Project Summary/Abstract Insufficient knowledge of brain-based risk factors for problem cannabis use is a key barrier to identifying individuals at risk for cannabis-related harm and designing targeted prevention and early intervention approaches to promote positive development for persons at risk. The current proposal combines multiple, large- scale, neuroimaging datasets with novel densely-sampled fMRI data to identify a neuromarker of problem cannabis use and characterize its development prior to cannabis use onset, and in the context of current cannabis use in adolescence. We will apply a whole-brain, machine-learning method – connectome-based predictive modeling – to identify a neural network predictive of problem cannabis use in a large sample of college students (Brain and Alcohol Research in College Students study; BARCS), and compare this network to canonical neural networks previously implicated in risk for cannabis use: frontoparietal, salience, and default mode networks. We will then examine the developmental trajectory of these cannabis risk networks in relation to other known risk factors for addiction prior to cannabis use onset in the Adolescent Brain Cognitive Development (ABCD) dataset, a nationally-representative sample of 11,875 children. We will examine whether individuals at high-risk for substance use problems (family history of substance use and exposure to early adversity) display a delay in the typical trajectory of network stabilization within cannabis risk networks. Finally, we will collect densely-sampled fMRI data (4 scans in 6 months) from 20 adolescents (age 15-17) who regularly use cannabis and 20 age- and sex-matched typically-developing adolescents. This study design will enable us to examine cannabis effects on short-term neural network dynamics and evaluate whether cannabis using adolescents are characterized by reduced stability of cannabis risk networks relative to their typically-developing peers. By combining multiple existing large-scale neuroimaging datasets with original, longitudinal, densely- sampled fMRI data, the proposed study design offers a unique opportunity to examine the interplay between neural mechanisms of risk for problem use and cannabis exposure effects on the developing adolescent brain and promises to yield important insights into the neural mechanisms of risk for problem cannabis use that can foster novel prevention and intervention initiatives to mitigate cannabis-related harms for individuals at high risk. The current proposal also aims to provide Dr. Sarah Lichenstein with expert mentorship by Drs. Pearlson, Casey, Yip, Scheinost and Stevens to build the skills necessary to develop into an independent clinical scientist applying multimodal neuroimaging methods to study the pathophysiology of addiction. Dr. Lichenstein will pursue specialized training in studying cannabis effects on brain and behavior, big data science, and the conduct of research with adolescent participants. Fur...