Relatore:
Sergio Beraldo
- Universita' di Napoli - Federico II
giovedì 29 aprile 2010
alle ore
12.30
Biblioteca DSE, Palazzina 32 ex caserma Passalacqua
The emergence of cooperation in a world subjected to evolutionary pressures has long been
represented as a puzzle, as acting cooperatively apparently puts the individual at a fitness
disadvantage. However, theorists may have set themselves an unnecessarily difficult
challenge by modelling cooperation problems as Prisoner’s Dilemmas. We present a new,
generally-applicable and realistic model of such problems, in which the benefits that each
player gains from cooperation are stochastic and known by the player in advance of the
game, and in which participation in the game is optional. We show that, even in an
anonymous, well-mixed population, a predator–prey mechanism can hold down the
frequency of cheating to a level at which mutually beneficial cooperation can be sustained.
