Social Motivation Is Correlated with Face Processing Skill in Children with ASD

Background: Numerous studies have found that children with autism spectrum disorders (ASD) have face processing impairments compared with typically developing controls (e.g. Wolf et al., 2008) yet, the precise etiology of these deficits remains unknown (Schultz 2005). Recent theories posit that early neurodevelopment is derailed due to diminished social interest (Grelotti et al., 2002). In line with this idea, eye tracking experiments have revealed that toddlers and children with ASD display a preference for inanimate objects over people (Nakano et al., 2010; Shic et. al., 2011).  Due to this diminished social interest, children with ASD would then fail to develop expert face processing skills like identity and emotion recognition. Despite the influence of the social motivation framework, however, the relationship between social motivation and face processing skills has not yet been directly addressed.

Objectives: Our primary goal was to test whether reduced social motivation is correlated with performance on face and emotion recognition tasks in children with ASD. We used gaze pattern in a passive viewing task as our measure of social motivation, and a computer task specifically designed to assess face processing skill. We expected to find a positive correlation between face processing skills and social motivation in children with ASD.

Methods: 14 children with ASD and 18 typically developing children matched on age (Mean=9.6 ± 2.95), IQ (Mean=119.93 ± 14.31) and sex ratio, passively viewed a series of scenes depicting dynamic images of facial expressions and moving objects. Two tasks from the Let’s Face it! Skills battery were used to assess face processing skill: one testing face recognition (LFI identity) and the other testing emotion recognition (LFI expression) (Tanaka et al., 2010). 

Results: Fixation count, fixation duration and time to first fixation to faces and objects were analyzed. Unlike TD children, children with ASD were not significantly more likely to fixate on faces before objects (ASD: t(13)=.80, p=0.44, d=.344; TD: (t(17)=3.70, p=.002, d=1.36) and had shorter fixation duration on faces compared to their TD counterparts (t(30)=-2.24, p=.03, d=.804 ). In the ASD group, fixation duration was positively correlated with accuracy on the LFI expression task (r=.689, p=0.006) while number of fixations on objects was negatively correlated with task accuracy (r=-.689, p=.006).  Additionally, children with ASD who performed better on the LFI identity task looked fewer number of times at objects of high autism interest (e.g., trains, airplanes) (r=-.539, p=0.047).

Conclusions: Consistent with previous studies, we found that children in the ASD group have differences in gaze pattern compared with TD counterparts, suggesting a tendency to prioritize objects over people (Nakano et al 2010). In the current study, the ASD group’s gaze patterns were correlated with facial recognition skills, lending evidence to the theory that social motivation relate to face and emotion processing skills.