Context effects and risk amplification: Why more is risky

Takao NoguchiUniversity of Warwick, Coventry, United Kingdom
Thomas HillsUniversity of Warwick, Coventry, United Kingdom

Abstract

Research on risky choice has been dominantly based on studies of choice between two alternatives, with the findings often generalized to environments with more than two alternatives. One prominent claim of this research is that choices differ with respect to risk when alternatives are described (the description paradigm) as opposed to experienced (the experience paradigm): Individuals appear to make decisions as if they over-weight small probabilities in the description paradigm, but under-weight the same probabilities in the experience paradigm. Here, we show that the under-weighting in the experience paradigm is sensitive to the choice set size in the gain domain. Two experiments show that as set sizes increase, choices systematically favour risky alternatives in the experience paradigm. Using simulations of three choice models, we further demonstrate that this risk-amplification is independent of choice and search strategies and is predicted by the statistical structure of pay-offs. The results suggest caution in generalising findings from two-choice environments to many-choice environments and further indicate a robust and systematic problem with increasing choice set sizes.

Files

Context effects and risk amplification: Why more is risky (1.1 MB)



Back to Table of Contents