Attention and Autonomic Arousal Toward Emotional Faces in Children with Autism Spectrum Disorder Who Are Minimally Verbal

Background: To our knowledge, no studies have examined attention and autonomic arousal toward emotional faces in children with autism spectrum disorder who are minimally verbal (ASD-MV). Children with autism spectrum disorder (ASD) and typically-developing children(TD) demonstrate a differential looking pattern toward faces based on emotion, including a preference for looking at the mouth region of happy compared to neutral or fearful faces (Eisenbarth & Alpers, 2011; Wagner et al., 2013). Children with ASD, compared to TD children, have a decreased pupillary response to unfamiliar fearful faces (Nuske et al., 2014) which may influence social communication. Utilizing eye tracking to measure allocation of attention and pupillary response to emotional faces in children with ASD-MV may provide insight into the social-communication challenges of this phenotype.

Objectives: This study investigated whether children with ASD-MV differentiate emotions based on allocation of attention and pupillary response to faces (happy, fear, neutral).

Methods: Participants (n=41) completed this eye tracking study at the baseline visit of a clinical trial targeting language through behavioral intervention combined with medication or placebo. Enrolled children were 6-11 years old, had a diagnosis of ASD, confirmed using the Autism Diagnostic Observation Schedule-2 (Lord et al., 2012), and fewer than 30 functional words used on a language sample. Eye tracking data was collected using the Tobii Eye Tracker. Following calibration, images of 10 individuals (5 female) were presented showing fear, neutral or happy facial expressions in a randomized order. Duration of fixations on eyes, mouth and face regions was calculated. An eyes-mouth index (EMI: ratio of looking time in the eye region to looking time in eye and mouth regions combined) was used to measure differences in gaze in these regions independent of total gaze (Key & Stone, 2011). Autonomic arousal was measured as a weighted average of pupil diameter when looking at the face region across same-emotion trials (Wagner et al., 2016).

Results: A repeated-measures ANOVA revealed significant differences in EMI by condition (F(2,80)=11.94, p=<.001). Post-hoc analyses identified a lower EMI in the fear condition (M=.746, SD=.203), (t(40)=-3.42, p=.001) and happy conditions (M=.725, SD=.237), compared to the neutral condition (M=.838, SD=.194), (t(40)=-4.619, <.001). No differences in autonomic arousal were found between conditions (F(2,80)=.459, p=.633).

Conclusions: Children with ASD-MV spent more time looking at the mouth of happy and fearful compared to neutral faces. Further investigation is needed to determine if the lack of behavioral differentiation between positive and negative emotional faces contributes to the increased social-communication challenges in children with ASD-MV relative to children with ASD and expressive language. There was no difference in autonomic arousal, measured by pupillary response, between emotions. As this is consistent with findings of atypical pupillary response in children with ASD (Nuske et al., 2014), this deficit may be related to general ASD characteristics, and not the specific social communication deficits of children with ASD-MV. Examination of the relation between looking patterns, autonomic arousal and behavioral and developmental measurements of functioning in children with ASD-MV is needed to provide insight into the clinical significance of these findings.