The Effects of Multimodal Behaviors from Parents on Children's Sustained Attention: A Dual Eye-Tracking Study in Naturalistic Child-Parent Social Interaction

Background: Sustained attention has traditionally been studied as an endogenous ability of a child (Mundy & Newell, 2007), but recent work suggests that it is socially modulated. Utilizing wearable eye trackers to study the dynamics of parent-child play, Yu & Smith (2016) found that typically-developing (TD) infants sustained their attention on toys for longer when the parent looked to the toy at the same time as compared to when the infants looked at the toy alone. Particularly, infants sustained their attention most when the parent looked at, touched, and talked about the toy to which they were attending (Suarez-Rivera et al., 2018). This paradigm has yet to be extended to study if and how sustained attention is modulated by parent behavior in children with autism spectrum disorder (ASD).

Objectives: The current study aims to determine whether young children with ASD (24-to-48-months-old) modulate their attention as a function of their parent’s attention compared to age-matched TD children while engaging in toy play. We will investigate (1) if coordinated parental attention extends sustained attention in both groups and (2) if different combinations of parent behaviors – looking, touching, and talking – relate to differential increases in sustained attention.

Methods: Child-caregiver dyads played with 24 toys in a toy room space while wearing head-mounted eye-trackers (Figure 1). Each dyad was given the toys to freely play with, and encouraged to play as they would at home. Sustained attention moments were defined as child looks to toys that were 3 seconds or greater. Sustained attention moments were “coordinated” if the parent joined the child in looking at the toy and were “solo” if the child looked alone. Coordinated sustained attention moments were further divided based on their co-occurrence with parent touch and talk.

Results: Data was acquired from 19 dyads in the ASD group and 17 dyads in the TD group. The groups did not differ in the amount of the play session spent in sustained attention (Figure 2a) or in the amount of sustained attention spent in coordination with a parent (Figure 2b) (all ps>0.47). Both groups showed an extension of sustained attention with coordinated attention from a parent compared to when attending alone (Figure 2c) (all ps>0.04), and there were no differences between groups in this extension (p=0.15). A linear mixed effects model revealed a significant main effect of parent behavior type (p<0.01). However, there was no effect of group and no interaction effect of group by parent behavior in the extension of sustained attention (all ps>0.15).

Conclusions: In the current study, we leveraged dual head-mounted eye-tracking to demonstrate the feasibility of studying naturalistic parent-child interaction with young children with ASD. Our preliminary results show present social modulation of sustained attention in children with ASD. Further, sustained attention is extended similarly by different parent behavioral cue combinations in the ASD group compared to the TD group. Future studies will examine how variability in social modulation during naturalistic play may relate to language abilities and symptom profile to inform individualized targets for intervention.