Monday, April 13, 2009
The neural correlates of many emotional states have been studied, most recently through the technique of fMRI. However, nothing is known about the neural substrates involved in evoking one of the most overwhelming of all affective states, that of romantic love, about which we report here. The activity in the brains of 17 subjects who were deeply in love was scanned using fMRI, while they viewed pictures of their partners, and compared with the activity produced by viewing pictures of three friends of similar age, sex and duration of friendship as their partners. The activity was restricted to foci in the medial insula and the anterior cingulate cortex and, subcortically, in the caudate nucleus and the putamen, all bilaterally.
Deactivations were observed in the posterior cingulate gyrus and in the amygdala and were right-lateralized in the prefrontal, parietal and middle temporal cortices. The combination of these sites differs from those in previous studies of emotion, suggesting that a unique network of areas is responsible for evoking this affective state. This leads us to postulate that the principle of functional specialization in the cortex applies to affective states as well.
Explorations into the biological foundations of behavior are throwing new light
on the way we organize our societies, and ourselves. Drawing on groundbreaking
studies of the brain activity behind monogamous and non-monogamous behavior of
animals, neuroscientists Insel and Young examine the implications for human
The 1990s witnessed a revolution in our understanding of how the brain mediates sensory processing, motor function, and memory. We have learned the importance of a small family of developmental genes for specifying normal neural development. At the same time, we have discovered the remarkable plasticity of both the developing brain and the adult brain. Sadly, this revolution in neuroscience has taught us less about the neurobiology of emotion and has revealed very little about the neural basis of love.
Although neuroscientists recognize the importance of love as a complex emotion, few have wanted to address the challenge of studying where or how the brain mediates love. The problem is not simply that love is complex. There is an abundant literature on the neurobiology of aggression, which is equally complex. Many have studied the neural consequences of loss or separation, which may be considered equally complicated and perhaps the inverse of love or attachment. The problem is not that love (or attachment) is a positive state and less likely to be studied than emotions, such as loss, which can be tied to a disorder. We have nearly 50 years of research on the neuroendocrinology of sex, which is also a healthy, positive behavior.
The relative absence of neuroscience research on love can be attributed to two fairly obvious and related limitations. One is definitional. Love, whether defined as attachment or a pair bond, is difficult to measure. Although most of us may recognize love as the most powerful psychological and biological experience of our lives, how do we quantify this experience? And how do we distinguish love from infatuation, craving, caring? A related problem for neuroscientists is that nothing remotely related to human love is observed in common laboratory animals, such as rats and mice. Yes, rat mothers will show intense devotion and defense of their young, but they are not selective in their maternal behavior, offering the same level of care to unrelated young placed in the nest.
In the past few years, a series of studies have begun to address both of these limitations. By studying natural populations of monogamous rodents, we and others have investigated the brain-based correlates of pair-bonding behaviors that resemble what we call "love" between humans. In non-human, monogamous species, pair-bonding involves a life-long selective attachment to one partner.
One caveat: It should be noted that the term "monogamy," as used by biologists, refers to a pattern of social not sexual organization. About 3 percent of mammals are considered monogamous. Bonded pairs in these species share a territory and a nest, exhibit both maternal and paternal care, and usually are similar in size and appearance. But few if any of these species exhibit sexual exclusivity. That is, biological monogamy refers to how you live not how you mate.
The availability of species that pair bond raises an option for addressing the second limitation of studying the neurobiology of love. While we cannot directly ask rodents how they feel, we can measure the behaviors essential for pair bonding. In monogamous rodents, the formation of a preference for a specific partner, the time spent in physical contact, and the response to separation can be easily quantified and standardized. Using these sorts of measures, we can begin to ask which genes, neurotransmitters, and neural systems are critical for pair bonding.
Monogamous mammals are found in many different species and in diverse environments. For example, among primates, gibbons, titi monkeys, and marmosets have been described as monogamous. But for neurobiological study, rodents are ideal and a group of microtine rodents or voles has proven especially informative. Prairie voles are highly social and show long-term pair bonds. Indeed, in field studies, prairie voles are usually found in communal burrows with a single breeding pair. If the male or female breeder dies, the remaining prairie vole will not accept a new mate, in 80 percent of such cases. Female prairie voles have an unusual pattern of reproductive development. They remain sexually immature while remaining in their birth group. Once exposed to an unrelated male, they ovulate for the first time, mate for 24-48 hours, and, apparently as a result of mating, develop a pair bond. Laboratory studies have demonstrated the importance of mating for the development of a partner preference.
Part of what makes prairie voles so informative for study is the availability of closely related species that differ so markedly in their social behavior. For instance, the montane voles, a near-relative to the prairie vole, is a non-monogamous species, living in isolated burrows, and showing no evidence of paternal behavior. Montane voles do not pair bond in the field and in laboratory studies they fail to develop partner preferences after mating.
Given the striking behavioral differences between these voles species, researchers have been attempting to define associated differences in brain anatomy or chemistry. Among the candidates for separating these two species are two neuropeptide hormones, oxytocin (OT) and vasopressin (AVP). These hormones are synthesized in the brain and have been previously shown to have important effects on social behavior in common laboratory animals, such as rats, mice, and even sheep. In mammals, OT has been shown to be important for labor and lactation. In rats and sheep, OT working within specific neural circuits is critical for the onset of maternal behavior. This is perhaps the best example in the behavioral neuroscience literature of a single neuropeptide influencing attachment behavior in non-monogamous mammals. In addition, OT is released during copulation and may influence female sexual behavior. AVP has been implicated in several aspects of male social behavior, including territorial displays, aggression, and social memory.
Given the evidence linking OT and AVP to complex social behaviors, it is perhaps not surprising that these hormones should also influence pair bond formation in the monogamous prairie vole. But there is an even better reason for investigating these two neuropeptides in pair bond formation. As noted above, in the normal life cycle of a prairie vole, mating is the critical event preceding pair bonding. Both OT and AVP are released with mating in other mammals and presumably these hormones are released in the vole brain during copulation.
Test show that in female prairie voles, if OT is given during exposure to a male, a partner preference can be induced without mating. Conversely, females allowed to mate, but treated with an antagonist or specific blocker of OT, fail to form a partner preference. Thus, OT appears necessary and sufficient for partner preference formation in the female prairie vole.
Surprisingly, OT is without effects in the male. For males, AVP seems to be the decisive element. An introduction of AVP in the absence of mating is effective in causing a pair-bond preference even in the absence of mating, while an AVP antagonist blocks partner preference formation if given during mating. In addition, AVP facilitates — and an AVP antagonist blocks — paternal behavior.
However, neither OT or AVP, given in an identical fashion, influence social behavior in the non-monogamous montale vole. Although these hormones can increase self-grooming or locomotor behavior, neither will induce a partner preference or paternal care in montane voles. The reason that these peptides have different effects appears to be due to species differences in the brain targets for both OT and AVP.
All neurohormones act via specific receptors. After release from nerve endings, the hormone binds to receptors that initiate a series of intracellular events. It is important to understand that where the receptors are located in the brain determines which cells are activated. In other words, the behavioral effects of any hormone are defined not just by the amount of hormone, but also by where its receptors are located in the brain. Both prairie voles and montane voles make OT and AVP, but these species have markedly different patterns of receptor distribution.
For example, OT receptors in the prairie vole brain are concentrated in "reward" pathways in the brain, target regions that are likely to lead to the conditioning of certain behaviors (such as a partner preference) after these targets are "activated" by a release of OT during copulation. In the montane vole, OT receptors are absent in these regions but are found in regions important for non-social behavior, and therefore, OT release with copulation would not be expected to induce a partner preference. Vole species do not differ in the brain distribution of several other hormone receptors that have been studied, suggesting that is the differences in the OT and AVP receptors that are crucial in the different social organization between the species.
It’s also important to emphasize that the difference between prairie voles and montane voles is not too much or too little OT (or AVP). The species have roughly equivalent amounts of both neurohormones. The difference is where the receptors are. Therefore, infusing micromolar quantities of either hormone into montane voles will not induce a pair bond. The receptors are in the wrong target areas of the montane vole brain for either hormone to influence social attachment. In a recent study, we developed a transgenic mouse with AVP receptors in the pattern normally found in the prairie vole brain. As a result, this mouse responded to AVP with an increase in social interaction, such as investigation and grooming of an unfamiliar mouse. Although these mice were not monogamous, these results demonstrated the importance of receptor localization for determining how hormones can influence complex social behavior.
Evolution and Monogamy
A central question is why such closely related species should exhibit these striking differences in chemical neuroanatomy. The answer may reside in the molecular structure of their OT and AVP receptor genes. Genes can be conceptualized as two main components: First, a coding sequence that provides the DNA information for the formation of one specific RNA which, in turn, directs the formation of one unique protein. The coding sequence determines if a gene will create a hormone, a receptor, or a structural piece of the cell. A second part of the gene is the promoter sequence. This area, which flanks the coding sequence, contains numerous response elements that provide the stop or go signals for the gene. If the coding sequence provides the information to make a given protein, the promoter provides the information of where and when this protein gets made.
There are few genomic differences between prairie and montane voles. When comparing the coding sequences for OT and AVP receptors, these species appear essentially identical. That is, there are no structural differences in their OT and AVP receptors. Not surprisingly, given that these receptors are found in different brain regions, there are differences in their promoter sequences. For OT receptors, prairie and montane voles show only subtle differences, but these differences appear to be substantial enough to direct different patterns of expression in the brain. For AVP receptors, the species differences are more extreme, with prairie voles showing a large segment of DNA inserted into their promoter regions — an insert lacking in montane voles.
Thus, if monogamy in prairie voles is associated with a DNA insertion in the AVP receptor gene, where does this insert come from? We don’t know the origin of these sequences, but similar regions have been reported in other receptor genes, leading to surprising differences in function. In the OT and AVP receptor genes, these inserted sequences include repeating patterns that are variable from one species to the next. It seems likely that as a result of these changing sequences in the promoters, OT and AVP receptors show marked species variation in their pattern of brain expression.
The existing data are consistent with the hypothesis that the promoter sequences for these two genes are "hot spots" for insertional events leading to different patterns of receptor expression, different functional effects of the hormones, and ultimately different patterns of social organization. Monogamy may have resulted from mutations in these "hot spots" and then survived under specific socio-ecological conditions in which pair bonding and paternal behavior were adaptive.
Of Human Bonding
We know very little about the neurobiology of human attachment. Humans have OT and AVP, and both hormones are released during copulation. Perhaps of greater importance, receptors for OT and AVP are found in the human brain, although the patterns are not quite like either the prairie vole or the montane vole. We do not know if OT or AVP is involved in the formation of pair bonds in humans. More specifically, we do not know if the differences found between species will be mirrored by differences within a species.
Obviously, there is still much work to be done; it’s been said that the purpose of experimental models is to focus our questions, not to provide final answers. But the use of non-human models to study attachment has provided a clear focus for beginning investigations into human attachment. Oxytocin and vasopressin are important candidate neural systems and their receptor fields provide potential nodes for processing the information necessary for attachment. We know that these receptor systems have substantial inter-species differences in regional expression ,based on variable regions in their genes. It seems possible, although still unproven, that these same regions will show intra-species variability and may correlate with individual differences in the capacity for human attachment.
THE CENTER FOR BEHAVIORAL NEUROSCIENCE,
Thomas R. Insel is Director of the Center for Behavioral Neuroscience, professor of psychiatry at Emory University School of Medicine, and adjunct professor of psychology at Emory’s College of Arts and Sciences. Larry J. Young is assistant professor of psychiatry at Emory University and faculty member of the Center for Behavioral Neuroscience.
Ferris C. (1992). Role of vasopressin in aggressive and dominant/subordinate behaviors. In C. Pedersen, J. Caldwell, G. Jirikowski, & T. Insel (Eds.), Oxytocin in Maternal, Sexual, and Social Behaviors, (Vol. 652, pp. 212-227). New York: New Pork Academy of Sciences Press.
Insel TR, and Hulihan TJ (1995): A gender-specific mechanism for pair bonding: oxytocin and partner preference formation in monogamous voles. Behavioral Neuroscience 109: 782-789.
Insel TR, Wang Z, and Ferris CF (1994): Patterns of brain vasopressin receptor distribution associated with social organization in microtine rodents. Journal of Neuroscience. 14: 5381-5392..
Modahl C, Green L, Fein D, Morris M, Waterhouse L, Feinstein C, and Levin H (1998): Plasma oxytocin levels in autistic children. Biological Psychiatry 432: 270-277.
Shapiro L, Austin D, Ward S, and Dewsbury D (1986): Familiarity and female mate choice in two species of voles (Microtus ochrogaster and microtus montanus). Animal Behavior 34: 90-97.
Winslow JT, Hastings N, Carter CS, Harbaugh CR, and Insel TR (1993a): A role for central vasopressin in pair bonding in monogamous prairie voles. Nature 365: 545-548.
Winslow JT, Shapiro LE, Carter CS, and Insel TR (1993b): Oxytocin and complex social behaviors: species comparisons. Psychopharmacology. Bulletin. 29: 409-414.
Young LJ, Winslow JT, Wang Z, Gingrich B, Guo Q, Matzuk MM, and Insel TR (1997): Gene targeting approaches to neuroendocrinology: Oxytocin, maternal behavior, and affiliation. Hormones and Behavior 31: 221-231.
Young, L.J. (1999): Oxytocin and Vasopressin Receptors and Species-Typical Social Behaviors. Hormones and Behavior 36: 212-221.