Behavioral and Electrophysiological Evidence of Impaired Social Orienting in ‘Unaffected' Siblings of Children with Autism Spectrum Disorder

Background: Failure to respond to name and impaired orienting to both social and non-social information has previously been found in infants and children diagnosed with or at risk for autism spectrum disorder (ASD).  Further, event-related potential (ERP) studies investigating selective attention have demonstrated reduced P3a amplitudes in individuals with ASD, which may be indicative of deficits in processing novel information and attentional capture in ASD.

Objectives: To investigate behavioral and electrophysiological indices of orienting to social and non-social information in high-risk children (defined as having an older sibling with ASD) with (HRA+) and without (HRA-) a diagnosis of ASD and low-risk typically developing (TD) children (defined as having a TD older sibling). 

Methods: Data collection for the current project is ongoing. To date, participants are nine 4 to 5 year old HRA- and 10 age- and IQ-matched low-risk, TD children. The study consisted of separate behavioral orienting and auditory odd-ball ERP experiments. For the behavioral orienting paradigm (modeled after Dawson et al., 1998; 2004), one experimenter engaged the child’s attention, while a second experimenter administered a series of 12 possible social or non-social sounds. Social sounds (hum, clap, laugh, yawn, snap, and the child’s name) and non-social sounds (toy rattle, car horn, toy squeak, vacuum, referee whistle, and a phone ring) were equated in intensity (70 dB) and duration (approximately 1 second).  Child’s orienting behavior, defined by either a head movement or a gaze shift was coded online by experimenters. For the auditory odd-ball task, auditory stimuli were identical to those used in the behavioral orienting paradigm. EEG was recorded using 128-channel high-density Geodesic electrode arrays, sampled at 500 Hz, and referenced to single vertex electrode. Data were segmented into 1200ms epochs (200ms pre- and 1000ms post-stimulus onset), trial/artifact rejected, and re-referenced to average reference. Mean amplitude and latency of the P3a component was measured for both social and non-social sounds.

Results: Paired t-tests showed that TD children oriented more frequently to social compared to non-social sounds, p < .01; In contrast, HRA- children shifted attention to both social and non-social sound at similar rates, p = .3.  Independent samples t-tests demonstrated that, compared to HRA- children, TD children shifted attention more frequently to social stimuli, p < .01, but not non-social stimuli, p < .1. Electrophysiological data revealed a similar pattern of results; for the TD group, there was a larger P3a response to social stimuli than to non-social sounds, p < 0.05. However, for the HRA- group there was no significant difference in P3a amplitude for social compared to non-social sounds, p > 0.5.

Conclusions: Behavioral and electrophysiological indices of orienting suggest social information is perceived as more salient and captures attention to a greater degree in TD but not HRA- children. Similar to prior findings of impaired social orienting in children with ASD, our preliminary results suggest that social orienting deficits may be present in clinically ‘unaffected’ siblings, and therefore represent a potential endophenotype.