(2004-12-20) Games Theory Culture

Jenna Bednar and Scott Page on "Can Games Theory Explain Culture?" They note that each player is playing multiple Games at the same time, and that different players are playing different sets of games, etc. We develop a model in which agents with cognitive constraints play multiple games. We find that the ensemble of games can influence behavior in individual games in ways that suggest the Emergence of Culture. We formally define four components of cultural behavior: intra-agent cultural behavior, inter-agent cultural behavior, contextual effects, and cultural bias and describe when each is likely to occur in the context of our model. We also use our model to demonstrate frequency dependence and behavioral rigidity.

The implications of our results are manifold. First, apart from our cultural interpretation, they are strong support for modeling behavior with Games Theory as opposed to Game Theory. Behavior in a specific context may well depend upon the other games people play. This is particularly likely to be true when some games have multiple equilibria. Second, our results have some fundamental implications for understanding the emergence culture. If people living in different climates or followers of different religions or citizens of various political institutions daily face different mixes of strategic situations, they may evolve different practices in common games and hence distinct behavioral cultures.

The Games Theory idea was informally present in Norton Long (1985), who called it an "Ecology Of Games" and in Samuel Bowles and Herbert Gintis (Democracy And Capitalism, 1986), who discuss the importance of the full spectrum of games---e.g. games played within the family, the state, and the economy---to understanding individual behavior.