Negative Versus Positive Emotion Identification in Children with Autism Spectrum Disorders

Background:  Prior research on emotion identification in autism spectrum disorders (ASD) has not produced consistent results. Although some studies have indicated no differences in emotion identification for individuals with ASD as opposed to controls (Capps, Yirmiya, & Sigman, 1992; Ozonoff, Pennington, & Rogers, 1990), other studies have reported deficits in emotion identification for those with ASD (Hobston, Ouston, & Lee, 1989; Macdonald et al., 1989).  Deficits related to specific emotions, such as fear, have also been identified (Pelphrey, Sasson, Reznick, Paul, Goldman, & Piven, 2002).  This study aims to clarify the ability of children with ASD to identify emotions in dynamic, talking videos, as many previous studies have used static photographs, which are less representative of real-life social interactions.

Objectives:  To examine the ability of children with ASD to identify emotional expressions in dynamic videos as compared to children with typical development, when matched on mental age.

Methods:  Participants included 8 children with diagnoses of Autistic Disorder (n = 4) or Asperger’s Disorder (n = 4) between the ages of 7 and 12 (7 boys, 1 girl), and 18 children with typical development between the ages of 4 and 12 (9 boys, 9 girls).  Participants observed 20 brief videos (under 5 seconds) of adult women talking while exhibiting one of five emotions (happy, sad, angry, scared, and excited).  Video clips were rated in advance by faculty and graduate students for their child-appropriateness and success in communicating the desired emotion.  Following each video clip, participants were asked, “How does she feel?”  Participants’ responses were documented and scored for accuracy, and mental ages were computed based on scores from the Kaufman Brief Intelligence Test, Second Edition (KBIT-2; Kaufman & Kaufman, 2004).  

Results:  Preliminary analyses indicated that the group with ASD identified positive emotions (i.e., happy and excited) with significantly less accuracy than did the group with typical development, t(24) = -2.982, p = .006, although the two groups did not differ in their ability to identify negative emotions (i.e., sad, angry, and scared), t(24) = -0.242, p = .811.  The group with ASD and the group with typical development evidenced comparable mental ages, t(23) = -1.320, p = .200.

Conclusions:  Results indicate that children with ASD were equally able to identify negative emotions as their counterparts with typical development, but did not perform as well as those with typical development when asked to identify positive emotions such as happy and excited. Therefore, children with ASD may be perceiving similar levels of negative emotions but fewer positive emotions in others. It is possible that the positive emotions are being misidentified as negative emotions by children with ASD, which could indicate a bias towards attributing negative emotions in others. This would be important to clarify in future research as such biases pose a risk for future emotional and behavioral difficulties (Meyer, Mundy, Van Hecke, & Durocher, 2006).