Move Your Mood

BILL DIDN’T KNOW he was missing out. When he turned fifty, he realized he was twenty pounds overweight, and he decided to go on a diet and start running. Before long, he began losing weight, and he noticed some dramatic side effects: he became less critical — of himself and of others — and less of a grouch. His wife and his children picked up on the shift and wanted to spend more time with him, which made him feel good and further improved his attitude. Bill had never had depression, per se, but there was no doubt he felt more passionate about life once he got into an exercise routine. He discovered, entirely by accident, that he could be happier.

Our understanding of depression followed a similar path. Pure serendipity led to our first antidepressants, when in the 1950s it was discovered that an experimental tuberculosis drug made people “inappropriately happy.” A few years later, a new antihistamine produced similar mood-elevating effects and spawned a class of drugs called tricyclic antidepressants. Suddenly there were medical treatments that could soothe symptoms of depression. It was the first real hint at the radical notion that there might be a biological explanation for what had been considered an entirely psychological problem. This led to a search for how the brain controls the mind, and the entire landscape of the field changed.

In the fifty years since then, mood disorders have been the focal point of psychiatric research. We still don’t know what causes depression, but we’ve made great strides in describing the brain activity underlying emotions. And the more we’ve learned about the biology of mood, the more we’ve come to understand how aerobic exercise alters it. In fact, it’s largely through depression research that we know as much as we do about what exercise does for the brain. It counteracts depression at almost every level.

In Britain, doctors now use exercise as a first-line treatment for depression, but it’s vastly underutilized in the United States, and that’s a shame. According to the World Health Organization, depression is the leading cause of disability in the United States and Canada, ahead of coronary heart disease, any given cancer, and AIDS. About 17 percent of American adults experience depression at some point in their lives, to the tune of $26.1 billion in health care costs each year. It’s impossible to know how many people try to commit suicide, but, tragically, in the United States someone succeeds about every seventeen minutes. For this reason, and also because 74 percent of depression patients experience some other disorder — including anxiety, substance abuse, and dementia — it’s an urgent problem. Unfortunately, it’s not getting any better.

One of the hurdles to conquering depression is that the disorder encompasses such a broad array of symptoms, most of which all of us experience at some point. Who doesn’t feel grouchy, irritable, pessimistic, lethargic, apathetic, self-critical, or melancholy on occasion? Sadness, for instance, is a normal aspect of the human condition — a response to a loss. But being sad isn’t the same as being depressed, unless the feeling never goes away or comes along with a certain number of the other symptoms.

And what’s the difference between a symptom and a personality trait? My patient Bill had been critical and negative most of his life. He wasn’t sick, technically, but he had what I would call a shadow syndrome of depression and was a perfect candidate for a lifestyle change to benefit his Eeyore-like outlook. Prescribing medicine to make people “better than well” is the subject of a long-standing ethical debate, and this is one area where exercise offers a huge advantage over antidepressants. Just because you don’t have all the symptoms of depression doesn’t mean you can’t feel better. Bill’s a happier guy because he started running. As I’ll explain, the same would probably hold true if he were clinically depressed. Aerobic exercise has a positive impact on the entire range of depressive symptoms, regardless of whether they come individually, in the form of a mild episode, or conspire to form a disorder. Overall, I think of depression as an erosion of connections — in your life as well as between your brain cells. Exercise reestablishes those connections.

Within the spectrum of symptoms, there are distinctly different types of depression. I’ve had patients who don’t eat and can’t sleep, and others who eat too much and are so fatigued they feel like they can’t get out of bed in the morning. Some can’t make even the simplest decision, and they quietly withdraw from the world in a posture of helplessness, while others shout and challenge everything and everyone. Such contradictions make treatment difficult. If you have breast cancer, a biopsy can determine the best treatment. If you have depression, you may take a psychological test, and then it’s a matter of trial and error to find a drug that might work; there is no blood test for depression.

Which brings us back to the quest to find a biological culprit. By reverse engineering our original, accidental antidepressants, we discovered that they increase the activity of the so-called monoamine neurotransmitters: norepinephrine, dopamine, and serotonin. When a research psychiatrist at Massachusetts Mental Health Center named Joseph Schildkraut found in 1965 that a breakdown product of norepinephrine, called methoxyhydroxyphenylglycol (MHPG), was reduced in depressed patients, we got excited about the idea that there was something to measure. If we could quantify the imbalance, we should be able to diagnose and attack the disease at the fundamental level of biology. His pioneering work led to the monoamine hypothesis, which holds that depression is caused by a deficit of these three neurotransmitters. Most of our treatments and research since then has been about trying to reverse that deficit.


In 1970, fresh out of college, I landed a job at the Massachusetts Mental Health Center and walked right into this sea change happening in psychiatry. Schildkraut was a mentor to me, and I was lucky enough to have a firsthand look at the scientific inquiry into the biological theory of mood disorders. Two years later, I was off to medical school at the University of Pittsburgh, where I started my own daily psychoanalysis and steeped myself in the emerging brain science. Everyone at Pitt was already working on MHPG, so I settled for measuring lithium uptake into red blood cells as a possible way of identifying different mood problems. I also froze urine samples of patients with schizophrenia, which were then sent off to Linus Pauling at Stanford; learned how to program computers to conduct data analysis; and presented findings at a psychophysiology conference. Through my research, I was utterly caught up in the passion of trying to make psychiatry a “real” science.

Around the same time, I came across an article about a hospital in Norway that was offering depressed patients the option of treatment with antidepressants or daily exercise. This was a shock: These drugs were really just being introduced, and the results were forcing us to change the way we thought about treatment, yet here was a hospital offering exercise for people with severe depression. And it was working! But the results were lost in the shuffle. At a time when we were just plumbing the depths of the brain, the need was for hard science.

When I settled in Boston for my residency at the Massachusetts Mental Health Center, I landed in another epicenter — the running craze — just as it was taking off. We had Olympic gold medalist Frank Shorter competing against the world’s best runners in his hometown marathon; we had Bill Rogers telling everyone to get out there; and we had a new phenomenon called the “endorphin rush.”

A postdoctoral neuroscientist from Johns Hopkins University named Candace Pert had recently discovered that there were opiate receptors in the brain, meaning the body had a built-in way of killing pain with molecules that worked like morphine. Endorphins, as they became known, dulled pain in the body and produced euphoria in the mind. When elevated levels of endorphins were detected in the blood samples of a group of runners, everything seemed to fit. The theory that exercise fills your brain with this morphinelike substance matched the good feeling everyone got. It gave us the expression “runner’s high,” an extreme version of the effect. For me, it was the first time I connected the idea of exercise and mood.

Endorphins are considered stress hormones — and there are forty types of them, with receptors throughout the brain and body — that calm the brain and relieve muscle pain during strenuous exercise. They are the elixir of heroism, helping us ignore pain when we’re physically overextended so we can finish the task at hand. Robert Pyles, the psychiatrist I mentioned in chapter 3, offers a good example. As a marathoner, he prided himself on always finishing, but that became a mighty challenge one year at Boston. His feet got tangled in a plastic trash bag someone had been using as a warm-up jacket near the starting line, and he hit the pavement, kneefirst. He popped back up and kept going, more shocked than anything. But as the miles ticked by, his stride started to feel funny, and by mile eighteen his swollen knee stopped working. He had fractured his femur. Every pounding step should have caused debilitating pain, but Pyles says he didn’t notice it. It had to be the endorphins.

Pain is related to depression, and after Pert’s discovery, others conducted experiments to see if endorphins were indeed the link between exercise and elevated mood. They expected to find that endorphin-blocking drugs would prevent runner’s high, but there were conflicting results. Then we found that endorphins produced in the body — the ones detected in the runners — cannot pass into the brain, and scientific enthusiasm for the endorphin rush faded. Endorphins obviously weren’t the single answer, so they were abandoned in the lab. Now we’re coming back around to them. Studies suggest that endorphins produced directly in the brain contribute to the general feeling of well-being that usually comes along with exercise. The truth is, how much they contribute to the magic is uncertain.

The problem with the strictly biological interpretation of psychology is that we sometimes lose sight of the fact that the mind, brain, and body all influence one another. In addition to feeling good when you exercise, you feel good about yourself, and that has a positive effect that can’t be traced to a particular chemical or area in the brain. If you’ve been feeling down and you start to exercise and feel better, the sense that you’re going to be OK and that you can count on yourself shifts your entire attitude. The stability of the routine alone can dramatically improve your mood. Clearly, there’s something going on.

One of the best examples is a landmark research project from the Human Population Laboratory in Berkeley called the Alameda County Study. Researchers tracked 8,023 people for twenty-six years, surveying them about a number of factors related to lifestyle habits and healthiness starting in 1965. They checked back in with the participants in 1974 and in 1983. Of all the people with no signs of depression at the beginning, those who became inactive over the next nine years were 1.5 times more likely to have depression by 1983 than their active counterparts. On the other hand, those who were inactive to begin with but increased their level of activity by the first interval were no more likely to be depressed by 1983 than those who were active to begin with. In other words, changing your exercise habits changes your risk for depression.

Several other sweeping studies have looked at the correlation from slightly different angles, and all of them came to the same conclusion. A massive Dutch study of 19,288 twins and their families published in 2006 showed that exercisers are less anxious, less depressed, less neurotic, and also more socially outgoing. A Finnish study of 3,403 people in 1999 showed that those who exercise at least two to three times a week experience significantly less depression, anger, stress, and “cynical distrust” than those who exercise less or not at all. This was a cardiovascular risk factor survey that included questions about mood, which is to say they were talking about a broader range of symptoms than clinical depression. Another study, from the epidemiology department at Columbia University published in 2003, surveyed 8,098 people and found the same inverse relationship between exercise and depression.


When the blockbuster drug Prozac came along, it was the first antidepressant that corrected the chemical imbalance of just one of the suspect neurotransmitters. Prozac is the mother of a class known as selective serotonin reuptake inhibitors (SSRIs), which prevent serotonin from being recycled at the synapse, leaving more available for use, and theoretically restoring normal transmission in the brain. Prozac was exciting because it worked for a lot of people and it pointed to a single problem that could be fixed. It had a huge impact because it didn’t just squelch negativity, it boosted self-esteem, which is a different dimension of the disease.

Twenty years later, it’s evident that Prozac and its SSRI daughters don’t work for everyone, nor do antidepressants that target norepinephrine, dopamine, or any combination of the three. One of the issues is side effects. To take an example, a good portion of my patients on SSRIs develop problems with their sex lives after a few months. Some estimates suggest that more than 50 percent complain of sexual side effects, ranging from lack of interest to lack of functionality. (It says something that SSRIs are often used as a treatment for premature ejaculation and for sexual offenders.) Sexual problems are easy for people to miss or dismiss, especially when they’re otherwise feeling good, but they can sneak up on us and lead to other issues. Sexual feelings and passions are primary drivers in all of us, and muzzling them can leave us with a general lack of passion for life or a lack of intimacy or maybe a list of missed opportunities. On balance, the side effects are certainly outweighed by the most serious consequences of depression, but they get in the way for a lot of people. Now SSRIs carry a warning that they may increase the risk of suicidal thoughts and actions in children and adolescents, a finding that is still in question. And stories are coming out about the difficulty of withdrawing from this class of drugs, especially from venlafaxine (Effexor).

Recently, I started treating a successful entrepreneur whose life was a mess. He and his wife were separated because he’d had an affair, and he’d also lost his business. He came to me to find out more about an issue that had cropped up in couple’s therapy, where it became clear that he suffered from ADHD.

Dead set against the idea of putting anything “unnatural” into his body, he resisted medication. But he finally agreed to try a stimulant, mostly because his wife was pressuring him, and he felt terribly guilty about cheating. We tried several different drugs, but we quickly stopped all of them. He got headaches, stomachaches, muscle pain.

His attention problems aside, the primary issue, I told him, was that he was depressed. He was inactive, unmotivated, and he felt hopeless. He had done nothing to sort out his work situation, and although it had been months, he denied there was a problem. Then one day, he came in, and I could see his demeanor had deteriorated. Typically very neat, he was unshaven and unkempt and told me he was having a hard time getting out of bed.

I insisted that he try an antidepressant, and I put him on the SSRI escitalopram (Lexapro). He had a severe reaction — he felt nauseated and started throwing up — and said he didn’t want to try another.

He’d been physically active in the past, and I told him that he should really be exercising every day. I’d been talking about it all along, but after the Lexapro episode, I explained what an enormous effect exercise has on the brain, and I gave him several relevant studies in an appeal to his professional sensibility.

Two weeks later, he looked like a different man. He was smiling and confident, and he felt good about the fact that he had been running nearly every day. Over the course of the next month, I saw him take up his job search in earnest and make strides in reconciling with his wife. And for the first time, he said he was hopeful that they might get back together. More than anything, he was amazed that he could feel so different and sustain that feeling.

Aside from elevating endorphins, exercise regulates all of the neurotransmitters targeted by antidepressants. For starters, exercise immediately elevates levels of Schildkraut’s favorite neurotransmitter, norepinephrine, in certain areas of the brain. It wakes up the brain and gets it going and improves self-esteem, which is one component of depression.

Exercise also boosts dopamine, which improves mood and feelings of wellness and jump-starts the attention system. Dopamine is all about motivation and attention. Studies have shown that chronic exercise increases dopamine storage in the brain and also triggers the production of enzymes that create dopamine receptors in the reward center of the brain, and this provides a feeling of satisfaction when we have accomplished something. If the demand is there, the dopamine genes get activated to produce more, and the overall effect is a more stable regulation of these pathways, which are important to controlling addictions.

Serotonin is equally affected by exercise, and it’s important for mood, impulse control, and self-esteem. It also helps stave off stress by counteracting cortisol, and it primes the cellular connections in the cortex and hippocampus that are important for learning.


We’ve known for a while that exercise influences the same chemicals as antidepressants do, but nobody had done a scientifically sound head-to-head comparison until researchers at Duke University took up the task in 1999. In a landmark study affectionately called SMILE (Standard Medical Intervention and Long-term Exercise), James Blumenthal and his colleagues pitted exercise against the SSRI sertraline (Zoloft) in a sixteen-week trial. They randomly divided 156 patients into three groups: Zoloft, exercise, or a combination of the two. The exercise group was assigned to supervised walking or jogging, at 70 to 85 percent of their aerobic capacity, for thirty minutes (not including a ten-minute warm-up and a five-minute cool-down) three times a week. The results? All three groups showed a significant drop in depression, and about half of each group was completely out of the woods — in remission. Another 13 percent experienced fewer symptoms but didn’t fully recover.

Blumenthal concluded that exercise was as effective as medication. This is the study I photocopy for patients who are skeptical of the idea that exercise changes their brain chemistry enough to help their depression, because it puts the issue in terms that are as black-and-white as psychiatry can hope to deliver, at least for now. The results should be taught in medical school and driven home with health insurance companies and posted on the bulletin boards of every nursing home in the country, where nearly a fifth of the residents have depression. If everyone knew that exercise worked as well as Zoloft, I think we could put a real dent in the disease.

Reading between the lines of the SMILE study gets at the complex issues that have kept exercise from being accepted as a medical treatment. Just as with Andreas Broocks’s comparison of exercise to the antianxiety drug clomipramine in 1997, although both groups ended up at the same point of improvement, the patients taking medication experienced more immediate relief. At first glance, this seemed to contradict the cautions from pharmaceutical companies that antidepressants might take three weeks to work. But those time estimates are based on statistics; I have had plenty of patients over the years who have responded within a few days.

Conversely, what about the studies showing that a single bout of exercise can improve mood? For instance, in 2001, Northern Arizona University psychology professor Cheryl Hansen showed that as little as ten minutes of exercise can immediately improve vigor and mood in healthy subjects. But if Hansen had reexamined mood several hours later, she probably would have found that it had returned to baseline. So while it’s important for us to realize that we can lift our mood with one session of exercise, it’s also important to keep in mind that it takes longer to change our mood day to day.

In Blumenthal’s study, he assessed mood once a week, before exercise. He did see immediate improvement in some patients, but not as dramatically as with medicine. A crucial aspect of recovering from depression is being able to predict that you’re still going to feel good five minutes from now, and then five hours from now. Eventually you’ll be confident that you’ll feel better tomorrow morning. That might take a little longer with periodic exercise.

Six months after the study, Blumenthal and his colleagues surveyed the patients to see how they were doing and found that exercise worked even better than medicine over the long term. About 30 percent of the exercise group remained depressed versus 52 percent of those on medication and 55 percent for those in the combined treatment group. (Blumenthal has some interesting theories about what happened in the combo group, which I’ll explain in a bit.) And of the patients who were in remission after the initial study, just 8 percent of the exercise group had a relapse versus 38 percent in the medication group — quite a significant difference.

After the initial four-month trial, the subjects were free to seek treatment (or not) however they chose, which complicated the results. Some signed up for psychotherapy, some in the medication group began exercising, and some in the exercise group started taking medicine, which brings a lot of variables into play. But Blumenthal’s group found that the most significant predictor of whether someone felt better was how much they exercised. Specifically, every fifty minutes of weekly exercise correlated to a 50 percent drop in the odds of being depressed. Blumenthal stopped short of concluding that exercise caused remission. Maybe the reverse was true: patients who continued to exercise did so because they were less depressed. It’s the classic chicken-egg problem scientists face when correlating physical activity and mood. Does it really matter whether you’re less depressed because you exercise or whether you exercise because you’re less depressed? Either way, you feel better.

But how to account for the combination group’s unexpected results? Blumenthal assumed those who exercised and took Zoloft would show the best results, but they had the worst relapse rates. He speculated that some of the participants didn’t like the idea of antidepressants, especially considering they had expressly signed up for a study to test the effects of exercise on depression. Several were disappointed when they found out they would also receive medication, and during the trial some said they thought the medication was interfering with the benefits of exercise. That’s unlikely from a physiological standpoint, but it’s possible that taking medication undercut the sense of self-mastery that exercise confers. As Blumenthal explained in the study, “Instead of incorporating the belief ‘I was dedicated and worked hard with the exercise program; it wasn’t easy, but I beat this depression,’ patients might incorporate the belief ‘I took an antidepressant and got better.’ ”


When we talk about depression, we don’t use the word cure because we only have subjective measures for behavior and emotions. About a third of patients with depression achieve full remission of their symptoms with antidepressants. Another third feel much better with medication, but they may continue to have problems with motivation, lethargy, and fatigue. The bad thoughts are gone, but even though they’re able to get out of bed, they’re still not knocking on the door for a new job or doing what they should be doing. They’re feeling less than well, lingering in the shadows of depression. The current Diagnostic and Statistical Manual of Mental Disorders lists nine symptoms for depression, and you need to have six to receive a diagnosis of depression. Say you can’t concentrate, can’t sleep, feel worthless, and aren’t interested in anything. That’s four. Technically, you aren’t depressed. What are you, then? Just miserable? My point is, with any degree of depression, you need to snuff it out completely. And exercise is starting to be taken very seriously in this regard.

Madhukar Trivedi, a clinical psychiatrist who is the director of the Mood Disorders Research Program at the University of Texas Southwestern Medical School, has been researching the effectiveness of using exercise to augment antidepressants. In 2006 he published a pilot study showing that patients who weren’t responding to antidepressants lowered their scores on a common depression test by 10.4 points on a 17-point scale — a huge drop — after twelve weeks of exercise. All seventeen patients were deeply depressed and had been taking antidepressants for at least four months. The drugs weren’t working, but they stayed on them for the experiment.

Trivedi enlisted the Cooper Institute to set up the exercise protocol. Then they let the patients, who were inactive to begin with, exercise at home, either walking or riding a stationary bike as often and as rigorously as they wanted. The only requirement was that they expend a certain amount of energy each week. Most chose to walk an average of fifty-five minutes three days a week. Nine people dropped out, which isn’t unusual, but of the eight who completed the exercise program, five achieved full remission. Even the patients who stuck with the program for only a few weeks showed an improvement in symptoms.

The numbers were small, but Trivedi’s results were huge. At least for some people, when drugs don’t work, exercise does. It begs the question, why wouldn’t you add exercise right from the beginning? Especially if you’re playing trial and error with a host of other medications? But the allure of the magic pill is powerful, and it takes a long time to overturn attitudes. Just ask T. Byram Karasu, who was in charge of the American Psychiatric Association’s work group on major depressive disorders. He pushed to get the APA to formally adopt exercise in its treatment guidelines for depression and suggested that psychiatrists tell every patient to walk three to five miles a day or do some other type of vigorous exercise. The APA balked, presumably because, while most doctors acknowledge the anecdotal evidence that exercise improves mood, they say there isn’t enough scientific evidence. In this age of deconstructing the brain and unlocking the secrets of the life and death of cells, it’s difficult for psychiatrists to consider such a holistic strategy as exercise a treatment.

Any doctor will tell you that the worst patients are other doctors. Imagine the difficulty, then, in convincing a patient of mine with a medical degree to exercise for her depression. Grace, who has a history of mild bouts of depression, also happens to be a psychiatrist with a sophisticated knowledge of medicine. Even so, we haven’t been able to find an antidepressant without side effects. SSRIs seem to work best for her, but she would try one and then stop because they all made her gain weight very quickly. She’s brilliant and well aware of the biology of exercise — at least some of it — but she just didn’t do it.

Last summer, she injured her back and ended up immobilized in bed for a period. Purely for physical rehabilitation, she started swimming. It was the only thing she could do, and it felt good because the water supported her body and eased her pain. She enjoyed it and began exercising in the pool for three hours a day. Not only did her pain decrease, but she also started seeing some long-lost muscle tone and she felt good about herself.

Then the pool shut down for the winter, her back flared up, and her mood took a dive again. This time she also got angry. Flat on her back with few options for exercise, she started lifting weights — just three-pound dumbbells that she would pump up and down fast enough to get her heart rate up several times a day. Even that minor activity helped, and more important, the experience sparked a shift in her brain and in her mind. I’ve been seeing Grace for years, but there was something about this situation that put exercise into perspective for her.

She’s worked out the kinks in her back and has been very steady with her swimming routine. She says she’s able to think and write more creatively, and she feels a newfound sense of vigor that hasn’t gone unnoticed by her family and friends. Not that it should have surprised her: She recalled that when she was training every day for the tae kwon do team in college, she did some of her best work. Then as a young doctor in Boston, she got into running marathons. Like many people, it was only after she started a family that she fell out of the habit of exercise. “I just got too busy and forgot the benefits of working out,” she says. “Now I feel like I have my brain back.”


It wasn’t until we were able to take good pictures of the brain that we really began to understand how various medical treatments and exercise break us out of the bonds of depression. In the early 1990s, using MRI, we noticed bright patches on brain scans of certain depressed patients. The hyperintensities, as they are known, appeared in the white matter, the portion of densely bundled axons that connect neurons in the gray matter of the cortex. Zooming in further, we found differences in the volume of the cortex — the gray matter was physically shrunken. Gray matter is the thin, wrinkly covering of the brain made up of cells that direct all of our complex functions such as attention, emotions, memory, and consciousness. The MRI scans pointed toward a radical notion: that chronic depression may cause structural damage in the thinking brain.

Related research showed that depressed patients also had measurable changes in the amygdala and the hippocampus, crucial players in the stress response. We knew the amygdala was central to our emotional life, but we were just discovering that the memory center was also involved in stress and depression. In 1996 Yvette Sheline of Washington University in St. Louis compared ten patients with depression to ten healthy controls of the same physical stature and educational background, and she found that the hippocampus of depressed patients was up to 15 percent smaller compared to that of the controls. She also found evidence that the degree of shrinkage was directly related to the length of depression, and this was news. It might explain why so many patients with depression complain of learning and memory trouble, and why mood deteriorates in Alzheimer’s, the neurodegenerative disease that begins with erosion of the hippocampus.

High levels of the stress hormone cortisol kill neurons in the hippocampus. If you put a neuron in a petri dish and flood it with cortisol, its vital connections to other cells retract. Fewer synapses develop and the dendrites wither. This causes a communication breakdown, which, in the hippocampus of a depressed brain, could partly explain why it gets locked into thinking negative thoughts — it’s recycling a negative memory, perhaps because it can’t branch out to form alternative connections.

Neuroimaging has cracked open an entirely new view of the biology of depression. The picture was — and still is — crude and fuzzy, but now positron-emission tomography (PET) scans and functional MRI (fMRI) scans allow scientists to go beyond snapshots and see the brain at work. At the same time, we learned that new nerve cells are born every day in the hippocampus and possibly in the prefrontal cortex — two areas shriveled in depression. The new tools and new discoveries led to a reformulation of the neurotransmitter theory.

Our understanding hasn’t junked the old theory, just expanded it. Now we see depression as a physical alteration of the brain’s emotional circuitry. Norepinephrine, dopamine, and serotonin are essential messengers that ferry information across the synapses, but without enough good connections in place, these neurotransmitters can only do so much. As far as the brain is concerned, it’s job is to transfer information and constantly rewire itself to help us adapt and survive. In depression, it seems that in certain areas, the brain’s ability to adapt grinds to a halt. The shutdown in depression is a shutdown of learning at the cellular level. Not only is the brain locked into a negative loop of self-hate, but it also loses the flexibility to work its way out of the hole.

Redefining depression as a connectivity issue helps explain the wide range of symptoms people experience. It’s not just a matter of feeling empty, helpless, and hopeless. It affects learning, attention, energy, and motivation — disparate systems that involve different parts of the thinking brain. Depression also affects the body, shutting down the drive to sleep, eat, have sex, and generally look after ourselves on a primitive level. Psychiatrist Alexander Niculescu sees depression as a survival instinct to conserve resources in an environment void of hope — “to keep still and stay out of harm’s way,” he wrote in a 2005 article in Genome Biology. It’s a form of hibernation: When the emotional landscape turns wintry, our neurobiology tells us to stay inside. Except that it can last much longer than a season. It’s as if our entire being has said, there’s nothing out there for me, so I may as well quit. Thus the atrophy, the halt of neuroplasticity, neurogenesis, and overall lack of connectedness. It’s no wonder we haven’t been able to define depression as a single problem.


If depression is primarily a communication breakdown, or a loss of the brain’s ability to adapt, it’s very good news for the value of exercise. We learned in the early 1990s that brain-derived neurotrophic factor (BDNF) protects neurons against cortisol in areas that control mood, including the hippocampus. It is the fertilizer that encourages neurons to connect to one another and grow, making it vital for neuroplasticity and neurogenesis. While extremely high levels of cortisol can decrease BDNF, antidepressants and especially exercise do just the opposite. BDNF is the rope in a tug of war between chronic stress and adaptability. The Miracle-Gro molecule became the new serotonin, and we began measuring, blocking, increasing, and otherwise tweaking it in every way imaginable, to see how it affected mood in mice and men.

We can’t ask a rodent if it’s depressed, but we can see how it reacts to inescapable stress. If its feet are shocked, does it attempt to escape or does it freeze? This is the experimental model for learned helplessness, a popular way of describing human depression that implies an inability to cope with adversity and to take the action necessary to survive and flourish. If the lab animal gives up, it’s considered to be depressed.

In one such experiment, when BDNF was injected directly into the hippocampi of mice, they were much quicker to try to escape than their nontreated counterparts. The injection seemed to have the same effect as exercise and antidepressants on the animals’ behavior. Conversely, scientists have bred mice with genes that produce 50 percent less BDNF and found that they don’t respond well to antidepressants, suggesting it’s a necessary ingredient for the drugs to work. Such mice were significantly slower to try to escape stress than their peers with normal BDNF function.

In people, scientists are limited to measuring BDNF in the bloodstream, which, at best, offers a rough approximation of levels in the brain. One study of thirty depressed patients showed that every one of them had lower than normal BDNF levels. In another, antidepressants restored depressed patients’ BDNF levels to normal, and yet another showed that higher levels correspond to fewer symptoms. In postmortem studies of people with depression who died of suicide, their brains had significantly decreased levels of BDNF. Even in healthy people, low BDNF levels have been correlated with personality traits that make them more vulnerable to depression, such as neuroticsm and hostility.

Exercise boosts BDNF at least as much as antidepressants, and sometimes more, in the rat hippocampus. One study showed that combining exercise with antidepressants spiked BDNF by 250 percent. And in humans, we know that exercise raises BDNF, at least in the bloodstream, much like antidepressants do.

As with norepinephrine in the 1960s, BDNF may be the tip of the iceberg. Today, research focuses on BDNF as well as vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF-2), insulin-like growth factor (IGF-1), and all the related chemicals involved in encouraging neuroplasticity and neurogenesis. At the same time, pharmaceutical companies are funding research to mark and measure all these factors, and to map the genes affected by them, so they can figure out how to mimic their actions. BDNF and its neurotrophic brethren are much farther upstream in the neurochemical cascade than serotonin, closer to the source. Ultimately, the genes have to turn on the flow.

The shift from the neurotransmitter hypothesis to the connectivity theory is a move from outside to inside the nerve cell. In addition to working at the synapse, as serotonin does, BDNF turns on genes to produce more neurotransmitters and neurotrophins, puts the brakes on self-destructive cellular activity, releases antioxidants, and provides the proteins used as building material for axons and dendrites. These gene-regulated adaptations of BDNF might go a long way toward explaining the delayed effect of antidepressants.

It can often take three weeks for antidepressants to work. Is it merely a coincidence that the process of neurogenesis — from the time a stem cell is born in the hippocampus until it plugs into the network — takes approximately the same amount of time? Many researchers think not. The latest twist to the connectivity theory is that a shutdown of neurogenesis might be a factor in depression. Some have shown that blocking neurogenesis in rats negates the effect of antidepressants, so it’s a possibility. This could provide an even stronger link between exercise and its antidepressant effect, given that exercise clearly elevates BDNF and its sibling growth factors, and that these are essential to the building process of neurogenesis. Nobody has shown that a deficit of BDNF causes depression, but it’s not for lack of trying. In 1997 Yale psychiatrist Ronald Duman published “A Molecular and Cellular Theory of Depression” in the Archives of General Psychiatry, and ever since he and others have been unraveling the story of BDNF. In 2006 he charted how various treatments influence BDNF, including all available antidepressants, as well as less common forms of treatment such as electroconvulsive therapy (ECT) and transcranial magnetic stimulation (TMS). They all boost BDNF in the hippocampus, and Duman pointed out that ECT, the most effective treatment, raises it by 250 percent.

But how is it that a blunt physical manipulation such as ECT — which sends an electrical current through the brain to induce a seizure — can work like drugs, therapy, and exercise? I think ECT provides a useful metaphor. If we look at depression as a sort of brain lock, then we can see a common thread between these approaches: They are all shocks of a sort. They send sparks flying to change the dynamic in the brain. Some parts are stuck in a constant whir, and others are locked in place. The key, I think, is to wake up the brain and the body so you can pull yourself out of the downward spiral. What makes aerobic exercise so powerful is that it’s our evolutionary method of generating that spark. It lights a fire on every level of your brain, from stoking up the neurons’ metabolic furnaces to forging the very structures that transmit information from one synapse to the next.


While molecular scientists approach brain lock with tools to pick it, Emory University neurologist Helen Mayberg wants to smash it wide open. Several years ago, she tested a radical therapy called deep brain stimulation (DBS), in which she inserted an electrode into the subgenual cortex in a half dozen severely depressed patients, for whom every other form of medical treatment had failed. “These people are stuck,” Mayberg says. “They have an inability to put thought to action because they’re not in gear. We need to find a way, metaphorically, to unstick them.” She used voltage, and the results were profound: All six patients spontaneously described sensations such as a “disappearance of the void” right there on the operating table, the second the electrodes were switched on. Four of them eventually achieved full remission.

The area of the brain Mayberg targeted is the tip of the anterior cingulate, which is the major way station between information coming down from prefrontal cortex and up from limbic system — the landing between upstairs and downstairs in your emotional stairwell. As the seat of executive function, it prioritizes what you pay attention to and indirectly regulates the limbic system, integrating cognitive and emotional signals. If it fails to shift attention from the negative, in the case of depression, you can’t think about anything else. “Not being able to initiate, not being able to think clearly, not being able to care about your family — those are secondary to something that your attention is directed to internally, but that is a false signal,” Mayberg explains. “You can fix it, and then other problems may be more accessible.”

The real goal of DBS is to bring the prefrontal cortex back online so you can use your executive function to tackle the core issues. It frees you up to attack the problem rationally — I’m not a bad person; my children love me; and my life is not beyond repair. This is also one of the effects of exercise. In 2003 a group of German neuroscientists did an experiment with twenty-four people with mild depression who were being treated with drugs, and ten healthy controls. They all took a battery of neuropsychological tests designed to gauge executive function. They repeated the tests after thirty minutes of stationary cycling at 40 percent of their maximum heart rate, then again at 60 percent — low enough to keep lactic acid out of the equation. The depressed patients showed significant improvement on two of the four tests at both intensity levels, which demonstrated that exercise immediately improves the highest form of thinking. Even one bout of exercise was enough to influence the prefrontal cortex. The ten nondepressed volunteers didn’t show much improvement, but then, there was nothing to fix.

Mayberg would be the first to say that executive function is still only part of the story. In comparing PET scans of patients who responded to antidepressants against others who responded to cognitive behavior therapy, she found that the two approaches change activity levels of the limbic system from opposite directions. Antidepressants seem to work through a bottom-up chain of events, meaning the activity begins in the brain stem and ripples through the limbic system until it reaches the prefrontal cortex. This might explain why antidepressants relieve the physical effects first — we feel more energetic before we feel less sad. With cognitive behavioral therapy and psychotherapy, we feel better about ourselves before we feel better physically. Therapy works from the prefrontal cortex down, to modify our thinking so we can challenge the learned helplessness and spring ourselves out of the hopeless spiral.

The beauty of exercise is that it attacks the problem from both directions at the same time. It gets us moving, naturally, which stimulates the brain stem and gives us more energy, passion, interest, and motivation. We feel more vigorous. From above, in the prefrontal cortex, exercise shifts our self-concept by adjusting all the chemicals I’ve mentioned, including serotonin, dopamine, norepinephrine, BDNF, VEGF, and so on. And unlike many antidepressants, exercise doesn’t selectively influence anything — it adjusts the chemistry of the entire brain to restore normal signaling. It frees up the prefrontal cortex so we can remember the good things and break out of the pessimistic patterns of depression. It also serves as proof that we can take the initiative to change something. This paradigm holds true for exercise’s effect on mood in general, regardless of whether we’re depressed or coping with some nagging symptoms. Or even if we’re just having a bad day.

The idea of attacking depression from both directions couldn’t be more important. “Even with the brain stimulation patients, once you get the system back online, they need mental rehabilitation,” Mayberg says. “Part of the initial rehab is simply to get them doing something. The best kind of behavioral therapy is to just go outside, go for a walk. Do something. It doesn’t require you to have elaborate planning. It doesn’t put you into a negative spin. If you do a physical activity, it’s very quickly reinforcing because, before, you couldn’t initiate to make yourself do something.”

If your prefrontal cortex has been offline for a while, you need to reprogram it, and exercise is the perfect tool. You begin to look at the world differently, and you see trees instead of a barren wasteland. When you see yourself moving, that alone is an achievement — proof that you can help yourself.


Science has come a long way since our search for the single culprit began, and the decades’ worth of research generated from the monoamine hypothesis has taught us volumes about the biology of emotions. The closer we get to the cause of depression, the more complex it appears. When we began, everyone was fairly certain that the problem was an imbalance of neurotransmitters at the synapses. Now we know for certain that it’s not so simple.

Ironically, I think this is precisely why exercise has yet to be embraced as a medical treatment. It doesn’t simply raise serotonin or dopamine or norepinephrine. It adjusts all of them, to levels that, we can only presume, have been optimally programmed by evolution. The same goes for exercise’s effect on BDNF, IGF-1, VEGF, and FGF-2, which provide the building material and oversight for the construction of new connections and neurons. In short, exercise affects so many variables in the brain that its nigh impossible to isolate its effect as we’d like — in the name of hard science. But the evidence is there, from the action of microscopic molecules to massive surveys of tens of thousands of people over the years. Yes, exercise is an antidepressant. But it is also much more.

Still, it’s not surprising that about half the patients in studies on exercise and depression drop out. Perhaps because most of them are inactive to begin with, getting started is that much more challenging. This is critical for doctors to keep in mind when recommending exercise. With people who are already feeling hopeless, it’s important to keep expectations reasonable, so as not to reinforce the negative. On the other hand, studies have shown that even people who find the act of exercise inherently unpleasant have a positive mood swing the minute the workout is over. If you know what’s on the other side, it’s easier to push through the difficulty.

Human beings are social animals, so if you’re depressed, it would be ideal to choose a form of exercise that encourages making connections and that can take place outside or in some environment that stimulates the senses. Asking someone to join you in exercise and putting yourself in a new setting will give those newly hatched neurons a powerful reason for being; new connections need to be formed to represent the sensory stimulation. Breaking out of the emptiness that the brain has been locked into provides a sense of purpose and self-worth that evokes a positive future. Once you develop the positive feeling, you need to devote it to something. Then you get the bottom-up motivation and physical boost combined with the top-down reevaluation of yourself. By motivating the body to move, you’re encouraging the mind to embrace life.


The first question people ask me when I suggest exercise as therapy is “How much should I do?” There is no firm answer — especially given depression’s wide spectrum of symptoms and severity — but Madhukar Trivedi has drawn some conclusions about the amount necessary to be effective. By quantifying exercise as a dose, he hopes to cast treatment in terms the medical profession might accept. This is all-important, because doctors have to spend time with their patients to figure out what sort of regimen is likely to keep them moving.

In one study, Trivedi and Andrea Dunn divided eighty depressed patients into five groups, four with exercise protocols of different intensities and frequencies and one control group assigned to supervised stretching only (to see if social interaction with the supervisors had any effect). They used calories burned per pound of body weight as the “dose” measure. The high-intensity groups burned an average of 1,400 calories (eight calories per pound) during the course of either three or five sessions per week. At the end of three months, regardless of frequency, the high-intensity groups cut their depression scores in half. Practically speaking, their symptoms dropped significantly. The low-intensity groups burned an average of 560 calories (three calories per pound) and lowered their scores by a third, about the same as the stretching group — and about as effective as a placebo. What this means for us is what I always tell people: some exercise is good, more is better (to a point).

Trivedi and Dunn based the high dose on public health recommendations for exercise, which suggest thirty minutes of moderate aerobic activity on most days. If you weigh 150 pounds, that would translate into about three hours at moderate intensity per week. The low dose would be equivalent to eighty minutes per week.

Just multiply your body weight by eight to figure out how much you should be burning for the high dose, and then head to the gym to find out how many calories you burn during a given workout (most of the aerobic machines track this for you). If you weigh 150 pounds and burn 200 calories in thirty minutes on the elliptical trainer, you’d want to do six sessions a week to meet the high dose.

I would recommend at least this much exercise for anyone who, like my patient Bill, has a shadow syndrome of depression. These are people who aren’t clinically depressed but who tend to look at life with a primarily pessimistic attitude, or who have the impression that nobody in the world, themselves included, meets their high standards. For Bill, when he got into running and lifting weights, he met up with a group of regulars at the gym every morning for a workout and coffee. His performance and relationships at work improved, and he felt like he didn’t look at every new project with knee-jerk resentment. He almost welcomed new challenges, which completely changed his wife’s view of him.

Then there are those people who are just moody. They tend to have low self-esteem, partly because they’re so uncertain about how they’ll feel from day to day — crabby or buoyant? One of my patients, Jillian, fit this description. We started exploring this issue after she got engaged to the man of her dreams and began feeling blue and irritable more often than not. Naturally, I had been pushing her to exercise, and she finally joined a gym near her office. Wisely, she teamed up with a co-worker who was already going, and they supported each other to work out every day at lunchtime. After a few months, she was pleased with herself, and she expressed how stabilizing the daily activity was. It helped her maintain a rhythm in her life, which added to her sense of stability.

Some people who have mood swings may describe themselves as manic-depressive or bipolar, but that’s a problem of an entirely different magnitude. I haven’t discussed bipolar disorder primarily because so little research has been done on exercise’s effects on it. Recently, though, a preliminary study clearly showed that hospitalized patients with bipolar disorder who participated in a walking program suffered fewer depressive symptoms and less anxiety compared to those who were unwilling or unable to participate. And giving bipolar patients a stable social routine has been shown to improve long-term outcomes. It’s only been very recently that exercise started making its way into treatment protocols for bipolar disorder.

In some ways, exercise is even more important for prevention than it is for treatment. One of the first symptoms of depression, even before your mood drops to new lows, is sleep disturbance. Either you can’t get up or you can’t get to sleep or both. I think of it as sleep inertia — trouble starting or stopping. First you lose your energy, then your interest in things. The key is to get moving immediately. And do not stop. Set up a schedule for a daily walk, run, jog, bike ride, or dance class. If you can’t sleep, go for a walk in the dawn light and do it every day. Take the dog out, change your schedule — run from the depression. Burn those 1,400 calories as if your life depended on it, and nip it in the bud.

If you have severe depression, you may feel like you’re at the bottom of a pit, in a state of slow death, and that it’s almost impossible to make it outside or go to the gym or even think about moving. First, see your doctor about medication and get yourself some omega-3 supplements, which are proven to have antidepressant effects. This will, I hope, loosen up the brain lock enough for you to at least go for a walk. Ask for help. Get a friend or family member to agree to come by every day, at the same time if possible, to escort you outside and around the block. In England and Australia, walking groups for patients with depression have been popular for years, and now they’re starting to make their way to the United States, so check the Internet to see if there is one in your area. If that’s not an option, and you have the means, set up a regular time with a personal trainer. This might sound crazy, I realize, if you feel like you can’t even lift yourself off the couch, but if that’s the case, activity is all the more urgent.

Exercise is not an instant cure, but you need to get your brain working again, and if you move your body your brain won’t have any choice. It’s a process, and the best strategy is to take it one step — and then one stride — at a time. Start slowly and build on it. At its core, depression is defined by an absence of moving toward anything, and exercise is the way to divert those negative signals and trick the brain into coming out of hibernation.