When Optimal Choices Feel Wrong: A Laboratory Study of Bayesian Updating, Complexity, and Affect

Abstract

We examine decision-making under risk in a laboratory experiment. The heart of our design examines how one's propensity to use Bayes' rule is affected by whether this rule is aligned with reinforcement or clashes with it. In some cases, we create environments where Bayesian updating after a successful outcome should lead a decision-maker to make a change, while no change should be made after observing an unsuccessful outcome. We observe striking patterns: When payoff reinforcement and Bayesian updating are aligned, nearly all people respond as expected. On the other hand, when these forces clash, around 50% of all decisions are inconsistent with Bayesian updating; a slight increase in the precision of the information and decrease in the complexity of the calculations does not lower the error rate. However, when a draw provides only information (and no payment), switching errors occur much less frequently, suggesting that the "emotional reinforcement" (affect) induced by payments is a critical factor in deviations from Bayesian updating. We also find considerable behavioral heterogeneity across the population. Finally, we see that people have a "taste for consistency", as voluntary draws are more likely to be repeated than draws that were required.

Keywords: Bayesian updating, Reinforcement, Affect, Experimental economics

JEL Classification: B49, C91, D80, D81