Detailed analysis of drift diffusion model parameters estimated for the ultimatum game

Abstract

Bargaining is fundamental in human social interactions and often studied using the ultimatum game, where a proposer offers a division of resources, and the responder decides whether to accept or reject it. If accepted, the resources are divided as proposed, but neither party receives anything otherwise. While previous research has typically focused on either the choice or response time, a computational approach that integrates both can provide deeper insights into the cognitive and neural processes involved. Although the drift diffusion model (DDM) has been used for this purpose, few studies have tested it in the context of the ultimatum game. Here, we collected participants' behaviors as a responder during the ultimatum game (n = 71) and analyzed them using a Bayesian version of DDM. The best (estimated) model included parameters for non-decision time, boundary separation, bias, and drift, with drift expressed as a linear combination of self-reward, advantageous inequity, and disadvantageous inequity. This model accurately replicated participants' choices and response times. Our analysis revealed that the drift parameter represents trial-by-trial choices and response times, while other parameters represent average rejection rates and response times. We also found that boundary separation and bias exhibited a more complex interaction than previously recognized. Thus, this study provides important insights into the application of DDM to studies on neural analysis during human bargaining behavior.