Picture this: You and your pal are out for a night of raucous adventure when you’re picked up by the law for some heinous crime that you allegedly committed.
You’re dragged downtown in separate cruisers, hand-cuffed to chairs in separate cells, and interviewed by the police. You haven’t had a chance to talk to your buddy, and you’re pretty sure you never did anything wrong in the first place, but a detective leans over you under the gently-swaying-but-far-too-bright light and offers you a deal. He says,
“Give up your friend. He’ll serve maximum time, but we’ll let you off the hook. If you don’t wanna play, we’ll hold you both. You have five minutes.”
He walks to the door, then slowly turns.
“Oh, and by the way? My partner is over with your friend right now, offering him the same deal. If I find out that you blame him and he blames you? Well, you’ll both do time—and lots of it.”
Welcome to the Prisoner’s dilemma, a thought-experiment bandied about for decades in an attempt to understand how cooperation and selfishness came to be. An extension of this game is called the “iterated prisoner’s dilemma,” where you have to go through this process not just once, but many, many times. Nature:
The simplest version of the game pits a pair players against each other. The players obtain particular pay-offs if they elect to cooperate or ‘defect’ (act selfishly). In a single bout it always makes sense to defect: that way you’re better off whatever your opponent does. But if the game is played again and again — if you have repeated opportunities to cheat on the other player — you both do better to cooperate.
This view, ‘sell out your friend if you only plan to get arrested once, but stick together if you’re facing a life of crime,’ was long thought to be the ideal strategy.
Earlier this year, however, two researchers, William Press and Freeman Dyson, published a study arguing that there’s another option. As described by Daniel Lende on his blog Neuroanthropology, there exist ways that a powerful player can abuse their friend over multiple rounds such that one comes out ahead and the other suffers. A simulated kingpin, if you will.
Press and Dyson call these “zero determinant” strategies, because the player can enforce a linear relationship of pay-offs that systematically favor the enforcer. Nothing the other player can do can change that result, so long as the original player chooses a unilinear strategy of their own that sets up this linear relationship.
But this domineering mode of playing the prisoner’s dilemma can’t last forever. Another piece of new research undoes some of the work of Press and Dyson, finding that though such exploitative strategies can work for a little while, more cooperatively-styled modes of interaction will persevere. The reason that selfishness and control breaks down, though, isn’t nearly so cheery. These domineering players, so-called “ZD players”, says Nature, “suffer from the same problem as habitual defectors: they do badly against their own kind.”
In other words, the only reason mafia dons don’t reign supreme is because their strategies fail when placed up against other mafia dons doing the same thing.
More from Smithsonianm.com:
A Game Where Nice Guys Finish First