Distortion of Outcome and Probability Information in Risky Decisions

Manuscript under review. Please do not quote without permission. Information Distortion in Risky Decisions 2 Manuscript under review. Please do not quote without permission. ABSTRACT—Substantial evidence indicates that information is distorted during decision making. However, no studies have assessed the distortion of outcome and probability information in risky decisions or the effects of ambiguity on information distortion. We report two studies involving six binary decisions (e.g., banning blood donations from people who have visited England, because of " mad cow disease "). In Study 1, participants distorted their evaluations of outcome and probability information in the direction of their preferred decision alternative and used these biased evaluations to update their preferences. Participants also evaluated the utilities of possible outcomes more positively when the outcomes could follow only from the preferred alternative and more negatively when they could follow only from the competing alternative. In Study 2, we manipulated ambiguity by describing outcomes and probabilities using either point estimates or ranges of values. Results replicated those of Study 1, with no effects of ambiguity on information distortion. Information Distortion in Risky Decisions 3 Manuscript under review. Please do not quote without permission. A central premise of normative decision making is that choices should be made on the basis of Winterfeldt & Edwards, 1986). Depending on the context, such information may consist of product attributes, legal evidence, medical symptoms, or the behavior of others in social or professional settings. In risky decisions, relevant information includes descriptions of possible outcomes and the probabilities of those outcomes. Although these considerations do inform many decisions, there is substantial evidence that inferences also operate in the opposite direction. Specifically, when one decision alternative is favored over another, information is often evaluated as being more consistent with the preferred option than is warranted. Such information distortion or bidirectional reasoning has been They are also consistent with research on the affect heuristic, which indicates that people's early affective reactions to hazards and other stimuli influence related judgments and decisions judgments of the risks and benefits associated with hazards appears to be mediated by initial affective reactions (Finucane et al., 2000). The above studies provide convincing evidence for the distortion of attribute information and evidence, but the possible distortion of outcome and probability information in risky decisions has been largely overlooked. Bond et al. (2005) reported one study involving a hypothetical gamble, in which participants' evaluations of the low …