Goeree & Holt (2001) observe that, for some parameter values, Nash equilibrium provides good predictions for actual behaviour in experiments. For other payoff parameters, however, actual behaviour deviates consistently from that predicted by Nash equilibria. They attribute the robust deviations...

Goeree and Holt (2001) experimentally study a number of games. In each case they initially find strong support for Nash equilibrium, however by changing an apparently irrelevant parameter they find results which contradict Nash equilibrium. In this paper, we study the five normal form games from...