"A second-class intellect but a first-class temperament" was Oliver Wendell Holmes' assessment of Franklin Roosevelt, reflecting an old and widespread notion that the smartest and most ingenious person in the ditch is probably not the one to lead everyone out of it. Human beings seem always to have preferred their leaders to be unsurprising and law-abiding, not clever or imaginative. Zeus runs Olympus, not tricky Hermes; Olòrún rules the Yoruba cosmos, not that smart-aleck Eshu. And it's Wotan who is in charge of the Norse pantheon, not wily Loki, who alternates "between charming and brilliant on one hand, and eccentric and borderline dangerous on the other." (Actually, that's a description of Newt Gingrich from this Politico piece, which quotes a number of people who buy Gingrich's smart-and-creative self-image and then make this point about how those traits don't necessarily make a leader). And now a pair of recent studies claim to confirm and explain this conventional wisdom. One found that higher IQ scores in childhood correlate with a greater chance of using drugs later in life. The other found that creative people were more likely to cheat on tests.

In the this latter study, authors Francesca Gino and Dan Ariely placed people in test-taking situations where they had the means to cheat and a motive as well (higher scores on the tasks resulted in higher pay for participating in the experiments). In one task, for example, volunteers had to record their test answers on a sheet that "accidentally" revealed what was correct for each question. In another, they had to say which side of a computer screen's line contained the most dots, and half the pictures were rather ambiguous, making it easy to make a wink-wink "mistake." In an experiment on working stiffs, the people who had jobs that required creativity proved quite a bit more likely to cheat. So did people who scored as more creative on measures that psychologists use to suss out such traits. In fact, even people who were simply nudged to think about creativity (by making up sentence that used words like "original" and "innovative") proved more likely to cheat. (They also proved to be more creative: Nearly half the people primed with the creativity words solved a subsequent puzzle, versus only about a quarter of those not primed. On the other hand, about half the creativity-primed people cheated, versus only about a quarter of the other group.)

The paper (to be published next year in the Journal of Personality and Social Psychology, online version available here), explains these results with a model of moral behavior in which selfish desires always battle with the desire to see one's self as a good person. People, according to Gino and Ariely, will depart from the ethical only to the extent that they can excuse themselves. Don't want to sit in a room with a handicapped person to watch a film? You'll say you'd rather see a different movie, and then you don't have to feel prejudiced. (In a striking study they cite, in fact, people who had to choose between two movie-rooms tended to pick the handicapped person when the films were the same, but chose an able-bodied partner when the films were different.)

So the idea is that creative people (even people who are just temporarily creative because they've been encouraged a minute ago) will be better at finding justifications for letting themselves off the moral hook. Gino and Ariely say their experiments confirm this model. Though they admit, of course, that they may have found that creativity undermines obedience to rules rather than morality itself. Sometimes the local rules are unethical (like "shred all emails about how we sold securities as good investments and then bet against them"), and there the creative person might find a way to ingeniously defend a moral principle where a plodder would keep his head down.