Jennifer Lerner on Emotions and Decision Making
Harvard Kennedy School
April 22, 2015
The complex role of emotion in human decision making and choice is a growing field of social science research. One of the leaders in the field isJennifer Lerner, professor of public policy and management at Harvard Kennedy School. Drawing insights from psychology, economics, and neuroscience, Lerner’s research examines human judgment and decision making.
Q. In a recent paper, you and three colleagues reviewed the data on the past 35 years of research on emotion and decision making. What was the first major take-away from that research?
Lerner: The first important take-away is that there has been a revolution in emotion science. Most of the 20th century was dominated by what we call the “cognitive revolution” in psychology, during which we gained tremendous insights into cognitive processes. Important insights in the new field of behavioral decision research led to Nobel Prizes for Daniel Kahneman and Herbert Simon. As the century drew to a close, we learned that although we know a tremendous amount about the nuances of cognitive processing, that knowledge does not go all the way toward explaining human behavior. In fact, emotion turns out to be an enormously powerful predictor of behavior, especially in the area of politics and policy.
Since the end of the last century, there has been a tremendous surge to understand the intricacies of emotion and decision making. The number of scholarly papers published annually on emotion and decision making doubled between 2004 and 2007, and doubled again between 2007 and 2011. In fact, for the first time in its history, the Annual Review of Psychology, which serves as the compendium of all the progress in psychology, requested a review paper on the field of emotion and decision making.
So, the first major take-away from my recent research paper is that there’s tremendous activity in this space, and it’s a terrifically exciting time for what we do here at the Kennedy School. Emotion has been the poorly understood driver of citizens’ behavior and policymakers’ behavior for far too long. Now we are starting to harness the insights needed to advance understanding of public policy and political behavior.
Q. What was the other important conclusion from this research?
Lerner: A second take-away is the fact that it’s useful to talk about two different kinds of emotion. In particular, we make a distinction between integral emotion and incidental emotion.
Integral emotions are those that are normatively relevant to the judgment or decision at hand. If you’re trying to decide whether to make a risky investment, for example, your sense of anxiety about possible stock volatility is integral to the judgment or decision at hand. We consider that emotion a rational input into the decision-making process. So, it would be a crude oversimplification to think of emotion as an irrational input.
One of the key findings in the literature is evidence revealing that when people have damage to important neural circuitry involved with the processing of emotion, they make sub-optimal decisions. For example, they tend to make decisions that lead to financial loss when they’re not able to appreciate the feeling of fear and anxiety that one should normally feel when considering a highly risky gamble. So in that sense, integral emotions are vital inputs to decisions.
What people more generally have in mind when they assume that emotions impair decisions are what we call incidental emotion. An incidental emotion is one that’s triggered by a particular situation that carries over and influences your subsequent judgments and decisions. My research group here at the Kennedy School is very actively involved in research on both kinds of emotion, but we have a larger emphasis on incidental emotion because of the problems that it causes for everyday decision making, particularly in the public sector.
Q. You have done quite a bit of research focusing on the impact of sadness on decision making. Please explain.
Lerner: There are key emotions, like sadness, that make us think quite systematically – but that doesn’t mean we think in an unbiased way. Specifically, we’ve found, for example, that sadness, owing in part to its underlying cognitive components, carries over and makes people extremely financially impatient. That tendency manifests in decisions where people must forego receiving greater amounts of money in the future to enjoy more money immediately. This desire to consume now rather than save is one of the key drivers of the major credit-card debt problem that exists in the United States and elsewhere. Sadness magnifies what is already a significant bias toward the present.
Thanks to our understanding of the cognitive, biological, and motivational components of a given emotion like sadness, we can selectively elicit a complementary emotion to try to impact decision making. In this case, we found that gratitude elicits opposing tendencies from sadness when it comes to economic behavior. We’ve shown that gratitude leads people to think more about future returns. Consequently, they become even more financially patient than when they’re in a neutral state.
This is a key finding because, as it turns out, it’s very easy to elicit a state of gratitude. My Kennedy School colleague Brigette Madrian and I are now working with retirement savings corporations to try to make an impact with citizens who are considering their retirement decisions. It’s not as simple as making people feel happy to help them make better decisions; happiness does not elicit this effect. Rather, the particular properties of gratitude counterbalances the components associated with sadness, leading it to have this effect on one’s long-term financial outlook.
Q. Your research has some very practical implications for public policy. Can you talk about that?
Lerner: In my past work, we have looked a lot at how people’s emotional responses are useful predictors across time, in terms of predicting responses to terrorism and of policy preferences regarding terrorism. That is a part of a broader body of work in which we find very often that emotional reactions actually happen to have more predictive power than people’s attitudes, in terms of driving ultimate behavior. There is something very powerful about a feeling that directly links to behavior, and it has to do, in large part, with the underlying neural circuitry and how close emotion neural circuitry is to motor activity and actual behavior.
We are moving now toward two other new policy areas. The first concerns how medical oncologists make end-of-life recommendations to their parents. End-of-life decision making is receiving a lot of attention lately, in part because we’re all keenly aware of how many choices people face about possible healthcare options near the end of life. Also, we’re aware of the expenses associated with those options and the inherent trade-offs associated with quality of life.
The Institute of Medicine recently came out with a significant report on end-of-life decision making, arguing among other things for the necessity of research to answer some of the thorny questions we are confronting. Most researchers are focusing on patients’ preferences and how to elicit them. My colleagues here at Harvard and at the Cancer Center at Mass. General Hospital and I are focusing instead on the oncologists, which is an unusual thing to do.
We’re doing this because when you do a careful analysis of the actual cases in which the decision is made to discontinue chemotherapy, over 90% of the time, the oncologist is the one who recommended this course of action, and then the patient agreed to it. So, the oncologist’s suggestion is the key driver, and that’s why we’re focusing on oncologists and their emotions.
Having received an initial grant for this work through the National Cancer Institute of National Institute of Health, we are assembling a team from across the country to apply for a larger grant in which we would examine the feelings oncologists have when recommending that patients discontinue chemotherapy. We know, of course, that this is a very difficult choice for the patients, but it is also a very difficult recommendation for the oncologists to make. To see how these decisions are made, we examine the interaction of emotion between the oncologist and the patient for a variety of measures, and we look at a variety of background characteristics as well.
Q. So, where do you see opportunities for future research in this field?
Lerner: Our future research plans in the area of emotion and decision making cluster into two categories: One I would call basic science, in which we look at underlying mind-brain behavior connections using a variety of measures. We’re presently using brain imaging at Massachusetts General Hospital while studying behavioral decisions here in the Harvard Kennedy School’s Decision Science Laboratory.
As we seek to understand some of those key mechanisms, we are engaged in a variety of other key projects looking at the implications for society and policy, such as how to improve the effectiveness of anti-smoking campaigns. The large settlements that have been reached with the tobacco industry in recent years have resulted in large sums of state and federal funding available for anti-smoking campaigns. You have probably seen some of these public service advertisements on television. They are typically terribly sad and focus on death related to smoking; their impact on smoking cessation has been mixed at best.
It turns out that these public service announcements, although they are well-intended, generally have not been informed by emotion or decision science. In collaboration with colleagues at the Harvard School of Public Health, my research group here at the Kennedy School is working on an extensive project in which we design anti-smoking public service announcements informed by science that we hope will be much more effective.