Impulse Control to Specific Interests in Children with Autism Versus Typical Development

Background: Deficits in impulse control and behavioral regulation are characteristic of autism spectrum disorder (ASD). Impulsivity negatively impacts learning and interferes with behavioral interventions in ASD. In typical development (TD), prior work has shown increased impulsivity to happy faces but in ASD, social cues have less significance. It is unknown how cues that are salient to children with ASD, such as special interests, influence impulsivity.

Objectives:  The goal of this study was to examine impulse control to interests versus social cues in ASD and TD. We hypothesized that impulsivity in children with ASD would be greatest to special interests versus faces and the opposite pattern would be observed in TD.

Methods: 34 children participated: 15 children with ASD (13 male, mean age 9 (7-12 yrs), mean VIQ = 97; mean NVIQ = 94; mean ADOS CSS Total = 8.4) and 19 TD children (18 male, mean age 9 (5-12 yrs), mean VIQ= 115, mean NVIQ = 115). Children performed a novel go/nogo task on an iPad with 5 different conditions. For non-social conditions, children chose their favorite hobby (interest) and their least favorite hobby (non-interest). For social conditions, children were presented with happy or calm faces. A control condition presented colored shapes. Children were instructed to touch ‘go’ cues and to withhold responses to ‘nogo’ cues. All 5 conditions served as both go and nogo cues and were counterbalanced across participants. D-prime was calculated as the normalized hit rate (go-accuracy) minus normalized false alarm rate. Reaction time variability was assessed with ex-Gaussian distribution functions.

Results: Go-accuracy did not differ by group, demonstrating all children attended to the conditions. For d-prime using linear mixed effects models there was an interaction between condition and group (F(105.84) =4.993, p = 0.001) demonstrating in TD there was lower d-primes (greater impulsivity) to happy faces relative to interests (p=0.023) and lower d-prime to neutral faces relative to interests (p=0.008). As hypothesized, children with ASD demonstrated a trend of lower d-prime to their interests relative to TD (p=0.093), with no differences between groups to faces or non-interests, suggesting that children with ASD may be more biased and distracted by their interests relative to TD. There were no significant differences between conditions in the ASD group and no significant interactions between condition and diagnosis on the ex-Gaussian functions. These parameters will be explored further as data collection is ongoing.

Conclusions: Children who were typically developing were more impulsive to socials cues than their interests, suggesting that interests elicited greater attention and facilitated inhibition. In contrast, there was a trend that children with ASD were more biased by their interests relative to TD, suggesting that interests may be particularly distracting for children with ASD and decrease inhibition. Ultimately identifying cues that are salient to children with ASD and thus increase impulsivity and distraction will be important for improving interventions that target attention and behavioral regulation.