Nonoptimal and counterintuitive choice

University Of California San Diego, La Jolla CA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): According to delay-reduction theory (DRT), the effectiveness of a stimulus as a conditioned reinforcer may be predicted most accurately by the reduction in length of time to primary reinforcement correlated with its onset compared to the length of time to primary reinforcement measured from the onset of the preceding stimulus. One set of proposed experiments would provide new tests regarding the issue of whether when the strength of a stimulus as a conditioned reinforcer is enhanced by correlating that stimulus with greater delay reduction, that enhancement carries over, at least temporarily, to a new context. The proposed experiments develop two ways of correlating a stimulus with greater delay reduction and propose two methods to assess strength of conditioned reinforcement in a new context. A counterintuitive prediction derived from DRT is that differences in delay to reward and not ratios of delay to reward should control choice when subjects are choosing between two schedules of reward as the outcomes, at least in the well-established concurrent-chains procedure. According to most theories, choice between two outcomes with different delays to reward is governed by the ratios of the delays, not by the differences between them. We propose to test the conditions under which ratios and/or differences control choice with two different choice procedures. We also propose to continue investigating the conjunction fallacy, displayed when subjects report that the conjunction of two events is more rather than less likely to occur than one of the events alone. We have developed a behavioral approach to the study of the fallacy that would permit us to investigate critical variables affecting its occurrence. This work takes us naturally to further investigation of pigeons', human adults' and children's response to novel compound stimuli. Together with our proposed research on problem solving these experiments would help clarify the role of rule-governed (instructed) and contingency (or experience)-shaped behavior on learning. Our interest in the factors governing reasoning in humans and non-humans extends to the sunk cost effect, a maladaptive economic behavior involving an increased tendency to persist in an endeavor once an investment has been made. The present aim is to develop a pigeon analog of the sunk-cost effect and to explore several variables that may contribute to the development and/or the persistence of commitment in the pigeon.