Economics portrays agents as choosing rationally. Many generalizations in economics concerning how people do in fact choose are also claims about how agents ought rationally to choose. This fact distinguishes economics from the natural sciences, whose quarks and polymers do not choose at all and whose theories have no comparable normative dimension.

The theory of rational choice that dominates economics derives from an everyday theory of human choice, which has been called “folk psychology.” This theory takes actions to derive jointly from beliefs and from a wide array of motivational factors such as urges, emotions, habits, and commitments. So, for example, when one rainy Friday night a hungry student named Ellen takes a frozen pizza out of the refrigerator, unwraps it, puts it in a stove, and turns knobs on the stove, we folk psychologists explain Ellen's action by Ellen's beliefs – including especially her beliefs that turning the knobs will cause the stove to heat the pizza – and by her desire to eat hot pizza.

This sort of explanation is familiar but not very satisfactory. Ellen might also like to eat her pizza frozen, or she might also have a desire to reheat some leftover meatloaf. Or she might rather skip dinner and keep studying decision theory. What explains her action is not merely wanting to eat hot pizza (plus possessing the requisite beliefs) but also wanting to do this as much or more than she wants to do any of the feasible alternatives.