(PS2-54) Do Autistic Children Experience and Express Joy Differently? A Study of Pupillometry and Facial Expression

Introduction: Although autistic children show differences from typically developing peers (Loveland, 2005), there is no consensus on their emotional presentation (Macari et al., 2018). Using complementary methodologies of measuring emotion arousal and expression may clarify presentation of emotional affect. Combining pupillometry, a measure of arousal, with facial expression coding allows us to examine whether autistic children experience arousal in response to joy-eliciting videos and whether this arousal is behaviorally expressed.

Objectives: We aimed to (1) assess whether participants experienced emotional arousal while watching emotion-eliciting videos and (2) examine potential differences and associations between emotional arousal and facial expressions of joy in autistic and typically developing children.

Method: Twenty-three autistic children (M = 40.91 months, SD =11.58) and 21 typically developing children (M = 44.1 months, SD = 10.88) watched sixteen 30-second joy-eliciting videos (i.e., bubbles, car races) presented with or without people (i.e., social or non-social) with orientation of target stimulus reversed (i.e. left or right), with fixation setpoints shown between clips. Facial expressions were video recorded. Pupil diameter was measured using the Tobii Pro Spectrum eye-tracking system. Data was cleaned to exclude outliers from blinking, single pupil tracking, or inattentiveness. Average pupil diameter was calculated for the first two seconds (onset) and the remaining 28 seconds (duration) to determine percent change in pupil diameter. Two researchers blinded to participant grouping coded facial intensity and frequency of joy expression, with discrepancies resolved by study principal investigator.

Results: Groups differed significantly in sex, race, cognitive ability, and adaptive behaviors (ps < .05). Significant differences (ts > 1.99, ps < .05) were found in pupil diameter between onset and duration across stimuli (Table 1). However, a one-way MANOVA showed no significant group differences in pupil diameter or behavior-coded positive facial emotion (all Fs < 2.5, all ps > .05, Table 2). No significant correlations were found between joy intensity, frequency, and percent change in pupil diameter, either in the total sample or within groups (all ps > .05).

Conclusion: Joy-eliciting videos successfully induced emotion arousal both in autistic and typically developing children, yet no group differences were found between internal arousal and facial expression of positive affect. However, differences may emerge in expressions of negative emotions, aligning with the negative emotionality bias theory (Macari et al., 2018). No significant relationship between pupil diameter and joy expression in either group suggests that facial expressions of joy may not be concordant with internal arousal while passively viewing stimuli. Pupillometry may serve as a complimentary method to measure change in affect alongside coding of facial expression.