by Paul Tough
Although the human stress-response system is highly complex in design, in practice it has all the subtlety of a croquet mallet. Depending on what kind of stress you experience, the ideal response might come from one of any number of defense mechanisms. If you’re about to receive a flesh wound, for instance, then it would be a good idea for your immune system to start producing copious antibodies. If you need to run away from an attacker, you want your heart rate and blood pressure to elevate. But the HPA axis can’t distinguish between different types of threat, so it activates every defense, all at once, in response to any threat. Unfortunately, this means you often experience stress responses that are not at all helpful—like when you need to speak before an audience, and suddenly your mouth goes dry. Your HPA axis, sensing danger, is conserving fluids, preparing to ward off an attack. And you’re standing there looking for a glass of water and swallowing hard.
Think of the HPA axis as a superdeluxe firehouse with a fleet of fancy, high-tech trucks, each with its own set of highly specialized tools and its own team of expertly trained firefighters. When the alarm bell rings, the firefighters don’t take the time to analyze exactly what the problem is and figure out which truck might be most appropriate. Instead, all the trucks rush off to the fire together at top speed, sirens blaring. Like the HPA axis, they simply respond quickly with every tool they might need. This may be the right strategy for saving lives in fires, but it can also result in a dozen trucks pulling up to put out a single smoldering trash can—or worse, responding to a false alarm.
5. Scared to Death
Nadine Burke Harris saw the results of this firehouse effect in her patients all the time. One day at the Bayview clinic, she introduced me to one of them, a teenager named Monisha Sullivan who had first come to the clinic when she was sixteen and a new mother. Monisha’s childhood was about as stressful as they come: She was abandoned just a few days after she was born by her mother, who was a heavy user of crack cocaine and other drugs. As a child, Monisha lived with her father and her older brother in a section of Hunters Point with a lot of gang violence until her father, too, got lost in a drug habit; when Monisha was ten, she and her brother were removed from their home by the city’s child protection bureau, separated from each other, and placed in foster care. Ever since, she had been ricocheting through the system, staying for a week or a month or a year in each foster or group home until, inevitably, tensions escalated over food or homework or TV, and she ran away or her caregivers gave up. Then it was on to another placement. In the previous six years, she had cycled through nine different homes.
When I met Monisha, in the fall of 2010, she had just turned eighteen, and three days earlier, she had been emancipated from the foster-care system in which she had spent almost half her life. Her most painful experience, she told me, was the day she was placed in foster care. Without any warning, she was pulled out of class by a social worker she had never met and driven to a strange new home. It was months before she was able to have any contact with her father. “I remember the first day like it was yesterday,” she told me. “Every detail. I still have dreams about it. I feel like I’m going to be damaged forever.”
As we sat in the therapy room at the clinic, I asked Monisha to describe for me what that damage felt like. She is unusually articulate about her emotional state—when she feels sad or depressed, she writes poems—and she enumerated her symptoms with precision. She had insomnia and nightmares, she said, and at times her body inexplicably ached. Her hands sometimes trembled uncontrollably. Her hair had recently started falling out, and she was wearing a pale green headscarf to cover up a thin patch. More than anything, she felt anxious: anxious about school, anxious about her young daughter, anxious about earthquakes. “I think about the weirdest things,” she said. “I think about the world ending. If a plane flies over me, I think they’re going to drop a bomb. I think about my dad dying. If I lose him, I don’t know what I’m going to do.” She was even anxious about her anxiety. “When I get scared, I start shaking,” she said. “My heart starts beating. I start sweating. You know how people say ‘I was scared to death’? I get scared that that’s really going to happen to me one day.”
The firehouse metaphor might help us understand what was happening with Monisha Sullivan. When she was a child, her fire alarm went off constantly, at top volume: My mom and stepmom are punching each other; I’m never going to see my dad again; no one’s home to make me dinner; my foster family isn’t going to take care of me. Every time the alarm went off, her stress-response system sent out all the trucks, sirens blaring. The firefighters smashed in some windows and soaked some carpets, and by the time Monisha turned eighteen, her biggest problem wasn’t the threats that she faced from the world around her. It was the damage the firefighters had done.
When McEwen first proposed the notion of allostatic load, in the 1990s, he didn’t conceive of it as an actual numerical index. But recently, he and other researchers, led by Teresa Seeman, a gerontologist at UCLA, have been trying to “operationalize” allostatic load, to produce a single number for each individual that would express the damage that a lifetime of stress management had imposed. Doctors use comparable biological-risk indicators all the time today, most notably blood pressure measurements. Those numbers are obviously useful as predictors of certain medical conditions (which is why your doctor takes your blood pressure every time you visit his or her office, no matter what ailment you might be there for). The problem is, blood pressure readings alone are not precise measures of future health risks. A more accurate allostatic-load index would include not just blood pressure and heart rate but other stress-sensitive measures: levels of cholesterol and high-sensitivity C-reactive protein (a leading marker for cardiovascular disease); readings of cortisol and other stress hormones in the urine and of glucose and insulin and lipids in the bloodstream. Seeman and McEwen have shown that a complex index including all those values would be a much more reliable indicator of future medical risk than blood pressure or any other single-factor measure in use today.
It’s an attractive and fascinating notion, and a slightly frightening one: a single number that a doctor could give you in, say, your early twenties that would reflect both the stress you had experienced in life to that point and the medical risks that you now faced as a result of that stress. In some ways it would be a more refined version of your ACE score. But unlike your ACE score, which relies on your own report of your childhood, your allostatic-load number would reflect nothing but cold, hard medical data: the actual physical effects of childhood adversity, written on your body, deep under your skin.
6. Executive Functions
As a medical doctor, Burke Harris was initially interested in the physiological effects that early trauma and unmanaged stress had on her patients: Monisha’s trembling hands and hair loss and unexplained pains. But Burke Harris quickly realized that these forces had an equally serious impact in other aspects of her patients’ lives. When she used a modified version of the Felitti-Anda ACE questionnaire with more than seven hundred patients at her clinic, she found a disturbingly powerful correlation between ACE scores and problems in school. Among her patients with an ACE score of 0, just 3 percent had been identified as having learning or behavioral problems. Among patients with an ACE score of 4 or higher, the figure was 51 percent.
Stress physiologists have found a biological explanation for this phenomenon as well. The part of the brain most affected by early stress is the prefrontal cortex, which is critical in self-regulatory activities of all kinds, both emotional and cognitive. As a result, children who grow up in stressful environments generally find it harder to concentrate, harder to sit still, harder to rebound from disappointments, and harder to follow directions. And that has a direct effect on their performance in school. When you’re overwhelmed by uncontrollable impulses and distracted by negative feelings, it’s hard to learn the alphabet. And in fact, when kindergarten teachers are surveyed about their students, they say that the bigge
st problem they face is not children who don’t know their letters and numbers; it is kids who don’t know how to manage their tempers or calm themselves down after a provocation. In one national survey, 46 percent of kindergarten teachers said that at least half the kids in their class had problems following directions. In another study, Head Start teachers reported that more than a quarter of their students exhibited serious self-control-related negative behaviors, such as kicking or threatening other students, at least once a week.
Some of the effects of stress on the prefrontal cortex can best be categorized as emotional, or psychological: anxiety and depression of all kinds. I kept in touch with Monisha in the months after our first meeting, and I saw a lot of those emotional symptoms in her. She was plagued by self-doubt—about her weight, her parenting ability, her prospects in general. She was assaulted one night by an ex-boyfriend, a sketchy character she had invited over, against her better judgment, to stave off her loneliness. And she struggled constantly to cope with a flood of emotions that always seemed on the verge of capsizing her. “Sometimes the stress is just too much for me to bear,” she told me one day. “I don’t see how people deal with it.”
For Monisha, the main effect of stress overload on her prefrontal cortex was that she had a hard time regulating her emotions. For many other young people, though, the main effect of stress is that it compromises their ability to regulate their thoughts. This has to do with a particular set of cognitive skills located in the prefrontal cortex known as executive functions. In wealthy school districts, executive function has become the new educational catch phrase, the most recent thing to evaluate and diagnose. But among scientists who study children in poverty, executive functions are a newly attractive field for another reason: improving executive function seems like a potentially promising vehicle for narrowing the achievement gap between poor kids and middle-class kids.
Executive functions, as we now understand them, are a collection of higher-order mental abilities; Jack Shonkoff, the head of the Center on the Developing Child at Harvard University, has compared them to a team of air traffic controllers overseeing the functions of the brain. Most broadly, they refer to the ability to deal with confusing and unpredictable situations and information. One famous test of executive-function ability is called the Stroop test. You see the word red written in green letters, and someone asks you what color that word is. It takes some effort to stop yourself from saying red, and the skills you’re drawing on when you resist that impulse are executive functions. And those skills are especially valuable in school. We’re constantly asking kids to deal with contradictory information. The letter C is pronounced like a K—unless it is pronounced like an S. Tale and tail sound the same but have different meanings. A zero means one thing on its own and an entirely different thing with a one in front of it. Keeping track of those various tricks and exceptions requires a certain amount of cognitive impulse control, and that is a skill that is neurologically related to emotional impulse control—your ability to refrain from punching the kid who just grabbed your favorite toy car. In both the Stroop test and the toy-car incident, you’re using the prefrontal cortex to overcome your immediate and instinctive reaction. And whether you’re utilizing your self-control in the emotional realm or the cognitive realm, that ability is crucially important to getting through the school day, whether you’re in kindergarten or your senior year of high school.
7. Simon
For a while now, we’ve known that executive-function ability correlates strongly with family income, but until recently, we didn’t know why. Then in 2009, two researchers at Cornell University, Gary Evans and Michelle Schamberg, designed an experiment that for the first time gave us a clear look at exactly how childhood poverty affects executive function. The particular executive-function skill they examined was working memory, which refers to the ability to keep a bunch of facts in your head at the same time. It’s quite distinct from long-term memory—working memory is not about remembering the name of your first-grade teacher; it’s about remembering everything you’re supposed to pick up at the supermarket. The tool that Evans and Schamberg selected to measure working memory was a kitschy one: the electronic children’s game Simon. If you grew up in the 1970s, as I did, you might remember this Hasbro game: it’s a UFO-looking disk about the size of an LP record but fatter, with four panels that light up and make distinct sounds. The panels illuminate in various sequential patterns, and you have to remember the order of the beeps and the flashes.
Evans and Schamberg used Simon to test the working memory of 195 seventeen-year-olds in rural upstate New York, all part of a group that Evans had been studying since they were born. About half the children had grown up below the poverty line and the other half in working- and middle-class families. Evans and Schamberg’s first discovery was that the amount of time that children spent in poverty when they were growing up predicted how well they would do on the Simon test, on average—kids who had spent ten years in poverty, in other words, did worse than kids who had spent just five years in poverty. This, on its own, was not too surprising; researchers had previously found correlations between poverty and working memory.
But then Evans and Schamberg did something new: They introduced some biological measures of stress. When the children in the study were nine years old, and again when they were thirteen, Evans’s researchers took a number of physiological readings from each child, including blood pressure, body mass index, and levels of certain stress hormones, including cortisol. Evans and Schamberg combined those biological data to create their own measure of allostatic load: the physical effects of having an overtaxed stress-response system. When they sat down with all the data and compared each child’s Simon score, poverty history, and allostatic-load reading, they found that the three measures correlated—more time in poverty meant higher allostatic-load numbers and lower scores on Simon. But then came the surprise: When they used statistical techniques to factor out the effect of allostatic load, the poverty effect disappeared completely. It wasn’t poverty itself that was compromising the executive-function abilities of the poor kids. It was the stress that went along with it.
This was, potentially at least, a big deal in terms of our understanding of poverty. Picture two boys sitting together playing Simon for the first time. One is from an upper-middle-class home, and one is from a low-income home. The kid from the upper-middle-class home is doing a lot better at memorizing the patterns. We might be inclined to assume that the reason for this effect is genetic—maybe there’s a Simon gene that rich kids are more likely to possess. Or maybe it has to do with material advantages in the upper-middle-class kid’s home—more books, more games, more electronic toys. Or maybe his school is a better place to learn short-term memory skills. Or perhaps it’s some combination of the three. But what Evans and Schamberg found is that the more significant disadvantage the low-income boy faces is in fact his elevated allostatic load. And if another low-income boy came along with low levels of allostatic load—if, for whatever reason, he had had a less stressful childhood, despite his family’s poverty—he would in all probability do just as well at the Simon competition as the rich kid. And why does a low Simon score matter? Because in high school, college, and the workplace, life is filled with tasks where working memory is crucial to success.
The reason that researchers who care about the gap between rich and poor are so excited about executive functions is that these skills are not only highly predictive of success; they are also quite malleable, much more so than other cognitive skills. The prefrontal cortex is more responsive to intervention than other parts of the brain, and it stays flexible well into adolescence and early adulthood. So if we can improve a child’s environment in the specific ways that lead to better executive functioning, we can increase his prospects for success in a particularly efficient way.
8. Mush
It is in early childhood that our brains and bodies are most sensitive to the effects of stress and trauma. But it is in
adolescence that the damage that stress inflicts on us can lead to the most serious and long-lasting problems. Partly, that’s just a practical fact of growing up. When you have trouble controlling your impulses in elementary school, the consequences are relatively limited: you might get sent to the principal’s office; you might alienate a friend. But the kind of impulsive decisions you are tempted to make in adolescence—driving drunk, having unprotected sex, dropping out of high school, stealing a wallet—can often have lifelong consequences.
What’s more, researchers have found that there is something uniquely out of balance about the adolescent brain that makes it especially susceptible to bad and impulsive decisions. Laurence Steinberg, a psychologist at Temple University, has analyzed two separate neurological systems that develop in childhood and early adulthood that together have a profound effect on the lives of adolescents. The problem is, these two systems are not well aligned. The first, called the incentive processing system, makes you more sensation seeking, more emotionally reactive, more attentive to social information. (If you’ve ever been a teenager, this may sound familiar.) The second, called the cognitive control system, allows you to regulate all those urges. The reason the teenage years have always been such a perilous time, Steinberg says, is that the incentive processing system reaches its full power in early adolescence while the cognitive control system doesn’t finish maturing until you’re in your twenties. So for a few wild years, we are all madly processing incentives without a corresponding control system to keep our behavior in check. And if you combine that standard-issue whacked-out adolescent neurochemistry with an overloaded HPA axis, you’ve got a particularly toxic brew.