July 3, 2014, by Suzanne
Sudeep Bhatia on Sequential sampling and paradoxes of risky choice
Abstract: The common-ratio, common-consequence, reflection, and event-splitting effects are some of the best-known findings in decision-making research. They represent robust violations of expected utility theory, and together form a benchmark against which descriptive theories of risky choice are tested. These effects are not currently predicted by sequential sampling models of risky choice, such as decision field theory (Busemeyer & Townsend 1993). This paper, however, shows that a minor extension to decision field theory, which allows for stochastic error in event sampling, can provide a parsimonious, cognitively plausible explanation for these effects. Moreover, these effects are guaranteed to emerge for a large range of parameter values, including best-fit parameters obtained from preexisting choice data.