Introduction to Game Theory – Prisoners’ Dilemma (Part 4)

The first three parts of this post have introduced the Prisoners’ Dilemma (Part one), given some real examples (Part two), and looked at what happens in reality (Part three).

Each of those posts has looked at a single game of the Prisoners’ Dilemma, where two players play only once. In this post we will look at what happens when the game is played lots of times in succession.

In a one-off game we have already shown that if the choices are to co-operate or defect then the best thing to do is defect. But when the game is played over and over again then the players can react to how they were treated in the previous round. So, if you were generous and decided to co-operate in the first round only to find out that the other player defected would you retaliate by also defecting in the next round?

When this game is played using various strategies the one that comes out on top is ‘tit for tat’. In this strategy a player co-operates in the first round and then in future rounds does whatever their opponent did to them in the previous round. Essentially, this strategy works because a player is willing to co-operate if their opponent is, but they also don’t allow themselves to be taken advantage of.

A new approach has beaten Tit for Tat, but this worked by entering lots of strategies in a competition with sequences embedded in them so that they could recognise each other and work together to get the best results. This is an interesting approach as it effectively gets around the problem that normally the two players in a Prisoners’ Dilemma can’t communicate. However, this trick won’t work for a single player as they can’t communicate and they won’t have a pre-arranged strategy with other players.

Today’s takeaway: Your strategy might need to change if you are going to deal with the same person again in the future.

This entry was posted in Game theory, Introduction and tagged , , , . Bookmark the permalink.