The PIs propose to investigate the apparently ubiquitous phenomenon that people are able to entertain only one hypothesis at a time. The PIs will examine a number of implications of this cognitive limitation. First, the PIs will study "pseudodiagnosticity," defined as the search for more information about one hypothesis as opposed to gathering information about rival hypotheses in a manner congruent with Bayes Theorem. The PIs will utilize the "information board" methodology to assess participants' information gathering behavior. Second, the PIs will investigate participants' information selection while performing a covariation estimation task in which participants attempt to assess the relation between two variables. Third, base rate neglect will be investigated. Finally, data selection in a complex hypothesis testing environment will be examined.
