game theory wouldn't be interesting if it was the study of the actions of perfect bayesian reasoners. that's the whole point, studying where we DON'T make the purely rational choice and why. Exposing cognitive bias is very important for every field of science.