“I WISH THAT I had magic answers and silver bullets where I could march into any group and say, ‘Here are the five things you need to do, and everything will be better,’” Stephen Dubner said during a recent speech I attended. Instead, he shared a few stories that had taught him and his coauthor, economist Steven Levitt, a different way to look at the world.
What Dubner learned was to Think like a Freak (the title of the duo’s new book), which means to observe, define, and recontextualize the elements of a problem. Then abandon all assumptions except the one that says “There must be a reason why” and reject all rejoinders except the one that says “I don’t know, but I’ll find out.”
Don’t Ask for Attention. Grab It!
To demonstrate his point, Dubner told a story about hand sanitation in hospitals. Ask a roomful of people whether they wash their hands after visiting a restroom, and a great number will lie. Why? “People produce an answer that they think you want them to give,” Dubner said, one reason that “self-reporting data can be very misleading.” For instance, in an Australian hospital, 73 percent of staff doctors claimed that they soaped up before interacting with patients. But a study based on observed behavior showed that only 9 percent of them did.
To improve the rate, hospital administrators at Cedars-Sinai Medical Center in Los Angeles tried everything from a strongly worded memo (which had zero effect) to an incentive of Starbucks gift cards (which the doctors loved, though it did not increase the hand-washing rate). Then a hospital epidemiologist came up with a perfectly absurd idea. At a lunch meeting, she had each administrator—the very people who were telling others how to behave—place one hand in a petri dish. She sent the handprinted dishes to the lab, where analysis turned up a who’s who of super-nefarious bacteria. An IT guy at the hospital got the idea to take a photo of those bacteria-laden samples to use as screen savers on computers all over the hospital. “Literally overnight, the hand-hygiene rate shot up to about 100 percent,” Dubner said. It worked so well, other hospitals started asking for the picture.
The moral of the story: “Human behavior is much harder to change than we think,” he said. To alter people’s habits, you almost always have to shift something in their environment as well.Reimagine the Simplest Acts
Another freaky thinker, according to Dubner: Takeru Kobayashi, the champion of Nathan’s 2001 hot dog eating contest in Brooklyn. Kobayashi deconstructed the whole process of eating a hot dog into elements of efficient consumption. His method went something like this: Separate the hot dog from the bun; break the hot dog in half so it can be swallowed more quickly. Soak the bun in warm water and turn it into a ball, which will help the bun go down faster. Jump up and down while you eat.
By breaking down the act of eating and recontextualizing it as a competition rather than an everyday act, Kobayashi doubled the contest’s previous record, eating 50 hot dogs in 12 minutes. The key to his success was understanding the nature of the project and calibrating his goal, not to past performance but to desired results. Twenty-six hot dogs would’ve gotten Kobayashi the record—50 got him worldwide recognition.
If It’s Legal, Try It!
Dubner said that Kobayashi’s approach to speed-eating also works in other areas of your life. That is, break down a problem into its most basic elements and explore each one from every angle, no matter how outlandish that angle might seem. Unfortunately, many people and businesses mistakenly approach situations in a linear way, trying one idea for six months to see if it works and, if it doesn’t, trying the next idea for another six months. “Experimentation would be to take 100 people and give them all six hours to come up with their best idea. Take those 100 ideas, throw out the 90 worst, and then take the ten that look viable, affordable, and legal, and put them all into play in some form at the same time in a random setting,” Dubner said. Then you might have something special.