Eye- Versus Mouth-Looking Patterns in Emotional Contexts in Children with ASD
Objectives: To examine looking at the eye versus mouth in children with ASD versus non-ASD during a social-information-seeking task depicting escalating emotional expression.
Methods: Thirty-six 4 to 8 year-olds, ASD n = 14 (MDQ= 88, SD = 20) and non-ASD n = 22 (MDQ= 109, SD=12), watched a video of an actress involved in a stressful activity (e.g. blowing up a balloon until it pops) that ends in a Resolution (e.g. actress sighs in relief after balloon pops). Previous eye tracking analyses had shown that children with ASD spent significantly less time looking at the face (Face%) during the Resolution than non-ASD children (p < .05) (Foster et al., unpublished). We examined the components of %Face to examine between-group differences in eye- and mouth-looking (%Eyes and %Mouth) within this escalating emotional context.
Results: Immediately prior to the Resolution, there was no difference in face, eye or mouth gaze allocations between ASD and non-ASD children (see Figure 1). During the Resolution, there was a main effect of diagnosis on %Mouth (F(1,36), p < .05, ηp2 = .157) with children with ASD spending less time on the mouth (M = .76, SD = .42) than non-ASD children (M = .13, SD = .06). No diagnosis effect on %Eyes was observed between non-ASD and ASD (Ms = .22 and .17, SDs = .03 and .03, respectively). Across all participants, there was a positive correlation between %Mouth and DAS-II Verbal Reasoning Standard Scores (r = .381, p < .05).
Conclusions: Children with ASD spend less time viewing the face than do non-ASD children in Emotional Contexts but the difference is driven by lower mouth- as opposed to eye-scanning. Given that the Resolution sets up the expectation of an emotional context, it could be that non-ASD children look to the mouth for additional input during periods of emotional expectation, a tendency that may not be developed in children with ASD. If so, then the results support studies that find mouth gazing to be a functional and adaptive behavior that can supplement other communicative visual cues, especially in situations that warrant context deduction (i.e. how is the actress feeling now). However, the correlation between %Mouth and verbal scores suggests an alternative interpretation which links insensitivity to cues during Resolution to communicative competence rather than social disability, specifically.