(Mary Cybulski/AP)

Six years after the financial crisis, little has been done to squash the caricature of the dishonest banker. Given the London Whale debacle, the LIBOR fixing scandal, the massive currency-rigging scheme and other rogue traders, it's hard not to think something is still very rotten in the state of banking.

Could it be that the culture of banking itself promotes such corrupt behavior? A newly published study in the top journal Nature suggests that might be the case.

The study, by three researchers at the University of Zurich and the University of Chicago, had 128 banking employees take part in a lab experiment that tests honesty. It found that the bankers who were reminded of their profession before the test were far more likely to be dishonest than those who were not.

The lab test was a simple one: It had participants toss a coin ten times in private and then self-report their results, with the potential to earn up to $200 if their outcomes were better than their peers'. (Participants were told in advance of a toss whether "heads" or "tails" would be the winning result.) Thus, the incentive to cheat was strong.

“The idea behind the coin toss is that if everyone behaves honestly, we should observe that roughly 50 percent of contests are heads and 50 percent are tails,” said Alain Cohen, one of the study’s researchers. “But if we observe statistically more than that, we know some people might have cheated.”

Each bank employee was asked a series of questions before the test began. One half, the control group, were asked questions that were irrelevant to their careers, such as how much television they watch. The other half were asked about their careers in finance, such as where they’re employed, what the advantages and disadvantages of a banking career are, and why they decided to be a banker. “By reminding them of their professional role, we put the emphasis on their occupational identity, and that makes the norms that come along with their role more prominent in their behavior.”

Indeed. When the bankers in the control group reported their results, just three percent of the coin flips were misrepresented in an effort to get more cash. But among the group of bankers who were asked about their jobs, 16 percent of the coin flips were misrepresented. That group was also significantly more likely to agree with the statement that a person's social status is primarily decided by his or her financial success.

To make sure a reminder of any professional identity—or association with making money—wouldn’t prompt similar results, the researchers then ran their study on a group of students and a group of employees from the general public who hailed from a wide range of professions. In those studies, unlike the one with bank employees, the people who were first asked questions about their careers (in the case of the workers) or about the banking industry in general (in the case of the students) were not significantly more dishonest than their peers.

So does this mean banking leaders should try to find ways to make bankers forget what they do for a living? Prohibit them from wearing suits to work? Pull down the wood paneling from the walls? Hold meetings in coffee shops rather than on trading floors?

No, said Cohen. But it is important for finance CEOs to work harder to change cultural norms, such as ensuring that managers model good behavior at the lowest levels of the bank and making sure performance incentives don’t inadvertently reward dishonesty. “It would also be good to have a professional code like the Hippocratic oath,” Cohen said. “Bankers would commit themselves to comply with the norms.”

The industry's image, after all, still has a lot of room for improvement. Cohen's study asked a general sample of people who they thought would cheat most on the coin flipping test—physicians, prison inmates, bank employees or people from the general population. Bankers were considered the most likely scammers, even more than prisoners.

In fact, the idea for the study, Cohen said, originally came from prior research by him and his co-authors, in which they asked inmates in prison to do a similar coin toss. “When you reminded criminals that they were criminals, they became more dishonest,” Cohen said. He was quick to note, however, that he was not drawing a parallel between bankers and criminals. “I don’t want to confuse people,” he said. “We should be careful. It was just the same methodology we used.”

Read also:

Why customers are less and less happy with their banks

How Wall Street should change its intern culture

A Hippocratic oath for bankers?

Like On Leadership? Follow us on Facebook and Twitter.