Published online by Cambridge University Press: 02 October 2003
Game theory poses problems for modeling rational belief, but it does not need a new theory of rationality. Experimental results that suggest otherwise often reveal difficulties in testing game theory, rather than mistakes or paradoxes. Even though the puzzles Colman discusses show no inadequacy in the standard theory of rationality, they show that improved models of belief are needed.
1. I do not, in fact, think that the standard theory of rationality is unproblematic (see e.g., my 1992 book, Chs. 2, 12, 13), but the difficulties I see are independent of those that Colman alleges.