Abstract

Although past theory and research have suggested that counterfactual thoughts (representations of alternatives to past outcomes) weaken the hindsight bias (after-thefact exaggeration of an outcome’s a priori likelihood), the present research shows the opposite (i.e., positive) relation. Experiment 1 demonstrated that counterfactual thinking can heighten the hindsight bias, and that the effect of counterfactuals on causal inferences can account for this relation. Experiment 2 indicated that postoutcome elaboration of the causal linkage between an antecedent and outcome is essential for thehindsightbias,and that this biasmay be redefinedto includepostoutcomecertainty regarding ‘‘what should have been’ ’ as well as what was. Experiment 3 provided more direct evidence that causal inferences mediate the facilitative effect of counterfactual thinking on the hindsight bias. © 1996 Academic Press, Inc. I just knew I should have picked door number two. L et’s Make a Deal contestant The above comment exemplifies a perception familiar perhaps not only to gameshowcontestants,but tomany ofus.Anunfortunateset of circumstances befalls us and we recognize instantly—alas, too late—that we might have This research was supported by a postdoctoral fellowship awarded to Neal Roese and a researchgrant awardedto James Olson,both fromthe SocialSciencesand HumanitiesResearch Council of Canada. We thank Dave Hamilton, Dale Miller, Mike Ross, Richard Sorrentino, Yaacov Trope, and two anonymous reviewers for providing valuable comments on various versionsof the manuscript.We are also grateful to Eileenda Pena for herassistance in runningthe first experiment. Correspondence and reprint requests should be addressed to Neal Roese,

Citations