Abstract
In a test of game theory as a descriptive model for behavior, a group of naive Ss played 2 × 2 zero-sum competitive games having small amounts of money as payoffs. The optimal strategies for each game were mixed. 5s were provided with a device which allowed them to choose a probability and play a given alternative with that probability, i.e., the device provided a ready means of choosing and playing a mixed strategy. The data indicate that Ss do not play, or learn to play, the optimal strategy. Furthermore, Ss’ use of the strategy-choosing device indicates that they have little tendency to choose a given alternative with a fixed probability.
Get full access to this article
View all access options for this article.
