The (Honest) Truth About Dishonesty

Although this was an intriguing result, we didn’t get too excited just yet. This first study showed that creativity and dishonesty are correlated, but that doesn’t necessarily mean that creativity is directly linked to dishonesty. For example, what if a third factor such as intelligence was the factor linked to both creativity and dishonesty?

The link among intelligence, creativity, and dishonesty seems especially plausible when one considers how clever people such as the Ponzi schemer Bernie Madoff or the famous check forger Frank Abagnale (the author of Catch Me If You Can) must have been to fool so many people. And so our next step was to carry out an experiment in which we checked to see whether creativity or intelligence was a better predictor of dishonesty.

Again, picture yourself as one of our participants. This time, the testing starts before you even set foot in the lab. A week earlier, you sit down at your home computer and complete an online survey, which includes questions to assess your creativity and also measure your intelligence. We measure your creativity using the same three measures from the previous study, and measure your intelligence in two ways. First, we ask you to answer three questions designed to test your reliance on logic versus intuition using a set of three questions collected by Shane Frederick (a professor at Yale University). Along with the correct answer, each question comes with an intuitive answer that is in fact incorrect.

To give you an example, try this one: “A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball. How much does the ball cost?”

Quick! What’s the answer?

Ten cents?

Good try, but no. It’s the seductive answer, but not the right one.

Although your intuition prods you to answer “$0.10,” if you rely on logic more than intuition, you’ll check your answer just to be sure: “If the ball were $0.10, the bat would be $1.10, which combine to equal $1.20, not $1.10 (.1 + (1 + .1) = 1.2)! Once you realize that your initial instinct is wrong, you enlist your memory of high school algebra and produce the correct solution (.05 + (1 + .05) = 1.1): 5 cents. Doesn’t it feel like the SATs all over again? And congratulations if you got it right. (If not, don’t worry, you would have most likely aced the two other questions on this short test.)

Next, we measure your intelligence through a verbal test. Here you’re presented with a series of ten words (such as “dwindle” and “palliate”), and for each word you have to choose which of six options is closest in meaning to the target word.

A week later, you come to the lab and settle into one of the computer-facing chairs. Once you’re situated, the instructions begin: “You’ll be taking part in three different tasks today; these will test your problem-solving abilities, perceptual skills, and general knowledge. For the sake of convenience, we’ve combined them all into one session.”

First up is the problem-solving task, which is none other than our trusty matrix task. When the five minutes for the test are over, you fold your worksheet and drop it into the recycling bin. What do you claim is your score? Do you report your actual score? Or do you dress it up a little?

Your second task, the perceptual skills task, is the dots test. Once again, you can cheat all you want. The incentive is there—you can earn $10 if you cheat in every one of the trials.

Finally, your third and final task is a multiple-choice general-knowledge quiz comprised of fifty questions of varying difficulty and subject matter. The questions include a variety of trivia such as “How far can a kangaroo jump?” (25 to 40 feet) and “What is the capital of Italy?” (Rome). For each correct answer, you receive 10 cents, for a maximum payout of $5. In the instructions for this last test, we ask that you circle your answers on the question sheet before later transferring them to a bubble sheet.

When you reach the end of this quiz, you put down your pencil. Suddenly the experimenter pipes up, “Oh, my gosh! I goofed! I mistakenly photocopied bubble sheets that are already marked with the correct answers. I’m so sorry. Would you mind using one of these premarked bubble sheets? I’ll try to erase all the marks so that they will not show very clearly. Okay?” Of course you agree.

Next the experimenter asks you to copy your answers from the quiz to the premarked bubble sheet, shred the test sheets with your original answers, and only then submit the premarked bubble sheet with your answers in order to collect your payment. Obviously, as you transfer your answers you realize that you can cheat: instead of transferring your own answers to the bubble sheets, you can just fill in the premarked answers and take more money. (“I knew all along that the capital of Switzerland is Bern. I just chose Zurich without thinking about it.”)