A commercial sponsored by Pfizer, the drug company that manufactures the antidepressant Zoloft, asserts, “While the cause [of depression] is unknown, depression may be related to an imbalance of natural chemicals between nerve cells in the brain. Prescription Zoloft works to correct this imbalance.” Using advertisements such as this one, pharmaceutical companies have widely promoted the idea that depression results from a chemical imbalance in the brain.

The general idea is that a deficiency of certain neurotransmitters (chemical messengers) at synapses, or tiny gaps, between neurons interferes with the transmission of nerve impulses, causing or contributing to depression. One of these neurotransmitters, serotonin, has attracted the most attention, but many others, including norepinephrine and dopamine, have also been granted supporting roles in the story.

Much of the general public seems to have accepted the chemical imbalance hypothesis uncritically. For example, in a 2007 survey of 262 undergraduates, psychologist Christopher M. France of Cleveland State University and his colleagues found that 84.7 percent of participants found it “likely” that chemical imbalances cause depression. In reality, however, depression cannot be boiled down to an excess or deficit of any particular chemical or even a suite of chemicals. “Chemical imbalance is sort of last-century thinking. It's much more complicated than that,” neuroscientist Joseph Coyle of Harvard Medical School was quoted as saying in a blog by National Public Radio's Alix Spiegel.

Indeed, it is very likely that depression stems from influences other than neurotransmitter abnormalities. Among the problems correlated with the disease are irregularities in brain structure and function, disturbances in neural circuitry, and various psychological contributions, such as life stressors. Of course, all these influences ultimately operate at the level of physiology, but understanding them requires explanations from other vantage points.

Are Your Chemicals out of Balance?
Perhaps the most frequently cited evidence in support of the chemical imbalance hypothesis is the effectiveness of antidepressants, many of which increase the amounts of serotonin and other neurotransmitters at synapses. Zoloft, Prozac and similar selective serotonin reuptake inhibitors (SSRIs) result in such an increase and can often relieve depression, at least when it is severe. As a result, many believe that a deficiency in serotonin and other neurotransmitters causes the disorder. But just because a drug reduces symptoms of a disease does not mean that those symptoms were caused by a chemical problem the drug corrects. Aspirin alleviates headaches, but headaches are not caused by a deficiency of aspirin.

Evidence against the hypothesis comes from the efficacy of a newly developed antidepressant, Stablon (Tianeptine), which decreases levels of serotonin at synapses. Indeed, in different experiments, activation or blockage of certain serotonin receptors has improved or worsened depression symptoms in an unpredictable manner. A further challenge to the chemical imbalance hypothesis is that many depressed people are not helped by SSRIs. In a 2009 review article psychiatrist Michael Gitlin of the University of California, Los Angeles, reported that one third of those treated with antidepressants do not improve, and a significant proportion of the remainder get somewhat better but remain depressed. If antidepressants correct a chemical imbalance that underlies depression, all or most depressed people should get better after taking them. That they do not suggests that we have only barely begun to understand the disorder at a molecular level. As a result, we must consider other, nonchemical leads.

This Is Your Brain on Depression
A possible clue lies in brain structures. Imaging studies have revealed that certain brain areas differ in size between depressed and mentally healthy individuals. For example, the amygdala, which responds to the emotional significance of events, tends to be smaller in depressed people than in those without the disorder. Other emotional regulatory centers that appear to be reduced in volume are the hippocampus, an interior brain region involved in emotional memory, the anterior cingulate cortex, which helps to govern impulse control and empathy, and certain sections of the prefrontal cortex, which plays an important role in emotional regulation. Nevertheless, the effects of these shrinkages on depression, if any, remain an open question.

Neuroimaging studies have revealed that the amygdala, hypothalamus and anterior cingulate cortex are often less active in depressed people. Some parts of the prefrontal cortex also show diminished activity, whereas other regions display the opposite pattern. The subcallosal cingulate gyrus, a region near the anterior cingulate, often shows abnormal activity levels in depressed individuals. These differences may contribute to depression, but if they do, scientists are not sure how.

In 2012 neurosurgeon Andres M. Lozano of the University of Toronto and his associates studied the effects of deep brain stimulation of the subcallosal cingulate gyrus in depressed patients who had not benefited from standard treatments. The intervention led to a significant reduction in symptoms of depression, supporting the idea that a dysfunction in this brain area may be involved in the illness.

Findings also point to a crucial role for psychosocial factors such as stress, especially when it arises from a loss of someone close to you or a failure to meet a major life goal. When someone is under a good deal of stress, a hormone called cortisol is released into the bloodstream by the adrenal glands. Over the short term, cortisol helps humans cope with dangers by mobilizing energy stores for flight or fight. But chronically high cortisol levels can harm some bodily systems. For example, at least in animals, excess cortisol reduces the volume of the hippocampus, which in turn may contribute to depression. Despite such data, we still do not know if stress alters the human brain in ways that can lead to depression.

Seeing the Elephant
Throughout this column, we have described associations between various brain changes and depression. We have not talked about “causes,” because no studies have established a cause-and-effect relation between any brain or psychosocial dysfunction and the disorder. In addition, depression almost certainly does not result from just one change in the brain or environmental factor. A focus on one piece of the depression puzzle—be it brain chemistry, neural networks or stress—is shortsighted.

The tunnel-vision approach is reminiscent of a classic story in which a group of blind men touch an elephant to learn what the animal looks like. Each one feels a different part, such as the trunk or the tusk. The men then compare notes and learn that they are in complete disagreement about the animal's appearance. To understand the causes of depression, we have to see the entire elephant—that is, we must integrate what we know at multiple scales, from molecules to the mind to the world we live in.