Separating response variability from structural inconsistency to test models of risky decision making

Abstract

AbstractIndividual true and error theory assumes that responses by the same person to the same choice problem within a block of trials are based on the same true preferences but may show preference reversals due to random error. Between blocks, a person’s true preferences may differ or stay the same. This theory is illustrated with studies testing two critical properties that distinguish models of risky decision making: (1) restricted branch independence, which is implied by original prospect theory and violated in a specific way by both cumulative prospect theory and the priority heuristic; and (2) stochastic dominance, which is implied by cumulative prospect theory. Corrected for random error, most individuals systematically violated stochastic dominance, ruling out cumulative prospect theory. Furthermore, most people violated restricted branch independence in the opposite way predicted by that theory and the priority heuristic. Both violations are consistent with the transfer of attention exchange model. No one was found whose data were compatible with cumulative prospect theory, except for those that were also compatible with expected utility, and no one satisfied the priority heuristic.