Walter Mischel came to study self-control by a circuitous route. Like Sigmund Freud and Billy Wilder, both of whom plumbed the murky depths of temptation and desire before him, Mischel was from Vienna and left under similar circumstances, although at a much younger age. Mischel was just a boy, and his is a classic story of its type: the prosperous and assimilated Jewish family impoverished and humiliated by the Nazis, the lucky escape to America, the sight of Manhattan breathtaking from the boat, the small store in Brooklyn, the unrelenting toil and sense of displacement, the brilliant son, the odd jobs in the garment business and in a department store, the awakening to literature. His overprotective mother, in Europe seen mainly on a divan with an ice pack on her head, was here transformed into a formidable businesswoman, shucking off her Old World vapors while instilling in the boy ( just as Freud’s mother did) the belief that he could do anything. He could paint, for example, and never mind that there was no money: “Oil colors were too expensive, but I found that Jell-O, my mother’s favorite fare, mixed the right way, especially the reds, could work reasonably, and a mostly Jell-O self-portrait at age 17 still hangs unfaded and uneaten in my Manhattan apartment.”
Mischel wrote those words as an old man at Columbia University, where he is still conducting research into self-control. But in the 1950s he was a young psychologist at Ohio State University when his then wife obtained a grant to study spirit possession in the Orisha religion of Trinidad. Trinidad seemed a lot more exciting than Ohio, so Mischel went along, spending summers there from 1955 to 1958. Orisha was then known as Shango, and Mischel reports that while the psychological tests he tried to administer were a waste of time, Shango and its practitioners were fascinating. In their religious ceremonies, he observed an intentional loss of control: “Laborers and domestic servants of the British by day became ‘possessed’ at night by the mix of saints and African gods whose spirits ‘rode’ them as they danced in hypnotized states, enacting their godlike roles, with the irresistible drums pounding and the rum bottles passing.”
Mischel spent his time in a section of Trinidad where the people were either of African or East Indian descent, both groups having found their way to the Caribbean as the result of British colonialism. Members of each group lived with their own kind, and each appeared to have its own values and stereotypes: “It did not take much listening to note a recurrent theme in their characterizations of each other. In the eyes of the East Indians, the Africans were just pleasure-bent, impulsive, eager to have a good time and live in the moment while never planning or thinking ahead about the future. Reciprocally, the Africans saw their East Indian neighbors as just working for the future and stuffing their money under the mattress without ever enjoying today.”
His curiosity piqued, Mischel went into the local schools and began to study children of both ethnic groups. He started interviewing kids to discern such things as their family situation, their motivation to achieve, and their intelligence. Then he did something hugely consequential.
When students completed Mischel’s questionnaire, he showed them two potential candy rewards, one costing just a penny and the other ten cents. He’d already learned from the students that, as expected, they much preferred the ten-cent candy, except the rewards now came with a catch: Mischel explained that he didn’t have enough of the better candies just then, so students would be given a choice of taking a penny candy now or the vastly more desirable ten-cent candy in a week. He then recorded their choices and correlated them with other factors he was studying.
In the first paper to emerge from this research, in 1958, the results were consistent with the cultural pattern he’d previously observed. Mischel studied children ages seven to nine. Of the black students, twenty-two out of thirty-five chose the smaller, immediate reward. Of the Indian students, these proportions were almost exactly reversed: twelve of eighteen held out for the larger candy. Mischel and later commentators speculated that these results might be a function of the students’ doubt that the later reward would ever materialize. Bolstering this notion was the remarkable finding that all eleven students without a father at home (ten of whom were black) chose the immediate reward, suggesting that the lack of a reliable male parent impaired confidence in the promises of grown-ups—and was a better predictor than ethnicity of what choice a child would make. (It’s also possible that men who live apart from their families are more prone to immediate gratification and that their children have inherited this impulsive bent.)
Subsequent research on race and delay of gratification has produced mixed results; a 1983 survey in the Journal of Black Psychology found no clear pattern. But an interesting study in Atlanta underscored the importance of trust. Bonnie R. Strickland reported in 1972 that black sixth-graders were much less likely to choose a larger, later reward than white students were—when the promise of a later reward came from a white experimenter. When the experimenter was black, however, black students were dramatically more willing to delay gratification, although still at a level well below that of white students, whose willingness was unaffected by the experimenter’s race.
Today Walter Mischel is a towering figure in the quietly sizzling realm of what might be called self-control studies. The technique that he used in Trinidad would come to be known as the “delay of gratification paradigm,” which basically means offering someone the choice of a small, immediate reward or a larger, later one—often much larger but not much later, so that the later reward is unequivocally a better deal—and has become the standard method of probing people’s ability to postpone rewards and exercise self-control. As a result of his summers in the Caribbean, Mischel wrote a series of papers looking at the relationship between delay of gratification and attention, intelligence, age, family structure, and income—papers, in other words, that addressed some of the most important questions in the field. More important still, his visits to Trinidad were the start of a lifelong effort to understand the problem of self-control experimentally. This effort would result in a body of astonishing research, most of it built on the foundations he established in the Caribbean. A lot of what we know (or suspect) about self-control today can be traced back to Mischel’s earliest efforts, which is one reason they’re worth exploring today.
Another is that one distilled nugget of his research, colloquially known as “the marshmallow test,” would eventually become a staple of preachers and pundits. The test presents kids with a choice of one treat now or two if they can just wait a while. Mischel’s findings with this test have prompted a series of motivational books, and DON’T EAT THE MARSHMALLOW was emblazoned on T-shirts in a Philadelphia charter school. The test would come to have political implications, too, like a modern version of Aesop’s fable of the ant and the scarab beetle (better known to us today as the story of a thrifty ant and an improvident grasshopper), but anointed by science. We’ll get to the marshmallow test in a minute. The point, for now, is that Mischel’s research shows with stunning clarity just how important self-control is in people’s lives.
What commentators rarely pay much attention to is the troubling question raised by Mischel’s research. Self-control matters, sure. But does his work show that we should exercise more of it? Or that we can’t, because our capacity for it is inborn? And if it is inborn, doesn’t that turn all the homilies citing Mischel’s work into parables about the hopelessness of predestination? “It’s highly likely to be like most things in life are turning out to be,” says Mischel, “which is, yes, the wiring makes a difference. And yes, the experience makes a difference. And the wiring and the experience are interacting and changing each other.”
But we’re getting ahead of ourselves. Let’s go back instead to Mischel’s time at wintry Harvard, that Puritan-inspired institution a world away from balmy Trinidad. It was at Harvard that he and his associates managed to complete some fascinating studies among Boston-area children. In this work, Mischel built on what he’d learned in Trinidad and used some of the same general techniques. In a study focusing on temptation, for instance, elementary school boys were invited to play an arcade-style game in which they were to shoot a toy ray gun at a disabled rocket. Each child was assigned to keep score for himself while the experimenter was out of the room, supposedly busy with paperwork. But the game was set up so that no matter how skillfully the weapon was fired, it was impossible to score enough points to win a badge at any of the three levels of highest expertise. The game, in other words, almost cried out for the kids to cheat. And cheat they did.
Mischel and Carol Gilligan, his coauthor (who would someday become famous for raising alarms about the supposedly inferior treatment of girls in school), also offered the children a series of choices between smaller, immediate rewards or larger, later rewards. The children’s responses enabled the researchers to grade their ability to delay gratification. Mischel and Gilligan found that the kids best able to do so were also least likely to cheat. And of those who did cheat, the ones best able to delay gratification waited longest before chiseling.
In other studies of Boston-area children, Mischel and his colleagues found that kids who were better able to delay gratification— typically, those who chose to wait a bit for an unequivocally superior reward instead of grabbing for the smaller one immediately—tended to be more intelligent, more socially responsible, and more ambitious for achievement. As Mischel put it in the deflating language of science, “The obtained concurrent associations are extensive.”
But in the early 1960s, Mischel began to find Harvard’s psychology department less and less hospitable. “The place kept getting crazier,” Mischel wrote later, and “it was impossible to work.” The reason, basically, was Timothy Leary, with whom Mischel got along well at first, since the two shared an interest in personality assessment. But Leary soon moved on to matters psychedelic, while the newly minted PhD Richard Alpert was on his way to becoming Baba Ram Dass. “Mattresses suddenly replaced several of the graduate student desks,” Mischel writes of those days, “and large packages from Ciba chemicals in Switzerland began to arrive in the department mail.”
In 1962, Mischel moved to Stanford, where he would undertake the experiments for which he is now famous. He spent nearly twenty years there with his second wife, with whom he had three daughters in rapid succession. Watching the girls grow up made him wonder all the more about how rewards enable people to delay gratification—and about the development of willpower, “a term that as a psychologist I still put into quotes.”
Stanford University’s new Bing Nursery School proved the ideal place to explore these issues. Not only did it put a group of young kids close at hand; it even came equipped with big one-way observation windows. Mischel already had a broad sense of how his Bing experiments would be carried out; he would use the techniques he’d employed in Trinidad. But this time he drew on his own daughters for some of his hypotheses. Perhaps more important, at Stanford he had time on his side, for over the succeeding decades Mischel and an ever changing cast of colleagues and graduate students would follow up on the Bing alumni, with remarkable results.
The experiments ran from 1968 to 1974, a time of turmoil outside the nursery walls—a period spanning Vietnam and Watergate, the assassinations of Robert Kennedy and Martin Luther King Jr. A total of 653 children participated during those years, of whom 316 were boys and 337 girls, and on average they were a few months past their fourth birthday. In a typical experiment, Mischel or one of his associates might play with a four-year-old for a few minutes and then sit the child at a table. On this table there might be three distinct things: a single marshmallow, a pair of marshmallows, and a bell. The experimenter would explain that he had to leave for a bit and that the child would have a choice. If he waited for the adult’s return, he could have the pair of marshmallows. But if he didn’t want to wait, he could summon the grown-up by ringing the bell—in which case he could only have one marshmallow. The experimenter would then leave the room for up to (usually) fifteen minutes. In some tests, the rewards were left exposed on the table, but in others they were removed from sight. In some, the children were even given advice (sometimes good, sometimes bad) on how to hold out. In all cases, the crucial measurement was: How many seconds would it take for the child to ring the bell?
The mostly middle-class Bing children in these experiments were around four and a half years old, an age at which it’s hard to delay gratification—in part because there’s still a lot of brain development ahead, but also perhaps because little kids haven’t yet learned any strategies for keeping temptation at bay. The experimenters tried various suggestions during the different experimental configurations and discovered that telling the children to contemplate the treat’s arousing properties—the luscious sweet flavor, for instance—not only cut delay times, at least when the rewards were exposed, but narrowed differences between children. And in the absence of any advice from the experimenters, the children tried their own strategies, some sensible and some not.
Self-distraction generally was effective. Some of the kids with the longest delay times covered their eyes or rested their heads on their arms. Others talked to themselves, turned their backs and sang songs, made up games with their hands and feet, tried to go to sleep to make the time pass, or climbed under the table to escape temptation. One boy took to kicking the table; “it’s a very male response,” says Mischel. Another boy, confronted with Oreos, seized one, pried it open, licked out the cream and then reassembled the cookie on the tray.
Mischel, an elfin figure who nowadays resembles a distant cousin of Yoda in Star Wars, spent hours watching the kids squirming and extemporizing from behind the nursery school’s one-way glass. He learned that what determined how long they could hold out before ringing the bell (if they rang at all) was how the children handled the challenge mentally. “We saw that what mattered was what was in their heads, and that trumped what was actually in front of them: The duration of delay depended on specific types of ‘hot’ or ‘cool’ mental representations and the ways in which attention was deployed during the delay interval.”
The idea that the brain is divided into “hot” and “cool” areas has gained some currency among researchers. The notion is that the cool areas, probably associated with the hippocampus and frontal lobes, correspond to what philosophers might call reason. They are rational and generate planned, advantageous behavior. The hot areas are more primitive, and develop earlier in young children. These areas, associated with immediate survival and operating more reflexively, take care of appetitive processing, fight-or-flight reactions, and other instant responses to stimuli. What Mischel and some other researchers think nowadays is that children held out more effectively when they were able to shift processing from the hot to the cold brain systems. In a recent interview, Mischel put it this way: “The same child who can’t wait a minute if they’re thinking about how yummy and chewy the marshmallow is can wait for 20 minutes if they’re thinking of the marshmallow as being puffy like a cotton ball or like a cloud floating in the sky.”
Clearly there’s a lot to be learned about self-control from these Bing Nursery experiments: attention management helps, focusing on the attractive qualities of the thing to be resisted does not, and keeping the desired object out of sight helps, too. So does thinking of the thing in an abstract way—the crunchy, salty pretzels might be construed as tasteless miniature logs. State of mind matters, sure, but the important point is that we have some control over our states of mind, even at the age of four.
Even more interesting is what Mischel and his colleagues discovered some years later, in follow-up studies of the same Bing four-year-olds: the best performers on the marshmallow test had developed into the best-adjusted adolescents.
“Those who had waited longer in this situation at four years of age were described more than ten years later by their parents as adolescents who were more academically and socially competent than their peers and more able to cope with frustration and resist temptation . . . Parents saw their children as more verbally fluent and able to express ideas; they used and responded to reason, were attentive and able to concentrate, to plan, and to think ahead, and were competent and skillful. Likewise they were perceived as able to cope and deal with stress more maturely and seemed more self-assured. In some variations of this laboratory situation, seconds of delay time in preschool also were related to their Scholastic Aptitude Test (SAT) scores when they applied to college.”
Mischel hasn’t published the size of the SAT difference between those who could wait and those who couldn’t, but he’s told some people that it was 210 points—which would be huge. And there was more. The kids with the shortest delay scores, according to Mischel, had lower grade point averages, more school suspensions, and “were most likely to end up as the bullies.” The ability to delay gratification also predicted body weight: kids with the higher delay scores were thinner. (This last finding was bolstered recently in two separate studies involving more than 1,800 children. Like Mischel, the researchers offered four- and five-year-olds a bigger treat if they could wait, discovering that kids less able to delay gratification were more likely to be fat by the age of eleven.)
By now Mischel knew he was really onto something, and he continued to follow his Bing Nursery subjects into adulthood. In another study, this time when the kids had reached an average age of twenty-seven, Mischel and his colleagues looked at sensitivity to rejection, which can be seen as a measure of social anxiety, and how it related to their performance years earlier as four-year-olds. The study, which covered 152 Bing alumni, found that rejection sensitivity caused problems for those found to have it—unless they had also had high delay-of-gratification scores at Bing, in which case rejection sensitivity didn’t seem to matter much. Highly rejection-sensitive individuals with low delay-of-gratification scores as tykes—these were the kids who just couldn’t wait—were found later in life to have lower educational attainment and more use of cocaine or crack.
About those four-year-olds: You’re probably wondering how long they managed to hold out in the face of all those marshmallows, pretzels, and so forth. In one paper that followed up on 185 of the kids, Mischel and colleagues reported that this group on average managed to delay gobbling their treat for 512.8 seconds, or less than nine minutes. On the whole, Mischel has said, depending on the goodies at hand and other circumstances, his four-year-olds held out for seven or eight minutes. But some could hold out for much longer—perhaps as long as twenty minutes.
Walter Mischel is a pioneer, but he isn’t the only one who’s noticed the many ramifications associated with self-control in the young. Take school performance, for example. A number of researchers have found self-regulatory abilities to be associated with grades.
Angela Duckworth and Martin Seligman are psychologists at the University of Pennsylvania’s dubious-sounding Positive Psychology Center, which actually isn’t dubious at all; Seligman is director and pretty much inventor of the positive psychology field, which studies the emotions, actions, and institutions that contribute to happiness. People do in fact have some control over their mental states. We know, for example, not only that emotions cause facial expressions, but that the arrow of causality runs in reverse as well, so that putting on a happy face really can make you happier. It doesn’t take a PhD to know that cultivating good thoughts helps, too. Long before we had a Positive Psychology Center, the Roman emperor Marcus Aurelius observed: “Such as are thy habitual thoughts, such also will be the character of thy mind; for the soul is dyed by the thoughts.”
In a 2005 study of 140 eighth-graders at an ethnically diverse magnet school, Duckworth (a former schoolteacher) and Seligman first graded the kids on self-discipline by using ratings from their parents and teachers and from the students’ own responses to a questionnaire that asked such things as, How hard is it for you to break bad habits? This all happened in the fall, near the beginning of the school year. In spring the researchers came back and found that self-discipline was correlated with school attendance, grades, standardized achievement test scores, and eventual admission to a competitive high school.
Duckworth and Seligman repeated the study with another 164 eighth-graders, this time adding a delay-of-gratification task and an IQ test. Self-discipline turned out to be a vastly better predictor of grades than was IQ—not surprisingly, since doing well in school requires sustained effort, putting off fun when it’s time to do homework, and steady work toward the long-range goal of a high grade at term’s end. Self-discipline also more accurately predicted attendance, how much time was spent on homework, and even at what time each evening homework was commenced. Self-discipline was also a much better predictor of how much TV the kids watched; the higher the self-discipline score, the less time in front of the tube. Afterward, the psychologists did not mince words.
“Underachievement among American youth is often blamed on inadequate teachers, boring textbooks, and large class sizes,” they wrote. “We suggest another reason for students falling short of their intellectual potential: their failure to exercise self-discipline . . . We believe that many of America’s children have trouble making choices that require them to sacrifice short-term pleasure for long-term gain, and that programs that build self-discipline may be the royal road to building academic achievement.”
Duckworth and Seligman found something else pretty interesting. They knew that girls consistently get better grades than boys in K–12 schooling, yet fail to outperform boys on achievement or IQ tests. So they wondered if self-control might play a role. After analyzing their data, they concluded that in their study, at least, it did. The girls were more self-disciplined, “and this advantage is more relevant to report card grades than to achievement or aptitude tests.”
These results are no flash in the pan, and some have speculated that, just as boys do worse in school for want of self-control, American students lag their foreign peers for similar reasons. Again, when it comes to school, self-mastery matters. Raymond Wolfe and Scott Johnson, for example, looked at thirty-two personality variables in a group of students and found that self-control was the only one with any real ability to predict grades in college. Self-control was a better predictor than SAT scores; only high school grades did a better job foretelling how students would fare at college, and the researchers recommended “that the global trait of self-control or conscientiousness be systematically assessed and used in college admissions decisions.”
Kids with good self-control also grow into adults who get along better with others. In one study, published in 1999, teacher ratings of kids’ self-control at ages four to six predicted which kids would be more popular. And work by Mischel and Yuichi Shoda showed that a four-year-old’s ability to defer gratification was associated with better relationships as a young adult. Other studies show that preschoolers with more self-control have more social competence and less negative emotional arousal; that high-schoolers with high self-control have fewer eating and drinking problems; that college students with high self-control get better grades; and that preteen and teenaged boys with low self-control are at greater risk of aggressive behavior and delinquency.
Shortfalls in self-control are also implicated in heroin addiction, alcohol abuse, and overspending among the young. The criminologist Travis Hirschi writes that among high school students, self-control “consistently predicts behavior analogous to crime: truancy, cheating on exams, being sent out of a classroom, driving while drinking, auto accidents, bike-skateboard-rollerblade accidents, broken bones, shooting dice for money, drinking alcohol, smoking tobacco, and smoking marijuana.”
Walter Mischel, meanwhile, the man who started it all, is still in hot pursuit of the mystery of self-control. With funding from the National Science Foundation, he and coinvestigators around the country have launched an ambitious effort to expand our knowledge about something, which, based on the marshmallow experiments, has already been established as a surpassingly important human characteristic. Using surveys and magnetic resonance brain scans of the now-grown Bing children, researchers are investigating the brain structures that enable people to exercise self-control. The scientists are also conducting genetic tests to try to figure out to what extent the ability to defer gratification is hereditary.
For there is unquestionably a genetic component, and modern technology makes it cheap and easy to tell which gene variants people have. The study will draw subjects from Mischel’s preschoolers dating back to the early 1970s, all of them in their late thirties and early forties now, and will focus especially on two groups: one that’s shown a high level of self-control over the years, and another that’s shown the opposite.
Mischel’s plan raises a crucial question. If self-control is so important for kids, can it be taught? There is every reason to believe that it can. Nobody seems to think it’s 100 percent hereditary, after all. Mischel contends that good parenting plays a large role—even in everyday matters like not snacking before meals. And researchers have already found some value in programs aimed at teaching children “executive functions,” which is psychological jargon for self-control. Adele Diamond evaluated a fascinating program called Tools of the Mind, developed by psychologists Deborah Leong and Elena Bodrova. The program asked students to plot their play intentions in writing, and their days were filled with various other games and activities designed to promote planning, focus, and delay of gratification. In a study of 147 kids roughly divided between those in Tools of the Mind and those not, Diamond found that after two years the program participants did quite a bit better on tests of executive function—suggesting they really did learn some self-control.
Other schools have embraced the gospel of self-control as well. The renowned KIPP (Knowledge Is Power Program) chain of charter schools, which has succeeded in inner-city neighborhoods where other schools have failed, has made self-control one of its core “character strengths.” The KIPP school in Philadelphia gave its students T-shirts with the words DON’T EAT THE MARSHMALLOW lettered across the front.
Diamond says kids today may need such training because old-fashioned styles of self-organized free play taught self-control skills in ways that modern, parent-organized sports and in-home electronic play may not. On the other hand, martial arts, music lessons, or other activities requiring sustained attention probably help. Says Diamond: “I think a lot of kids get diagnosed with ADHD now, not all but many just because they never learned how to exercise self-control.”