Tit for Tat game
From the point of view of evolution of behavior, it is very important that the same individuals repeatedly come into mutual interactions. These individuals are able to adjust their behavior according to the response to the behavior in the past and, at the same time, they have to expect that their behavior will influence the future behavior of a partner (opponent). In this case, even in the prisoner’s dilemma game, a number of strategies exist that are far more advantageous (and friendly) than the always betray strategy. Among the first recognized and yet relatively most successful strategies is the "Tit for Tat" strategy (Axelrod & Hamilton 1981). It always begins with cooperation and in every next step the individual repeats the opponent’s strategy in the last step. If two individuals following this strategy meet, they can, in the long-term, profit from mutual cooperation while, if they meet an opponent who follows the always betray strategy, they lose in the first step but in the next ones they give the chronic betrayers no advantage. In the overall balance, the simple Tit for Tat strategy wins.
The above-mentioned conclusions are valid only under one vitally important condition: neither opponent is not allowed to know when their interactions will finish, i.e. how many steps (moves) are left in the game. As soon as it would be obvious that the game is ending and the opponents would know they are not going to meet in the future, for any of them the most advantageous thing to do would be to betray in the last step and get the extra reward for one-sided betrayal. So the last step would be determined and immediately the question would arise as to how to act in the previous move – the most advantageous solution would be betrayal again. From the beginning, the game would be about who will be the first to betray. The situation is completely different when the players do not know which step will be the last, which is much more favorable for spreading of cooperative game strategies.
Email this page