Modeling individuals’ cognitive and affective responses in spatial learning behavior

Activity-based analysis has slowly shifted gear from analysis of daily activity patterns to analysis and modeling of dynamic activity-travel patterns. In this paper, we describe a dynamic model that is concerned with simulating cognitive and affective responses in spatial learning behavior for a multi-agent system. By implementing activity-travel choices, agents observe the differences between actual experience and expectation, which may give rise to negative or positive emotions that influence the awareness of alternatives and, furthermore impact the evaluation of alternatives and hence choice behavior. As such, it provides an approach to model the dynamic process of rational and emotional mechanisms in the formation and adaptation of an individual's location awareness set, distinguishing habitual, exploitation and exploration modes of dynamic, context-dependent choice behavior. Principles of reinforcement learning and Bayesian perception updating are used. We demonstrate model properties using numerical simulation with a case study of shopping activity.