Given the existence of risky events, how do individuals make decisions under risk? First, they must evaluate the risk itself. As seen in Chapter 2, probability assessments provide a way of characterizing the nature and extent of individual risk exposure. In this chapter, we will assume the risk has been assessed and that the corresponding probabilities have been estimated. The next issue is, given an assessment of risk exposure, which decision should the individual make? This is a nontrivial issue. Indeed, human decision-making under uncertainty can be extremely complex for at least two reasons. First, the number of risky events facing an individual is typically quite large. Second, the way information is processed to make decisions under risk can be quite complicated.

Given these complexities, we will start with simple hypotheses about decision-making under risk. As you might expect, while simple models have the advantage of being empirically tractable, they may provide unrealistic representations of human decision-making. This identifies some trade-off between empirical tractability and realism. The analysis presented in this chapter will be limited in scope. We consider only the case of uncertain monetary rewards, and we focus our attention on the expected utility model developed by von Neumann and Morgenstern in the mid 1940s. It has become the dominant model used to represent decision-making under uncertainty. Further extensions and generalizations will be explored in later chapters.