We study how players in a local interaction hawk dove game will learn, if they can either imitate the most succesful player in the neighborhood or play a best reply versus the opponent's previous action. From simulations it appears that each learning strategy will be used, because each performs...