Predictors of variable language development in autism

Project Summary Impairments in language development are common in children with autism and can have a profound influence on their future developmental outcomes and quality of life. There is great variability in both language ability at the time of autism diagnosis and in the rate of language gains during preschool years. Importantly, language

ability by early school age is one of the best predictors of future academic success, behavioral functioning, and independence. However, our understanding of factors, including underlying neurobiological mechanisms, impacting early language development in autism is limited, reducing our ability to develop effective

interventions and improve outcomes. Through harmonization of longitudinal and cross-sectional data collected from 165 autistic children across four studies, the proposed secondary analysis aims to identify early EEG indices of delays in language acquisition in ASD, and identify neurobiological, behavioral, and

environmental factors that prevent or promote further language gains. The goal of using this approach is to identify specific mechanistic factors that impede language early language acquisition in autism. To do this we will use neuroimaging data, language assessments, and natural language samples collected during early

development to (1) Characterize resting state electrophysiological differences in autistic children with and without language impairment at 2–3-years of age; (2) Identify early neural markers that predict limited language gains in ASD; and (3) Identify and characterize neurobiological, behavioral, and environmental factors that

predict greater language gains in 2-year-olds with ASD. In alignment with the goals of the NIH Tackling Acquisition of Language in Kids (TALK) initiative, this proposal leverages existing longitudinal and cross-sectional data sets to understand developmental trajectories of late talking children with autism. Research activities include harmonization of EEG processing and behavioral

measures across multiple data sets, additional transcription and behavioral coding of parent-child interactions, and re-consenting participants to make data available to NIH data repositories. Results from this study will advance our understanding of neurodevelopmental pathways preventing or promoting language development in autism and inform new methods for early detection and therapeutic

intervention.