Proceedings of the Annual Meeting of the Cognitive Science Society

This paper provides empirical evidence that human decision- makers use prospective regret minimization as their dominant decision strategy when regret calculations are cognitively easier to perform, and use expected utility maximization when they aren't. We designed a simple decision problem wherein utility maximization and expected regret minimization yield distinctly difference choices, and manipulated the cognitive effort involved in making regret calculations across respondent samples to arrive at our results. While previous research has associated ecological considerations like sense of responsibility and familiarity with this difference, we show that, at least in experimental settings, cognitive calculability in regret space appears to predominantly drive this difference. We also show that this preference for regret minimization can be countermanded by changing the distribution of options presented to the respondent, posing a challenge to simple sequential accounts of strategy selection learning which sequence strategy selection and application in order.