Behavioral Differences in Reward Salience but Not Motivation in Toddlers with ASD: Results from a Visual Search Task

Background: Visual search is consistently documented as an area of superior skill in adults and adolescents with ASD (Kaldy et al., 2013); however, few studies have been conducted in younger samples or samples with a broad range of functioning.  It has been hypothesized that children with ASD find social stimuli less rewarding than do people with typical development (Dawson et al., 2004; Dawson et al., 2005).  Based on current evidence, it is unclear whether differences in reward processing are specific to social stimuli.  Questions also remain about what aspects of reward processing are different in ASD. 

Objectives: There were three main goals of this study:  (1) to develop a visual search task for very young children with ASD that could be completed without any verbal instructions by using gaze-contingent eye-tracking; (2) to test differences between children with ASD and typical development in the motivation for and saliency of social and nonsocial rewards; and (3) to examine the effect of social and nonsocial reward on learning.

Methods: Two groups of two year olds participated in this study:  ASD (N =21) and typically developing (TD = 23). During the visual search task, participants were rewarded with either a social or nonsocial video when they fixated on the target shape among a circular array of distractor shapes.  Key dependent eye-tracking variables were reaction time (RT) to finding the target and looking time (LT) at the reward video when it played.

Results: Two sets of multilevel models were examined with RT and LT as the dependent measures, group (ASD vs. TD) as a main effect, condition (social vs. object) as a repeated measure, and block of trials as the time variable.  Three models were tested: no growth, linear, quadratic.  In the best fitting model of RT the ASD group demonstrated faster RT.  There was no significant interaction between group and reward condition, indicating no difference in motivation based on reward type.  Both groups demonstrated learning on the task as indicated by a significant decrease in RT across the task. The group by time interaction was not significant.  Learning did not vary in either group depending on reward condition (i.e., there was no group by time by condition interaction).  In the best fitting model of LT, a significant group by condition interaction was found, where children in the ASD group looked less at the social videos than nonsocial videos; whereas, the TD group had the opposite pattern of visual attention. 

Conclusions: Results replicate findings of superior visual search in children with ASD compared to TD children (Kadly et al., 2011). These results also add support to previous reports of lower saliency of social stimuli for toddlers with ASD compared to TD controls (Shic et al., 2011).  Although there was a group difference in saliency, there were no differences in motivation or learning based on reward type for either group.  According to these results, children with ASD may work to obtain social rewards to the same degree as TD children, but engage more with nonsocial stimuli.