On Prior Confidence and Belief Updating

Abstract

We experimentally investigate how confidence over multiple priors affects belief updating. Theory predicts that the average Bayesian posterior is unaffected by confidence over multiple priors if average priors are the same. We manipulate confidence by varying the time subjects view a black-and-white grid, the proportion representing the prior in a Bernoulli distribution. We find that when subjects view the grid for a longer duration, they have more confidence, under-update more, and place more (less) weight on priors (signals). Overall, confidence over multiple priors matters when it should not, while confidence in prior beliefs does not matter when it should.