Default or reactance? : identity priming effects on overconfidence in Germany and Japan

This experimental study measures three types of overconfidence in the decision behavior of participants from Germany and Japan. In the first stage of the experiment subjects completed a Raven Progressive Matrices test and subsequently assessed their test performance in absolute and relative terms. During the second stage subjects provided probability forecasts by confidence intervals to artificially generated price paths. Furthermore subjects ́better-than-average bias was assessed during a post-experimental questionnaire. We find that monetary incentives as a reason to honestly self-evaluate reduce cultural differences in overplacement and in overestimation of own performance. Over-precision in probability judgment accuracy on the other hand significantly differed across ethnical groups. To analyze national identity priming effects on overconfidence, a pre-experimental questionnaire made this identity salient to a randomly selected treatment group. Previous studies found that primes of certain identities can trigger behavior that is consistent with the stereotypes associated with that identity, but can also cause psychological reactance leading to counter-stereotypical behavior. We find that in a setting where there are no incentives to provide honest performance self-evaluations the identity prime reinforces behavior consistent with a default strategy which helps adapting to social norms. In incentivized decisions on the other hand the prime causes counter-stereotypical self-perception. Reasons for this phenomenon are discussed.