Proceedings of the Annual Meeting of the Cognitive Science Society

Numerous experiments have suggested that extreme outcomesare disproportionately influential when we make decisions in-volving risk, but there is less consensus on what it actuallymeans to be extreme. Existing accounts broadly fall into twocategories: those that suggest that the best and worst outcomesare uniquely influential and those that suggest that outcomesbecome more influential with increasing deviation from thecentre of the distribution. We conducted two experiments thataimed to tease apart these explanations. Although there wassome evidence that the distance from the centre influencesmemory, neither account was able to fully explain the choicesmade by participants. This finding has implications for the vi-ability of these explanations as well as for the generalisabilityof the effect and the interpretation of the method used to assessmemory.