Implicit Social Evaluations in Toddlers with Autism Spectrum Disorder

Background: Children with autism spectrum disorder (ASD) show pervasive social and communicative deficits. The current study investigates what social abilities may underlie these difficulties. Specifically, we test whether toddlers with ASD form implicit expectations about third-party social interactions.

Objectives: The goal is to clarify whether toddlers with ASD construct social evaluations similarly to their typically developing (TD) peers, using two methodologies that were originally designed for use with preverbal infants (Kuhlmeier, Wynn, & Bloom, 2003; Hamlin, Wynn, & Bloom, 2008).

Methods: Both the ASD (n = 14; 10 males; M = 24.13 months, SD = 4.46) and TD (n = 21; 8 males; M = 23.44 months, SD= 3.04) participants were recruited from a larger study using the Modified Checklist for Autism in Toddlers, Revised and Follow up (M-CHAT-R/F; Robins, Fein, & Barton, 2009). Each child habituated to a live display in which a red circle started at rest at the bottom of the hill and tried repeatedly to climb to the top (see Figure 1). Sometimes the red circle was pushed up to the top by a helper, and other times it was pushed down to the bottom by a hinderer; the helper/hinderer identity and order of events was counterbalanced between children. In six subsequent test trials, children saw the red circle move toward either the helper or the hinderer and looking time was recorded. After the looking time procedure, children were presented with wooden versions of the helper and hinderer and were asked to choose one.

Results: A repeated-measures 2x2 ANOVA of looking times during test trials revealed a significant main effect of actor, F(1, 33) = 4.84, p = .035 (see Figure 2). Children across groups looked longer when the circle approached the hinderer than when it approached the helper. There was no main effect of diagnostic group, F(1, 33) = .10, p = .69, and no interaction, F(1, 33) = .16, p = .75. In the choice task, TD children showed a reliable preference for the helper over the hinderer. Of the 19 who made a choice, 15 chose the helper (one-tailed binomial p < .001). Children with ASD did not show a reliable preference (5 of 12 chose the helper; p = .39).

Conclusions: Across diagnostic groups, children showed surprise when the red circle approached the shape that had previously hindered it, as evidenced by prolonged looking. Future larger samples will allow for more detailed analyses of within-group effects. Results from the choice task paint a clearer picture; TD children reliably preferred the helper, whereas children with ASD did not. These results suggest that TD two-year-olds are capable of making implicit social evaluations after viewing live displays, and 2-year-olds with ASD have at least some ability to do so. However, unlike their TD peers, children with ASD fail to integrate these judgments in their explicit behaviors. This study will add to the greater knowledge of the underlying social processes affected in children with ASD.