Attention and Brain Response during Simulated Social Interactions in ASD

Thursday, May 12, 2016: 11:30 AM-1:30 PM
Hall A (Baltimore Convention Center)
A. Naples, M. Rolison, J. H. Foss-Feig, K. K. Stavropoulos, T. Winkelman and J. McPartland, Child Study Center, Yale School of Medicine, New Haven, CT

Difficulties with eye-contact represent some of the earliest observable symptoms of autism spectrum disorder (ASD). While difficulties with eye-contact are clinically diagnostic, experimental investigations of gaze processing have yielded mixed results. Prior work has focused primarily on experiments in which individuals with ASD passively observe faces. Our preliminary work assessing response to mutual gaze has revealed ERP markers of eye-contact. The N170 and P300, temporally early electrophysiological components, show variability in response to shared gaze that track with clinical characteristics. Further, eye movements to interactive faces distinguish between groups and reflect variability in the clinical phenotype.


To investigate neural processing of interactive eye-contact; investigate attention by measuring eye-movements during interaction with an onscreen face; and evaluate the relationship between neural and attentional markers of eye-contact and clinical characteristics.


ERPs were recorded from 44 children with ASD (M =14.5 years, mean IQ =103) and 44 controls (M =13.7 years, mean IQ=107) using a 128 electrode Geodesic Net. Eye movements were recorded with an SR-Research Eyelink 1000 integrated with EEG recording. Through co-registered and time-locked ET and EEG, the experiment was controlled by participant gaze. Trials began with a centrally presented face displaying direct or averted gaze. Contingent upon participant fixation, the face responded by either looking at (eye contact) or away (averted gaze) from the participant. ERPs were time-locked to face movement.


Preliminary analyses of eye-tracking data using linear mixed models indicated that, when initially looking to faces, individuals with ASD show longer gaze latency than controls [B = 83.10, p = .02], but an interaction between condition (gaze direction) and group [B = 49.5  p = 0.08] revealed that control participants looked more slowly to faces showing averted gaze. Region of interest analyses indicated that children with ASD looked less to the left eye of the face [B = 40.0, p = .06]. Analyses of pupil size revealed an interaction [B = 4.1, p = .06], such that individuals with ASD exhibited pupil dilation following direct gaze whereas controls exhibited constriction. Preliminary analyses of ERP data revealed that averted gaze elicited a more positive P300 in both groups [p < .05], although an interaction revealed that increasing levels of anxiety predicted more positive P300s for direct gaze [B = 0.063, p= 0.06]. Analyses in progress investigate brain response to mutual gaze and the relationship between attentional dynamics and clinical characterization.


We demonstrate an electrophysiological marker of reciprocal gaze perception that is sensitive to clinical characteristics and a pattern of attenuated attention, but increased arousal, to direct gaze in individuals with ASD. Our ERP findings suggest that direct gaze perception is associated with anxiety regardless of group assignment. Relationships between pupillary dilation and anxiety suggest that mechanisms of arousal associate with social brain response and reflect meaningful variability in the clinical phenotype. By linking both eye-tracking and EEG during social interactions, we can gain insight into the complex interrelationship between attention and brain response to social information during the course of an interaction.