Bayesian Nonparametric Modeling of Individual Differences: A Case Study Using Decision-Making on Bandit Problems

We develop and compare two non-parametric Bayesian ap-proaches for modeling individual differences in cognitive pro-cesses. These approaches both allow major discrete differ-ences between groups of people to be modeled, without mak-ing strong prior assumptions about how many groups are re-quired. Instead, the number of groups can naturally grow as more information about the behavior of people becomes avail-able. One of our models extends previous work by allow-ing continuous differences between people within the same group to be modeled. We demonstrate both approaches in a case study using a classic heuristic model of human decision-making on bandit problems, applied to previously reported be-havioral data from 451 participants. We conclude that the abil-ity to model both discrete and continuous aspects of individual differences in cognition is important, and that non-parametric approaches are well suited to inferring these types of differ-ences from empirical data