Project Summary/Abstract Early initiation of alcohol use (EIAU) during adolescence is a key contributor to development of lifetime alcohol (AUD) and/or substance use disorders (SUDs). A critical component for preventing and intervening EIAU is early detection using behavioral or cognitive risk precursors; and understanding better their underlying neural substrates before the onset of substance use may aid in developing more effective and targeted interventions. Prior studies suggest sleep disturbance, internalizing behaviors, externalizing behaviors, and cognitive performance are potential risk precursors of EIAU. However, mixed or inconsistent findings have been reported which may reflect study design and methodological issues, and the neural substrates subserving co- development of these frequently occurring problems remain to be identified. In this application, we propose to use data from the longitudinal Adolescent Brain Cognitive Development (ABCD) study to investigate the neural substrates that are predictive of EIAU and its risk precursors. A number of advanced analytical approaches will be used to extract novel features at both structural and functional levels. For example, the Joint and Individual Variance Explained (JIVE) method will be used to extract biologically meaningful cortical-subcortical covariation patterns. Existing literature and our preliminary results led us to hypothesize that the cortical-subcortical covariation pattern including brainstem, thalamus, prefrontal cortex and other regions underlies the co- development of sleep, internalizing behaviors, externalizing behaviors, and poor cognitive function from late childhood to mid-adolescence. Specifically, this project has three complementary aims. Aim 1 will assess the relationships between the baseline cortical-subcortical covariation patterns and EIAU and its four risk precursors in mid-adolescence. Aim 2 will characterize the developmental trajectory of cortical-subcortical covariation pattern and assess its dynamic relationship with the four risk precursors longitudinally. Aim 3 will explore whether functional connectivity predicts sleep disturbance, cognitive performance, behavioral problems, and/or age at first alcohol use. Results from this application will significantly advance the field of addiction neuroscience.