Sex Differences in the Neural Processing of Interactive Eye Contact in Individuals with Autism Spectrum Disorder

Background: Sex differences in autism spectrum disorder (ASD) exist in the manifestation of social deficits (Lai, Lombardo, Auyeung, Chakrabarti, & Baron-Cohen, 2015), including differences in neural processing of static faces (Coffman, Anderson, Naples, & McPartland, 2015). However, sex differences in the neural correlates of social processing during dynamic social perception have not yet been examined, which may better reflect the interactive demands of social activity.

Objectives:  To identify sex differences in neural processing of interactive eye contact in individuals with ASD and typical development.

Methods:  Individuals with ASD (n=55; male=41) and typically developing (TD) individuals (n=44; male=26) participated in a gaze-contingent EEG experiment. Sex by diagnosis subgroups were matched for age and IQ. EEG data were collected using a 128 electrode Geodesic net, and eye-tracking data were collected using an SR-Research EyeLink 1000. Each trial began with a face displaying direct or averted gaze. When the participant looked at the face, the face responded by either looking at (direct gaze) or away from (averted gaze) the participant. P100 and N170 ERPs, components reflecting lower-level visual processing and face processing, respectively, were examined.

Results:  A significant interaction was found between sex, diagnosis, and hemisphere for N170 peak amplitude (F(1, 93)=9.10, p<.01). Specifically, left hemisphere N170s were more negative than right hemisphere N170s for females with ASD, while males with ASD and TD females and males displayed the opposite lateralization pattern. There was also a significant interaction between sex, diagnosis, and condition (direct or averted gaze) for N170 peak amplitude (F(1, 93)=5.26, p=0.02). More negative N170s were evoked to direct gaze compared to averted gaze for females with ASD and males in both groups, while TD females displayed the reverse pattern. Eye-tracking results revealed a significant interaction between sex, diagnosis, and condition for amount of time looking at the eyes (F(1, 93)=5.71, p=0.02). Females with ASD spent a significantly longer time looking at the eyes in the averted compared to the direct condition (t(15)=2.61, p=0.02). There was no significant difference in looking patterns between conditions for other sex by diagnosis subgroups.

Conclusions: These findings suggest that there are important differences in how females with ASD respond to dynamic social interactions in both behavior and neural processing. Females with ASD look longer at eyes displaying averted compared to direct gaze, and, during these interactive tasks, this group displays atypical left-lateralization in neural response that differs from males with ASD and TD peers. Sex-specific brain activity and looking patterns in individuals with ASD highlight the importance of considering sex as a variable in understanding ASD and suggest distinct mechanisms underlying social perception in females with ASD.