Near the end of the movie A Beautiful Mind, the schizophrenic mathematician John Nash decides he wants to stay off his medication and out of the hospital, despite experiencing incessant delusions. "If I can just think through this," he explains to his long-suffering wife, "I can make it better." She takes his hand and puts it against her head. "Maybe the part that knows the waking from the dream—maybe it isn't here," she says. She moves his hand down to her heart. "Maybe it's here."
It's an ancient trope, but a false one. Several decades from now, when TV watchers encounter A Beautiful Mind on a classic movies channel, the idea of love residing in the heart as opposed to the head will seem as absurd as bloodletting. That's because the physiological reality of love belongs no more to the heart than it does to the liver. Like all emotions, love originates in the brain as surely as brilliant mathematical theorems do. We feel the passions of love because our brains contain specific neurochemical systems that create those feelings in us. We are not torn between the heart and the brain but rather between different parts of the brain, parts that specialize in the cornerstones of rational thought, such as long-term planning, and parts that give our lives emotional color.
Scientists who study the brain have traditionally spent far more time exploring the neural pathways of negative emotional responses: On our current map of the mind, the regions of fear are clearly delineated. Not so the kingdom of love and attachment, which has been a vast terra incognita until recently. But a new portrait of love has begun to emerge, and at its center lies a fascinating hormone called oxytocin that may well follow in the footsteps of serotonin, which shot into the popular consciousness a dozen years ago as Prozac was introduced. We are entering an age of brain biochemistry that can grasp the undecipherable—love.
You Tarzan, Me Jane In March of 1998, a psychology professor at the University of California at Los Angeles named Shelley Taylor attended a guest lecture at the university's Westwood campus. The topic was stress and the fight-or-flight instinct, a subject she knew a thing or two about, having studied human stress response for 20 years. At one point in the lecture, the speaker told a story about the levels of aggression he had witnessed in laboratory rats placed in stressful situations. After they had been repeatedly shocked with an electrical charge, the rats began to bite and claw each other to death.
"That went off like a lightbulb in my head, because it's not at all descriptive of what we typically see in human studies," Taylor recalls, sitting in her campus office, a Los Angeles cityscape hovering behind the pine branches outside her window. "I went back to my lab group, and I said, 'What do you make of these disjunctions between the animal studies and what we see in humans?' And one of them said, 'You know, the animal studies are all based on males. They hardly ever include females, because females cycle so rapidly.' And then someone else said, 'You know, I think that's true for the human literature as well.' So we started looking through the literature to see how well female responses to stress were represented, and the answer was very poorly. Prior to 1995, females constituted 17 percent of participants. There were virtually no studies where you had enough female participation to do a comparative study."
The lack of gender parity was not just a political issue. For decades, the scientific literature on stress response revolved around a fundamental causal chain: Introduce a stressor—a lunging predator, say, or a rival stealing your food supply—and the body initiates the now-famous fight-or-flight response. Confronted with stress, the theory went, our bodies instinctively primed themselves to strike back or run away. Fight or flight was compatible with the old Darwinian nature-red-in-tooth-and-claw stereotypes, but it didn't leave much room for an equally common human response to traumatic events: reaching out to loved ones. A parent reacting to a sudden threat will often put himself or herself in greater danger if it means protecting a child or a partner. That selfless behavior makes perfect sense to anyone who has felt parental or romantic love, but under the fight-or-flight paradigm, the behavior seems anomalous.
Taylor suspected that the fight-or-flight response was only half the story, and that gender differences might help shed light on the other half. "I said to my group, 'OK, let's start from scratch. What are women doing? Is fight or flight a reasonable description of women's response to stress?' And within seconds, all of us had an immediate response: No. Because what differentiates female responses to stress from those of males is that female responses have to incorporate the protection of offspring, at least for the period of time that there are offspring. Our idea was that fight behavior works fine if you're an individual, but if you're trying to protect young, fighting just isn't going to work. The same goes for flight—only ungulates like deer have offspring that are capable of fleeing shortly after birth." Two years after attending that guest lecture, Taylor had formulated a new theory, in the form of an essay published in Psychological Review titled "Behavioral Responses to Stress in Females." Fight or flight was one way of dealing with stress, she acknowledged, but there was another option: tend and befriend. You can combat threats by literally going to combat with them, or you can lean on your friends and family for support.
Taylor believes the tending instinct is more commonly expressed in women. "There was recently a meta-review of 28 different studies, and 26 of them found that women sought social support more than men. Short of childbirth, there is no sex difference in humans that looks like that. With most sex differences—men have a slight spatial advantage, women have a linguistic advantage—when you actually look at the curves, there's an enormous overlap." But when it comes to seeking out social bonds in the face of stress, the contrast is emphatic.
Taylor and her team even had a solid hunch about the brain chemistry behind the tending instinct. Researchers had long since detected the release of the peptide oxytocin during some of the key life experiences that involve intense emotional attachment: birth, breast-feeding, and sexual climax. In recent years, higher oxytocin levels had been linked to stressful experiences as well. While oxytocin was present in both male and female brains, evidence suggested that estrogen enhanced the peptide's effects, making it less powerful in males because of testosterone levels. If there was a biologically grounded tending instinct, oxytocin probably played a role.
Taylor's "tend or befriend" enjoyed its 15 minutes as a media sound bite following the publication in 2002 of her book The Tending Instinct, but the underlying concept was more than just a passing slogan. The idea that brain circuitries devoted to affiliation and social bonds may well be as sophisticated as our fear mechanisms had been percolating for almost a decade. And while an aggressive pack of lab rats might have provided Taylor with her eureka moment, the search for the brain science of attachment began with a more unlikely test subject.
Load up on adrenaline, or cool down with oxytocin About 20 years ago, neuroendocrinologist Sue Carter began examining the brains of prairie voles to understand why the small rodent indigenous to the midwestern plains of the United States is one of the natural world's great romantics. After mating, most voles remain monogamously attached for life, raising pups together in a rodent version of domestic bliss. This is, to say the least, an unusual practice in nature: Less than 5 percent of all mammals show monogamous, biparental behavior.
"I became interested in oxytocin then because I knew that oxytocin was released during sexual behavior," Carter says. "There was already research coming out showing that oxytocin facilitated parent-child bonding in sheep." When Carter injected oxytocin into the brains of voles, they formed bonds more quickly than usual. She and her colleagues also explored the effects of oxytocin from the reverse angle, by injecting chemicals that blocked the oxytocin receptors, cutting off the supply of the hormone. The voles' lifestyle began to look less like Leave It to Beaver and more like Woodstock: indiscriminate mating without any lasting attachment. "The most compelling evidence for oxytocin's role in bonding is simply that when you block the oxytocin receptors, the animals don't form pair-bonds," Carter says.
Several years later, Tom Insel, a former colleague of Carter's who is now president of the National Institute of Mental Health, began a comparative study analyzing the brains of prairie voles and their less monogamous cousins, the montane voles. Insel discovered a remarkable difference between the two species: In the faithful voles, the oxytocin receptors overlapped with dopamine receptors in an area of the brain called the nucleus accumbens; in the nonmonogamous voles, the oxytocin receptors were located elsewhere. The nucleus accumbens is generally regarded as one of the brain's essential pleasure centers. Dopamine coordinates many seeking and appetitive behaviors. In the monogamous voles, oxytocin receptors were planted firmly in the reward circuitry of the brain. The architecture suggested that behaviors associated with oxytocin release would feel good in the brains of the prairie voles but leave the montane voles relatively unaffected. If oxytocin encouraged the animals to stay attached to a partner, it was no wonder the prairie voles turned out to be so committed. Their brains were wired to make forming attachments pleasurable.
For the first time, researchers glimpsed the underlying circuitry that made pair-bonding desirable. The temptation to extrapolate the vole studies to the brain chemistry of humans was irresistible. Could the voles give us an answer to the age-old question, Why do fools fall in love?
By most accounts, exploring the neurochemical basis of love in the human brain has proved to be a thorny enterprise. When I ask Carter about the difficulty of designing experiments with human subjects, she laughs. "Well, do you want to be in the study? I mean, this is powerful stuff. It's extremely powerful and important to human behavior. Suppose you agree to participate in a love study, and I pick out a random partner for you? They do that on TV, but you can't do it in universities."
Despite the inherent difficulties, a number of recent studies have shed light on the brain chemistry of human love. Like those of the monogamous prairie vole, human oxytocin receptors are located in several dopamine-rich regions of the brain, suggesting that oxytocin is embedded in our reward circuitry. One study compared the brain activity of people looking at pictures of loved ones or at pictures of non-romantic friends. The pattern of activity in the cortex was markedly different depending on which type of face the subject was exposed to. FMRI scans of brains processing a romantic gaze bear a striking resemblance to the brain activity of new mothers listening to infants' cries. They also resemble brain images of people under the influence of cocaine.
The face-recognition studies are of particular interest because a number of animal studies have convincingly linked oxytocin to the formation of social memory. One hypothesis is that oxytocin release during key pair-bonding events like sexual climax or childbirth helps cement the image of a partner or a newborn in the mind's eye. Mothers who breast-feed their children often describe powerful memories of infants gazing up at them during nursing. The vividness of those memories, and their association with warm feelings, may well be the imprint of oxytocin.
Oxytocin has also been widely implicated in human responses to stress. "Because of my whole personal history of feeding my children, I became interested in breast-feeding as a protective mechanism," Carter says. "We were able to compare the effects of stress on lactating and non-lactating women. With the lactating women, we know they have more oxytocin, and we know they manage stress better." A number of studies have convincingly demonstrated that oxytocin is what scientists call a down regulator of the body's stress-axis system, which creates the bleak, gut-tightening feeling you experience when you get the news that the promotion didn't come through. People under the influence of oxytocin have smaller, briefer stress responses than others do; bad news seems to roll off them more readily.
The link between stress response and social attachment is at the heart of Taylor's idea of the tending instinct. You can fight your way out of stress by destroying your enemies, or you can reduce stress by reaching out to loved ones. In terms of brain chemistry, you can load up on adrenaline and fight or flee, or you can cool down with oxytocin and tend and befriend.
Love Potion No. 9 Doesn't Exist While the range and potency of oxytocin make for a fascinating story, Taylor cautions that its effect on human emotions is far from simple: "A lot of people say, 'Oxytocin is the cuddly hormone,' or 'Oxytocin's the love hormone.' Oxytocin is much more evasive than that, and it doesn't have one-to-one correspondence with psychological states. It's real risky trying to map these molecules onto specific states.
"For instance," Taylor says, leaning forward in her chair for emphasis, "older women who are living with husbands and finding those husbands to be nonsupportive have chronically higher levels of oxytocin. Now it's not clear what the direction of causality is. But a tentative conclusion that I would make is that when social-support needs are not being met, oxytocin levels go up as a signal to seek out social contact. And then once found, oxytocin may be restored to normal levels. So oxytocin isn't the 'feel good' hormone. At times, it may be the 'feel crummy' hormone that leads you to take steps to feel better."
Some scientists believe oxytocin works in tandem with the body's natural opiates, with oxytocin triggering the drive for social attachment and the opioids supplying the warm, fuzzy feeling of being in the company of loved ones. "The oxytocin story has come on like gangbusters, and it's certainly a big-ticket item," says Jaak Panksepp, a neuroscientist at Bowling Green State University in Ohio whose laboratory began working on opioids and social attachment in the 1970s. "But unfortunately, the oxytocin people forgot about the earlier opioid story."
Panksepp believes that one of the effects oxytocin has is to reduce the tolerance effect that plays such a devastating role in drug addiction. Just as addicts develop a tolerance to heroin that causes them to take ever-larger doses, the brain develops an identical tolerance to naturally occurring opiates. In tests with animal subjects, oxytocin injections dramatically reduced tolerance to opiates. In other words, oxytocin may not create the visceral pleasure of love and attachment, but it does enable that pleasure to last for a longer period of time.
All of which suggests that the phrase "addicted to love" may be more than poetry. Drugs like heroin do their damage because they tap directly into the brain chemistry that regulates the bonds of love. When people become addicted to drugs, one of the most common reactions expressed by close friends is a sense of bewilderment at the addict's ability to turn his or her back on family and friendship. Not knowing firsthand the tremendous force of addiction, it seems monstrous to us that someone could trade a child's love for the prick of a needle. But that needle contains the very drug that helps make the child's love appealing. We understand intuitively why someone might sacrifice a life for a child. When drug addicts make comparable sacrifices, it seems positively inhuman. Yet neurochemically, those sacrifices are laid at the same altar.
The link between opioids and oxytocin underscores Taylor's point about the "love drug" reductionism. Folklore and literature are filled with tales of love potions, but the story is far more complicated than that. There is a biologically grounded brain system that creates and maintains the feeling we call love, but its cause can't be reduced to a single molecule. There are undeniable interactions between oxytocin and opioids, and the prairie vole's brain anatomy suggests a strong connection between dopamine and oxytocin. More important, oxytocin's effects are heightened by estrogen and dampened by androgens like testosterone, which may help explain differences between male and female stress responses. Love may not reside in the heart, as folk wisdom would have it, but neither does it reside in a single molecule. When we feel the stirring of romantic love or parental attachment, we are sensing a complex interplay of brain chemicals, triggering activity in specific regions of the brain. Oxytocin is critical to that interplay, but it is not the whole story.
Why do voles fall in love? The complexity of the human brain—and the ethical problems of experimenting with humans—may mean that the scientific understanding of attachment will not proceed quickly. When I ask Taylor what breakthrough she'd most like to see in this field, she doesn't miss a beat. "I'd like to have a human equivalent to the prairie-vole model," she says, looking wistful. "The prairie-vole model is a beautiful one."
But while our knowledge of human neurochemistry is finite, the extent to which the chemistry repeats itself in other mammals suggests that love is as much a part of our evolutionary heritage as heartbeat regulation or stereovision. If we had evolved as a species with different mating and child-rearing habits—abandoning our children at birth and moving indiscriminately from partner to partner, like most reptiles—it's likely our brains would be incapable of feeling love.
Reptiles lack our neocortex, the seat of language and higher learning, and have a very primitive limbic system, the part of the brain that plays a key role in regulating emotional response. Reptile brains produce only a very preliminary version of the oxytocin molecule. If some accident of evolution had led reptiles to develop a neocortex while maintaining their nonexistent child-rearing habits, they might have ended up writing powerful verse about some other deep-seated biochemical urge—temperature regulation, say—but there would be no love sonnets in the reptile canon.
The biological capacity for love is one way the brain prepares us for offspring who are born young and helpless and need tending to have the slightest hope of survival. That tending comes in the form of social bonds—between parent and child, between parents, among the extended social family members who help raise the child. The glue that keeps those bonds strong is the feeling of pleasure and reward and satisfaction that our brains concoct for us when we enter into loving relationships.
It can be daunting to think that the core ingredients of that glue are shared by humans and prairie voles. Because love is the source of so many of humanity's highest creative achievements, we like to think that the feeling itself is just as unique. But the commonalities of brain chemistry—and the commonalities of behavior—suggest that at least some part of love's intoxication is experienced by other mammals. "I think reasonable people have to open their minds to the possibility that very similar basic feelings occur," Panksepp says. "Other mammals can't make movies or art or other great things with their feelings the way we can. But it would be foolish for us to deny the continuity of the foundational elements."
When asked how the subjective experience of prairie-vole love might differ from the human experience, Panksepp is speechless. "I don't think that question can be answered," he says, then starts to formulate an answer. "We know much more about the chemistry of attachment and love in prairie voles than we do about that chemistry in humans. I mean, the animal work has yielded such riches in terms of the underlying details. We just don't know that much about our own species. But everything we're learning from the human and animal genome projects, about the conservation of neurochemistries and the neuroanatomies, all of this points me to the conclusion that we are learning about ourselves when we study these little critters."