Striatal Reward Sensitivity Predicts Substance Use for Adolescent Boys but not Girls

• In Event: 3-141 - Neural Reward Reactivity and Adolescent Psychopathology and Substance Use: fMRI and ERP Studies

Integrative Statement

Adolescence is a time of significant development of reward processing brain networks (Casey et al., 2008). In addition, during adolescence there are increases in risk-taking behaviors including substance use. Several investigators have hypothesized that changes in reward system function in adolescence contribute to heightened risk behaviors. However, few studies have examined associations between neural reward system function and actual adolescent risk behaviors. Initial studies have found that elevated caudate and putamen response to monetary reward and elevated NAc response to risk-related decision-making predict future substance use in adolescents (Morales et al., 2018; Stice et al., 2013), but more work is needed.

The present study examined associations between early adolescents’ striatal responses to monetary reward and substance use behavior one year later. We hypothesized that heightened reward reactivity would predict substance use. Further, given theories of sex differences in predictors of substance use (Hammerslag & Gulley, 2016), we examined sex differences in reward reactivity-substance use associations.

78 12-14 year olds (41 boys; 74% White) were recruited from a larger longitudinal behavioral study to participate in an MRI session at Time 1. In the MRI session, adolescents completed a monetary reward task in an event related design (Forbes et al., 2008). In each trial, adolescents guessed whether the value of a card would be higher or lower than 5 (3s), were told the trial type (possible win, possible loss) (3 s), and saw the card and were told the outcome (win, loss, or neutral) (1s). Youth won $1 for each win outcome. fMRI responses in three striatal regions of interest (ROIs; nucleus accumbens [NAc], caudate, putamen) to win outcomes and neutral outcomes were extracted using the FSL Harvard-Oxford Atlas. Analyses examined win outcome > neutral outcome contrasts to reflect response to reward. Adolescents reported on lifetime substance use and completed urine and breath screens at Time 1 and one year later.

Logistic regressions found significant ROI response to reward X sex interactions predicting substance use (yes/no) at one year follow-up, controlling for Time 1 substance use, for R caudate (β = 1.96, p = .04), L caudate (β =2.01, p = .04), and R putamen (β = 3.09, p = .02) (see Figure). Results were not significant for NAc. For boys, heightened R caudate response to reward predicted substance use (β = 1.49, p = .04) and heightened L caudate response to reward tended to predict substance use (β = 0.97, p = .09), with large effect sizes. For girls, lower R putamen response to reward tended to predict substance use, with a large effect size (β = -2.50, p = .08).

While limited by small sample size, findings indicated sex differences in associations between caudate and putamen response to reward in early adolescence and future substance use with large effect sizes. Findings suggest that boys may be more likely than girls to take a pathway to substance use characterized by high striatal (particularly caudate) responsivity to reward. These sex differences in reward system pathways suggest a need for gender-sensitive prevention programs.  

Authors

• Tara Chaplin, George Mason University
  Presenting Author

• Jennifer (Jenny) Ann Poon, George Mason University
  Non-Presenting Author

• Caitlin Turpyn, George Mason University
  Non-Presenting Author

• James Thompson, George Mason University
  Non-Presenting Author