Learning inhibition in the Monty Hall problem: the role of dysfunctional counterfactual prescriptions

Despite repeated trials, people consistently fail to learn the solution to the Monty Hall problem (MHP). This research examines the links between learning, counterfactual thinking, and memory for decision/outcome frequencies. Study 1 participants completed 60 MHP trials and listed their thoughts following losses. Results showed that participants tended to counterfactualize switch losses more than stick losses, adhered to the prescriptions of their counterfactuals more frequently following switch losses than any other decisions/outcomes, and were less likely to learn the solution as counterfactuals increased. Furthermore, memory for switch losses was significantly overestimated, and the relationship between counterfactuals and learning was mediated by misestimation of decision/outcome frequencies. In Study 2, counterfactual salience was manipulated. Learning was less likely to occur when counterfactual salience was high than when it was low, a relationship that was again mediated by memory of decision/outcome frequencies. Findings are discussed in light of their theoretical and applied implications.