This is from the events calendar of Howard University, where Freeman Dyson gave a talk on October 12, 2012.
Just in case this disappears from the web, here is the abstract of his talk:
“The Prisoner’s Dilemma: Is it a model for the evolution of cooperation in the real world, or is it only a mathematical toy?”
Prof. Freeman Dyson, Institute of Advanced Study
Prisoner’s Dilemma is a very simple game which is supposed to be a model for the evolution of cooperation. Two guys are picked up after a crime and interrogated separately by the police. Each player has the choice, either to say the other guy did it or to remain silent. If two players play the game repeatedly, then each does better in the short run by squealing but does better in the long run by staying silent. So a good strategy should evolve into a cooperation with both players remaining silent.
To the amazement of the experts, Bill Press this year discovered a new set of strategies for Prisoner’s Dilemma which allow one player to dominate the other. I did the math to figure out why his strategies work. The question, whether this mathematical trickery is relevant to the evolution of cooperation in the real world of biology, remains open. This question is connected to the much deeper question, whether group selection is as important as individual selection in biological evolution. I will argue that group selection is important, contrary to the prevailing dogma among biologists. If group selection is important, Prisoner’s Dilemma is not a good model for evolution. It is still an amusing toy for mathematicians and game-theorists to play with.
In a word, this is brilliant! I love that Dyson contributes to the first major breakthrough in strategy formulation for the Iterated Prisoner’s Dilemma put forth in over a decade and then is perfectly happy to suggest that this model is a “toy”. I agree. The model is important and instructive, but all the bluster about how Press and Dyson have discovered the basis for human enlightened self-interest is an over-reach.
A Minor Post, Altruism, Cooperation, Evolution, Evolutionary Modeling, Game Theory, Group Selection, Web