Due to genetic and phenotypic heterogeneity, the etiology of autism spectrum disorder (ASD) is not well understood. Endophenotypes of ASD at the neural systems level may offer insight into the pathophysiology and psychopathology of ASD by indicating components of complex social behaviors that lie closer to specific genetic factors conferring ASD risk. This project investigates the neural correlates of imitation -- a social function impaired in ASD -- in 4- to 6-year-old children with ASD, unaffected siblings of children with ASD, and their typically developing peers. Unaffected siblings represent a unique and potentially essential study population for understanding ASD risk and development, as these children are likely to share many of the genetic risk factors that contribute to ASD in their siblings, despite themselves exhibiting typical social functioning. This research training project integrates imaging and behavioral methods to identify 1) neural correlates of imitation unique to the state of having ASD; 2) neural correlates of social dysfunction shared by children with ASD and unaffected siblings, and 3) neural correlates of imitation unique to unaffected siblings. Genetic factors or regions associated with these endophenotypes represent ripe avenues for future research into factors that protect against ASD development. This project thus aims to identify and investigate biomarkers of immediate clinical utility in the earlier diagnosis and prediction of ASD outcomes. The results of this study may also contribute to the design of more targeted and potentially effective interventions to increase the social functioning and improve the outcomes of children at risk for ASD.