Likelihood judgment based on previously observed outcomes: the alternative-outcomes effect in a learning paradigm

Probability judgment and subadditivity: The role of working memory capacity and constraining retrieval

“…One noteworthy finding was that we failed to find an effect of distribution on the judged probability of the freq-20 items, a finding that is inconsistent with recent research by Windschitl et al (2002). One possible explanation for this failure to replicate Windschitl et al is that our judgment task differed from their standard paradigm, in that the participants in Windschitl et al judged the freq-20 items prior to judging any of the other items in the distributions.…”

“…The latter finding was surprising, especially given that Windschitl et al (2002) showed an effect of distribution in a similar learning-based paradigm. We discuss possible reasons for the failure to find such an effect in the General Discussion section.…”

“…1 In addition to examining overall subadditivity, we were also interested in examining participants' judgments for the items that occurred 20 times in each distribution. Windschitl et al (2002) revealed that participants tended to give higher probability judgments when the focal events (i.e., items with a frequency of 20) are embedded in the context of weak alternatives. Thus, in addition to expecting an effect of distribution on subadditivity, we expected an effect of distribution on judgments of the items with a frequency of 20.…”

“…Windschitl and Wells argued that judgments were lower for the second distribution because the participants compared the number of tickets they held with the number of tickets held by the dominant alternative. Thus, according to Windschitl and Wells, the judged probability of the focal hypothesis depends on a comparison with the strength of the strongest alternative (see also Windschitl & Young, 2001;Windschitl, Young, & Jensen, 2002). Windschitl and Wells's (1998) comparison heuristic is a special case of Tversky and Koehler's (1994) support theory.…”

Diagnostic hypothesis generation and human judgment.

“…Judged probability of an event is sensitive to the strength of its alternatives. Windschitl and Wells (1998;Windschitl & Young, 2001;Windschitl, Young, & Jensen, 2002) showed that the judged probability of a focal hypothesis is sensitive to the strength of its alternatives, a finding they termed the alternative outcomes effect. Dougherty andHunter (2003a, 2003b) extended this finding to show that the degree to which participants' probability judgments are subadditive is affected by the strength of the alternatives.…”