The Neuroeconomics of Behavioral Therapies for Adolescent Substance Abuse

Little is known about the role of adolescent neurodevelopment in adolescent substance abuse treatment outcomes. The development and evaluation of adolescent substance abuse treatments rarely includes consideration of varying cognitive capacities and their neural representations as determinants of individual variation in treatment response. This proposed R21 project would address this gap by identifying associations among decision making, task-related neural processing, and treatment outcome among adolescents participating/have participated in two randomized clinical trials for adolescent substance abuse. Increasing our understanding of neural processes that underlie decision making in adolescent marijuana and alcohol users would inform the development of future intervention and prevention efforts.

This complementary project draws subjects from two studies investigating contingency-management (CM) based treatments "Behavioral Treatment of Adolescent Marijuana Abuse" (DA015186), and "Family Based Contingency Management for Adolescent Alcohol Abuse" (AA016917). Both trials compare a unique CM intervention that involves an abstinence-based reinforcement program to a standard, state of the art cognitive behavioral intervention. Analyses would be performed separately for the two samples, with hypotheses tested first using the Marijuana sample, and assessed for replication/specificity using the Alcohol sample. The proposed project would explore novel neurobiological predictors of response to CM interventions.

Adolescents recruited into the Marijuana Trial (n=69; 23 per treatment arm) and the Alcohol Trial (n=54; 27 per treatment arm) during the period of this R21 project would participate in a single neuroimaging session as soon after the intake session as possible, within 7 to 30 days. During the neuroimaging session, adolescents would make intertemporal choice decisions in a Delay Discounting task regarding gains of $100 and $1,000. Preliminary data from the Marijuana trial demonstrates significant association between performance on this behavioral delay discounting task and abstinence achieved during treatment over and above the significant effect of treatment condition. The investigators seek to understand the neural processes that underlie performance on this laboratory task, and the degree to which variation in these neural processes relates to and predicts adolescent substance abuse treatment outcomes. The conceptual framework for the proposed project is based on the neuroeconomics of addiction in general, and the competing neural systems hypothesis in particular (Bechara, 2005a; Bickel, et al., 2007; Daw, et al., 2005; Jentsch & Taylor, 1999). This behavioral choice hypothesis postulates a biased competition between an "impulsive" (or ''reflexive") neural system (including the striatum, amygdala, ventral pallidum, and related structures) and the "executive" (or "reflective") neural system (including the prefrontal cortex) in understanding patterns of suboptimal decision making among substance-dependent individuals. An overarching hypothesis is that differing responses to distinct treatment approaches (CM vs. CBT) are partially determined by the pattern of activation or functional connectivity within and across these competing neural systems. These project hypotheses would be tested by the following specific aims:

Primary Aims:

• Determine the degree to which performance on a laboratory delay discounting task correlates with activity in impulsive and/or executive neural systems.
• Determine the degree to which performance on a laboratory delay discounting task and neural processing predict adolescent substance abuse treatment outcome over and above the effects of treatment condition.