This paper reports laboratory experiments and simulations on a minority game. The minority game is the most important example for a classic non-zerosum-game. The game can be applied on different situations with social and economic contests. We chose an elementary traffic scenario, in which subjects had to choose between a road A and a road B. Nine subjects participated in each session. Subjects played 100 rounds and had to choose between one of the roads. The road which the minority of players chose got positive payoffs. We constructed an extended reinforcement model which fits the empirical data.