The pupil: a window on social automatic processing in autism spectrum disorder children

• Published in: Journal of child psychology and psychiatry 2020-07, Vol.61 (7), p.768-778
• Main Authors: Aguillon‐Hernandez, Nadia, Mofid, Yassine, Latinus, Marianne, Roché, Laetitia, Bufo, Maria Rosa, Lemaire, Mathieu, Malvy, Joëlle, Martineau, Joëlle, Wardak, Claire, Bonnet‐Brilhault, Frédérique
• Format: Article
• Language: English
• Subjects: Autism; autism spectrum disorder; Autism Spectrum Disorders; Autistic children; Child & adolescent psychiatry; development; dynamism; emotion; Emotion recognition; faces; Facial expressions; Life Sciences; Neurons and Cognition; neutral; Physiological arousal; Physiology; Pupil; Stimuli; Stimulus; Students
• ISSN: 0021-9630; 1469-7610
• DOI: 10.1111/jcpp.13170

Background Faces are crucial social stimuli, eliciting automatic processing associated with increased physiological arousal in observers. The level of arousal can be indexed by pupil diameter (the ‘Event‐Related Pupil Dilation’, ERPD). However, many parameters could influence the arousal evoked by a face and its social saliency (e.g. virtual vs. real, neutral vs. emotional, static vs. dynamic). A few studies have shown an atypical ERPD in autism spectrum disorder (ASD) patients using several kinds of faces but no study has focused on identifying which parameter of the stimulus is the most interfering with face processing in ASD. Methods In order to disentangle the influence of these parameters, we propose an original paradigm including stimuli along an ecological social saliency gradient: from static objects to virtual faces to dynamic emotional faces. This strategy was applied to 186 children (78 ASD and 108 typically developing (TD) children) in two pupillometric studies (22 ASD and 47 TD children in the study 1 and 56 ASD and 61 TD children in the study 2). Results Strikingly, the ERPD in ASD children is insensitive to any of the parameters tested: the ERPD was similar for objects, static faces or dynamic faces. On the opposite, the ERPD in TD children is sensitive to all the parameters tested: the humanoid, biological, dynamic and emotional quality of the stimuli. Moreover, ERPD had a good discriminative power between ASD and TD children: ASD had a larger ERPD than TD in response to virtual faces, while TD had a larger ERPD than ASD for dynamic faces. Conclusions This novel approach evidences an abnormal physiological adjustment to socially relevant stimuli in ASD.