Project Summary Current gold-standard diagnostic techniques for Autism Spectrum Disorder (ASD) rely on behavioral observation, and diagnosis is often not conferred until after age 4. This delay is unfortunate, because early intervention can drastically improve outcomes. Given the rapid and sweeping changes in neural architecture, cognitive ability, and behavioral repertoire that occur within the first year of life, identifying early-emerging neurobiological markers that can predict abnormal neurodevelopment before behavioral symptoms manifest is a public health priority. Following the onset of behavioral signs and symptoms, there is great need for stratification biomarkers that can dissect ASD heterogeneity, thereby informing individualized treatment plans. Such a precision-medicine approach necessitates greater understanding of the longitudinal pathways of development in domains. This will set the stage for biomarkers that are sensitive to, and predictive of, changes in behavior resulting from effective interventions. This proposal aims to identify and validate features of temperodynamic brain signal variability that can serve as diagnostic, risk, and/or treatment response biomarkers of social dysfunction in ASD. Traditionally modeled out of analyses as mere “noise”, measures of temperodynamic neural variability capture the inherently fluctuating nature of the brain, which is increasingly understood to play a valuable functional role in the establishment of neural networks and in the transfer of information throughout the brain. Temperodynamic neural variability has been linked to cognitive performance, development, and autism. The current proposal will capitalize and expand upon this promising neurobiological marker to 1) identify and optimize metrics for assessing temperodynamic neural variability by conducting the most comprehensive comparison of time-series analytics of any study to date, and 2) establish these metrics as biomarkers for neurodevelopmental outcomes by leveraging large clinical and longitudinal data sets consisting of multilevel genetic, neural, and behavioral data. The proposed research extends the candidate’s prior work through new training in autism research, translational developmental cognitive neuroscience, and timeseries and predictive analytics applied to large- scale datasets necessary to advance biomarker development. These training goals will support the candidate’s ultimate career goal of developing an independent research program dedicated to the use of interdisciplinary, multidimensional, collaborative, and cutting-edge approaches to understanding the neurobiological and developmental factors that contribute to individual differences in social behavior across the full continuum of abilities – from healthy to disordered. The University of Virginia is committed to high caliber, collaborative research and scientific preeminence in neuroscience, autism, and data science and will provide the ideal environment for ...