A Behavioral Approach To Decision Making Under Uncertainty: Implications And Lessons For Expected Utility Theory

Experimental investigations by psychologists have uncovered many instances where decision makers consistently and persistently violate the postulates of classical expected utility theory. Modification and weakening of the expected utility assumptions/axioms have resulted in the development of a class of non-expected utility models which can account for some but not all observed violations. Introduction of psychological decision dimensions has resulted in theories of choice designed to describe observed behavior (e.g., regret theory and portfolio theory) and in models that formalize such psychological variables as perceived riskiness or ambiguity. Finally, at a metatheoretical level, realization of the nature and extent of human cognitive processing and memory limitations has resulted in a reevaluation of rationality or optimality criteria ("bounded rationality"), in process-model modifications of expected utility models (i.e., prospect theory), and in the study of cost-benefit tradeoffs of task-contingent judgment and decision strategies. This paper reviews recent work in these areas with an emphasis on implications for prescriptive theories for risky decision making.