Imitation Impairments in Autism Spectrum Disorder: A Social Motivation or Motor-Execution Problem?

Background: Imitation is an important early developmental skill; it provides a means to acquire practical knowledge and to practice and develop interpersonal skills (e.g., Munson, Meltzoff, & Dawson, 2006). Imitation deficits are a characteristic feature of autism spectrum disorder (ASD) and, while several explanatory theories have been proposed, the mechanisms underlying these difficulties remain unclear (Edwards, 2014).

Objectives: Within a mixed quasi-experimental and individual differences design, this study examined the contribution of reduced social motivation and motor-execution difficulties for propensity to imitate among children with ASD. 

Methods: Among 55 child participants – 35 with ASD, 20 typically developing (TD) – a novel imitation task was delivered via . This included four experimental conditions arising from manipulations of social and motor domains; to-be-imitated actions were presented by a social and an asocial model, and under conditions of low- (single action) and high motor-demand (multiple actions in sequence). Imitative performance in each condition was coded such that higher total score indicated more accurate imitation. Several standardized assessments were also conducted; fine motor coordination was assessed using the Vineland Adaptive Behavior Scales (VABS-II; Sparrow, Cicchetti, & Balla, 2005) and Mullen Scales of Early Learning (MSEL; Mullen, 1995), and social motivation was assessed among children with ASD using the Autism Diagnostic Observation Schedule, Social Affect domain (ADOS-2 SA; Lord et al., 2012).

Results: There was a nonsignificant trend towards overall reduced imitation by children with ASD vs. TD children. Imitative performance was significantly better in the low- relative to high motor-demand condition, for each group with no between-group differences in performance in either motor demand condition. There were no between-group imitation differences in response to the social vs. asocial model. However, a significant three-way interaction term revealed that children with ASD imitated more poorly during the high- relative to low motor-demand condition, when demonstration was by a social model. Among children with ASD, propensity to imitate was not associated with ADOS-2 SA score, and showed a conflicting pattern of correlations with fine motor skills as assessed via the MSEL (significant positive association) and VABS-II (no significant association).

Conclusions: While there was a trend for poorer imitation by children with ASD, the lack of clear group differences on this novel task of spontaneous imitation did not provide robust replication of previously-reported imitation deficits in this population. Moreover, findings of a) a lack of between-group differences in imitation in response to a social vs. asocial model, and under conditions of low- vs. high motor-demand, and b) nonsignificant correlations between imitation performance and standardized measures of sociability and fine motor skill suggest that neither social motivation nor motor-execution capacities decisively underpin the propensity to imitate by children with ASD. Interestingly, these data suggest that imitative performance in ASD may arise from a complex interplay between these factors, as imitation was poorer for children with ASD as both the social-processing and motoric task demands increased. Further research is warranted to understand precisely how social motivation and motor execution capacities combine to influence spontaneous imitation in ASD.