Iterated Games and the Evolution of Cooperation

Posted on March 16th, 2005 by

Seminar by Tom LoFaro

Tuesday, March 22, 2005 at 3:30pm in Olin 320

We will consider a discrete model of two interacting species each of whose fitness is determined by the outcome of an iterated game. This problem was first explored by Doebeli et al where the game was the Iterated Prisoner’s Dilemma and the two competing strategies are known as Always Defect (AD) and Tit-for-Tat (TFT). The strategy AD is competive whereas TFT is cooperative. Doebeli et al show that in a deterministic model the TFT strategy cannot become established. They go on to show that if stochasticity is introduced to the model then numerical simulations suggest that TFT (and hence cooperation) can become established. We will review the deterministic aspects of this work in a more general setting. In particular, we will establish conditions under which any strategy can become established. In the process we will also learn why stochasticity is essential in the TFT versus AD problem.

As is the custom, there will also be beautiful pictures of New Zealand.

Refreshments will be served.
Tom LoFaro is an Associate Professor in the Mathematics and Computer Science Department.

For the MCS seminar schedule see
If you or someone you know is interested in giving a talk this year, contact David Wolfe at or Tom LoFaro at


Comments are closed.