Wednesday, April 12, 2006

Is Anger Optimal?

I have been auditing Steven Pinker's course at Harvard on the Human Mind. In a recent lecture, Pinker discussed the Prisoners' Dilemma, which we covered in ec 10 when we studied oligopoly and game theory. As we have seen, this simple game provides a prototype for understanding many strategic situations.

In my Principles text, I describe a tournament run by Robert Axelrod, in which computer programs played repeated Prisoners' Dilemma against one another. The winning program was the simple tit-for tat strategy: a player should start by cooperating and then do whatever the other player did last time. In other words, a tit-for-tat player cooperates until the other player defects; he then defects until the other player cooperates again.

Pinker pointed out an intriguing angle on the Axelrod finding: Perhaps our emotions have adapted through evolution to induce us to play the tit-for-tat strategy without explicitly thinking about it. The emotion of anger tells us to punish uncooperative players by defecting after they have defected. The emotion of gratitude tells us to reward cooperative players by cooperating after they have cooperated. Axelrod’s prisoners' dilemma tournament suggests that tit-for-tat may be a good rule of thumb for playing some of the games of life. Evolution has given us the psychological make-up to play this game optimally.

In some ways, the logic here turns behavioral economics on its head. Behavioral economists bring psychology into economics to explain why people are not fully rational, at least in the standard neoclassical sense. By contrast, Pinker was bringing game theory into psychology to explain why emotional reactions may be rational.