A disconfirmation bias in the evaluation of arguments.

Abstract

Two experiments provided evidence for a disconfirmation bias in argument evaluation such that arguments incompatible with prior beliefs are scrutinized longer, subjected to more extensive refutational analyses, and consequently are judged to be weaker than arguments compatible with prior beliefs. The idea that people are unable to evaluate evidence independently of prior beliefs has been documented elsewhere, including in the classic study by C. G. Lord, L. Ross, and M. R. Lepper (1979). The present findings contribute to this literature by specifying the processes by which prior beliefs affect the evaluation of evidence. The authors compare the disconflrmation model to several other models of how prior beliefs influence current judgments and present data that provide support for the disconfirmation model. Results indicate that whether a person's prior belief is accompanied by emotional conviction affects the magnitude and form of the disconfirmation bias. When evaluating an argument, can one assess its strength independently of one's prior belief in the conclusion? A good deal of evidence indicates the answer is an emphatic no (e.g., Batson, 1975; Chapman & Chapman, 1959; Darley & Gross, 1983; Geller & Pitz, 1968; Nisbett & Ross, 1980; Sherif & Hovland, 1961). This phenomenon, which we refer to as the prior belief effect, has important implications. Given two people, or groups, with opposing beliefs about a social, political, or scientific issue, the degree to which they will view relevant evidence as strong will differ. This difference, in turn, may result in a failure of the opposing parties to converge on any kind of meaningful agreement, and, under some circumstances, they may become more extreme in their beliefs. Perhaps the most renowned study documenting the prior belief effect is one conducted by Lord, Ross, and Lepper (1979); this study served as the starting point for our work. Lord et al. were concerned with people's evaluations of arguments about