Rational Choice Theory

Elements and Structure of Rational Choice Theory: Expectations and Game Theory

In rational choice theory, agents are described by their unchanging sets of preferences over all conceivable global outcomes, such as whether candidate Smith, Davidson, or Nelson will win an election, whether dinner will consist of chicken, fish, or tofu, or whether a public policy is one of waging war, negotiating a settlement, or relying on the international community of nation states to provide leadership. Agents are said to be rational if their preferences are complete, that is, if they reflect a relationship of superiority, inferiority, or indifference among all pairs of choices; and logically ordered, that is, they do not exhibit any cyclic inconsistencies of the sort: Chicken is preferred to fish, fish is preferred to tofu, and tofu is preferred to chicken. In addition, for choices in which the probabilities of outcomes are either risky, or uncertain, rational agents exhibit consistencies among their choices much as one would expect from an astute gambler. All of these consistency relations among preferences over outcomes are stated in mathematical axioms; a rational agent is one whose choices reflect internal consistency demanded by the axioms of rational choice. Rational choice theory holds that all considerations [Page 786]pertinent to choice (that may include attitudes toward risk, resentment, sympathy, envy, loyalty, love, and a sense of fairness) can be incorporated into agents' preference rankings over all possible end states. Social scientists only have indirect access to agents' desires through their revealed choices; therefore, researchers infer back from observed behavior to reconstruct the preference hierarchy that is thought to regulate a rational agent's decisions.

It is generally not appreciated, but important, that the consistency constraints defining rational choice theory are not equivalent to those specifying maximization of marginal utility under a budget constraint, although formal bridging conditions may be added to achieve congruence. It is also the case that even though many social scientists that use rational choice theory adopt one canonical axiomatized form specified as “expected utility theory,” the research paradigm sustains alternative views as well. There are subtleties in probability theory that divide researchers: Are probabilities objective features of the world or are they best regarded as subjective features of individuals' psychology, and what sorts of consistency conditions apply to decision problems that incorporate both attitudes toward risk and unknowable probabilities? The intractability of decision making in uncertain circumstances has lead to the formulation of bounded rationality that grounds rational choice in manageable rule-of-thumb calculations in a series of one-time circumstances. As well, psychologists have observed several prominent and predictable empirical deviations from rational choice theory that has made it possible to identify patterns of what may be termed “folk psychology.”

...