Crime and Punishment: Why Do We Conform to Society?

Scientific American
October 05, 2007

By Nikhil Swaminathan

A pair of brain regions work together to assess the threat of punishment and override our selfish tendencies

Whether you subscribe to the Ten Commandments, the Golden Rule or some instinctive moral code, society functions largely because most of its denizens adhere to a set of norms that allow them to live together in relative tranquility.

But, why is it that we put a vast amount of social resources into keeping stealing, murdering and other unfair (not to mention violent and illegal) acts to a minimum? Seems it all comes down to the fact that most of us don't cotton to being punished by our peers.

"The reason why punishment for norm violations is important is that it disciplines the potential norm violators," says Ernst Fehr, an economist at the University of Zurich and the senior author of a paper on the issue published this week in Neuron.

In the new study, Fehr and colleagues uncovered activity in two areas of the brain underlying the neural mechanism involved in conforming to society's values. They further determined that subjects with Machiavellian personalities—a strong sense of self-interest, opportunism and manipulation—have heightened activity in one of these regions, which the authors believe is related to assessing the threat of punishment.

During the study, which also involved scientists at the University of Ulm in Germany, 23 male students were instructed to play a version of the "ultimatum game" while their brains were scanned via functional magnetic resonance imaging (fMRI). Each participant was given a sum of money (100 monetary units) to split however he chose with an anonymous partner. In some cases the recipient simply had to accept any offer made. Other times, after an offer was made, the recipient had the option penalize the giver by taking some or all of their money, if the latter had not shared generously.

The subjects' brains were only scanned when they played the giver role. Before each trial, both players were told whether the recipient would be allowed to exact a punishment if he felt he got too slim a slice of the pie. Two areas of the cortex (the brain's primary processing unit) were particularly active during the trials when punishment was an option: the lateral orbitofrontal cortex, a region below the temples of the head that had, in previous research, been implicated in processing a threat stimulus, and a section just behind it called the dorsolateral prefrontal cortex.

"The lateral orbitofrontal cortex [activity] represents the punishment threat here," says Fehr, citing previous research that fingered it in threat assessment. "More specifically, how bad does the brain interpret this punishment threat?"

Alternatively, he says, "[the dorsolateral prefrontal cortex] is an area that is involved in cognitive control and overriding prepotent impulses. Here, we have a design where the prepotent impulse is not to share the money—at least to the extent that player B wants it shared."

Interestingly, the research team also had their subjects fill out a questionnaire to determine their degree of Machiavellian behavior. Those who proved to be the most ruthless of the bunch offered little to nothing when there was no threat of punishment, but within the punishment paradigm, they were generous enough to stave off retribution.

"These are socially intelligent, selfish people," Fehr says about the more calculating subjects. "They escape the punishments that are inherent in social interactions, because they seem to have a fine sense of when punishment is in the air."

Jorge Moll, principal investigator of the cognitive and behavioral neuroscience unit at the Rede Labs-D'Or Hospitals in Rio de Janeiro, says the most interesting findings were that individual scores on Machiavellianism predicted "how much a given subject will change his behavior depending on the presence of punishment," and "that the level of activity within the lateral orbitofrontal cortex is strongly related to Machiavellian personality style."


Researchers say the results could have wide-reaching implications, potentially paving the way to understand—and perhaps one day reverse—the neurobiology behind psychopathic and sociopathic personalities. They intend to repeat the study with patients suffering from antisocial anxiety and personality disorders to determine if their behavior can be explained by a lack of impulse control or a poor assessment of punishment.

Fehr argues the results could also impact the criminal justice system since the dorsolateral prefrontal cortex does not fully develop until after a person is around 20 years old.

"This area seems to be critically important in overriding self-interest," he says. Thus, "you just can't treat an immature adolescent the same way as a mature adult—that's at least my view of doing justice." It's unclear whether judges and juries see it that way, however.