The experiment presented in this paper employs 3x3 games to analyze how perception of a game affects behavior in the presence or absence of a minimal framing effect and of uncertainty about the values of some game payoffs. We vary the harmony of practice stage games, and explain how this changes later behavior. We employ techniques, such as payoff integration and similarity evaluations, that could be used in further research to open the black box of framing effects. Game harmony is a measure summarizing how harmonious the interests of the players are in the game. It is associated with cooperation.