Joy leads to Overconfidence – and a Simple Remedy

This study examines the influence of joy on absolute and relative overconfidence in a decision experiment with monetary incentives. We used a general knowledge task of medium difficulty to measure overconfidence. Participants’ moods were measured at the beginning of the experiment using subscales from the Positive and Negative Affect Schedule (PANAS-X). On average, participants were in a slightly joyful mood. In the experimental treatment, we made participants aware of the irrelevant source of their joyful mood by showing them a humorous film clip and subsequently asking them about their mood again. Our results show that self-reported joy increased overconfidence in the control group, whereas the participants in the treatment group had well-calibrated judgments, on average. Our results are consistent with the mood-as-information hypothesis, which suggests that affective states with a non-salient and irrelevant cause have an informative function that can lead to overconfidence. However, if the cause of the affective state is salient and obviously irrelevant (i.e., a humorous film clip), the informative function is deactivated, leading to better judgments and decisions.