Since this book was first published there has been no shortage of new programs designed to help people live happier and more productive lives. Alas, some have not taken the lessons of this book about story-editing into account. Even worse, many have committed the cardinal scientific sin of failing to test their interventions properly. This has been the case, unfortunately, with some high-profile programs that are costing taxpayers millions or even billions of dollars.
But there is plenty of good news as well. New story-editing interventions have been developed that are simple, cheap, and effective. And plans are under way to scale them up so that they are available nationwide. I’ll start with some laments about the poorly tested programs and end with applause for the recent successes.
Can you guess which war is the longest in U.S. history? If you said the war in Afghanistan, you are correct. American troops entered Afghanistan soon after the terrorist attacks in September 2001. As I write more than a dozen years later, the conflict is winding down, but hostilities and casualties continue. Because of the length of this war many U.S. soldiers have served multiple tours in Afghanistan, even if they experienced symptoms of post-traumatic stress disorder (PTSD). Countless others experienced marriage difficulties, family conflict, and communication problems but return again and again. Some, sadly, are so scarred by their experiences that they are unable to return to active duty. Estimates of the rate of PTSD among soldiers who fought in Iraq or Afghanistan range from 10 to 18 percent, which translates to a great deal of suffering in thousands of men and women who served their country.1 It is thus imperative to find ways to treat PTSD. Even better would be to prevent it in the first place, instilling resilience and psychological strength in such a way that would prepare soldiers for the stress of combat. This is the goal of the Comprehensive Soldier Fitness (CSF) program, one of the most ambitious psychological interventions ever attempted. The program originated in conversations between Martin Seligman, the father of the positive psychology movement, and military officers, including the chief of staff of the army, who were rightly concerned about the high incidence of PTSD, suicide, and divorce among combat veterans. Seligman and colleagues developed a program to increase resilience among soldiers using the principles of positive psychology, which attempts to increase human strengths and flourishing, rather than waiting for mental health problems to develop and then treating them.
CSF involves giving a “psychological fitness” assessment to soldiers, followed by resilience training classes, individualized sessions if needed, and repeated assessment. The goal is to increase emotional, social, spiritual, and interpersonal strengths among soldiers and their families. These interventions are facilitated by “master resilience trainers,” sergeants who take a ten-day training course. The program obviously has admirable goals, and Seligman and his colleagues are to be lauded for attempting to help those who are putting their lives at risk for their country. But good intentions don’t always result in good practices, and if we dig a little deeper some problems become clear. One concerns how evidence-based the program is. It turns out that CSF is modeled on an earlier program designed to increase resilience in schoolchildren aged eight to fifteen that itself had only modest benefits. Even if this school program worked well, there is the issue of using techniques developed to help kids bounce back from the travails of adolescence, such as not being invited to a birthday party, to help soldiers who witness their best friend blown up by a roadside bomb or who have to amputate the leg of a fellow soldier to keep him from bleeding to death. That seems like quite a leap.2
Of course, we have to start somewhere, and maybe we could use the intervention program for kids as a starting point, tailoring it, as best we can, into one more appropriate for soldiers. That would be a reasonable approach, as long as we took these steps: first, we shouldn’t simply assume that our newly adapted intervention is going to work, especially given how many well-intentioned behavioral interventions have failed or backfired over the years. To find out if it does, we should conduct an experiment with a relatively small group of participants—large enough to be able to detect our effects statistically but no larger, given the possibility that our intervention might backfire. If the intervention doesn’t work, we could redesign it and try it again. If it does work, we could then take steps to scale it up to larger groups, continuing to test it as we go with experimental designs.
Sadly, the designers of the Comprehensive Soldier Fitness program took a different approach: they delivered their untested intervention to all soldiers in the U.S. Army and members of their families—with no randomly assigned control group. A $125 million program, based on research with adolescents, was put in place with little hope of knowing whether it would work, do nothing, or even cause harm.
But at least we can tell whether soldiers who get the resiliency training show improvements over time in psychological health—right? Indeed, the program is being evaluated in this manner. But as the careful reader of chapter 2 knows, pre-post designs of this type are inadequate tests of whether an intervention works and instead can be quite misleading. For example, early reports suggest that soldiers taking part in CSF are doing no better over time, which some are taking as evidence that the program isn’t working.3 It is possible, though, that an “untreated” group of soldiers would show a drop in resilience over time, as they experience the stress of military life, and that the CSF program is successfully preventing this drop? That is, the failure to include a control group could be masking the fact that CSF is having a preventive effect. But it is also possible that an “untreated” group would show an increase in resilience over time as they adapt to military life, and that the CSF program is having the harmful effect of inhibiting this natural gain in psychological strength. A third possibility, of course, is that a control group would look the same as those who took part in the CSF program, indicating that it is a waste of time and money. Without a randomly assigned control group, we will never know which of these three possibilities is true. As far back as 1935, Richard Cabot Lodge recognized the critical importance of including control groups when testing behavioral interventions when he set in motion the Cambridge-Somerville Youth Study project discussed in chapter 6. It is a shame—one might even say shameful—that the designers of CSF did not heed his wisdom.
Unfortunately, they are not alone. A little-known provision of the Affordable Care Act (“Obamacare”) funds research on how best to deliver health care so that we can find out what works and what doesn’t. This program, called the Patient-Centered Outcomes Research Institute (PCORI; http://www.pcori.org/), has a budget of $10 billion for ten years to study such questions as how best to encourage people to take advantage of preventive-care opportunities. Although this sounds very promising, the new institute has been criticized for failing to fund random-controlled trials of different health care approaches and instead funding “demonstration projects” without control groups. “Does it look like it is working?” asks Dr. Patrick Conway, the director of the institute. “If it does not look like it is working, we can stop.” But without a randomly assigned control group, it is unlikely that Dr. Conway will know what is working.4
But enough of the bad news. There is also reason to be optimistic, thanks to exciting examples of interventions being done the right way. One is the Behavioural Science Unit created by the British government in 2010 to conduct random-controlled tests of ways of improving human behavior and making government more efficient. Using principles of social psychology and behavioral economics, this unit has succeeded in getting businesses to pay their taxes in a more timely manner, increasing donations to charities, and increasing the number of people who register as organ donors—all by conducting experiments to see what works best. I am hopeful that such units will spread to other countries, states, cities, and organizations (including universities).5
Further, more and more researchers are discovering important ways of addressing behavioral and social problems using the story-editing principles discussed in this book and testing these principles experimentally. There are too many new studies to summarize them all, but I will mention a few.
A college education can be a ticket out of poverty for students of limited means. Unfortunately, first-generation students—those who are the first in their family to go to college—often struggle and have high dropout rates. Why is this, and how might we help such students succeed? Well, you might think the problem is that poorer students are less prepared for college. They are less likely to have attended top-notch high schools, less likely to have taken college-preparatory classes, and more likely to have been taught by inexperienced teachers. All of this is true, suggesting that the answer is to advocate for structural changes in American public education or, even more fundamentally, figure out how to reduce poverty once and for all.
These are laudable goals and there is no doubt that such changes are needed. But that is a tall order. Are there other causes of the social-class achievement gap that we could target with the story-editing approach? Psychological factors that might be much easier to change in the short run? In chapter 9 we saw that some exciting story-editing interventions reduced the academic achievement gap between African Americans and white students. Two recent studies have extended this work to first-generation college students. One of these, designed by Judith Harackiewicz and colleagues at the University of Wisconsin–Madison, hypothesized that first-generation students might experience stereotype threat in the same manner as minority students. As discussed in chapter 9, stereotype threat is psychological stress that individuals feel when they think they are at risk for confirming a negative stereotype about a group to which they belong. We saw there that stereotype threat can lower academic performance among African Americans because they (compared to their white counterparts) have the additional worry that they will confirm the stereotype that African Americans are less intelligent. We also saw that a simple “values affirmation” writing exercise, in which middle-school students wrote about nonacademic values that were important to them, reduced stereotype threat among African Americans and improved their grades.
Dr. Harackiewicz and her colleagues showed that a similar story-editing intervention works with first-generation college students. Students in an introductory biology class for biology majors and premedical students were randomly assigned either to a group that completed a values-affirmation writing task twice during the semester or to a control group that did not do the writing exercise. As predicted, this minimal intervention improved grades among the first-generation college students, reducing the gap between their grades and other students’ grades by 50 percent. The writing exercise also increased the percentage of first-generation students who took the next course in the biology sequence from 66 to 86 percent.
Nicole Stephens and her colleagues at Northwestern and Stanford Universities took a different approach. A sample of first-generation college students, in their first semester in college, attended a panel discussion in which older students talked about their college experiences. In the treatment condition, the panelists talked about how their backgrounds influenced their adjustment to college. One of the panelists, for example, was a first-generation college student who said, “Because my parents didn’t go to college, they weren’t always able to provide me the advice I needed.… But there are other people who can provide that advice, and I learned that I needed to rely on my adviser more than other students.” That is, students in the treatment condition heard older students discuss their social-class backgrounds and how they had overcome barriers they encountered. Students in the control condition also attended a panel discussion, but in that case the older students did not discuss their class backgrounds or how that affected their performance. As predicted, first-generation college students in the treatment condition received significantly better grades at the end of the semester than did first-generation students in the control condition. The former students also reported that they fit in better at their university and felt a greater satisfaction with their lives than did the latter students. In short, two independent investigations found relatively simple ways of improving the academic performance of first-generation college students, using story-editing techniques.6
In chapter 3 we reviewed what it means to be “happy” and how people can increase their sense of well-being, particularly through writing exercises. Recent research has corroborated the usefulness of these exercises and revealed new ways in which they can improve our lives. Suppose, for example, that you have an important presentation coming up at your job. You will be making a speech to a group of three hundred people at an annual meeting, talking about a project that you have worked on for the past year. There is a lot at stake: your boss and her boss will be seated in the first row, and this is your chance to impress them and the rest of the firm. Getting nervous yet?
If so, you are not alone. Fear of public speaking is one of the most common sources of anxiety—so common that when research psychologists want to study social anxiety in the laboratory, they often ask people to make an impromptu speech to a bunch of strangers. It is a surefire way of getting people’s bodies revved up (for example, their heart rate increases) and making them feel anxious. But research psychologists are not sadists who enjoy seeing fellow humans reduced to bowls of quivering jelly. The point of studying social anxiety is to find ways of alleviating it, and two recent studies have had remarkable success in doing so using the story-editing approach.
One technique is to say out loud, right before giving a presentation, “I am excited!” As simple (and possibly silly) as this sounds, research by Alison Brooks at the Harvard Business School shows it will help you give a better speech. The reason is that most people’s bodies are revved up right before a speech, with an elevated heart rate, increase in perspiration, and so on, and how people label that arousal is crucial. Most of us label it as nervousness, which is not very helpful. Nor does it work to say to ourselves, “Oh come, chill out,” because it is nearly impossible to convince ourselves that we are calm when our heart is beating so rapidly. Research shows, though, that relabeling our arousal as excitement is effective. People who were randomly assigned to say “I am excited” right before giving a speech, or having to sing a song in front of a stranger, performed better (that is, gave better speeches or sang better) than people who said nothing, said “I am calm,” or said “I am anxious.”7
Another approach to the fear of public speaking involves a writing exercise similar to one we encountered in chapter 3. University of Michigan psychologist Ethan Kross and his colleagues asked people to prepare for a speech by writing about how giving the speech would make them feel. Some (randomly assigned, of course) were asked to write from a third-person perspective, calling themselves by name as much as possible. “In other words,” the instructions read, “if your name was Jane, you would ask yourself, ‘Why will Jane feel this way? What will be the underlying causes and reasons for Jane’s feelings?’ ” The other participants were asked to take a first-person perspective, answering the question, “Why will I feel this way?”
In chapter 3 we saw that analyzing the reasons for a past upsetting event from a third-person perspective helped people feel better about that event (the step-back-and-ask-why exercise). Would it also work to help people deal with an imminent anxiety-provoking event, namely, giving a speech? Indeed it did: taking the third-person perspective helped people process the event more objectively and reappraise it in less threatening terms (for example, “You have done this many times before; come on, Jane, it won’t be that bad.”). People who wrote from this distanced perspective, as opposed to the first-person perspective, reported lower anxiety about the upcoming speech and actually gave better speeches, as judged by objective observers.8
Even better, taking a distanced perspective can improve people’s marriages. Northwestern University psychologist Eli Finkel and his colleagues asked a sample of married couples to think about a significant disagreement they had had in recent months. Then, half of the couples were asked to write about the disagreement “from the perspective of a neutral third party who wants the best for all involved,” to think about the obstacles preventing them from taking a third-party perspective, and to try to take this third-party perspective when future disagreements arose between them and their spouse. This writing exercise—which couples did three times over the course of a year—reduced the couples’ distress about their disagreements and increased their satisfaction with their marriages, compared to couples in the control condition who did not do the writing exercise.9
These and many other story-editing interventions show great promise. The next step is for researchers and policy makers to find ways of scaling them up so that they are widely available to people who might benefit from them. That can be a tricky process because there is no guarantee that an intervention developed in one setting with a particular type of participant will work in the same way in a different setting with a different type of participant (see the Comprehensive Soldier Fitness program above). The solution is to continue to test the intervention as it is scaled up and applied in different settings. Fortunately, social psychologists are doing just that with some of the interventions we have discussed in this book, notably the mind-set interventions designed to help kids change their views of their own intelligence in ways that improve their academic performance (see chapters 4 and 9). Researchers at Stanford University, for example, have developed a web-based program that can be used by schools and universities throughout the United States, whereby students get web-based materials and are randomly assigned to a mind-set intervention condition or control condition (see perts.net). Schools that take part agree to share with the researchers anonymized data on students’ academic progress, allowing the researchers to see how well a given intervention works, with whom, and in what types of schools.
In short, I think the tide is turning in people’s attitudes about social, behavioral, and educational interventions. These interventions can and do work, including story-editing approaches honed in the social-psychological laboratory. As we have seen, relatively simple and cheap interventions have been found to have powerful, long-term benefits. We also know that many popular and expensive programs do not work or even do harm. Policy makers and stakeholders should demand that evidence-based programs be adopted, and be clear what “evidence-based” means—programs tested with the gold standard of the experimental method. Progress often seems slow, but I am convinced that we are headed toward a world where at-risk students get the latest story-editing interventions that have been shown to reduce the achievement gap; where parents of troubled teens know to avoid scared-straight programs and try functional family therapy instead; and yes, where soldiers can count on getting resilience training that has been shown to work. May we get there quickly.