It takes two to play the Ultimatum game. One player is called the proposer. Every turn, they are given a small sum, say 20€. That sum is theirs to keep, provided that they offer part of it to the other player, and that their offer is accepted. If the other player (the "responder") rejects the offer, no one receives anything.

The fundamental intuition in this game is that the responder will refuse offers perceived as "unfair", amounts of less than half the total sum. Which is exactly what happens. If you, the proposer, offer to give me 2€ while you keep 18, I will in all likelihood tell you to go hang. That notion of "fairness" is what gives the Ultimatum Game its spice.

A team of Swiss neuroscientists have recently used TMS to investigate the brain processes involved in the game. They posit that players' decisions in the Ultimatum Game reflect a balance between the contradictory impulses of self-interest (I want that money, so I'll accept a 2$ offer) and fairness (if I can't profit, then neither will that greedy pig on the other side of the fence).

They go further by hypothesising that a very popular part of the left hemisphere's frontal cortex, the dorso-lateral bit of the prefrontal (DLPFC) is involved in balancing self-interest with fairness judgment. This was suggested to them by a number of recent imaging studies, mostly fMRI on human subjects.

A word about the method used : in TMS (Trans-cranial Magnetic Stimulation) experimenters use electricity to whack out entire brain areas, on a - hopefully - temporary basis. My friends tell me TMS experiments are even worse than contrast-detection psychophysical experiments, which says a lot, at least to those of you who have ever sat in a dark room for two hours, fighting off sleep, and trying to decide on which side of the goddamn monitor that flicker of light could have been.

But I digress. The basic idea behind the use of TMS is to create artificial lesions : if you disrupt a specific brain area, and subsequently the subject proves himself incapable of accomplishing a task he had no problems with before, then there's a chance that brain area you just buzzed silly is involved in the task.

Knoch et al. use this technique in a straightforward manner. They reason that if the left DLPFC is involved in balancing self-interest with fairness, then if it is prevented from operating normally, players will accept unfair offers more often, because they'll just be "selfish".

And that's exactly what happens. Compared to controls (subjects who had received TMS to the right DLPFC, and the lucky subjects who had received only sham TMS), subjects whose left DLPFC was non-functioning accepted unfair offers with very high probability. And they seemed to "think less" about it, as indicated by mean reaction times that did not vary much between really unfair offers and less unfair ones (control subjects considered unfair offers for longer before reaching a decision).

That fits nicely in with their hypothesis, but I don't feel entirely convinced. The Ultimatum Game involves an aspect of signalling, and maybe of retribution. One possible motivation for rejecting low offers is to let the other player know that you will be rejecting similar offers in the future, and that he or she will have to be more generous. Another would be to punish a bad deed : we won't let the other profit from a lack of generosity. Yet another would be to avoid displays of weakness (accepting low offers puts the adversary in a dominant position). All of these could be construed as examples of planning/strategy, social cognition, moral reasoning, tasks that are generally thought to be the province of the frontal cortex.

Knoch et al.'s experiment, although interesting, fails to distinguish between these possibilities : it is not quite clear exactly what is being disturbed.

It's still a cool game, though.

Original link from Mindhacks. Link to original paper in Science.