فصل 9

9

LIFE IN THE FAST LANE

Jen slams her car door and rushes toward the office building where she works, nearly tripping over a crack in the sidewalk as she checks her watch. She spent forty-five minutes in a traffic jam, and now she’s almost late for a meeting with her boss. She reviews her talking points to herself as she strides down the office hallway. Jen likes her boss—and her job—but today is her yearly performance evaluation, and she’s nervous about it. She arrives just in time for the meeting, which goes about how she expected. The feedback is good overall, but she gets chastised for a careless mistake she made with a client recently. Her job is secure, for now. Although at times stress pushes Jen to perform her very best, at other times, it undermines her performance, saps her quality of life, and drives her to overeat unhealthy food.

Psychological stress is nearly ubiquitous in the modern world. Careers, finances, home ownership, parenthood, chronic health problems, social isolation, and many other life stressors add up to a potent brew for many people. Since 2007, the American Psychological Association has used nationwide surveys to track many aspects of stress and published the results in its yearly Stress in America reports. These reports show that three-quarters of Americans regularly suffer negative physical or psychological effects from stress, such as headaches, fatigue, upset stomach, insomnia, and irritability.

Are we maladapted to our current level of stress because our ancestors led more serene lives than we do today? Although it makes a good story to say that modern cultures suffer from more chronic stress than our distant ancestors did, convincing evidence is hard to come by. It seems to me that current nonindustrial cultures have plenty to worry about, including high infant mortality rates, deadly infectious diseases, accidents, homicide, starvation, witchcraft, and sometimes even predation. Perhaps we’ve lost our traditional coping strategies, like close community bonds and daily physical activity. In any case, whether or not we’re more stressed than our ancestors, stress is a powerful force in the modern world that makes a lot of people miserable.

Given the profound effects stress exerts on many brain systems, it’s no surprise that it impacts eating behavior. What is surprising, however, is that the textbook response to stress is actually a loss of appetite, reduced food intake, and weight loss. This is certainly true for many types of stress, particularly physical stress—such as the flu—and powerful, acute psychological stress resulting from things like a car accident. But the full story is more complicated. This complexity is well illustrated by the 2007 Stress in America report, which includes data on how Americans react to and cope with stress in their lives. One of the most commonly reported effects of stress was a change in eating behavior—79 percent of people reported such a change. However, people reported conflicting effects. Forty-three percent reported overeating in response to stress, while 36 percent reported skipping a meal, and this result has been repeatedly confirmed by other studies. So it appears that stress can cause you to overeat or undereat, depending on who you are and what type of stress you experience.

In my view, this is an extraordinary finding that begs us to dig deeper. It would make sense if we all overate, to varying degrees, in response to stress. But to see diametrically opposed responses in different people is just odd. Solving this puzzle will require us to push into the neuroscience and psychology of stress.

FEARLESS MONKEYS

What is stress? Broadly, it’s a coordinated set of physiological and behavioral responses that the brain engages to meet a challenging situation. These responses limit the damaging consequences of a threatening scenario and make us perform our best when the stakes are high. For many animals, stress is a way of life. From avoiding predators, to finding scarce food, to competing for mates, the ability to mount a fast and effective stress response is critical to survival and reproduction in a natural environment. Because of this, stress responses are deeply wired into many of our brain systems.

Yet stress isn’t just one thing. For example, if you get into a car accident and you’re losing blood quickly, that’s a different kind of stress from when you’re in a meeting with your boss that could end with a pink slip. The first scenario is an example of stress arising from physical damage to your body, while in the second, the stressor is an abstract concept. You’re stressed about the uncertain prospect of losing your job, and all the downstream effects that might have on your future life, such as defaulting on your mortgage or rent, having to explain the situation to your partner, and having to go through the job search rigmarole. This is psychological stress, and it’s the type that affects us the most in affluent nations today. Certain types of psychological stress have the peculiar ability to provoke overeating.

Ultimately, although different types of stress provoke different brain responses, they all engage a partially overlapping collection of brain circuits I’ll call the threat response system. The threat response system activates a number of processes that together serve to protect you from danger and help you make the most of a challenge.

In 1939, an unusual experiment by the German American psychologist Heinrich Klüver and the American neurosurgeon Paul Bucy took the first steps toward unraveling the core brain circuitry of the threat response system. As part of an investigation into the effects of the psychedelic drug mescaline, Klüver obtained a group of rhesus monkeys and arranged for Bucy to surgically remove specific parts of the monkeys’ brains, including the temporal lobe of the cortex, the hippocampus, and an area called the amygdala. He called these surgically transformed animals “bilateral temporal monkeys.” Although the surgery didn’t have much impact on the monkeys’ response to mescaline, it produced pronounced behavioral changes. Most relevant to us is the fact that it made them fearless. According to Klüver and Bucy: The typical reaction of a “wild” monkey when suddenly turned loose in a room consists in getting away from the experimenter as rapidly as possible. It will try to find a secure place near the ceiling or hide in an inaccessible corner where it cannot be seen. If seen, it will either crouch and, without uttering a sound, remain in a state of almost complete immobility or suddenly dash away to an apparently safer place. This behavior is frequently accompanied by other signs of strong emotional excitement. In general, all such reactions are absent in the bilateral temporal monkey. Instead of trying to escape, it will contact and examine one object after another or other parts of the objects, including the experimenter, stranger or other animals … Expressions of emotions, such a vocal behavior, “chattering” and different facial expressions, are generally lost for several months. In some cases, the loss of fear and anger is complete.

These monkeys also failed to mount a normal fear reaction in threatening social situations, leading them to fight with larger, more dominant monkeys and often sustain severe injuries. This odd collection of behaviors was dubbed Klüver-Bucy syndrome. Later studies found that the fearlessness of Klüver-Bucy syndrome could be largely replicated simply by disrupting the amygdala and adjacent brain tissue.

In the seventy-five years since Klüver and Bucy’s seminal findings, researchers have gradually uncovered a complex network of brain structures that play a role in the threat response system. These structures range from the oldest parts of the brain stem to the most recently evolved parts of the frontal cortex, emphasizing how deeply rooted the threat response system is in the brain, and its critical importance to the survival of our ancestors.

At the heart of this network of brain structures lies the amygdala—roughly the size and shape of a large almond. (The amygdala and other closely associated brain areas are often lumped together under the term “extended amygdala.” Unless otherwise stated, when I say “amygdala,” I’m referring to the entire extended amygdala region.) Decades of research in animal models and humans suggest that the amygdala is the central node of the threat response system that responds to external and psychological threats, including the everyday psychological stressors we commonly call “stress.” In broad strokes, here’s how it works: The amygdala cooperates with many different brain regions, both conscious and nonconscious, to scan for signs of a threat. Some of these regions process concrete sensory information, such as objects moving rapidly toward you, things that look like spiders, or loud sounds, while others process abstract concepts like being laid off, carrying debt, or arguing with a loved one. When the amygdala detects a threat, it communicates with other brain regions to activate a coordinated suite of threat responses designed to minimize the potential damaging consequences of the situation. This is illustrated schematically in figure 45.

The amygdala increases your level of arousal by activating many of the same brain regions that are involved in regulating the sleep/wake cycle. This bathes broad swaths of your brain in dopamine, serotonin, noradrenaline, and other chemicals that focus your mind on the problem and motivate you to do what it takes to resolve it. That’s why it’s hard to sleep when you’re stressed.

Depending on the threat, the amygdala may send signals to your brain stem that activate fast defensive reflexes, such as startling, freezing, protecting your head, or closing your eyes. These signals can also instinctively generate facial expressions of fear or anxiety—recognizable across all human cultures.

At the same time, the amygdala sends signals to activate your sympathetic nervous system. The sympathetic nervous system is a network of nerve fibers that runs throughout the body and participates in the fight-or-flight response. Your pulse and breathing rate quicken, and your blood pressure rises. Your palms begin to sweat. Digestion slows, and in extreme cases, your bladder and rectum may expel their contents. Blood flow to your muscles increases. The levels of sugar and fat in your bloodstream begin to climb, providing your muscles with more energy for fight or flight. This process prepares your body for vigorous action, but it happens even in response to psychological threats that don’t require you to meet physical challenges. Your body is preparing to fight off a grizzly bear, even if the actual threat is an Excel spreadsheet.

Simultaneously, the amygdala activates a critical part of the threat response system called the hypothalamic-pituitary-adrenal axis (HPA axis). It does so by sending a signal that causes the hypothalamus to release a chemical called corticotropin-releasing factor (CRF). CRF and a few related molecules turn out to be key players in the threat response.

The release of CRF is the first step in the signaling cascade of the HPA axis, which culminates in your adrenal glands producing the stress hormone cortisol. Some of the effects of cortisol, such as increasing blood levels of sugar and fat, are similar to those of the sympathetic nervous system, but on a slower, longer time scale. These sustain the high metabolic demands of dealing with a stressor for a long time. Other effects of cortisol include suppressing immune function and, as we will soon see, increasing food intake. Because of its slow kinetics, cortisol is a key player in the long-term response to chronic stress.

The amygdala also makes its own CRF, which plays an important role in increasing your arousal level, your anxiety-related behaviors, and the activity of your sympathetic nervous system. CRF stimulates the brain as a whole to shift from normal, everyday behaviors like eating and socializing, to threat response behaviors like running away and (perhaps) thinking about how you’ll pay the bills. To bring it back to the basal ganglia, CRF seems to increase the bid strength of threat-related option generators, and weaken the bid strength of option generators that are unrelated to the threat. CRF is such a powerful mediator of stress that several pharmaceutical companies are investing large amounts of money into trying to block it in humans. The final act of the brain’s threat response is learning. Once you’ve survived a threatening situation, your amygdala learns how to recognize and respond to similar situations more effectively in the future. With practice, your brain gets better at protecting you. But even though the threat response system evolved to keep us safe—and it often does—at times it can do things that seem quite counterproductive, like making us eat too much.

THE EVERYDAY STRUGGLE … OF A RHESUS MONKEY

At the Yerkes National Primate Research Center in Atlanta, Georgia, neuroscience researcher Mark Wilson observes five dun-colored female rhesus monkeys in an enclosure. On one side of the enclosure, two monkeys are seated peacefully, one grooming the other. On the other side, one monkey suddenly slaps another on the side of the head and, a moment later, chases the slapee into a corner, where it cowers until the aggressor relents. The fifth monkey watches the spectacle and then saunters over to a feeding station, where it reaches into a tube in the center of a complicated-looking machine, pulls out a food pellet, and eats it. Unbeknownst to the monkey, there is a tiny electronic tag implanted in its wrist that the feeding station reads each time it reaches into the feeding tube for a pellet. The machine also precisely weighs each pellet, and in this way it accurately registers each monkey’s food intake. These data are helping Wilson and his team understand how, and when, stress makes us overeat.

How can rhesus monkeys teach us about stress eating? Like many animals, rhesus monkeys spontaneously organize into social hierarchies, and dominant monkeys maintain the hierarchy by harassing and occasionally hitting or biting subordinates. That means Wilson’s team doesn’t have to do anything to create stress: The monkeys do it to themselves. “The primary stressor,” explains Wilson, “is just getting harassed, day in and day out.” Most of the harassment in stable social hierarchies is what Wilson calls noncontact aggression—the monkeys are threatening and chasing one another but not making physical contact. This isn’t too far off from the chronic psychological stress most humans experience, where the stressor is most often a threat of harm (e.g., a fear of being laid off) rather than harm itself (e.g., actually being laid off). Even though no one gets hurt, it’s still unpleasant. And it’s uncontrollable.

This lack of control is a key factor in modeling the most harmful types of human psychological stress. In the modern world, we’re often subjected to stressful events we can’t easily control, like traffic, bullying, nagging, illness, deadlines, and debt. Research in psychology and neuroscience suggests that uncontrollable stressors have a stronger effect on the threat response system, and are much more harmful to our health and mental state, than stressors we believe we can control.

How do these uncontrollable, often social, stressors affect food intake? Remarkably, it depends entirely on the food that’s available. When Wilson’s team feeds its monkeys healthy, unrefined, high-fiber chow, stressed subordinate monkeys eat less and lose weight while dominant monkeys maintain weight. Yet when the researchers give the monkeys a choice between standard chow and a very rewarding high-fat, high-sugar diet, the monkeys’ eating behavior changes dramatically. First of all, not surprisingly, both dominant and submissive animals prefer the rewarding diet and eat it at the expense of the healthy diet. Yet the dominant animals keep eating the same amount of food as before. In contrast, the stressed subordinates double their daily calorie intake. So in the context of a strict healthy diet, stress makes monkeys undereat, whereas when they have a choice between healthy fare and junk food, they overeat spectacularly.

Following up on this finding, when Wilson’s team blocks the effects of CRF in the monkeys’ brains, knocking out a key component of the threat response system, the subordinate monkeys stop overeating. This confirms that overeating in stressed monkeys, and perhaps ourselves as well, is a direct result of the activation of the threat response system in the brain. Wilson’s research suggests that the magic formula for overeating is the combination of chronic uncontrollable stress and a choice of highly rewarding food. This specificity may go a long way toward explaining why some people overeat when they’re stressed and others don’t. Each person experiences a different combination of stressor type and food environment, and only some of these combinations are the magic formula.

This may also explain part of the reason why traditionally living cultures don’t seem to overeat and gain weight when they’re chronically stressed, like many of us do. Even though traditional cultures have plenty to be stressed about, their food tends to be much simpler, less refined, and less rewarding. They may be more like the stressed monkeys who don’t overeat because they only have access to healthy, unrefined food. In contrast, many of us are like the stressed monkeys who overeat because we live in an extravagant food environment.

Why does chronic uncontrollable stress push us to overeat, and why does this only seem to happen when there’s highly rewarding food around? To answer these questions, we’ll have to turn to the effects of uncontrollable stress on the endocrine system.

HORMONE HUNGER

In 1910, the neurosurgeon Harvey Cushing saw a twenty-three-year-old patient named Minnie G. who was suffering from a peculiar and distressing disorder. Minnie G. had stopped menstruating, showed excessive hair growth, and most important for our purposes, she had abdominal obesity. Because of the fact that she also had increased fluid pressure in her brain case (hydrocephalus), Cushing suspected that her disorder was caused by a tumor in her pituitary gland that had disrupted her hormone levels. After seeing more patients with a similar ailment, Cushing determined that pituitary tumors were indeed often the cause. These tumors increase the volume of the portion of the gland that secretes adrenocorticotropic hormone (ACTH), a critical part of the HPA axis that mediates part of the body’s stress response. In turn, excess ACTH tells the adrenal glands to secrete too much of the stress hormone cortisol, which causes the disorder that was eventually named Cushing’s disease. More generally, it’s called Cushing’s syndrome when a person shows signs and symptoms of excess cortisol but we don’t know the specific cause. Cushing was unable to perform an autopsy on the brain of Minnie G., so we’ll never know whether or not she actually had a pituitary tumor, but today we do know that most people who spontaneously develop high cortisol levels do so precisely because of such tumors.

Yet there’s another cause of Cushing’s syndrome, which helps us shed more light on the connection between cortisol and overeating: This form is actually caused by medical treatment. Doctors often prescribe drugs similar to cortisol (such as prednisone) due to their remarkable ability to suppress the immune system. This can be useful if a person suffers from severe immune-mediated problems, such as rheumatoid arthritis or organ transplant rejection. Yet when taken at high doses, these drugs cause people to develop the abdominal obesity characteristic of Cushing’s syndrome.

In 1987, Eric Ravussin and his research team set out to learn why this happens. They recruited twenty healthy young men, assigned them to either methylprednisolone (a cortisol-like drug) pills or placebo, and carefully monitored each group’s food intake. Over a four-day period, Ravussin’s team found that the methylprednisolone group ate a whopping 1,687 Calories more per day than the placebo group. This showed unequivocally that engaging a key part of the threat response system, albeit to an abnormally strong degree, causes overeating. Ravussin’s team concluded: Our data suggest that therapeutic doses of [cortisol-like drugs] induce obesity mostly by increasing energy intake, an effect which may be related to the ability of [cortisol-like drugs] to act directly or indirectly on the [brain] regulation of appetite.

To understand why cortisol can have such a profound effect on food intake and adiposity, we need to return to the system in the brain that regulates adiposity and appetite, the lipostat. As we discussed previously, leptin tells the hypothalamus how much fat the body carries, and the hypothalamus uses that signal to regulate food intake and energy expenditure. When the hypothalamus becomes resistant to the leptin signal, adiposity increases.

With this in mind, there’s a tidy explanation for why overeating and obesity are key features of Cushing’s syndrome: Cortisol and related compounds cause leptin resistance in the hypothalamus. In 1997, a Swiss research team led by Katerina Zakrzewska determined that removing cortisol from the circulation of rats makes them exquisitely leptin-sensitive, and also lean (technically, these studies manipulated levels of corticosterone, the rodent equivalent of cortisol). Incrementally increasing their cortisol levels makes the rats incrementally less leptin sensitive and incrementally fatter. Research from other groups showed that cortisol-like compounds interfere with the ability of leptin to activate its signaling pathways in cells of the hypothalamus, and increase levels of the hunger-promoting substance NPY. Together, this research shows that a key component of the threat response system can interfere with a key component of the lipostat, driving the nonconscious parts of the brain that regulate adiposity and appetite to favor overeating and fat gain.

Although this is interesting stuff, we haven’t quite made it from saying that cortisol can cause leptin resistance, overeating, and abdominal fat accumulation in extreme cases, to saying that it does cause these things in people living their everyday stress-filled lives. The latter is much harder to demonstrate, and there is no smoking gun yet. However, there is quite a bit of evidence that’s strongly suggestive.

Several large studies have shown that people with higher stress levels tend to gain more body fat over time than people with lower stress levels—and this fat gain occurs particularly in the abdominal area, resembling a mild form of Cushing’s syndrome (they also exhibit metabolic changes that are reminiscent of Cushing’s). Elissa Epel, professor of psychiatry at the University of California, San Francisco, believes that this phenomenon may be more prevalent than we think. “It’s really insidious and common, and we often don’t see it,” explains Epel, because people can accumulate abdominal fat while still appearing fairly lean overall. These people are at a high risk of health problems because fat inside the abdominal cavity (where your digestive organs are) is more dangerous than fat under the skin, but they may be hard to identify because they aren’t always classified as obese on the body mass index scale that doctors commonly use to diagnose obesity. At any level of overall adiposity, chronic stress seems to shift the distribution of body fat toward the visceral cavity, where it tends to wreak metabolic havoc.

Epel’s group has found that some people secrete high levels of cortisol in response to experimental stressors under lab conditions, while others don’t secrete much cortisol at all. The high secretors tend to eat more food when they’re stressed, while the low secretors don’t. Together, this paints a fairly coherent picture suggesting that cortisol may be a key reason why everyday stress can cause overeating and fat accumulation around the midsection. And it also helps explain why only certain people overeat in response to stress.

As it turns out, uncontrollable stress—like being hassled all day by four rhesus monkeys or your boss—has a particularly potent cortisol-raising effect. In contrast, when you’re facing a challenge but you have a chance to determine your own fate, the situation feels less threatening, and your cortisol response is proportionally smaller. This may be part of the reason why uncontrollable stress is the most effective at driving overeating and adiposity.

Let’s return to Jen’s story, which we began at the opening of the chapter. Before and during her workday, Jen’s amygdalae were powerfully activated by brain regions that understand the abstract concepts of traffic jams, being late for work, and performance evaluations—and all their potential consequences. As her amygdalae stimulated her sympathetic nervous system and HPA axis, her heart rate accelerated, her blood levels of sugar and fat increased, her palms began to sweat, and she became more alert. Jen doesn’t know it, but she tends to secrete a lot of cortisol when she’s stressed—particularly when the stressor feels outside of her control. On her way to work, she faced a maddening traffic jam, and there was nothing she could do but wait it out. When she walked into her boss’s office, she didn’t know what he was going to say, but again there wasn’t much she could do to change the outcome. In the end, she made it to work on time, and the evaluation went fine, but by that time her cortisol levels were soaring. This cortisol traveled to her brain, causing her hypothalamus to become less sensitive to the appetite-restraining effects of leptin. At her meals that day, she noticed that her appetite was unusually large. Each time this happens, Jen gains a bit of weight, especially around her midsection.

Yet there’s another, more common-sense reason why some of us overeat when we’re stressed, and research is increasingly suggesting that it could be important: Junk food simply makes us feel better emotionally.

THE COMFORT FOOD CONNECTION

“When I first got into HPA axis research,” reminisces Mary Dallman, professor emeritus of physiology at the University of California, San Francisco, “people didn’t know what the hell was going on.” Dallman waded into physiology in the 1960s—a time when our understanding of HPA axis biology was superficial, and the field was virtually off-limits to women. She began as a technician at Harvard and eventually became the first female research faculty member at UCSF. Her subsequent work played a critical role in uncovering how the HPA axis is regulated—in particular, how it shuts itself off. Yet in recent years, Dallman’s research has expanded from studying basic HPA axis biology to include matters of more direct relevance to human health. As is often the case in the research world, Dallman explains that “almost everything I’ve gone after has been from an accidental observation.” One of these came from her husband, Peter Dallman, who happened to be a pediatric hematologist, a doctor who treats blood disorders in children. He frequently used dexamethasone, a cortisol-like drug, to treat specific blood disorders that benefited from suppressing the immune system. “He said the first thing you see when you give dexamethasone,” recalls Mary Dallman, “is that the kids would start eating like crazy—and you knew the treatment was going to work.” This led her to study the effect of cortisol and related compounds on calorie intake in rodents.

Although this was a productive line of research, Dallman knew there must be more to the story. Human studies were showing that stress doesn’t just change the amount of food we eat; it also markedly alters the types of food we eat. Depending on how you’re wired, calorie intake can go up or down during stress, yet regardless of this, most of us gravitate toward calorie-dense “comfort foods” like chocolate, ice cream, macaroni and cheese, potato chips, and pizza. Dallman suspected that people might actually be using food to self-medicate their stress, dampening the activity of their own threat response systems.

To test this hypothesis, Dallman’s team turned to a tried-and-true model of stress in rodents called restraint stress. In this model, researchers place rodents in a small confined space for a short period of time, which activates their threat response systems. Researchers can then measure this activation by looking for increased activity of the sympathetic nervous system and the HPA axis, as well as defensive behaviors such as freezing and hiding.

To study whether tasty foods can dampen the stress response, Dallman’s team gave one group of rats unlimited access to sugar water for ten days, while another group only got plain water. At the end of this period, researchers subjected both groups to restraint stress and then measured the stress-induced activation of their HPA axis. As predicted by Dallman’s comfort food hypothesis, the group that drank sugar water showed a smaller stress response than the group that drank plain water. It appeared that sugar helped them feel better in the face of stress. Yet it’s not just sugar: Dallman and others have since shown that giving rats access to a high-fat food does the same thing.

What is it about comfort food, exactly, that dampens the stress response? Is it the body’s metabolic response to sugar, fat, and/or calories, or is it the food’s reward value itself? Yvonne Ulrich-Lai, a neuroscience researcher at the University of Cincinnati, set out to answer this question. She and her team offered rats intermittent access to small volumes of sugar water, water sweetened with the calorie-free sweetener saccharin, or plain water, and then measured the activation of their HPA axis following restraint stress. She hypothesized that sugar’s stress-busting effect is due to its metabolic impact on the body, not its rewarding effects in the brain, and therefore that saccharin would be ineffective. “The result,” explains Ulrich-Lai, “was completely opposite my hypothesis. The saccharin worked just as well as sugar.” Subsequent experiments confirmed that the sweet taste itself was responsible.

In 2010, her team went further by seeing if rewarding behaviors besides eating also dampen stress responses. To do this, they deployed the only natural reward that can compete with tasty food: sex. Ulrich-Lai’s team gave one group of male rats daily access to a sexually receptive female and let another group of males see and smell a female but not touch her. Then they tested the rats’ stress responses to restraint stress. The sex worked. In fact, according to Ulrich-Lai, it was a little bit more effective than the sugar.

Peering into the rats’ brains, her team was also able to determine that natural rewards may attenuate stress by changing how the amygdala processes stress-related information. Together, Ulrich-Lai’s results suggest that rewarding behaviors directly oppose the activation of the threat response system in the brain. Although the finding still needs to be confirmed in humans, it offers a compelling explanation for why we seek highly rewarding junk food (and drugs like alcohol) when we’re stressed: The reward value itself helps us feel better by dampening the activity of the threat response system. It may also explain why stress only causes overeating when there’s highly rewarding food around: When we want to self-medicate our stress, bland food just doesn’t cut it.

CONSTRUCTIVE COMFORT

Ulrich-Lai’s findings have a valuable practical implication: The brain contains a natural stress relief pathway that we can tap into using a variety of everyday behaviors. “If we can understand that [built-in] stress relief pathway,” explains Ulrich-Lai, “then we might be able to find other ways to target it, either behaviorally or using drugs.” Why would we want to find other ways to target it besides comfort food? Because eating calorie-dense comfort food too often can make us gain weight and degrade our health. If we might gain the same stress-busting benefit from any rewarding behavior, then why not choose something more constructive, like calling a friend, jogging, gardening—or romance?

Let’s return to Jen’s story once again. Like the rest of us, Jen doesn’t like to feel stressed, and she takes steps to manage it. She knows that when she’s stressed, eating something tasty helps her feel better. On her way home from work, she stops by the grocery store to buy food for the next few days. In addition to the (mostly healthy) food she puts into her cart, she grabs a box of chocolate chip cookies. Between her heightened hunger and her desire for comfort food, she ends up eating a third of the box—far more calories than she needs. She resolves that the next time she feels that way, she’ll take a walk or a bubble bath instead.