A comparison of five models that predict violations of first-order stochastic dominance in risky decision making

Five descriptive models of risky decision making are tested in this article, including four quantitative models and one heuristic account. Seven studies with 1802 participants were conducted to compare accuracy of predictions to new tests of first order stochastic dominance. Although the heuristic model was a contender in previous studies, it can be rejected by the present data, which show that incidence of violations varies systematically with the probability distribution in the gambles. The majority continues to violate stochastic dominance even when two of three branches have higher consequences in the dominant gamble, and they persist in mixed gambles even when probability to win is higher and probability to lose is lower in the dominant gamble. The transfer of attention exchange model (TAX) was the most accurate model for predicting the results.