A new look at risk-taking: using a translational approach to examine risk-taking behavior on the balloon analogue risk task.

Models of risk-taking typically assume that the variability of outcomes is important in the likelihood of making a risky choice. In an animal model of the Balloon Analogue Risk Task (BART), within-session variability, or the coefficient of variability (CV), was found to be a novel predictor of behavior (Jentsch et al., 2010). Human studies have not investigated how BART performance differs when using the CV versus a traditional BART measure (e.g., number of pumps). This study sought to determine whether the CV provides a unique and valuable alternative index of risk-taking on the BART, and to determine the relationship of the CV to self-reported alcohol consumption. Young adult heavy drinkers (n = 58, 72% male, mean age 21.53) completed an assessment of drinking patterns and a modified version of the BART. Multiple regression results indicated that CV is a unique predictor of total explosions and total money earned on the BART. Higher levels of variability were associated with fewer explosions but less money earned, whereas more pumps was associated with more explosions but more money. Higher CV was also associated with lower lifetime and past 3 months peak drinking quantity, higher levels of self-efficacy to control drinking, and lower levels of drinking acceptability (i.e., injunctive norms). Total pumps was associated with higher lifetime peak drinking, lower self-efficacy to control drinking, and higher levels drinking acceptability. Overall, the CV can provide an alternative method of assessing BART performance and the association of risk-taking with drinking patterns.