ABSTRACT

Unlike most conventional theories in other branches of the social and biological sciences, game theory is incapable of being empirically tested and falsified; in this respect it resembles probability theory or statistics rather than, say, psychophysics or relativity theory. The function of game theory is to provide an abstract framework for modelling situations involving interdependent choice. In some cases, including all finite, strictly competitive games, it helps us to discover how rational decision makers ought to behave in order to attain certain clearly defined goals; but about how people actually do behave it says nothing. The ways in which human beings typically respond to problems of interdependent choice is an essentially empirical question, and it is investigated through experimental games in which decision making is observed under controlled conditions.