The use of alternative reasons in probabilistic judgment

We examine a behavioral measure of actively open-minded thinking (AOT), based on generation of alternative reasons or contradicting evidence. We also developed and tested a short online module to train people in actively open-minded thinking, based on the same idea. In three studies reported here, subjects made probabilistic judgments in three-choice almanac questions. Subjects were overconfident in their probability judgments, but overconfidence was lower (in two studies) in subjects who scored higher on a measure of AOT beliefs, and on trials when the behavioral measure was higher. Study 2 showed that forcing subjects to think of alternative reasons reduced overconfidence. Study 3 tested the effectiveness of new online AOT training modules we designed for adults, in a pre-test/post-test design. The training module, relative to a control condition, increased both the behavioral and belief-based measures of AOT and reduced overconfidence. Otherwise, the behavioral and belief-based AOT measures did not correlate with each other.