Toward a Mechanistic Account of Gender Differences in Reward-Based Decision-Making
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2016 American Psychological Association. Gender differences in reward-based decision-making have been extensively researched, yet the mechanisms underlying these differences remain poorly understood. We sought to develop a mechanistic account of how men and women differ in their decisionmaking strategies. We examined gender differences in performance on the Iowa Gambling Task (IGT; Experiment 1) as well as the Soochow Gambling Task (SGT; Experiment 2). Expectancy valence and prospect valence learning computational models were fit to the data for both tasks to assess specific strategies that men and women utilized during the decision-making process. Our results replicated the behavioral gender difference finding on the IGT. Women selected the disadvantageous Deck B more than did men. We extended these findings to the SGT. Modeling results revealed that women's data were best fit by higher recency, or learning rate, parameter values than were men's data in the IGT and SGT. This suggests that women gave greater weight to recent events than did men and that they tended to ignore large, infrequent losses in both experiments. Overall, our results suggest that the mechanism accounting for how men and women differ in reward-based decision-making is that women tend to focus on the relative frequency of gains and losses and attend to recent reward outcomes, whereas men focus more on the extreme gains and losses associated with each alternative and attend to long-term decision outcomes. Implications for these gender differences in reward-based decision-making strategies are discussed.
