Heuristic use or information integration in the estimation of subjective likelihood?

Abstract

Use of the representativeness heuristic [described by Kahneman & Tversky (1972)] in estimating subjective likelihoods was investigated in three experiments. Procedural factors (number of problems per subject, information available to subjects) were found to account for differences in data from the two-urn situation collected by Kahneman and Tversky (1972) and Leon and Anderson (1974). Subjects use the representativeness heuristic in such problems but seem to modify it by using other sources of information when these are available or demanded by the situation. © 1975, The Psychonomic Society, Inc.. All rights reserved.