I was sitting in a rocking chair in the nursery of the hospital where my son had just been born—this wonderful, delicate, perfect little human being cradled in my arms. I could barely contain my joy as I waited to rejoin my wife, who was in the operating room recovering from the C-section birth. I guess my son wasn’t quite as thrilled, because after a few minutes he began to fuss and cry. “Uh-oh,” I thought. “What do I do now?” Just then a nurse walked by and handed me a pacifier, which my son readily accepted. He was happily sucking away when another nurse walked by, glanced at me disapprovingly, and said, “So—you’ve made the choice to use a pacifier, have you?”

“Oh my God!” I thought. “I’ve already made a critical parenting decision without even knowing it! You mean some parents don’t give their babies pacifiers? They didn’t cover this in our Lamaze class. What does the research say? Have I doomed my son to a life of oral dependency? What if he refuses to breast-feed? What am I going to tell my wife?”

As I learned, parents have to make a constant stream of decisions from the moment their babies are born. I also learned that it is easy to exaggerate the importance of these decisions. I don’t think it mattered that I plopped a pacifier in my son’s mouth; he easily learned to nurse, and my wife and I couldn’t be prouder of the adult he has grown up to be (we are equally proud of his younger sister). Surely whether we give our children pacifiers, wrap them in disposable or cloth diapers, or put them down to sleep at 7:00 p.m. or 8:00 p.m. does not determine whether they live fruitful, happy lives or become juvenile delinquents. Some researchers even argue that parents don’t make much difference at all. According to this viewpoint, our kids’ futures are determined largely by their genes and their peers, and as long as we don’t abuse them, they’ll do just fine.¹

But I think the pendulum has swung too far over to the “parents don’t matter” camp. There is plenty of research showing that parents exert a huge influence on their kids’ cognitive and emotional development. Consider, for example, the fact that children born into families of high socioeconomic status (SES) have higher IQs, on average, than children born into families of low socioeconomic status. Some researchers argue that this is due to genetic differences in intelligence, namely, the possibility that people with high IQs are more likely to “rise to the top” economically, to marry people with high IQs, and then to pass these “high IQ” genes on to their children. The social psychologist Richard Nisbett, however, has argued persuasively against this strict “hereditarian” view of intelligence, pointing out that one reason for the class difference in IQ is different child-rearing practices. Higher-SES parents talk a lot more to their children than do lower-SES parents—on the order of two thousand words per hour versus thirteen hundred words per hour. By the age of three, children of higher-SES parents know 50 percent more words than do the children of lower-SES parents. And it’s not just the volume of words: higher-SES parents talk to their kids in a way that encourages them to question, analyze, and think about the physical and social world, which sharpens their thinking and prepares them for school.²

There is also plenty of research showing that parents exert a huge influence on their kids’ social and emotional development. Effective parents are attentive to their children’s needs and form close attachments with them, which allows the children to internalize healthy narratives about relationships—narratives that increase the likelihood that they will form trusting, loving relationships with others throughout their lives. Effective parents also know how to shape their children’s behavior in ways that encourage them to internalize desired values and attitudes. But how do good parents learn how to do this?

Many mothers and fathers trust their instincts and hope for the best. Others rely on the wisdom of their elders, including their own parents and grandparents. Still others seek advice from the dozens of parenting books that are available. Indeed, the number of books advising parents on how to raise their kids is rivaled only by the tomes in the self-help section of bookstores. There are books on “scream-free” parenting, “simplicity” parenting, and “playful” parenting, as well as Parenting by the Book: Biblical Wisdom for Raising Your Child and The Complete Idiot’s Guide to Fatherhood. There are countless books on how to discipline kids, including Parenting with Love and Logic, Making Children Mind Without Losing Yours, and Parenting Your Out-of-Control Teenager.

For many years, one of the most popular parenting books was Benjamin Spock’s Baby and Child Care, originally published in 1946. Although Dr. Spock died in 1998, the book is still in print and has sold millions of copies. One reason the book has been so popular is that it conveys the message that parents should relax, trust their instincts, not worry so much about rigid feeding schedules, and by all means pick up a crying baby without fear of spoiling her or him. This message was novel in the 1940s, when parents were advised to stick to strict schedules and to avoid spoiling their children. Countless parents of baby boomers—mine included—had dog-eared copies of Baby and Child Care that were trusted sources of advice. (When I was in college, I had the opportunity to hear Dr. Spock give a talk to a large audience of students. He began by opening his arms wide and proclaiming, “My children!”)

But Dr. Spock also said that trusting our instincts isn’t enough: “There may have been a time when parents knew exactly how to bring up their children or thought they did. But for most of us now there simply aren’t clear-cut rules to follow. We have more options, but there are also more choices that have to be made… Everywhere you turn, there are experts telling you what to do. The problem is, they often don’t agree with each other!”³

This conundrum is by now familiar to readers of this book—grandmothers and other experts give us advice, but much of it conflicts and little of it has been adequately tested. Many parenting practices can’t be tested experimentally, of course; no scientist in his or her right mind would randomly assign families to a condition in which they were asked to hit their children or to a control condition in which they were not. But many other techniques can be tested, and when they have been, the answers are often surprising.

In 1997, for example, a company named Baby Einstein released videos for young children that were designed, as the name implies, to enhance the intelligence of the toddlers who watched them. The Baby Einstein website says that the videos “are designed as interactive tools for parents to use with their babies… allowing a parent to have two free hands while enjoying and experiencing the video with their little one.” Many parents, however, interpreted this to mean that their child would benefit from watching the video by him- or herself—kind of like a PBS documentary for toddlers. Who wouldn’t feel good letting their child watch this video while they got some housework done or made a few phone calls? Several companies now market videos for kids and millions of children have watched them. As a result, these kids are exceedingly smart, right?⁴

Unfortunately not, because it turns out that toddlers do not learn very well from disembodied voice-overs on videos. Consider a clever study done by colleagues of mine at the University of Virginia, in which parents of twelve-to-eighteen-month-old infants were randomly assigned to one of four conditions. In one condition, the infants watched a commercial educational video by themselves, at least five times a week for four weeks. The video was designed to increase toddlers’ vocabulary by introducing them to new words. The parents were in the room with the infants but did not watch the video with them. In a second condition, the infants watched the video for the same amount of time, but a parent watched with them. In a third condition, the infants never saw the video. Instead, the parents were given a list of the vocabulary words featured in the video and were instructed to teach their kids the words in whatever way seemed natural to them. The fourth condition was a control group in which the parents neither showed the video to their children nor taught them the words from it.

At the end of the four weeks, the researchers tested the children’s knowledge of the words featured in the video. The testers, who did not know what condition the children had been in, showed the kids a pair of objects and asked the kids to point to the one the testers named (e.g., “Can you show me the clock?”). If the child pointed to the correct object, and “clock” was one of the words featured in the video, then bingo, the child got credit for having learned that word. Which group of kids learned the most words?

It turns out that there is no substitute for good old parent-child interaction: the kids whose parents tried to teach them the words, without showing them the video, learned the most words. The kids who saw the video, with or without their parents watching with them, didn’t learn the words any better than did the kids in the control condition. Interestingly, many of the parents thought their children had learned a lot from the videos. In fact, the more the parents liked the videos, the more they thought their children had learned from them. Unfortunately, our instincts about what works best for our kids are not always correct.⁵

Okay, maybe we’re not experts in language development. But surely we have good instincts about how to nurture our kids and make them feel good about themselves: all we need to do is to shower them with love and praise, right? Well, yes and no. It is important to form close emotional bonds with our children, but showering them with praise is not always the best thing to do.

We will see why in this chapter, which focuses on how parents can shape the narratives of their kids in healthy ways—namely, with story-editing interventions that have been tested scientifically. This is not a chapter on general parenting advice: there won’t be any tips on how to cure diaper rash, get your child to sleep through the night, or help him or her deal with bullies. These are important topics, to be sure. But perhaps the most important thing parents do is redirect their children’s views of themselves and the world—the personal interpretations that are the topic of this book and the blueprints by which children will live their lives. I’ll begin with how parents can get their kids to develop a sense of autonomy and purpose, which, as we saw in the previous chapter, is a key element of a healthy adult narrative.

AUTONOMY AND PURPOSE

A few years ago, when I was chair of the psychology department at the University of Virginia, I experienced the phenomenon of “helicopter parents” firsthand. As I entered the building one day, the receptionist took me aside. “There’s a man here to see you,” she said. “He says he flew down from New York this morning to talk to you about his daughter.” There he was, sitting impatiently with his daughter, a fourth-year psychology major. When I invited them into my office, the man wasted no time getting started: “I think it is outrageous,” he said, “that my daughter can’t get into the seminar of her choice.”

By way of background: all of our psychology majors are required to take at least one small seminar, usually in their fourth year. But psychology is a popular major, and the seminars are popular courses, and competition for them is fierce. Not all students get their first choice. Such was the case with the student in my office; she had wanted to take a particular seminar but it had filled up quickly and there were no more spaces available. “I pay a big chunk of change in out-of-state tuition,” her father continued. “And I think my daughter should be able to take the course she wants.”

I explained that we do the best we can to fulfill all students’ requests but that we can’t give everyone their first choice. But the man wasn’t satisfied: after he left my office he marched over to the main administration building and demanded to see the president of the university. Officials there politely told him that there was nothing they could do, and, alas, his daughter had to take her second choice of seminar.

Incidents such as this are not uncommon on college campuses. A few years ago, Judith Shapiro, then the president of Barnard College, wrote an op-ed piece in The New York Times asking parents to back off. She gave these examples of helicopter parenting on her campus: one father called his daughter’s career counselor to get a list of prospective employers so that he could contact them and bring them up to date on her sterling qualities. A mother accompanied her daughter to a meeting with a dean so that she could help with her daughter’s “independent” research project. When it came time for one young woman to apply to colleges, her father took a year off from his job to shepherd her through the process.⁶

Clearly, these parents were just trying to help. But what message were they conveying to their kids? The father who flew down to see me was a take-charge kind of guy, used to getting what he wanted, and he probably thought he was teaching his daughter a valuable lesson—namely, that it is a dog-eat-dog world and you have to push to get what you want. But I wonder if she learned something else entirely: “Dad will step in when I need help, so there’s no need for me to figure out things for myself.”

In the previous chapter, we saw that a key to happiness is having a sense of purpose and autonomy—to be able to set goals for oneself and decide how to pursue them. One of the best things we can do for our children is to help them develop this sense of autonomy—even if that means that they experience a few bumps along the way, such as not getting their first choice of seminar. Hovering over our children from birth through the college years, making every choice for them, is hardly the way to accomplish this.

But let’s look back at more formative ages, when children’s narratives about autonomy and control are developing. What can parents do when their kids are taking tentative steps out of the nest and becoming more independent, as they do during the important middle-school years? I’ll begin with what parents shouldn’t do, namely, adopt a highly domineering style in which they attempt to control every step their children take. Several studies have found that an intrusive, controlling parenting style is associated with low autonomy in children and low overall well-being.

In one study, students who were starting seventh grade, in the United States and China, were asked how decisions were typically made in their families—namely, whether they were most likely to decide things for themselves or whether their parents decided for them. The questions covered several areas that are important to seventh graders, such as who they make friends with, how much time they spend with their friends, what time they go to bed, how much time they spend studying, and what they do in their free time. In both countries, some kids reported that they had considerable latitude in making these choices, some reported that their parents decided for them in most of the areas, and some reported that it was in between—i.e., that they decided what to do in certain areas and their parents decided for them in others. The researchers then tracked the kids for the next two years and measured their emotional well-being. The main result was that the more controlling the parents were (at least in the eyes of their kids), the worse off their children were emotionally over the next two years. In other words, if kids reported at the beginning of the seventh grade that their parents made most of their decisions for them, they were more likely to show a decrease in emotional functioning over the next two years.

Interestingly, this pattern of results was found among both the American and Chinese adolescents. The issue of parental control in different cultures has been a matter of debate: some have argued that a controlling style is the norm in Asian countries, and that because this parenting style is expected and accepted there it is less harmful. In the study just mentioned, however, no such difference emerged: both the Chinese and American kids who had controlling parents were likely to suffer emotionally in subsequent years. I wouldn’t say that this issue is fully settled; researchers continue to look at the role of parenting styles in different cultures. Further, we need to be careful in interpreting correlational studies such as this one, in which researchers measure the association of two variables rather than conducting an experiment. The parents who were more controlling might have differed in any number of other ways from parents who were less controlling (see the discussion of unmeasured third variables in chapter 2). At a minimum, though, the results show that controlling parenting styles are predictive of problems down the road for adolescents, at least among kids from the United States and China.⁷

Other studies have shown the dangers of another type of controlling parental style—namely, doling out love and affection only when our kids do what we want them to do. In a study of ninth graders in Israel, researchers examined the extent to which parents use conditional negative regard, a technique in which they attempt to control their kids’ behavior by withholding their affection when their kids do something wrong. The ninth graders reported their level of agreement with statements such as “If I do poorly in school, my mother will ignore me for a while” and “If I show my fear, my father will express less warmth toward me for a while.” Those who agreed with statements such as these were classified as having parents who use conditional negative regard. The downside of this strategy is that kids come to view their parents as punitive and overly controlling and tend not to internalize the goal to do well on their own. Indeed, kids whose parents used this style reported greater resentment toward their parents, and their teachers rated them as less engaged in school, as compared to kids whose parents did not use this style.

“Well,” you might think, “I would never punish my kids by ignoring them or withholding love. Rewards work better than punishments, so better to shower my kids with love when they do well. What better way to teach them good habits?” If that’s what you do, you are using conditional positive regard, a technique in which parents respond to their kids’ good behavior with love and affection. In the Israeli study, kids who endorsed statements such as “I feel that when I’m studying hard, my father appreciates me much more than usual” and “If I am afraid but do not express my fear, my mother will express more love for me” were classified as having parents who use conditional positive regard.

This strategy is better than withholding affection, but the danger is that kids learn the lesson that they can earn their parents’ love only when they do what their parents want them to do. In the academic realm, for example, kids might focus on getting good grades to please their parents but fail to develop an intrinsic interest in their studies. This is what the study of Israeli kids found: the kids whose parents used conditional positive regard were more interested in grades than in the academic material (as reported by their teachers).

Well, if it’s not advisable to withhold affection or make it contingent on our children’s successes, what are parents to do? Actually, there is a better approach, called autonomy support, in which parents try to see things from their children’s perspective, helping them understand the value of the different alternatives with which they are faced and convey a sense that they, the children, are ultimately the ones choosing which path to follow. The idea is to gently guide one’s kids in the right direction while giving them the sense that they are making the choices themselves. To measure this parenting style, the researchers in the Israeli study asked the ninth graders what typically happens when they disagree with their parents; e.g., when “I think that my investment in school is adequate, but my father thinks it is not.” Teens who reported that their parents understand them, but also that their parents provide a rationale for their (the parents’) perspective, were classified as receiving a high amount of autonomy support. As the researchers predicted, these kids were better off than the kids whose parents used the other two parenting styles. They reported the highest amount of control over their feelings and behavior, were better able to regulate their emotions, and their teachers reported that they had the most intrinsic interest in academics. Another study (done in the United States) assessed parenting styles by interviewing the mothers and fathers rather than relying on how their kids saw it. Consistent with the Israeli study, the parents who scored high in autonomy support were more likely to have kids who were well behaved in school and got high grades.⁸

“I’D BE A LITTLE ANNOYED…”

So, autonomy support is preferred to controlling parenting styles. But as any parent knows, it isn’t always possible to reason with one’s child and let her make her own decisions. Especially when kids are young, we have to set limits and make it clear what the consequences will be if they do something wrong. And when our kids do act out, we need to decide whether to punish them and, if so, how. If our five-year-old is hitting his little sister with a toy train, or trying to pry the safety cover off an electrical outlet with a fork, reasoning with him only goes so far.

Many parents believe that an effective strategy in such situations is to give their kids a swat on the rear. In a survey conducted by the nonprofit organization Zero to Three, 61 percent of parents of young children endorsed spanking as a regular form of punishment, with 57 percent of the parents reporting that spanking helps children develop a better sense of self-control. Another survey found that 94 percent of parents have spanked their child at least once by the time the child is four years old. Spanking in the home is legal in all U.S. states, and some states allow corporal punishment in schools.⁹

The American Academy of Pediatrics and the National Association of Social Workers disagree, recommending against the use of corporal punishment by parents. Many countries have outlawed corporal punishment in public and private, including Sweden, Norway, Germany, Austria, and Hungary. What does the research say? Corporal punishment does succeed in getting kids to stop doing whatever it was that brought about the spanking. But it is also associated with a number of negative outcomes, including an increase in aggressive and antisocial behavior, a lower-quality relationship with parents, poor mental health, and an increased likelihood that the child will grow into an adult who abuses his or her child or spouse. Now, I should hasten to add that these are correlational findings, which makes it impossible to determine whether corporal punishment causes these negative outcomes. We can’t rule out the possibility that there is some third variable, such as family pathology, that increases the likelihood of both corporal punishment and negative outcomes.

It is worth noting, though, that kids who have been spanked or slapped are bad at what psychologists call “moral internalization.” Rather than learning that “I shouldn’t hit my little sister because that’s the wrong thing to do,” kids learn that “I shouldn’t hit my little sister because Mom will slap me if I do.” This means that the child won’t hit his little sister when his mom is around, but what happens when she is out of the room or at work? Watch out—the kid who behaves only to avoid Mom’s wrath is now free to unload on his sister. Further, his mother has taught him that violence is a reasonable way to try to control someone else’s behavior, so why shouldn’t he use it himself? Studies have shown that kids are excellent imitators of the techniques other people use to get what they want, including aggression. In short, many parents who use corporal punishment focus too much on controlling their children’s behavior and too little on what they are doing to their kids’ narratives. Ultimately, we want our kids to internalize appropriate values and attitudes, rather than obeying in order to avoid being punished.¹⁰

Issues such as these concerning discipline are some of the most difficult ones parents face and are the focus of many parenting books. Dr. Spock stated the dilemma well. On the one hand, he said, “You can create a harsh system of rewards and punishments so that, like good little robots, your children will behave perfectly most of the time. But what would be the effect on the child’s spirit, on his sense of self-worth, on his personal happiness, on his feelings toward others?” On the other hand, he continued, “Imagine a child whose every whim is indulged, whose every action, good or bad, is lavishly praised. Such a child might have a certain measure of happiness, but most people wouldn’t want to be within ten feet of him.” Most of us would agree that there should be a happy medium between these two extremes, but exactly how do we accomplish this? The answer is to use the story-editing approach, which isn’t in any of the many parenting books I’ve perused, including Dr. Spock’s.¹¹

To understand how parents can use this approach, consider this example: you and your eleven-year-old daughter are making some cookies for a bake sale at her elementary school. The two of you have a good time mixing the dough, spooning it onto a baking sheet, and smelling the wonderful aroma that wafts from the oven. When the batch is done, you and your daughter treat yourselves to a cookie with a glass of milk. “Can’t I have another one?” your daughter pleads. “No,” you say. “It would spoil your dinner, and besides, if we eat more we won’t have enough for the bake sale. Why don’t you start your homework while I go pay some bills?” Your daughter sits there sulkily, casting none-too-subtle looks at the cookies cooling on a rack on the counter.

You are about to leave your daughter in a highly tempting situation; she could easily sneak a cookie while you are out of the room. Say you’ve been having a little trouble getting her to obey your wishes of late; just the other night she snuck out of bed and got on her computer after you told her to turn out the light, and a couple of days before that she spent the evening texting her friends instead of doing her homework. You decide it’s time to crack down. “Look,” you say. “If you take another cookie while I’m paying the bills, you’ll be in big trouble—I’ll take your cell phone away for a week and let you use your computer only to do your homework—no more Facebook.” “That ought to do it,” you think. “It’s high time she learned to obey the rules.” Just to be sure, you count the number of cookies on the rack, and when you return later you are relieved to see that they are all still there.

It might seem like Mission Accomplished. Your daughter didn’t spoil her dinner and learned a valuable lesson about avoiding temptation and doing what she was told. Or did she? Was your threat commensurate with the potential transgression, or did you go a little overboard? Even if you did, is that so bad? After all, you succeeded in getting your daughter to obey the rules. But again, the issue isn’t just one of compliance but of internalization. Have you directed her narrative in a good or bad direction?

The answer is suggested by a two-session experiment that was done with children in the second grade. At the first session, a researcher left a child alone in a room in a tempting situation, not unlike the one with your daughter and the cookie. In this case, there were a bunch of toys in the room and the researcher told the child not to play with one of the best ones. He then varied the severity of the threat he delivered to the child. In one condition he was harsh, not unlike the mother threatening to take away her daughter’s cell phone and ban her from Facebook: “Now, while I’m gone,” he said, “you can play with all of the toys except the robot. It would be wrong to play with the robot, and if you play with it, I would be very upset and very angry with you.” In another condition, the researcher dialed back his comments, giving only a mild threat. Instead of saying he would be “very upset and very angry,” he said he would be “a little bit annoyed” if the child played with the forbidden toy. Then he left the room for a few minutes.

A key result is that both threats were successful: no child actually played with the forbidden toy while the researcher was gone. From a story-editing perspective, however, the critical question is how the children explained to themselves why they weren’t playing with the toy. In the severe threat condition, the kids were likely to attribute their behavior to the threat. “I’m not sure why that guy was so bent out of shape about the robot,” they probably thought, “but I’m sure not going to find out by playing with it. He’d be really mad at me.” In the mild threat condition, however, the children could not as easily use the threat to explain why they didn’t play with the robot. After all, the researcher only said he’d be a little annoyed, so what’s the big deal?

The fascinating thing about the mild threat is that it was strong enough to prevent the kids from playing with the toy, but not so strong as to convince them that it was the reason they weren’t playing with it. Instead, the kids inferred that they must not like that robot as much as they thought. That is, they assumed that they weren’t playing with it because they didn’t really like it very much—not because of the researcher’s threat. We know this because at the end of the experimental session, the kids in the mild threat condition reported that they liked the robot significantly less than did kids in the severe threat condition.

The kids in the mild threat condition didn’t stop with devaluing the robot. They found another reason to explain their good behavior, namely, that they must be especially honest kids who are good at avoiding temptation. After all, they could have played with the robot—the researcher would only have been a little annoyed—so why didn’t they? In addition to deciding that the robot wasn’t all that great, they concluded that they were good, honest kids who obey the rules.

The way we know this is that the researchers brought the kids back to the lab three weeks later and put them in another tempting situation. A different researcher, who didn’t know which condition the children had been assigned to in the first session, asked the kids to play an electronic bowling game in which they could win prizes such as a doll, flashlight, or magnifying glass if they got a score of thirty-five or above. Then the researcher left the room and told each child to keep score him- or herself until the bowling game turned itself off. Lo and behold, the game ended when the child had earned only thirty-three points, just shy of a winning score. The researcher returned and asked the children how many points they had gotten. Were they honest or did they add a couple of points to their score in order to win a prize?

As the researchers expected, it depended on what condition the kids had been randomly assigned to in the initial session three weeks earlier. Very few kids who had received the mild threat cheated on the bowling game, whereas a substantial number of kids who received the severe threat did. This supports the idea that kids in the mild threat condition saw their good behavior in session 1 as a sign that they were honest kids who obeyed their elders, whereas kids in the severe threat condition simply said to themselves, “Hey, I didn’t play with the robot because that guy would be really upset—not because I’m a particularly honest kid.”

As discussed earlier, these kinds of changes in people’s narratives are not fully conscious. The kids in the toy study did not sit there and scratch their heads, verbalizing to themselves the reasons they weren’t playing with the robot. Instead, the mild threat seems to have led to a change in the kids’ implicit self-narratives, which made them behave honestly when left alone with the bowling game three weeks later.¹²

The toy study and others like it have clear implications for what our hypothetical parent should have done in the cookie example. The threats to take the daughter’s cell phone away and ban her from Facebook were probably too severe. Yes, these threats succeeded in getting her to obey, but in all likelihood they were so strong that they prevented any internalization. The daughter probably resisted temptation in order to avoid the punishment, rather than convincing herself that she is an honest kid who believes in doing the right thing. It would have been better to dial back the threat, keeping it strong enough to work but weak enough so that the daughter attributed her compliance to her own honesty. For example, the parent might have said, “I’d be annoyed if you snuck a cookie,” though he or she would need to determine the exact level of threat that is both mild and effective.

Some parents might question the use of threats at all. Isn’t it better to shower our kids with praise and to reward them for good behavior than to threaten them? Instead of telling our daughter that we would be annoyed if she disobeyed, for example, might it have been better to reward her for obeying? It depends on how strong those rewards are and how they are used, as we will now see.

SHOW ME THE MONEY

Consider another parenting dilemma: you would like your ten-year-old son to read in his free time instead of playing video games and watching television. You take him to the bookstore and buy him some age-appropriate books, and that works to some extent. He enjoys How to Eat Fried Worms by Thomas Rockwell, but he goes back to playing video games and the other books collect dust on his nightstand. Feeling a little exasperated, you decide on a new approach: rewarding your son for reading. You tell him that for every ten books he reads, you will buy him a new video game of his choice. And it works! Your son devours The Boys from Brooklyn, Lawn Boy, and the other books you bought him. Mission accomplished, right?

Well, maybe not. There is no doubt that rewarding kids can change their behavior in dramatic ways. Indeed, parents use incentives all the time. “If you eat two more bites of your carrots you can have dessert,” moms and dads say, or, “If you practice the piano I’ll take you out for ice cream.” As long as the incentives are strong enough, then bingo, the carrots disappear and the piano gets played. Reward programs for kids have become institutionalized, such as the long-standing Book It! program sponsored by the Pizza Hut restaurant chain. According to the Book It! website (http://www.bookitprogram.com/), the goal of the program is to “motivate children to read by rewarding their reading accomplishments with praise, recognition, and fun!” When kids read a number of books set by their teacher (twenty books in my kids’ elementary school), they receive a coupon for a free personal pan pizza. “The restaurant manager and team congratulate every child for meeting the monthly reading goal and reward them with a free, one-topping Personal Pan Pizza, Book It! card, and backpack clip,” the website says—noting that more than ten million students participate in the program each year.

Programs such as these follow the most basic law of economics, namely, that people do what they do because of incentives. How can you go wrong by offering kids incentives to do what you want them to do, such as practicing the piano or reading more books? But again, it depends on whether our goal is compliance or internalization. If we want kids to read more, then rewarding them can work—as long as the incentives continue to be available. Rewards can produce compliance, just as threats of punishment can. But as in our example of teaching kids to be honest, we want our kids to internalize desired attitudes and values, in this case a love of reading. After all, we can’t reward them for reading a book for the rest of their lives. What happens when summer comes and the Pizza Hut program ends?

“Well,” an economist might answer, “we hope that kids will learn to like the activities after doing them for a while, so that they keep doing them after the rewards stop. But if they don’t, the worst that can happen is that when the rewards aren’t available anymore, kids go back to their original level of reading. No harm, no foul.”

But it turns out that there is a possible foul: rewards can undermine kids’ intrinsic interest in an activity by convincing them that they are doing it for the reward and not because the task is enjoyable in and of itself. This “overjustification” effect has been found in dozens of experiments. One of my favorites took place in an elementary school with fourth and fifth graders. The teachers wanted to increase students’ interest in math, and they figured that a reward program was worth a try. First, the teachers introduced four new math games during class and observed how much the students played with each one during this baseline period. The answer was some but not much. Next, the teachers introduced a reward program. For every three hours the kids spent on the math games, they earned credits that they could use to get trophies and other prizes. These rewards were popular and effective: the time the kids spent on the math activities shot up.

So far so good. But what happened after the reward program ended? To find out, the teachers announced that, to be fair to the rest of the students in the school, they had to terminate the reward program, but that the math games would still be available for kids to play with, just as before. Now that there was nothing in it for the kids, their interest in the games plummeted to a level below where it had been during the pre-reward baseline period. In other words, it didn’t just go back to where it had been before the reward program was instituted, as an economist might have predicted—the kids were now less interested in the games than they were when the program started. (The researchers determined, by comparing these results to those in a control condition, that it was the rewards that undermined kids’ interest in the math games, not boredom with the games as time went by.) This study and others like it show that giving people strong rewards for an activity convinces them that they are “doing it for the money” and not because they have any intrinsic interest in the activities, and that in fact rewards can undermine interest that was there at the outset.¹³

Does this mean that parents and educators should abandon all rewards, throwing out their drawers full of gold stickers, prizes, and lollipops? Not at all. First, there isn’t much risk of undermining kids’ intrinsic interest when it is zero to begin with. If your child shows absolutely no interest in reading—he hasn’t touched a book in months—then a small reward to get him started can’t do any harm. Parents are in a position to know what their kids like and dislike, and thus to know when to use rewards. The problem with programs such as Book It! is that they are implemented throughout entire schools, where some of the kids will already be interested in reading and thus will be at risk for having that interest go down as a result of being rewarded with pizzas for something they already like to do. (In chapter 8, we will discuss programs that have tried—without much success—to close the achievement gap by paying disadvantaged kids to do well in school.)

Second, note that in the math game experiment the kids earned rewards simply for playing with the math games, regardless of how well they performed. Many rewards, such as good grades in school, are contingent on doing well—people get them only if they perform at a high level. Rewards of this type are less likely to undermine intrinsic interest because they convey to people that they are good at the activity being rewarded. But even rewards like grades can be problematic, to the extent that they focus people’s attention on the incentives and raise anxiety about being evaluated.

Another approach is to use rewards but downplay their power and significance. Now, this might seem odd: if the goal is to get kids to do what we want, why wouldn’t we use powerful incentives? But, as we saw, the problem with strong rewards is that they can undermine the very interest we are trying to foster. What if we adopted the same approach that I recommended for the use of threats, namely, dialing them back so that they are just strong enough to be effective? The goal would be to make them strong enough to get kids to perform the desired activity but not so strong that the kids latch onto the rewards as the sole explanation for their behavior.

A study with preschool children in Canada did just that. Four-year-old children were taken to a room and told that they could play with a drum, which is a highly popular activity with kids that age. In the salient reward condition, the researcher told the kids that they would receive a prize if they played with the drum, and the prize was introduced in such a way that kids could hardly think about anything else: the researcher said that the prize was under a box that was directly in front of the child, and told the kids that they couldn’t lift up the box to see what the prize was until after the drum-playing time was over. Think about what this would be like for four-year-olds—they are sitting there playing the drum and staring at the box, constantly wondering what the prize is, their excitement mounting as time goes by. In the nonsalient reward condition, the kids were also told that they would get a prize after they played the drum but the researcher didn’t say anything more about it; there was no box with a prize under it to capture kids’ attention. Finally, there was a control condition in which the kids played the drum with no mention of any rewards.

The researchers measured the children’s subsequent interest in the drum in a variety of ways, including asking them to name their favorite toy in the room and observing how much time they spent playing with the drum during a “free time” period both right after the initial session and several weeks later. On each of these measures, the kids in the salient reward condition showed a drop in intrinsic interest relative to the kids in the control condition. For example, they were less likely to name the drum as their favorite toy and they spent less time playing with it during the free play times. “I did it to get the prize,” they seemed to have said to themselves, “and now that no prize is being offered, I’ll play with the tinker toys instead of the drum, thank you very much.” What about the kids in the nonsalient reward condition? They, too, got a prize for playing with the drum, but because their attention was not as focused on the reward, they seem not to have assumed that that was the only reason they played with it. In fact, compared to kids in the control condition, their subsequent interest in the drum was slightly higher—the prize added some excitement to playing with the drum but was not viewed as the sole reason for playing with it.¹⁴

The moral of all of this is what the social psychologist Mark Lepper calls the minimal sufficiency principle. If the goal is to get kids to internalize desired attitudes and values, then parents should use threats and rewards that are minimally sufficient to get kids to do the desired behaviors, but not so strong that the kids view the threats or rewards as the reason they are acting that way. Minimally sufficient threats and rewards are an effective story-editing technique, convincing kids that they are doing the right thing because they believe in doing the right thing.

Although this is great in theory, I have to say that in practice it isn’t always easy to pull off. The problem is that we don’t always know in advance what a minimally sufficient threat or reward will be. How much of a reward is enough to get a preschooler to practice the piano? How much of a threat is sufficient to get a teenager to avoid smoking? Further, there is a danger to erring on the low side. If the threats or rewards are too weak, our kids won’t do what we want them to do. This can backfire, because the kids might become even more enamored of the undesirable behavior. If teens decide to smoke even though their parents would be annoyed, for example, they are likely to infer that smoking must be a really attractive thing. Similarly, if they decide not to practice the piano, even though they could have gotten some candy for doing so, they are likely to infer that piano playing must be really boring. The best approach is to start with threats or rewards that are strong enough to get kids to do the desired behavior, but then, on future occasions, dial them back a bit, making it harder for kids to attribute their actions to the threat or reward.

Another possible objection to the minimal sufficiency principle is that it might seem devious or controlling. Is it right to “trick” our kids into thinking they are doing something because they want to? Parents will have to decide this one for themselves, though in my view it is our job to get our kids to internalize desired attitudes and values. The irony is that this can be done more effectively with mild threats and small rewards than with severe threats and large rewards. Isn’t it more controlling to take an overly authoritarian stance, one that will ultimately lead to less internalization of desired attitudes and values?

“YOU ARE SUCH A HELPFUL CHILD…”

There is a more straightforward way to get kids to internalize desired values, one that uses the story-prompting approach: simply providing kids with the right label for their behavior. Let’s go back to the cookies example. Say you tried the minimal sufficiency principle, telling your daughter that you would be a little annoyed if she snatched a cookie while you were out of the room. But it turns out that your threat wasn’t strong enough. When you return there is one less cookie on the plate and some telltale crumbs on the edge of your daughter’s mouth. She sees the look of disappointment on your face and clearly feels bad for disobeying you. But what lesson has she learned, exactly? If she interprets her feelings as guilt, she might internalize the lesson that she should avoid future transgressions—after all, she doesn’t want to disappoint you again. But if she says to herself, “Darn, I can’t believe I got caught,” she has learned a very different lesson—that next time she should do a better job of rearranging the cookies on the plate and brushing the crumbs off her mouth.

It turns out that adults can help kids label feelings such as these. Consider a rather devious experiment conducted with second graders. A researcher took each kid to a room in which a toy race car was moving slowly around an oval track and asked him or her to make sure that the car didn’t jump the track. He showed the child how to stop the car if it started going too fast, and then left the room. Unbeknownst to the child, there was an accomplice in another room who was controlling the race car, and, lo and behold, when the child’s attention wandered to other toys, he sped up the car and made it jump the track.

Soon thereafter, the first researcher returned, looked at the errant car, and noticed with dismay that it was broken. Now imagine you are the second grader. The man had just asked you to do something but you really screwed up. You are probably feeling pretty bad. But how exactly do you interpret these feelings? Do you feel guilty or annoyed over getting caught? For the kids in the race-car study, it depended on how the researcher labeled their feelings. Some of the children were randomly assigned to a guilt condition, in which the researcher communicated that they probably felt guilty for doing something wrong. “I bet you feel a little bad now that the car fell off,” he said. “I’ve seen other kids feel bad when they weren’t able to do exactly what they were supposed to do.” Other kids were randomly assigned to what we might call a dang, I can’t believe I got caught condition. Here, the researcher altered his comments slightly, labeling the kids’ feelings as disappointment over being discovered. “I bet you feel a little bad now that I found out the car fell off,” he said. “I’ve seen other kids feel bad when someone found out they weren’t able to do exactly what they were supposed to do.”

To see if his comments had sunk in, the researcher gave the kids a new race car, showed them how to put it back on the track if it came off, and asked them to keep an eye on it while it went around the track. As he left, he emphasized that no one would come into the room for several minutes. Then the researchers surreptitiously measured how much time the kids spent focusing on the car while they were alone, making sure that it didn’t jump the track, and how much time they spent looking at the other toys in the room.

It turned out that the label the researcher had provided did sink in. The kids who had been encouraged to attribute their earlier transgression to guilt spent significantly more time watching the race car than did the kids who had been encouraged to attribute their behavior to disappointment over being caught. The former kids seemed to be thinking, “I’d better keep an eye on the car, because I sure would feel guilty if I screwed up again, even if no one finds out.” But the latter kids seemed to be thinking, “So what if the car jumps the track—no big deal. I can just put it back on the track before that guy comes back.” If you want your kids to internalize the motivation to be good, then when they screw up, label their feelings as guilt.

What about when kids do the right thing? Here again, the way in which we label the behavior can make a difference, as shown in a study that tried to get fifth graders to be good citizens and not throw their trash on the floor. In one randomly chosen classroom, over a period of eight days, the teacher and principal consistently labeled the kids as people who did not litter. On one day, for example, the teacher pointed to a piece of trash that had been left by students from another class and noted, “Our class is clean and would not do that.” On another day, the teacher put up a poster in which a cartoon character was saying, “We are Andersen’s Litter-Conscious Class.” On yet another day, the principal came into the class and commented on how clean the room was and how orderly the students were. In another randomly chosen classroom, the teacher and principal took a different tack. Instead of labeling the kids as nonlitterers, they used standard persuasive appeals to convince them to keep the room clean. On one day, the teacher lectured the students on ecology and the dangers of littering, and on another she told the students how important it was to put candy wrappers in the trash instead of throwing them on the floor. On still another, the principal visited the classroom and told the students how important it was to keep the classroom tidy. Finally, a third randomly chosen classroom served as a control group and didn’t receive any communications about littering.

After the intervention, the researchers observed how much the kids in each classroom littered. For example, two weeks later, the teachers gave each student a puzzle that was wrapped in a small disposable container. The researchers kept track of the number of kids who threw the containers on the floor and the number of kids who put the containers in the trash. It turned out that kids in the first, “labeling” condition internalized the message the most; they were the least likely to litter on the follow-up tests, even when they didn’t know that their behavior was being monitored. For example, only about 15 percent of the kids in this classroom threw their puzzle containers on the floor, whereas in both the classroom that got the persuasive messages and the control classroom, about 70 percent of the kids threw the containers on the floor.¹⁵

Does this mean that we should lavish our kids with praise, as a way of giving them positive labels for their behavior? If you are like me, you have a strong desire to compliment your kids when they excel at something, such as doing well in school. “Great job on the multiplication test,” you might say. “You sure are good at math.” Nothing wrong with a verbal pat on the back, right?

Well, it depends on the message that the pat conveys. As we’ve seen, the labels adults provide matter, and the danger with praise is that it might convey the wrong label—what social psychologist Carol Dweck calls a fixed mindset, which is the idea that we are born with a fixed amount of talent in a specific area, such as math or soccer or music. We either have that talent or we don’t, according to this viewpoint, which is fine as long as we continue to do well (e.g., get As on math tests) but can be devastating when we don’t. Suppose that in middle school we do poorly on our first algebra test—what are we to conclude? If we believe that everyone is born with a fixed amount of mathematical ability, we can only surmise that we have reached the limit of ours and that there is no use in studying hard for the next test. “I’m just not a math person” is the logical conclusion.

But suppose we adopt what Dweck calls a growth mindset about math, the idea that we can learn to be good at something if we try hard and practice, that abilities are like muscles that get stronger with use. When we have a setback, such as that poor grade on the algebra test, we are more likely to take it as a sign that we didn’t study hard enough and that we need to try harder next time.

Note that fixed versus growth mindsets are narratives that kids adopt and do not necessarily reflect how talented they actually are. Talent isn’t irrelevant, of course. Some people, like former basketball great Michael Jordan, have far more natural ability than the rest of us. But the importance of practice should not be underestimated, and this is determined largely by people’s mindsets. You might be surprised to learn, for example, that Michael Jordan was cut from his high school basketball team the first time he tried out. But rather than giving up, he redoubled his efforts, leaving home at 6:00 a.m. in order to practice for an hour before school started. Eventually, he did make his high school team and played well enough to earn a scholarship to the University of North Carolina, which has one of the premier college basketball programs in the country. But even then he didn’t stop practicing. When North Carolina’s season ended with a disappointing loss in the NCAA tournament, Michael Jordan’s response was to go to the gym right after the game and practice his jump shot. Clearly, he had a growth mindset about his basketball abilities.

Where do these mindsets come from? Parents play a key role. When Michael Jordan’s mother heard that he hadn’t made his high school team, she told him to “go back and discipline yourself”—which is classic “growth mindset” feedback. Rather than saying, “You were robbed, I’m going to talk to the coach” or “Maybe you should try soccer instead,” his mother conveyed the message that you have to work at something in order to be good at it—even if you are Michael Jordan.¹⁶

By the same token, parents should be careful about how they label their kids’ successes. As tempting as it is to praise our children’s outstanding talents—after all, we want to think of them as gifted—doing so can convey exactly the wrong message. In the second panel of the cartoon below, for example, Sally makes a classic mistake: she conveys a fixed mindset to her daughter, the idea that people do well in school because they are “gifted.” What will her daughter conclude if she doesn’t do well on her exams? It is better to compliment our kids for the hard work they put in, to send the message that success comes from hard work, not innate abilities. “Great job on the multiplication test,” you might say. “I really admire how much time you spent studying for it. It really paid off.”¹⁷

CORE ATTACHMENT NARRATIVES

What else can parents do to promote healthy narratives in their children? In the previous chapter, we discussed the importance of having a coherent core narrative that answers key questions about our existence and place in the world. An important feature of core narratives we haven’t yet encountered is the way they portray our relationships with other people. What are other people like? Can they be trusted? If we get close to people, will they be there when we need them, or will they eventually leave us for somebody else? Many people have core relationship narratives that provide heartening answers to these questions; they believe that most people are trustworthy and can be counted on. Other people aren’t so sure—their core relationship narratives portray people as untrustworthy and not to be depended on.

Where do these relationship narratives come from? According to attachment theory, these narratives—or “attachment models”—have their roots in our relationships with our primary caregivers in the first years of life. Say that your primary caregiver was your mother. If she was attentive and responsive to your needs, and provided consistent, dependable, and prompt care, you likely developed a secure model of attachment. If she wasn’t, you might have developed a less healthy model of attachment. If she tended to ignore you when you were crying or in distress, you are likely to have developed an avoidant attachment style, whereby you have difficulty getting close to and trusting others. If she was inconsistent in her attention, vacillating between attentiveness and neglect, you are likely to have developed an ambivalent/resistant attachment style, whereby you fear that your loved ones will abandon you and therefore you tend to smother them with affection. If you felt both frightened and comforted by your primary caregiver, because he or she vacillated between threats and reassurance, you are likely to have developed a disorganized attachment style.

These attachment styles form the blueprints by which we understand our relationships with other people, the narratives that we use to interpret other people’s behavior and guide our behavior toward them. One study found that infants who were securely attached to their mothers at twelve months of age were more skilled at making friends in elementary school (as rated by their teachers), were more secure about their friendships at age sixteen, and had more positive and supportive romantic relationships in their early twenties, than were infants who were not securely attached at age twelve months.¹⁸

Does this does mean that if we were insecurely attached as infants we are doomed to unhappy relationships for the rest of our lives? Well, it does raise the odds, but, as we have seen, people’s narratives can change. Children with insecure models of attachment can, through successful interactions with caregivers, acquire more secure models. Some studies show that the quality of a person’s adult romantic relationships is largely related to his or her current attachment model—the way he or she views relationships as an adult—which is not necessarily the same as the model he or she had as an infant.

This raises the question of how best to redirect attachment narratives. Perhaps we could target mothers and infants who seem to be having difficulty forming close attachment bonds and help them in some way. This might seem like a daunting task, because mothers and their children can be caught in a vicious cycle in which the mother is insensitive to her baby’s needs (perhaps because of her own insecure model of attachment), which lowers the probability that her baby will develop a secure model, and so on—part of a long chain of insecurity breeding insecurity. How can we possibly break this chain, short of major psychotherapy?

As we have seen, a fundamental assumption of the story-editing approach is that small interventions can reap huge benefits, if they succeed in redirecting a key element of a person’s narrative. There is evidence that this can work to break the insecurity chain of attachment. One intervention, which involved only three home visits by a researcher, targeted mothers in the Netherlands who had firstborn infants who were high in irritability. Highly irritable infants were selected because it can be difficult for parents to establish secure bonds with babies who fuss and cry a lot. In fact, parents of these kinds of babies are especially likely to get caught in a vicious cycle in which they respond more and more passively to their babies—partly because the babies are less likely to smile and coo than nonirritable babies and partly because the more passively parents respond, the less likely they are to trigger a crying spell in their babies. Consequently, parents become less attentive to their children’s needs and are less likely to become securely attached to them.

The researcher tried to interrupt this negative cycle, beginning when the infants were six months old. During each of three two-hour visits, she attempted to get the mother to be more attentive to the signals the infant was giving about his or her needs, to interpret the signals correctly, and to respond appropriately. When the baby fussed and cried, the mother was encouraged to try a variety of techniques to soothe him or her, because some babies prefer close contact and others do not. Mothers were encouraged to engage their babies in positive ways, rather than respond only to negative signals. For example, mothers were taught to make eye contact with their babies and coo at them when their babies held their gaze (because eye contact is a sign that the baby is engaged and interested in interacting), but to be silent when the babies looked away. The intervention was tailored to each individual mother-infant pair, with the overall goal of getting the mother to be more attentive to her baby’s signals and needs.

Compared to mothers randomly assigned to a control group, those who received the intervention did indeed become more responsive to their infants’ needs. As a result, their infants became less irritable, more sociable, engaged in more exploration, and were better able to soothe themselves when upset. At twelve months of age, only 22 percent of the infants in the control group were classified as securely attached to their mothers, but this number nearly tripled to 62 percent in the intervention group—an amazing difference given that the intervention consisted of only three home visits.

Even better, mothers in the intervention group were still more responsive to their children, compared to mothers of children in the control group, three years later. At age three and a half, children in the intervention group were more secure and cooperative than were children in the control group. These positive effects, it turned out, were brought about by the change in the child’s attachment relationship with the mother, which supports the idea that establishing a secure internal working model of relationships is key to child development, and that at least for some parents (e.g., those with highly irritable children), a simple, short-term intervention can reap large benefits in attachment. And, like many other story-editing interventions, this one worked in part because it was self-sustaining. Once a mother learned to read her baby’s signals better, the baby fussed less and was more responsive to her. This positive outcome encouraged the mother to stick with the program and try harder to be sensitive to her baby’s needs.¹⁹

PREVENTING CHILD ABUSE

“That’s all well and good,” you might think, “but would story editing work with families that are even more at risk than the ones studied in the Netherlands?” There is a lot of social pathology out there and some parents do terrible things to their children. Surely they can’t be helped with one of these tiny little story-editing interventions. Or can they?

Well, consider child abuse, which is surely one of the most heinous of crimes. Unfortunately, it is far from uncommon. Approximately 900,000 children are found to be the victims of abuse or neglect each year in the United States. In 2007, 1,760 of these children died as a result of abuse or neglect.²⁰

In the 1990s, the United States Department of Health and Human Services declared that child abuse was a national emergency and urged that prevention programs be implemented. One organization, Prevent Child Abuse America (PCA America), believed that a home visitation program founded in Hawaii would work. The Hawaiian Healthy Start Program, which had been in operation since the 1980s, screened new parents in the hospital and offered counseling and home visits to those deemed at risk for abusing their children. The program was thought to be so effective that it was adopted as a national model by PCA America. It was renamed Healthy Families America and now has chapters in more than 440 communities in the United States. Parents of newborns who are believed to be at risk for abuse are referred to the program and receive home visits by trained staff who provide support, parenting tips, and anger management training. Federal and state governments spend millions of dollars a year on this well-intentioned program.²¹

There is only one problem—it doesn’t work. Like other commonsense interventions we have encountered, Healthy Families America was implemented on a mass scale before anyone tested it with rigorous experimental techniques. Once it was tested, the results were disappointing. In one study, for example, Hawaiian families were screened as usual for the Healthy Start Program. Those deemed at risk were randomly assigned to participate in the program or to a control group that did not participate. It turned out that there was no difference between the two groups in the percentage of children referred to child protective services because of child abuse (these percentages were very low in both conditions); no difference in mothers’ reports of corporal punishment or physical aggression; no difference in the percentage of mothers who relinquished their role as primary caregiver (8 percent in both groups); and no difference in the percentage of children who required emergency health care, such as emergency room visits. In short, there was little or no evidence that the Healthy Start Program was having any impact at all on the behaviors it was designed to prevent. Other reviews of home visitation programs have reached similar conclusions.²²

As well-intentioned as the Healthy Start Program is, and as likely as it seems that it would work, there are a number of reasons why it doesn’t. For example, it ignores some of the key variables that increase the risk of child abuse, such as spousal abuse, parental depression, and parental substance abuse. Another possible reason it doesn’t work comes right out of the story-editing approach: the program fails to target the way in which parents interpret their children’s behavior, which may be a key contributor to child abuse.

In fact, social psychologists have noticed something curious about parents who abuse their children: they seem to blame the children for being difficult. When their kids act up, these parents often think, “It’s his fault”—he has a bad disposition, he’s stubborn, he was born that way, or he is deliberately trying to provoke them. Parents who don’t abuse their children are more likely to attribute their kids’ crankiness to something about the situation that is easily addressed—the child is hungry, tired, or needs a hug. Further, researchers suggest that these attributions are not necessarily conscious, and in fact may be part of parents’ implicit narratives that they cannot fully verbalize, even to themselves.²³

If so, then it might work to get mothers to redirect these narratives with a story-prompting intervention. To find out, researchers at the University of California, Santa Barbara, recruited families to participate in the Healthy Start Program, but gave some of them prompts to change their explanations about why their kids were acting up.

Here’s how it worked: as with the Healthy Start Program, the Santa Barbara researchers targeted parents who were deemed at risk for being abusive to their children. For example, in one of their studies, 50 percent of the parents reported that they had been physically abused when they were growing up. In another, the participants were low-income women whose children were born with medical problems. These families were randomly assigned to one of three conditions. Some were in a control condition that received information about available community services but did not receive any home visits. Others took part in the standard Healthy Start Program, in which they received an average of seventeen home visits from trained staff over the course of the first year of their babies’ lives. The visits followed the procedures of the Healthy Start Program; for example, they included parent education, anger management training, and parenting tips.

The third group also took part in the Healthy Start Program, but in addition received a story-editing intervention that tried to reduce the extent to which the mothers blamed their babies for caregiving problems and to increase the extent to which they attributed these problems to causes that could be fixed. Rather than simply giving the parents a new explanation for their babies’ conduct, the home visitors encouraged parents to come up with these explanations themselves. First, they asked parents to think of examples of times when their children were difficult to care for and why they thought this was the case. Typically, the parents would attribute the problems to factors that were difficult to change, and would often blame their children. For example, a mother might say that her baby cried inconsolably after being fed in the evening and that the reason was that the baby was mad at her or trying to provoke her. At that point, the home visitor asked the mother whether she could think of any other reasons why her baby might cry, continuing with this line of questioning until the mother came up with an explanation that did not blame the child and was easier to remedy. After a while, for example, the mother might say, “Well, maybe my baby gets indigestion in the evening.” This line of questioning would continue until the mother came up with a new strategy for solving the problem; for example, “I could try a different kind of formula and spend more time burping him after he eats.” At the next visit, the home visitor inquired about the success of the strategy the mother came up with, discussed changes in the strategy if it wasn’t working, and then repeated this process with a different caregiving problem. The hope was that after several such interactions, parents would blame their children less and identify solutions to behavioral problems that they could easily implement.

At the end of the yearlong program, parents reported on the health of their children and on how often they used abusive or harsh parenting techniques, such as hitting, shaking, or slapping. Consistent with previous studies, the Healthy Family home visits alone had little impact. Twenty-three percent of the children in this condition were physically abused, compared to 26 percent in the control condition—a difference that is not statistically significant. But, as the researchers predicted, the story-editing intervention worked: only 4 percent of the children in this condition were physically abused. Similar results were found on measures of corporal punishment that do not meet the legal definition of abuse, namely, slapping and spanking: 42 percent of parents in the control and Healthy Families group reported using these punitive techniques, compared to only 18 percent of parents in the story-editing condition. In follow-up studies, the story-editing intervention was found to reduce child injuries, lower the levels of cortisol in the children (cortisol is a hormone that is an indicator of stress), and improve cognitive functioning in the children. Consistent with many other story-editing interventions, redirecting people’s narratives—in this case, encouraging parents to reinterpret why their children act up—reaped huge benefits down the road.²⁴

USING IT

If you are a parent, grandparent, or someone who spends a lot of time with children (a teacher, say), I hope you learned some practical lessons from this chapter. Here are some techniques to remember when trying to redirect your children’s narratives.

Avoid an Overly Controlling Parenting Style: Parents who are too controlling prevent their children from developing a sense of autonomy and purpose. I certainly wouldn’t recommend that you go to the opposite extreme, becoming so lax that your kids fail to learn any limits or self-control. We’ve all seen parents who are incapable of using the “no” word and who end up at the mercy of their kids’ every whim and wish. One way to find a happy medium is to use the minimal sufficiency principle, whereby you dole out rewards and punishments that are strong enough to get your kids to do what you want them to do, but not so strong that the kids think the rewards and punishments are the reason for their behavior. If you use this principle, your kids are likely to internalize good values and character traits, rather than believing that they are “doing it for the money” or to avoid the wrath of Mom and Dad. As I mentioned, this is sometimes easier said than done, because it can be hard to tell in advance what will be “minimally sufficient” to get our kids to do what we want them to do. What level of threats will be sufficient to keep our teenagers from texting while they are driving, for example? Saying we would be “a little annoyed” might be too weak, with disastrous results. Better to err on the side of too strong a threat about such dangerous actions, at least at first, while recognizing the risks of going too far overboard.

Label Your Kids’ Behavior Appropriately: Once your kids are behaving appropriately, it can help to provide them with a favorable label for those behaviors. “You sure are a safe driver,” you might say, or “Looks like you’re the kind of person who loves to read.” Again, don’t go overboard. My kids were fond of making fun of my clumsy attempts to label their behavior, repeating in a singsong voice, “We know, Dad, we’re not eating our vegetables in order to get dessert, we love vegetables.” Further, be mindful of the kinds of labels you use. As Carol Dweck’s work shows, it is harmful to label your kids’ successes in ways that imply a fixed mindset (“You are such a talented athlete!”). Better to label them in ways that imply a growth mindset (“Your practice really paid off!”).

Foster Secure Attachment Models: In the last part of this chapter, we saw the importance of core attachment narratives. Beginning the moment your child is born, be attentive to his or her needs and respond consistently with appropriate love and care. Doing so will foster a secure attachment model in your children, which will serve them well as they grow older and help them form bonds with others in their lives. Along with a sense of autonomy and purpose, there may be no greater gift we can give our children than the narrative that they can form close, trusting relationships with others.

CHAPTER 4

Shaping Our Kids’ Narratives