Abstract
We test the assumption that people desire to be accurate when making predictions about their own future. Results revealed that, across four different scenarios and three manipulated variables (commitment to a decision, agency over the decision, and control over outcomes), participants thought it was better to make optimistically biased predictions than accurate or pessimistically biased predictions. Additionally, participants thought that they and others would be optimistic in the scenarios they read, but insufficiently so. We argue that prescriptions can serve as one standard by which the quality of predictions can be judged, and that this particular standard strongly endorses optimism.