Decreased Fear Response in Toddlers with ASD Is Linked with Attenuated Changes in Physiological Arousal

Background: Compared to typically developing toddlers (TD) and controls with developmental delay (DD), toddlers with autism spectrum disorder (ASD) express less intense negative emotional behaviors in response to real-world fear-eliciting challenges (Macari et al., 2017;Abstract #1)). It is not clear if this phenomenon is due to differences in behavioral display of emotions or in arousal levels in response to these events. Physiological arousal is thought to be altered in ASD; however, research findings have been inconsistent, likely due to differences in induction techniques, intensity and valence of emotional triggers, and measurement methods.

Objectives: To examine concurrently for the first time, physiological arousal (changes in skin conductance level, ΔSCL) and intensity and valence of behavioral affective responses to structured real-world challenges in toddlers with ASD. (2) To test whether ASD toddlers show greater ΔSCL in response to Anger- and Fear- but not Joy-eliciting probes as seen in typical development (Kreibig et al., 2010). (3) To test whether the magnitude of ΔSCL is associated with intensity of emotions exhibited during Fear and Anger but not during Joy probes.

Methods: Participants included 26 toddlers with ASD (age: M=23mo, SD=3) and 24 age-matched TD controls (age: M=22mo, SD=4). ΔSCL was measured in response to multi-trial conditions eliciting Fear, Anger, or Joy, using an Affectiva Q-Sensor placed on the child’s ankle. ΔSCL was calculated between the start and end of each trial, standardized and averaged within each condition. Intensity of emotional behavioral responses was coded offline.

Results: At baseline, the ASD and TD groups showed comparable SCL scores (p>.40). A linear mixed model analysis of ΔSCL indicated no effect of diagnosis (p=.74), a significant effect of condition, (p<.001), and a diagnosis by condition interaction (p=.012)(Figure1). ΔSCL in the ASD group was comparable to the TD group in the Anger (p=.58) and Joy (p=.17) conditions, but was lower in the Fear condition (p=.016). In the two groups combined, ΔSCL was positively correlated with intensity of Anger (r(36)=.30, p=.07) and Fear (r(35)=.45, p=.008), but not with Joy (p=.75). Regression analysis of ΔSCL in the Fear condition revealed a significant contribution from parent-reported ECBQ Fear score (p<.002) and observed intensity of anger and fear (both p<.002, Adjusted R²=.42), but not from autism severity, NVDQ, or VDQ.

Conclusions: Physiological arousal increased in response to negatively- but not to positively-valenced challenges in both groups. The magnitude of changes in physiological response to Anger and Joy probes was comparable between ASD and TD toddlers. However, ASD toddlers showed an attenuated physiological response to Fear, mirroring the differences observed at the behavioral level (Macari et al., 2017). Changes in physiological arousal level were predicted uniquely by both parent-rated fearful temperament and intensity of behavioral fear responses, but not by severity of autism or levels of functioning. These findings show the usability of measuring physiological arousal when examining emotional response in toddlers with ASD and inform both clinical and theoretical approaches regarding response to threat in ASD.