Descriptive and prescriptive models of decision-making: implications for the development of decision aids

Experimental investigations by psychologists have revealed significant deviations of actual human decision behavior from classical rational theories of judgment and decision-making. Weakening the assumptions of the latter has led to the development of such new theories such as prospect theory or rank-dependent subjective expected utility theory. Recent work in this area is reviewed with an emphasis on its implications for prescriptive modeling. These implications manifest themselves at every step of the decision analytic process, potentially limiting the effectiveness of these tools. It is proposed that it may be possible to enhance the effectiveness of decision analytic tools or numerical optimization if they are coupled with some symbolic reasoning capability. The relevance of artificial intelligence techniques for enhancing existing decision tools is also discussed. Some links between the descriptive decision research of psychology, prescriptive models of operations research, and symbolic reasoning research of artificial intelligence are shown