Proceedings of the Annual Meeting of the Cognitive Science Society

Procrastination is prevalent. Empirical studies of procrastination have identified various contributing factors underlyingprocrastination. Models of procrastination, however, have only considered temporal discounting and have ignored otherfactors. Moreover, existing models of procrastination are mostly conceptual, and there is a lack of process models toexplain why people procrastinate. Here, we use reinforcement learning theory to build a process model of procrastination.The model assumes that people maximize expected utility while minimizing the total cost of the effort. Our model makesseveral predictions: 1. Strong temporal discounters will delay working early and rush to work near the deadline; 2. If atlow effort cost, cost is sensitive to increases in effort, people will delay working until the last minute; 3. If time pressure oreffort cost is high, perfectionists will not work at all. We designed a behavioral experiment to study the factors underlyingprocrastination and to test our model predictions.