Following recommendations from public and private correlation devices in a game of chicken

We consider a specific parametric version of Chicken and two different correlation devices, public and private, with the same expected payoffs in equilibrium, which is also the best correlated equilibrium payoff for the game. Despite our choices of parameters (payoffs) in the game, in an experiment with these two correlated equilibria, we find that the rate of "following recommendations" vary significantly within and between two treatments using these devices.