Proceedings of the Annual Meeting of the Cognitive Science Society

Leading theories of risky choice predict that decision makersare sensitive to the variability of payoff distributions. Yet, lit-tle is known about how experience affects perceived variabil-ity. Existing empirical research on risky choice provides onlyinconclusive evidence about this issue because choices are notonly affected by perceived variability but also perceived valueand (unobserved) risk preferences. In re-analyses of experi-mental data and survey data from two nationally representativepanels, we show that perceived variability strongly depends onsample variability. In a new experiment, we also demonstratethat perceived variability systematically depends on samplesize, a result consistent with the predictions of a recent the-oretical paper by the authors (Konovalova & Le Mens, 2017).