Testing the Generalizability of the Priority Heuristic to Two-Stage Decision Making Tasks: A Comparison to Quantum...

Our goal was to test the ability of the priority heuristic (PH) to generalize to a two-stage decision making task designed to investigate dynamic inconsistency, a phenomenon whereby decision makers deviate from their plans. The PH is a non-compensatory decision strategy in which features are evaluated sequentially in order of importance until one option is determined to be superior. We extended the PH by incorporating reference point dependence into the valuation process and tested it against data from a previously published two-stage decision task (Barkan, R., & Busemeyer, J. R. Journal of Behavioral Decision Making, 16(4), 235-255 2003). Although we demonstrated that the extended PH model can produce dynamic inconsistency, further analysis using a true and error model revealed two deficiencies: (1) the model systematically underestimated risk taking preference, and (2) the data violated a critical inequality derived from the model. In a second analysis, we developed a variant which incorporates individual differences, and compared it to two existing models-one based on prospect theory and another based on quantum cognition. Our results show strong evidence for the quantum cognition model over the extended priority heuristic and the model based on prospect theory. Taken together, our results cast doubt on the ability of the PH to generalize to two-stage decision making.