Split or Steal: Radiolab Goes Game Theory

A couple weeks ago, Radiolab did a segment on a British game show that lasted […]

Ethan Richardson / 8.5.15

A couple weeks ago, Radiolab did a segment on a British game show that lasted three seasons, called Golden Balls. Not joking. The host’s name was Jasper Carrott. It had all the game show tropes of its time—lighted floors, long silences, hot seats. But unlike Millionaire or Deal or No Deal, the money in Golden Balls is only won after two contestants have made an agreement with each other.

Here’s how it works. After the game is played down to two players, the final jackpot is compiled from both player’s earnings. The two contestants onstage must then make an individual decision about how the money will be divided. They each have two choices: they can split or they can steal. Split means precisely that—the contestant is choosing to split the pot with their opponent. Steal means the contestant is choosing the whole pot for themselves. Here’s where things get interesting:

  • If both contestants choose “split,” the money is split 50/50.
  • If both contestants choose “steal,” neither party gets the pot, and instead each go home with nothing.
  • But if one contestant chooses to “split” and the other contestant chooses to “steal,” the contestant who steals takes everything.

This moment is a fabulous insight into human motivation. And, as you might guess, deception is the name of the game. As Jad and Robert discuss on Radiolab, at this moment there is an incentive to share, but there is also an incentive—a greater monetary incentive—to lie. To posture yourself as a generous sharer on the outside—to garner the trust and cooperation of your opponent—while simultaneously scheming to steal it all away. Exhibit A:

[youtube=https://www.youtube.com/watch?v=p3Uos2fzIJ0&w=600]

This is a phenomenon well known in game theory as the prisoner’s dilemma, where two crime suspects are separately given two options: to either cooperate with the other suspect and stay quiet, or to take a lesser sentence by defecting. And, in a scenario where one cannot know what the other is doing until its already been done, it’s easy to see how humanity’s capricious and selfish qualities could reign supreme. In other words, even if we say we believe that human beings have good souls, the theory tends to explain that any slight justification for offing an opponent/friend seem pretty reasonable in exchange for a big payoff.

But going a little deeper, the Radiolab guys also think it has to do with shame. “Fear of being the sucker far overwhelms any desire to do good.” More than cooperating, more than being good, contestants (and humans) are concerned primarily with not being the sucker. It’s not simply having a natural impulse to lie; it’s wanting to lie for fear of being fooled.

We all know this one, don’t we? This is not just a phenomenon of games or of economics—it is one of the profound dynamics of human relationships. This is the heart behind much of Brené Brown’s work. The experience of open honesty and vulnerability is excoriating. We may know that vulnerability is the important element to communication in a relationship, or that honesty will lay the groundwork for love. But rather than extend the hand of our own cooperation, we would permanently fold our arms in fear that this is how somebody gets duped. Sadly, it is often the case. And so hands of cooperation are bluffed with crossed fingers and ulterior motives.

As long as the show ran, there was a striking correlation between a contestant’s willingness to ‘split’ and his/her trust of the opponent’s reciprocation. It had very little to do with whether or not they wanted to split on their own. If they weren’t going to split, he/she wouldn’t split either. As the O-Town song goes, “I want it all, or nothing at all.”

But an interesting wrench got thrown into all of this sucker fear with a man named Nick. Nick came with a plan (start at the two-minute mark):

[youtube=https://www.youtube.com/watch?v=S0qjK3TWZE8&w=600]

Despite being unbearably stubborn, and completely hijacking the cooperation conversation, Nick’s plan works. And what’s crazier, in the Radiolab interview, Ibrahim was lying. Despite saying he was going to split over and over again, Ibrahim admits that his plan was to “steal” all along, because he didn’t want to be the sucker. Nothing was going to keep him from stealing, he said. But Nick’s brutal honesty, in fact, did.

Sure it’s an odd parallel, but Nick’s gambit is not too far off from the gambit of honesty in our relationships. In any environment where cooperation is often chosen second to acquisition, where fear preempts trust, brutal honesty is our only step forward. To be frank about our own plans for deceiving, to be frank about our fear of being deceived, is a good starting point for trust, precisely because it deconstructs our façade of goodness. After all, Ibrahim says he went into the game saying, “Don’t trust anybody. Don’t trust no one.” Precisely by being the untrustworthy guy he was, Nick endeared himself to Ibrahim. “And now I’m really glad we split.”

Of course these are all horizontal dynamics. The vertical dynamics, thank God, seem to work sans Golden Balls. There’s nothing to split—it is all given. Hit it, O-Town.

[youtube=https://www.youtube.com/watch?v=TG8IkUoZ6j0&w=600]

subscribe to the Mockingbird newsletter

COMMENTS


Leave a Reply

Your email address will not be published. Required fields are marked *