Do human beings act purely out of self-interest? The Prisoners’ Dilemma is a classic puzzle designed to test and explore this question:

Two members of a criminal gang are arrested and imprisoned. Each prisoner is in solitary confinement with no means of communicating with the other. The prosecutors lack sufficient evidence to convict the pair on the principal charge. They hope to get both sentenced to a year in prison on a lesser charge. Simultaneously, the prosecutors offer each prisoner a bargain. Each prisoner is given the opportunity either to: betray the other by testifying that the other committed the crime, or to cooperate with the other by remaining silent. The offer is:

If A and B each betray the other, each of them serves 2 years in prison.
If A betrays B but B remains silent, A will be set free and B will serve 3 years in prison (and vice versa).
If A and B both remain silent, both of them will only serve 1 year in prison (on the lesser charge).

What would you do? If both prisoners act purely in self-interest, they will betray each other, yet if they cooperate and keep silent, they will get a greater reward.

In one experiment, in which many participants played the game repeatedly over several days, researchers found the following:

Roughly 60% of the player population is “rational” in the sense [that they act in their own perceived self-interest once burned by attempts at cooperation] and thus is susceptible to “unravelling” dynamics [in which all players begin to defect from cooperation]. Second, however, roughly 40% of the player population is not rational in this sense, instead [cooperating] for the duration of the experiment even as they are exploited by the rational majority. Finally, the existence of these resilient cooperators appears to stabilize the unravelling dynamics after several days, thereby conferring long-run benefits on both the resilient minority and the rational majority. Strikingly, the overall rate of cooperation stayed above 84% throughout the experiment, meaning that players collectively extracted roughly 84% of the maximum average payout possible. Our results therefore cast prospects for long-run cooperation in a hopeful light; as long as a sufficiently large minority of people are determined to act as conditional cooperators, high levels of cooperation can be sustained indefinitely even when the majority is willing to cooperate only when it is in their pragmatic self-interest to do so. (source)