The Frontal Cortex

Nudges and Decision-Making

David Leonhardt has an interesting column on the importance of using subtle environmental cues – Leonhardt calls them “nudges” – to encourage good decision-making. He begins with a fascinating anecdote about patients in hospital beds:

For more than a decade, it turns out, medical researchers have known that people on ventilators should generally have their heads elevated. When the patients are lying down, bacteria can easily travel from the stomach, up to the mouth and breathing tube, and ultimately into the lungs, causing pneumonia. When people are propped up, gravity becomes their ally.

But hospitals have had a hard time translating this scientific knowledge into better medical care. Patients frequently need to be put on their backs, to be bathed or to receive treatment, and once they are lying down, doctors and nurses — busy worrying about dozens of other things — don’t always remember to move the bed back up.

“When you have to rely on someone to do it, it’s not going to happen every time,” said Dr. Michael Gropper, the director of critical care medicine at UCSF Medical Center, the hospital I was visiting.

So Dr. Gropper made a new rule. Unless there was a written order from a doctor saying that a patient should be lying down, every patient on a ventilator had to be sitting up.

The rule was one small part of a common-sense campaign to reduce infections in the intensive care unit over the last two years. None of it was cutting-edge science, but it has made a big difference: the incidence of ventilator-associated pneumonia has fallen more than 40 percent since 2005. There are people walking around Northern California this morning who otherwise would not be alive.


I’m not sure if the failure to sit people up in hospital beds is the status quo bias or just plain old laziness. But the work of behavioral economists has made it clear that it’s much easier for people to make the right decision when that decision is the default choice. More people would enroll in 401(k)’s if they were automatically enrolled, but could choose to opt out. People eat significantly less food from an all-you-can-eat buffet when their plate size is reduced, even though they can go and refill the plate. Such reforms skirt the usual libertarian objections, since they don’t diminish choice, but simply frame it in a new way.

Of course, redefining the default will only work in a limited number of situations. In many cases, it’s even more important to present the information in as clear and simple a manner as possible, so that people can make a decision based on relevant facts. Our brain is easily intimidated, and too much information is aversive.

Leonhardt notes that almost 25 percent of low income seniors eligible for a subsidy for the new Medicare drug benefit haven’t signed up, because the program is simply too complicated. There are too many options and variables. “It’s sufficiently complicated that people sort of throw up their hands and say, ‘I can’t deal with it,’ ” said Joseph Newhouse, a Harvard economist.

The same principle is at work elsewhere. Sheena Iyengar, a psychologist at Columbia, has shown that people are less likely to invest in a 401(k) when they are given a larger number of investment options. (For every 10 investment plans on offer, participation in 401(k) plans drops by 2 percent.) Car shoppers are more likely to buy a fuel-efficient vehicle when the fuel economy is translated into yearly fuel costs, as opposed to miles-per-gallon. Leonhardt cites the Charlotte school district as another example of excessive information:

Five years ago, the Charlotte-Mecklenburg Schools, the largest district in North Carolina, started a school choice program, giving parents a bigger say over where their children went to school. But finding good information about schools, like their average test scores, sometimes seemed like an unpleasant exam in its own right. Parents often had to wade through a Web site filled with acronyms like EOC, EOG, ABC and AYP.

Last year, at the urging of an economist named Justine Hastings and two other Yale researchers, the Charlotte schools conducted a little experiment to see if this complexity mattered. Along with their school choice applications, a few thousand parents were also mailed a sheet of paper listing a single test score — the average of the math and reading scores — for each school they could apply to.

And guess what? These parents were much more likely than others to apply to schools with high scores. They were starting to create the feedback loop that is the whole point of school choice.