Let’s revisit the prisoner’s dilemma. In a nutshell, two prisoners are being interrogated separately. If they both talk to the cops, they both serve hard time. If neither talks, they take a light sentence. However, if one talks and the other doesn’t, the talker walks free and the other does the hardest possible sentence. If a prisoner is looking out for number one, talking is the smart play—but the other guy might talk too, so it’s best for both if they keep their mouths shut. Hence, the dilemma.
Some researchers published in the PNAS decided to take things a step further, however, because they wanted to know what would happen to a large population, each subjected to the prisoner’s dilemma, over a long period of time. So, the researchers modelled a large population of evolving players and put them through multiple trials of the game.
In the long term, players who favoured selfish strategies ended up losing more over time than generous players. Even though they could get an ideal outcome once in a while, over time and over a large population, generous players won out.
In other words, cooperation allowed our ancient ancestors to outlast things like blizzards and angry grizzly bears. However, we must point out that sometimes, in one specific instant, for one specific player, it pays to be selfish, like our ancient ancestors, who figured out that you didn’t need to outrun a bear, just outrun the slowest member of the team.