Can wishful thinking explain evidence for overconfidence? An experiment on belief updating
实验发现，一厢情愿（对结果的乐观）只能解释最乐观人群的过度自信，而保守主义可能更令人担忧，若忽略它，过度自信可能被低估。
Abstract Recent theoretical work shows that the better-than-average effect, where a majority believes their ability to be better than average, can be perfectly consistent with Bayesian updating. However, later experiments that account for this theoretical advance still find behavior consistent with overconfidence. The literature notes that overoptimism can be caused by either overconfidence (optimism about performance), wishful thinking (optimism about outcomes), or both. To test whether the better-than-average effect might be explained by wishful thinking instead of overconfidence, we conduct an experiment that is similar to those used in the overconfidence literature, but removes performance as a potential channel. We find evidence that wishful thinking might explain overconfidence only among the most optimistic subjects and that conservatism is possibly more of a worry; if unaccounted for, overconfidence might be underestimated.