Physiological reactivity to faces via live and video-mediated communication in typical and atypical development

Abstract

The human face is a powerful elicitor of emotion, which induces autonomic nervous system responses. In this study, we explored physiological arousal and reactivity to affective facial displays shown in person and through video-mediated communication. We compared measures of physiological arousal and reactivity in typically developing individuals and those with the developmental disorders Williams syndrome (WS) and autism spectrum disorder (ASD). Participants attended to facial displays of happy, sad, and neutral expressions via live and video-mediated communication. Skin conductance level (SCL) indicated that live faces, but not video-mediated faces, increased arousal, especially for typically developing individuals and those with WS. There was less increase of SCL, and physiological reactivity was comparable for live and video-mediated faces in ASD. In typical development and WS, physiological reactivity was greater for live than for video-mediated communication. Individuals with WS showed lower SCL than typically developing individuals, suggesting possible hypoarousal in this group, even though they showed an increase in arousal for faces. The results are discussed in terms of the use of video-mediated communication with typically and atypically developing individuals and atypicalities of physiological arousal across neurodevelopmental disorder groups.

Keywords:

• Williams syndrome
• Autism spectrum disorder
• Emotion
• Physiological arousal

Acknowledgments

This research was supported by a grant from the Economic and Social Research Council (RES-062-23-1365) to G.D.S. and D.M.R. We are grateful to the Williams Syndrome Foundation and to the “Database for Children with Autism Spectrum Disorders Living in the North East of England” (Daslⁿe) for their support. We are especially grateful to all individuals who participated in the study.

Notes

¹With regard to accuracy in the recognition phase, participants were asked to first spontaneously label the emotional expressions. This was correctly completed by 67% of the ASD, 92% of the ASD–TD matches, 42% of the WS group, and 83% of the WS–TD matches. Critically, all individuals who had problems spontaneously labeling the emotional expressions were then given the forced-choice labels for the emotions, and at this point all participants (100% in each group) were able to match the emotional labels to the face stimuli.

²The lack of significant interaction may be driven by relatively low power to detect this interaction. When studied individually, the TD group showed a significant difference of SCL for the baseline and LF conditions (p < .05) and between the LF and VM conditions (p < .01). However, the ASD group showed no significant difference between any of the conditions (difference baseline to live face is p = .53).

³It is worth noting that, corroborating the suggestion of lower baseline arousal reported in Doherty-Sneddon et al. (2009), at baseline the WS group were significantly lower in terms of their SCL than the TD group, t(24) = 2.64, p < .05 (mean TD 9.02, WS 7.04). This goes along with the trend towards significance when looking at the overall effect of group in the ANOVA.

⁴As seen in both Figure 2 and Table 2, there was no clear difference in SCL or SCR for the older and younger TD control groups—those matched to the ASD group were significantly younger than those matched to the WS group, t(23) = 3.34, p < .01. To ensure that any effects in the current data were not driven by effects of chronological age on physiological arousal in typical development, we explored the relationship between chronological age and both SCL and SCR. There were no significant relationships (ps > .05), and therefore we conclude that any effects seen here were not driven by chronological age.

⁵Note that no individuals from the current study had participated in the previous research reported by Doherty-Sneddon et al. (2009). Although the sample size of both studies is reasonably small due to the prevalence of the disorder, together the findings support the notion of low arousal levels associated with WS.