Autism spectrum disorders (ASD) are extremely heterogeneous, presenting with a wide range of phenotypic characteristics. High twin concordance and increased sibling recurrence suggest that genetic risk factors play a significant role in development of the disorder. Efforts to identify ASD causative genes have yielded a handful of associated risk variants, each contributing to only a small fraction of ASD cases. It is evident that many more such variants exist, but have yet to be discovered. Collectively, these individual gene variants may account for a large portion of ASD cases. Gene-hunting efforts to date have relied heavily on coordinated efforts of networks such as the Autism Genome Project (AGP), and the availability of large-scale sample collections, such as the Autism Genetics Resource Exchange (AGRE) and the more recently available Simons Simplex Collection (SSC), which together provide a current sample set of approximately 4,000 ASD families. Albeit a tremendous asset, the existing sample size is insufficient to identify the vast majority of ASD-related risk variants. Growth of existing genetic collections is limited by traditional research methods requiring individuals with ASD to undergo time-consuming, costly, clinic-based diagnostic assessment and exhaustive phenotyping as part of the research protocol. In this study, the investigators aim to develop and test a rapid phenotyping protocol (RPP) that will be reliable and feasible for ascertainment of non-verbal ASD (NV-ASD) for large-scale genetic research. Existing literature which shows good correlation between parent-provided measures and clinical assessments, and preliminary analysis of IAN data provide supportive evidence that an RPP can ultimately be successful for NV-ASD. By utilizing the Interactive Autism Network (IAN), the nation’s largest online autism research effort, an internet-mediated recruitment and ascertainment strategy will allow researchers to quickly recruit a large number of subjects for genetics analysis at a fraction of the usual costs. This project seeks to further demonstrate that a rapid phenotyping approach executed via the internet can be developed to diagnose nonverbal children with ASD with demonstrated agreement with “gold standard” diagnostic assessments.