Natural Freedom

This is a video about the most famous problem in Game Theory, the Prisoner’s Dilemma.

Robert Axelrod, a political scientist interested in game theory and conflict resolution, designed a unique computer tournament to uncover effective strategies in the repeated Prisoner's Dilemma game. He invited experts and game theorists to submit computer programs embodying different strategies. Each program would play against every other program, including itself, for 200 rounds, earning points based on the payoff structure of the Prisoner's Dilemma. The goal was to identify strategies that accumulated the most points, indicating successful conflict resolution.

The key findings from Axelrod's tournament revealed that the most successful strategies shared four distinct qualities:
1. Niceness: Successful strategies were "nice" in the sense that they did not defect first. They would cooperate initially and only defect in response to the opponent's defection. This quality ensured that the strategy didn't contribute to escalating conflict and allowed for the possibility of mutual cooperation.
2. Forgiveness: Forgiveness was a critical trait. These strategies could retaliate when the opponent defected but didn't hold a grudge indefinitely. They would respond to the opponent's defection but didn't let past actions influence their current decisions too strongly. This balance between retaliation and forgiveness prevented endless cycles of retaliation.
3. Retaliation: While forgiveness was important, successful strategies also needed to be retaliatory. They wouldn't be pushovers; they would strike back when the opponent defected. This ensured that the strategy wasn't taken advantage of and maintained a certain level of deterrence.
4. Clarity: Effective strategies were also clear and straightforward. They weren't overly complex or difficult to understand. This clarity allowed other programs to establish patterns and predict their behavior to some extent, fostering a level of trust and cooperation.

The winning strategy from the tournament, "Tit for Tat," embodied all four of these qualities. It started by cooperating and then simply mirrored the opponent's previous move. If the opponent cooperated, it would cooperate, and if the opponent defected, it would defect in response, but only once. This combination of niceness, forgiveness, retaliation, and clarity made "Tit for Tat" highly effective in the tournament, and it served as a model for understanding successful conflict resolution strategies.

The ultimatum game (UG) is widely used to investigate our sense of fairness, a key characteristic that differentiates us from our closest living relatives, bonobos and chimpanzees. Previous studies found that, in general, great apes behave as rational maximizers in the UG. Proposers tend to choose self-maximizing offers, while responders accept most non-zero offers. These studies do not rule out the possibility that apes can behave prosocially to improve the returns for themselves and others. However, this has never been well studied. In this study, we offer chimpanzee and bonobo proposers the possibility of taking into account the leverage of responders over the offers they receive. This leverage takes the form of access to alternatives for responders. We find that proposers tend to propose fairer offers when responders have the option to access alternatives. Furthermore, we find that both species use their leverage to reject unequal offers. Our results suggest that great apes mostly act as rational maximizers in an UG, yet access to alternatives can lead them to change their strategies such as not choosing the self-maximizing offer as proposers and not accepting every offer higher than zero as responders.