Harvard psychology professor Daniel Gilbert is widely known for his 2006 best seller, Stumbling on Happiness. His work reveals, among other things, the systematic mistakes we all make in imagining how happy (or miserable) we’ll be. In this edited interview with HBR’s Gardiner Morse, Gilbert surveys the field of happiness research and explores its frontiers.
HBR: Happiness research has become a hot topic in the past 20 years. Why?
Gilbert: It’s only recently that we realized we could marry one of our oldest questions—“What is the nature of human happiness?”—to our newest way of getting answers: science. Until just a few decades ago, the problem of happiness was mainly in the hands of philosophers and poets.
Psychologists have always been interested in emotion, but in the past two decades the study of emotion has exploded, and one of the emotions that psychologists have studied most intensively is happiness. Recently economists and neuroscientists joined the party. All these disciplines have distinct but intersecting interests: Psychologists want to understand what people feel, economists want to know what people value, and neuroscientists want to know how people’s brains respond to rewards. Having three separate disciplines all interested in a single topic has put that topic on the scientific map. Papers on happiness are published in Science, people who study happiness win Nobel prizes, and governments all over the world are rushing to figure out how to measure and increase the happiness of their citizens.
How is it possible to measure something as subjective as happiness?
Measuring subjective experiences is a lot easier than you think. It’s what your eye doctor does when she fits you for glasses. She puts a lens in front of your eye and asks you to report your experience, and then she puts another lens up, and then another. She uses your reports as data, submits the data to scientific analysis, and designs a lens that will give you perfect vision—all on the basis of your reports of your subjective experience. People’s real-time reports are very good approximations of their experiences, and they make it possible for us to see the world through their eyes. People may not be able to tell us how happy they were yesterday or how happy they will be tomorrow, but they can tell us how they’re feeling at the moment we ask them. “How are you?” may be the world’s most frequently asked question, and nobody’s stumped by it.
There are many ways to measure happiness. We can ask people “How happy are you right now?” and have them rate it on a scale. We can use magnetic resonance imaging to measure cerebral blood flow, or electromyography to measure the activity of the “smile muscles” in the face. But in most circumstances those measures are highly correlated, and you’d have to be the federal government to prefer the complicated, expensive measures over the simple, inexpensive one.
But isn’t the scale itself subjective? Your five might be my six.
Imagine that a drugstore sold a bunch of cheap thermometers that weren’t very well calibrated. People with normal temperatures might get readings other than 98.6, and two people with the same temperature might get different readings. These inaccuracies could cause people to seek medical treatment they didn’t need or to miss getting treatment they did need. So buggy thermometers are sometimes a problem—but not always. For example, if I brought 100 people to my lab, exposed half of them to a flu virus, and then used those buggy thermometers to take their temperatures a week later, the average temperature of the people who’d been exposed would almost surely be higher than the average temperature of the others. Some thermometers would underestimate, some would overestimate, but as long as I measured enough people, the inaccuracies would cancel themselves out. Even with poorly calibrated instruments, we can compare large groups of people.
A rating scale is like a buggy thermometer. Its inaccuracies make it inappropriate for some kinds of measurement (for example, saying exactly how happy John was at 10:42 AM on July 3, 2010), but it’s perfectly appropriate for the kinds of measurements most psychological scientists make.
What did all these happiness researchers discover?
Much of the research confirms things we’ve always suspected. For example, in general people who are in good romantic relationships are happier than those who aren’t. Healthy people are happier than sick people. People who participate in their churches are happier than those who don’t. Rich people are happier than poor people. And so on.
That said, there have been some surprises. For example, while all these things do make people happier, it’s astonishing how little any one of them matters. Yes, a new house or a new spouse will make you happier, but not much and not for long. As it turns out, people are not very good at predicting what will make them happy and how long that happiness will last. They expect positive events to make them much happier than those events actually do, and they expect negative events to make them unhappier than they actually do. In both field and lab studies, we’ve found that winning or losing an election, gaining or losing a romantic partner, getting or not getting a promotion, passing or failing an exam—all have less impact on happiness than people think they will. A recent study showed that very few experiences affect us for more than three months. When good things happen, we celebrate for a while and then sober up. When bad things happen, we weep and whine for a while and then pick ourselves up and get on with it.