Modeling Psychological Refractory Period (PRP) and Practice Effect on PRP with Queuing Networks and Reinforcement Learning Algorithms

PRP (Psychological Refractory Period) is a basic but important form of human information processing in dual-task situations. This article describes a queuing network model of PRP that successfully modeled PRP without the need of setting up complex lock/unlock performance strategies employed in the EPIC model of PRP or drawing complex scheduling charts employed in the ACT-R/PM model of PRP. Further, by integrating queuing networks with reinforcement learning algorithms, the queuing network model successfully simulated practice effect on PRP, which has not been modeled in existing PRP models. The current research indicates that depending on an individual’s degree of practice, cognition can be either serial or parallel at the level of production or response selection. Extensions of queuing network model in modeling other tasks and its easiness in modeling concurrent tasks in practice are also discussed.