The Madoff Case: Is Cheating Contagious?

Sometimes cheaters do get caught. New York financier Bernie Madoff faces a possible life sentence after pleading guilty to 11 felony counts in one of history's largest investment frauds. Unfortunately, this is just one of many recent stories about captains of industry and finance behaving badly with other people's money. In fact there have been so many revelations like this recently, it's enough to make the average tax-paying citizen wonder whether anyone actually plays by the rules any more. And if people who already have plenty of money can't help skimming (or just hauling away wads of cash), it's fair to ask how an ordinary person is supposed to resist the temptation to skim on their taxes or pad their time sheets at work when times are tight.

So why do some people cheat and others don't? The classical explanation is that it's a rational choice, a cold calculation of cost and benefit. Can I get away with it, and how much can I get away with before I risk getting caught? But some scientists have begun questioning this cynical view of human ethics and suggest that the decision is much more complex than this simple calculation.

Three psychologists recently decided to explore these knotty ethical questions in the laboratory. Francesca Gino of the University of North Carolina and Shahar Ayal and Dan Ariely of Duke University set up an elaborate hoax to see if they could actually make people cheat—in order to illuminate the psychological forces at work in the dishonest mind.

Here, briefly, is what they did: They asked a large group of university students to solve a set of complex math problems in a very short time. They made it hard enough that none could realistically solve all the problems, and they paid them for whatever ones they did solve. The math exercise was just a pretense for the real experiment: shortly after the students began on the math problems, one of them (actually a paid actor) loudly announced to the room: "I've solved everything. What should I do?" Everyone in the room knew this was impossible, so the student-actor was a clear example of blatant cheating. He also took all of the cash, as if he had a perfect score and—very important—left without any consequences.

The idea was to see how many of the students followed the cheater's example—to see if blatant dishonesty boosted cheating among students generally. And it did, dramatically. But the psychologists added another twist to the experiment: sometimes they had the actor wear the T shirt of a rival university, other times not. They wanted to see if the cheater's group identity—classmate or outsider—influenced the level of copycat cheating. That is, would students cheat more (or less) when they saw a rival cheat, as compared to seeing a compatriot cheat?

The results were unambiguous. As reported in the March issue of Psychological Science, fellow classmates had much more influence than outsiders. Indeed, seeing a rival cheat actually lowered the level of overall cheating slightly—compared to students who simply cheated on their own initiative, without any prodding. These findings argue against the "cold calculation" theory of cheating. After all, if the students only weighed the can-I-get-away-with-it factor, then they would have been influenced equally by the successful cheating of both compatriot and outsider. And they weren't.

The psychologists decided to double-check these findings with another small experiment. It was basically the same setup, but in this scenario the actor didn't do anything; he simply asked out loud of the proctor: "Is it OK to cheat?" I know, stupid question. Nobody would really do that. But the idea was simply to nudge the inner moralist in the students' minds, to bring the issues of cheating and dishonesty front and center. And when they did this, the students cheated noticeably less. No role models, good or bad, just priming the idea of unethical behavior—that was enough to keep the students honest.

So it appears our inner moralist doesn't really want to cheat. Yet it also appears that dishonesty can be contagious—if we witness one of our own committing the public act of dishonesty. These findings point to a possible strategy for preventing a wave of unethical contagion. If cheating in general declines when cheaters are perceived as outlaws, then it should help to stigmatize public cheaters as just that—outlaws, bad apples. Of course, Bernie Madoff and the rest of Wall Street's alleged fraudsters have already done a lot of that work for us.

Join the Discussion