The dark side of creativityOriginal thinkers can be more dishonest.

Picture a blank answer sheet and a six-sided die. Imagine someone alone in a room, with no camera and no proctor, filling in scores they didn't earn, rolling a die, and reporting a number that conveniently comes up high. This is the laboratory that Francesca Gino and Dan Ariely built to catch cheaters. The twist, the finding that makes this research genuinely uncomfortable, is that the people most likely to cheat weren't the cleverest in the room. They were the most creative. Creativity is one of those traits that everyone wants more of. Organizations try to hire for it, and schools try to cultivate it. The case for creativity is almost always framed in pure upside: original thinking solves hard problems, drives innovation, and helps people adapt. Gino and Ariely don't dispute any of that. What they propose is that the same cognitive flexibility that generates novel ideas also makes it easier to develop novel excuses. They call this moral flexibility — the ability to produce self-serving justifications that make unethical choices feel acceptable. The question they set out to answer across five experiments is whether that flexibility translates into actual dishonest behavior. The first experiment tested dispositional creativity — that is, creativity as a stable personality trait rather than a temporary state. One hundred eleven students completed established creativity scales, two measures of intelligence, and three behavioral tasks that let them overreport their performance for monetary gain: a matrix problem-solving task, a multiple-choice general-knowledge test, and a perceptual task. In each case, participants scored themselves. Cheating was defined as the gap between what they claimed and what they actually did, aggregated into a composite dishonesty index using z-scores. The correlations were striking. Gough's Creative Personality scale correlated with cheating on the matrix task at point five three. The Creative Behavior Inventory correlated at point four two. Creative Cognitive Style came in at point three five. All significant, all pointing in the same direction. Intelligence, by contrast, showed no reliable relationship with dishonesty at all. When Gino and Ariely ran a regression that included both creativity and intelligence in the model simultaneously, creativity was a strong predictor of cheating, with a standardized beta of point five one and a p-value below 0.001. Intelligence's beta was point zero six, nowhere near significant. The same minds that score high on creative personality scales are the ones inflating their scores. That pattern holds up even after accounting for how smart they are. This is the first blow to the straightforward belief that creativity is universally good. But it raises an obvious question: is this just about who creative people are, or can you temporarily make someone more creative and observe their honesty drop in real time? Experiment two answers that directly. Here, the manipulation was a scrambled-sentence task. Participants rearranged sets of five words into grammatical four-word sentences. In the creative-mindset condition, twelve of the twenty sentences included words like original, inventiveness, ingenious, and imagination. The control condition had none. This is a subtle prime — it doesn't instruct anyone to be creative; it simply saturates their working environment with creativity-related concepts for a few minutes. The result showed up clearly in the manipulation check: forty-seven percent of primed participants solved the Duncker candle problem, a classic measure of creative insight, compared to twenty-seven percent of the controls. Then came the matrix task. Participants in the creative-mindset condition overstated their performance by an average of two point seven one matrices. Controls overstated by one point zero nine. Forty-nine percent of the primed group cheated at all, compared to twenty-seven percent of the controls. Actual performance on the matrices was identical across conditions, so the prime didn't make people smarter or better at the task; it made them more likely to lie about how well they did. That finding is significant because it shifts the argument from personality to causality. You don't have to be a dispositionally creative person for this to happen to you. You just need to spend a few minutes thinking creatively, and the effect follows. Experiments three and four investigate the mechanism. The proposed explanation is that creativity expands people's ability to justify their own behavior — to construct reasons why, in their particular situation, bending the rules is acceptable. To test this, Gino and Ariely introduced a private die-roll paradigm. Participants rolled a six-sided die alone, reported the outcome, and received payment in dollars matching whatever number they reported. No one could catch an individual liar. But statistically, if a group reports suspiciously high numbers, you can detect inflation. Experiment three used a two-by-two design: creative prime versus control, crossed with low versus high justification. In the low-justification condition, participants rolled the die once and reported it — no wiggle room. In the high-justification condition, they rolled multiple times but were instructed to report the first roll, which sounds constrained. However, in practice, it gave them a reason to convince themselves they were reporting something legitimate. The interaction was significant, with an F-value of four point zero five and a p-value of 0.046. In the low-justification condition, creatively primed participants reported substantially higher die outcomes than controls. In the high-justification condition — where everyone had room to rationalize — the creative prime made almost no difference. When external justification is available to everyone, the creative advantage in self-justification disappears. Experiment four replicated this logic using dispositional creativity instead of the prime. Ninety-seven students completed the same die-roll task with the same justification manipulation, while the researchers measured their creative personality and controlled for narcissism, since narcissism is associated with dishonesty and some creativity measures, so it needed to be ruled out. The moderated regression found a significant interaction between creative personality and justification condition, with a beta of negative point two zero. In the low-justification condition, dispositional creativity strongly predicted higher reported die outcomes, with a simple slope beta of point five six. In the high-justification condition, the slope was weaker but still present, with a beta of point two seven. Read together, experiments three and four pinpoint the mechanism. Whether you temporarily prime creativity or measure it as a stable trait, the effect on cheating is strongest when external justification is scarce. Creative people generate their own justifications internally. Take away the environmental excuse, and they still find a way. When everyone has a ready-made excuse, the creative advantage shrinks because less creative people can use it too. Experiment five adds a final dimension by asking: what happens when you combine a high-creativity personality with a creativity prime? One hundred eight students completed the same scrambled-sentence prime from experiment two, plus a battery of dispositional creativity measures taken in an online survey beforehand. The interaction between the creative-mindset prime and dispositional creativity predicted cheating, both the decision to cheat and the extent of cheating reached significance. The simple slopes tell the real story. For participants low in dispositional creativity, the prime was powerful: it raised the likelihood of cheating with a standardized slope of point five two and the extent of cheating with a slope of point four eight. For participants high in dispositional creativity, the prime had essentially no effect — slopes near zero, p-values near point nine. Highly creative people are already operating near their ceiling for self-justification; they don't need a situational nudge to get there. So, across five studies, the picture is consistent. Creative personality predicts cheating better than intelligence does. Temporarily priming creativity increases cheating even when actual ability stays the same. The effect runs through justification — remove the ability to rationalize, and the creativity effect shrinks. Highly creative people are already so well-supplied with internal justifications that an external prime adds nothing. Gino and Ariely are careful about what this means in practice. They acknowledge that their studies were laboratory tasks with monetary incentives, and they explicitly call for field research to test whether these effects hold in real workplaces with real stakes. They also acknowledge that self-reported performance and self-reported die rolls have limitations — motivated perception, not conscious fraud, might explain some results. But the organizational implication is hard to sidestep. Creativity carries genuine economic value. Gino and Ariely note that creative products generate higher average returns; yet the evidence suggests that the people most likely to produce those creative returns may also be the most capable of constructing reasons why bending the rules is fine in their particular case. The same divergent thinking that generates breakthroughs also generates the best excuses. These two things are not separate; they are the same cognitive capacity pointing in different directions. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.