Testing p-beauty contest game: An experimental study

This paper reports the results of an experiment designed to test the strategic behavior of a sample of students from Shahid Beheshti University. Their behavior is tested against the Nash equilibrium prediction of p-beauty contest. For this purpose, the game is repeated with six different parameters. To insure incentive compatibility, we used salient financial incentives  as 30x minimum wage. The behavior of a sample of 73 students is in sharp contrast from the Nash equilibrium prediction of the game. While responses can be explained by level-k reasoning, the findings indicate that most responses are rather random indicating level-0 reasoning. While there are some significant correlations between responses to cognitive mathematical questions and guesses closer to Nash equilibrium, there is no significant correlation between the results of Raven IQ test and better performance of players