and shorter looks than TD children (ps< .005), particularly for social, but not nonsocial events. Children with ASD showed reduced overall looking to social positive events compared with TD children (p =.02), but comparable looking times overall.

Analyses of symptom severity (measured by the Social Communication Questionnaire) revealed a significant negative correlation between severity and looking time to social positive events (r =-.74, p <.01), with less looking associated with more symptoms. Children with ASD with more symptoms (according to a median split) showed shorter looks (p=.02) and more disengagements, particularly for social positive events (p=.007), and less overall looking to social positive events than children with ASD with fewer symptoms (ps<.01). Furthermore, children with more symptoms differed from TD children on all measures (ps<.01), whereas those with fewer symptoms did not.

Conclusions: Results demonstrate that children with ASD show social orienting impairments with less looking time, more disengagements, and shorter looks to social events, particularly affectively positive speech, than TD children. Moreover, greater symptom severity is associated with greater impairment. Affectively positive events provide high intersensory redundancy, an important basis for social orienting in typical development. Measures of disengagement and look length reflect important individual differences in attention underlying social orienting. These findings suggest a link between symptom severity and intersensory processing disturbance in autism.