Effect of Eye Contact and Anxiety Symptoms on ERPs in Autism Spectrum Disorder and Schizophrenia

Background: Individuals with Autism Spectrum Disorder (ASD) exhibit high levels of comorbid anxiety that can impede treatment, but the neural basis of this anxiety is under-characterized. Previous work studying early time course brain activity in anxiety and ASD has identified that the P100 and N170 ERP components are atypical in response to social stimuli. However, few have examined this brain activity transdiagnostically in interactive social contexts that better capture the reciprocal nature of social interactions.

Objectives: We examined brain responses to faces across diagnostic categories with vulnerabilities in social and anxious symptomologies. We assessed the relationship of categorical and continuous clinical characterization on brain activity.

Methods: 22 individuals with ASD, 23 with Schizophrenia (SCZ), and 31 TD adults (aged 18-48) participated. Eye tracking and EEG were used to measure brain response to reciprocal gaze. Participants looked at the mouth or eyes of a face; the mouth (“Mouth”) or eyes (“Eye Contact”) then opened. EEG data was recorded using an EGI 128 channel net at 500Hz. P100 and N170 components were recorded from temporal-parietal electrodes. Anxiety symptoms were measured using the State-Trait Anxiety Inventory (STAI) and Beck Anxiety Inventory (BAI). ASD symptoms were measured with the Autism Quotient (AQ), ADOS, and Social Communication Questionnaire (SCQ).

Results: ANOVAs were run with ERP component as dependent, and diagnosis, hemisphere, and condition as independent, variables. Main effects of condition for N170 amplitude, latency, and P100 amplitude were significant (p<0.01): Eye Contact increased N170 amplitude and decreased P100 amplitude and latency relative to the Mouth condition. There were no main effects of diagnosis. Across groups, increased STAI [r(40)=-0.38, p<0.05] and BAI total scores [r(42)=-0.43, p<0.01] correlated with faster P100 in the right hemisphere (RH) during Eye Contact. Greater anxiety correlated with smaller differences in P100 amplitude between Eye Contact and Mouth Movement [STAI r(40)=-0.33, p<0.05; BAI r(42)=-0.32, p<0.05]. Increased ASD traits on the AQ correlated with decreased P100 to Eye Contact [r(71)=-0.32, p<0.01]. Among individuals who did not show eye contact on the ADOS, BAI score was associated with decreased P100 latency [r(5)=-0.93, p<0.05]. On the SCQ, participants that looked at faces had slower N170s with increased STAI score [r(10)=0.68, p=0.01]. Those that did not had a marginally significant correlation between faster N170s and increased STAI score [r(9)=-0.58, p=0.056]. These correlations differed from each other (p<0.01).

Conclusions: Symptoms of ASD and anxiety modulate the P100 and N170. Consistent with previous research, greater anxiety related to faster P100 in the RH. Greater anxiety was also associated with smaller differences between P100 amplitudes in the Eye Contact and Mouth conditions. Previous research has found eye contact evokes stronger N170 responses than no eye contact. This indicates that individuals with greater anxiety ascribe similar amounts of attention to both Eye Contact and Mouth conditions. Increased symptoms of ASD correlated with slower P100 during Eye Contact. Lack of eye contact, an important social symptom in ASD, and anxiety significantly altered latency of both N170 and P100, suggesting that anxiety symptoms alter early processing of social information in this group.