This paper studies a game of strategic experimentation in which the players learn from the experiments of others as well as their own. <br> We first establish the efficient benchmark where the players co-ordinate in order to maximise joint expected payoffs, and then show that, because of...