WikiFindings

A clever game player can choose a strategy than ensures it beats an opponent in the Prisoner’s dilemma game

Background

Cooperatively behaviour is a puzzle for evolutionary biology because we might expect animals to look after themselves. Researchers have understand how cooperation can evolve by using game theory, especially the Prisoner’s dilemma game. In this game players can do well by cooperating a lot, but there is always the temptation to defect on the deal for a better payoff. For a long time it seemed that the best way to play used tit-for-tat: punishing defection but then forgiving the partner.

Findings

There are ways to play the game that enable a player to always win. A game with many opportunities to cooperate or defect can be played with a strategy of four probabilities, one for each of the last interactions: both cooperated, focal player cooperated and partner defected; focal player cooperated and partner defected; both defected. A player that appreciates that their partner will respond to their behaviour can extort them to ensure they win by choosing probabilities from certain ranges. The more the partner cooperates, the more both their scores increase, but the extortioner always wins.

Implications

Until now, it was believed that defecting against a sensible strategy like tit-for-tat will reduce everyone’s fitness. In a population of lots of competing strategies, tit-for-tat usually wins, especially ‘generous tit-for-tat’ (see left figure) does well because it punishes defection but avoids long stretches of mutual defection (by being forgiving: green arrows). Extorting strategies, by slightly increasing the chance of defection (red arrows on right), beat even generous tit-for-tat in one-on-on interactions.