Excerpted from Touching a Nerve: The Self as Brain, by Patricia S. Churchland. Copyright © 2013 Patricia S. Churchland. With permission of the publisher, W. W. Norton & Co., Inc.

Diverging Paths in Human Sexual Development
The basic account of a typical brain-hormone interaction for XX and XY fetuses has been outlined. But not all cases conform to the prototype. Variability is always a part of biology. For example, unusual chromosome arrangements can occur. Rarely, an egg or a sperm might actually carry more than one chromosome, so the conceptus ends up with more than just a pair of sex chromosomes. About 1 in 650 males are born with XXY, a condition known as Klinefelter’s syndrome (as I mentioned in Chapter 1, my brother has this condition). The outcome can be quite variable, but basically what happens is that the testosterone supply dependent on the Y chromosome gets swamped by the estrogen production linked to the two X chromosomes. This affects gonad development, musculature, and fertility. There are cognitive costs, too, mainly having to do with the role of the prefrontal cortex in impulse control and the capacity to delay gratification.

Other chromosomal variations are also seen: XYY occurs in about 1 in 1,000 male births. It frequently goes unnoticed because there are no consistent symptoms. At one time it was claimed, mainly on a priori grounds, that XYY persons are especially aggressive, but this turns out not to be correct. XXYY, which is much more rare (about 1 in 20,000 male births), has many deleterious effects. This condition is linked to seizures, autism, and developmental delays in intellectual functions. In about 1 in 5,000 cases, a fetus may have only a single chromosome—an X—a condition known as Turner’s syndrome. The damaging effects are very broad, including short stature, low-set ears, heart defects, nonworking ovaries, and learning deficits. If a conceptus has only a single chromosome—a Y—it probably fails to implant in the uterus and never develops.

So just at the level of the chromosomes, we see variability that belies the idea that we are all either XX or XY. What about variability in the genes that leads to variability in brain development? Various factors, both genetic and environmental, can deflect the intricate development of a body and its brain from its typical course.

Consider an XY fetus. For the androgens (testosterone and dihydrotestosterone) to do their work in its brain, they must bind to special receptors tailored specially for androgens. The androgens fit into the receptors like a key into a lock. The receptors are actually proteins, made by genes. Even small variations in the gene (SRY) involved in making androgen receptors can lead to a hitch. And small variations in that gene are not uncommon. In some genetic variants, the receptor lacks the right shape to allow the androgens to bind to it. This prevents the process of masculinizing the gonads and the brain. In other genetic variants, no receptor proteins are produced at all, so the androgen has nothing to bind to. In these conditions, the androgens cannot masculinize the brain or the body, despite the XY genetic makeup. In consequence, the baby, though a genetic male, will probably have a small vagina and will be believed to be female when born. This baby will grow breasts at puberty, though she/he will not menstruate and has no ovaries. This is sometimes described in the following way: the person is genetically a male, but bodily (phenotypically) a female. These individuals usually lead quite normal lives and may be sexually attracted to men or to women or in some cases to both.

If an XX fetus is exposed to high testosterone levels in the womb, her gonads at birth may be rather ambiguous, with a large clitoris or a small penis. This condition is known as congenital adrenal hyperplasia, or CAH. This usually results from a genetic abnormality that causes the adrenal glands to produce extra androgens. As a child, she may be more likely to engage in rough-and-tumble play and to eschew more typical girl games such as “playing house.” When at puberty there is a surge of testosterone, she may develop a normal penis, testicles will descend, and her musculature may become more masculine. Though raised as a girl, persons with this history tend to live as heterosexual males. A male XY fetus may also carry the genetic defect, but in that case, the extra androgens are consistent with the male body and brain, and the condition may go unrecognized.

Some of these discoveries begin to explain things about gender identity that otherwise strike us as puzzling. A person who is genetically XY with male gonads and a typical male body may not feel at all right as a man. So far as gender is concerned, he feels completely female. Conforming to a male role model may cause acute misery and dissonance, sometimes ending in suicide. Conversely, a woman who is genetically XX may feel a powerful conviction that she is psychologically, and in her real nature, a man. Sometimes this disconnect is characterized as a female trapped in a man’s body or a male trapped in a female’s body. Statistically, male-to-female transsexuals are about 2.6 times as common as female-to-male transsexuals.

Is this conviction on the part of the person not just imagination run amok? In most cases, the matter is purely biological. Once we know something about the many factors, genetic and otherwise, that can alter the degree to which a brain is masculinized, it is a little easier to grasp a biological explanation for how a person might feel a disconnect between his or her gonads and his or her gender identity.

For example, in fetal development, the cells that make GnRH (gonadotropin-releasing hormone) may not have had the normal migration into the hypothalamus. If this happens, the typical masculinization of the brain cannot occur. For some individuals, the explanation of the chromosome-phenotype disconnect (XY but has a female gender identity) probably lies with GnRH, despite the existence of circulating testosterone in the blood. It is noteworthy that the data indicate that most male-to-female transsexuals do have normal levels of circulating testosterone, and most female-to-male transsexuals do have normal levels of circulating estrogen. This means that it is not the levels of these hormones in the blood that explains their predicament. Rather, we need to look at the brain itself.

Examining brains at autopsy is currently the only way to test in humans whether an explanation for a behavioral variant in terms of sexually dimorphic brain areas is on the right track. Here is some evidence that it is. First, there is a subcortical area close to the thalamus called (sorry about this) the bed nucleus of the stria terminalis (BNST). The BNST is normally about twice as large in males as in females. What about male-to-female transsexuals? Because the data can be obtained only at autopsy, they are limited. Nevertheless, in the male-to-female transsexual brains examined so far, the number of cells in the BNST is small. The number looks closer to the standard female number than to the standard male number. For the single case coming to autopsy of a female-to-male transsexual, the BNST looks like that of a typical male.

What are the causal origins of this mismatch of gonads and brain? The answers are still pending, but in addition to the many ways in which lots and lots of genes could be implicated, various drugs taken by the mother during pregnancy are possible factors. What drugs? Among others, nicotine, phenobarbitol, and amphetamines.

Like many others of my generation, I first learned about transsexuality when the British journalist and travel writer James/Jan Morris was interviewed on the BBC in 1975. A highly gifted and clearheaded writer with a wonderful sense of humor, Morris discussed her long struggle with the dilemma of feeling undeniably like a female from the age of 5 and yet possessing a male body and presenting to the world as a male. In Morris’s forthright book on the subject, Conundrum, she provides one of the deepest and most revealing narratives on what it is like to be unable to enjoy that calm sense of being at home in your own skin, of being one with yourself. Morris had married a woman to whom he was deeply attached and had five children. By all accounts, he was a wonderful and devoted father. But as the years went by, he became ever more miserable until in his 50s, and with the blessing of his wife, he underwent the long and difficult process of a sex change. Here are Morris’s heartfelt words describing the change:

Now when I looked down at myself I no longer seemed a hybrid or chimera: I was all of a piece, as proportioned once again, as I had been so exuberantly on Everest long before. Then I had felt lean and muscular; now I felt above all, deliciously clean. The protuberances I had grown increasingly to detest had been scoured from me. I was made, by my own lights, normal.

Gender identity is one thing, sexual orientation another, and the vast majority of homosexuals have no issue with gender identity at all. They just happen to be attracted to members of their own sex. This tends to be true also of cross-dressing males, who are fully content in the male gender identity but who enjoy dressing up in women’s clothing. How is sexual orientation related to the brain? There are undoubtedly many causal pathways that can lead to homosexuality or bisexuality, most involving the hypothalamus in one way or another. Sometimes sexual orientation can be affected by the chromosomal and genetic variations discussed earlier. The hypothalamic changes are likely to be quite different from those found in someone who is transsexual or transgendered.

Sexual Attraction and Its Biology
Homosexuality is one area where understanding even a little about the brain and the biological basis of a behavior has had a huge social impact. Although there are still those who maintain that homosexuality is a life choice, over the last 30 years, the credibility of this idea has largely crumbled. One significant reason was the discovery by Simon LeVay in 1991 concerning a small region of the hypothalamus. In postmortem comparisons of brains, LeVay found that a small hypothalamic region in the brains of male homosexuals was anatomically different from that of male heterosexuals; the region was smaller, more like that seen in the female brain. In and of itself, this discovery could not point to this brain area as the source of sexual orientation, nor could it claim with any certainty that gay men were born gay. And LeVay, tough-minded scientist that he is, was entirely clear on that point. Nevertheless, given what else was known from basic research about the role of the hypothalamus in sexual behavior, and given what else was known about fetal brain development, it seemed a good guess that being gay was simply the way some people are. More recent research strongly supports this idea.

To many people, the simple fact revealed in LeVay’s anatomical discovery diminished the authority of theological arguments that homosexuality is a life choice and a direct route to damnation. Although it does not follow from the data themselves that sexual orientation is basically a biological feature, LeVay’s discovery was a powerful element in ushering in a change in the culture of attitudes toward homosexuality, especially in the young, but not only in them. Antisodomy laws changed, fathers openly embraced their gay sons, and the hypocrisy of homophobic theologians and priests was exposed when they themselves were “outed.” To be sure, these changes did not happen overnight, and well-organized campaigns, such as that of Harvey Milk in San Francisco, were extraordinarily important as well, but the discoveries about the brain grounded the change in attitudes in a special way.

There appears to be very little evidence that environment has much to do with sexual orientation. It is not “catching,” so far as anyone can tell. Children of homosexual parents are no more likely to be homosexual themselves than children of heterosexual parents. This implies that somewhere in the thicket of causality involving genes, hormones, neurochemicals, and fetal brain development, certain of the hypothalamic cell groups are masculinized or not, to some degree or other, with the result that sexual orientation is largely established before birth. This is biology, not a choice made in adulthood. No 5-year-old child makes a life choice to be gay.

The main point of this section has been to discuss what is known about differences between male and female brains, especially as they pertain to testosterone. This main point benefits from being set in a broader context of the impressive variability in the biology of masculinizing or feminizing a human fetal brain. The simple idea that we are all either just like Adam or just like Eve is belied by the biological subtleties of orchestrating genes, hormones, enzymes, and receptors. One of the truly brilliant achievements of neuroscience in the last 40 years has been to make remarkable progress in explaining the link between brains and sex.

Testosterone and Aggression
Unlike voluntary cooperation in social birds and mammals, the roots of aggression reach deep, deep into our biological past. Even crayfish and fruit flies show aggressive behavior. In mammals, aggressive behavior depends on many neurobiological and hormonal elements. Testosterone is one element, albeit an important element, among many others. Not only are there other elements, but there are big interactive effects. Tracking interactions means tracking a dynamic system, and that quickly gets very complicated. It is a little like tracking the movement and development of a tornado. The distribution and degree of high and low pressures, warm and cold fronts, determines whether and where a tornado forms. Once it forms, the tornado itself has an effect on those very factors, which in turn continue to affect the tornado, and so back and forth. Despite the interactive nature of elements regulating aggressive behavior, some general features of the phenomenon emerge fairly clearly.

Here is one important and consistent finding. The balance between testosterone and stress hormones is a strong predictor of the profile of aggression in a particular male. A disposition to be especially aggressive seems to be associated with the imbalance between testosterone and stress hormones. It is not just the level of testosterone per se that predisposes to aggression; it is the ratio between stress hormones (mainly cortisol) and testosterone.

More precisely, male animals with high levels of testosterone and low levels of cortisol display more aggression than those with high levels of testosterone and high levels of cortisol. These high-high males are more vigilant and calculating. The high-low males, by contrast, are less responsive to harmful consequences that may ensue. Roughly speaking, those with higher levels of cortisol are more likely to evaluate consequences and to appreciate the nature of the risks. When high levels of testosterone balance with high levels of cortisol, the male is likely to be courageous but not reckless. Consistent with the hypothesis that the balance between testosterone and cortisol matters enormously, when levels of testosterone (but not cortisol) were experimentally raised in human males, fear levels decreased, even though this was not consciously recognized by the subjects themselves.

Those with low levels of testosterone and high levels of cortisol are more apt to be fearful and avoid combative encounters altogether. They are particularly sensitive to risk and possible damage. They are on the reserved end of the aggression spectrum.

What controls the balance of testosterone and cortisol? Many things, including genes, other hormones, neuromodulators, age, and environment. For example, in male chimpanzees, testosterone is released when the females go into estrus.

The account is complex also because stress hormone levels can vary as a function of a host of factors. Some factors are internal, such as a genetic variant, while some are external, associated with local social practices. For example, suppose a male happens to have a very large and muscular build. In certain social conditions, he may be less stressed than a slighter male because other males will fear provoking him. In these social conditions, such individuals are treated with respect and are seldom provoked, so they come to be known as gentle giants.

If, however, more macho social customs prevail, the very same man may be tested by others precisely because of his build. In these “prove yourself” conditions, the man’s stress levels may be higher. He is often provoked, so he is regularly alert for provocation and he responds accordingly. He may seem more aggressive than gentle in the stressful environment. So add the external social conventions as a third dimension.

Aggression has a fourth dimension: the neuromodulator serotonin. To a first approximation, levels of serotonin affect whether aggression is impulsive (low serotonin) or planned (high serotonin). Emotional responses are in play here, too, since impulsive actions tend to be motivated by very strong emotions, such as rage, whereas planned actions seem to involve more emotional control—resolve rather than rage, cagily waiting for the right moment to strike rather than heedlessly slashing away. In high-serotonin conditions, assuming that the man has a standard balance of testosterone and cortisol, then anger and fear are less prominent than intense vigilance.

Consider a male with this combination: high testosterone, low cortisol, low serotonin. This man may be especially problematic; he may be fearless and quick to anger, and the anger may be poorly controlled. Change his ratios so that the serotonin levels become high, and his aggressive impulses will be under more control. Now consider a different male with that same profile—high testosterone, low cortisol, and high serotonin. Suppose, in addition, he rarely feels empathy. For the sake of argument, let us suppose that he happens to have low oxytocin levels, influencing his low empathy response. This man may be predisposed to psychopathy in his behavior. He can be aggressive, but he plans carefully and feels little anxiety during his preparations. Nor, in the aftermath, does he feel remorse for the injury inflicted. American Psycho portrayed such a man.

We are not yet done. Here is another, and completely unexpected, dimension (we are up to five) to aggression: a gas, nitric oxide. It is released from neurons and dampens aggression. It is made by an enzyme, nitric oxide synthase. Male animals that happen to lack the gene for making that enzyme are particularly aggressive compared with those who have the gene and hence have normal amounts of nitric oxide. In contrast to males, female animals lacking that gene are not hyperaggressive. There is some evidence for an interaction between an important neurotransmitter in the reward system, dopamine, and nitric oxide. A balance of the two is associated with control over aggression, but exactly how nitric oxide interacts with elements such as testosterone or serotonin is not understood.

There is still more. For males, defense of offspring is testosterone mediated. Vasopressin is also essential for expression of aggressive behavior against those who threaten mates and offspring. Interestingly, even with their very low testosterone levels, castrated male mice will attack nest intruders if vasopressin is administered. As discussed earlier, vasopressin is an ancient hormone that, among other things, is associated with affiliative behavior in mammals and is believed to be more abundant in males than in females. In the brain, vasopressin operates mainly on subcortical neurons, especially those associated with fear and anger. As we have come to expect, there are interactive effects with other elements in the portfolio.

To have an effect on neurons, vasopressin has to bind to receptors on the neurons and fit into those receptors, key in lock. No receptors, no effect. Vasopressin receptor density in the amygdala, a complex subcortical structure, affects the nature of the responses to fear and anger and hence affects aggressive behavior. Moreover, vasopressin receptor density itself depends on the presence of androgens, including testosterone. Additionally, androgens stimulate the genes that express vasopressin itself. Given the right balance in the hormonal and receptor orchestra, the appropriate behavior will appear at the appropriate time. The right music flows forth. But there is lots of room for dissonance and disharmony.

For females, the neural basis for defense of offspring turns out to be somewhat different from that of males. Greatly simplified, here is the story. In females, progesterone inhibits aggression. Progesterone is produced in the ovaries, the adrenals, and, during pregnancy, in the placenta. Immediately after giving birth, the mother’s progesterone levels fall, but her oxytocin and vasopressin levels remain high. Oxytocin is essential to the expression of aggression, and especially in high-anxiety females, higher vasopressin levels correlate with strong aggressiveness. So much for labeling oxytocin “the cuddle hormone.” Biology is so much more complicated than that. The balance between progesterone and oxytocin is crucial if the mother is to be ferocious in defense of her brood.

The male and female brains differ also in the density of projections from the prefrontal cortex to subcortical areas involved in aggression, such as the amygdala and the BNST. In addition, the male brain has a higher density of receptors for androgens (male sex hormones) in the amygdala and the BNST than is seen in females. Bear in mind, too, that within a population, there can be a lot of natural variability in levels of hormones, density of receptors, sensitivity to environmental stimuli, gene-environment interaction, and so forth.

Does the circuitry for aggressive behavior in predation, defense, and sexual competition overlap? Multiuse wiring is probable. Circuitry in other areas appears to be multiuse, so we should not be surprised if it is here, too. If so, a consequence is that in mammals, aggression can be sloppy. It is not always strictly suitable to the stimulus or situation.

The summary point of this section is that yes, testosterone is a major factor in male aggressive behavior, but high levels of testosterone alone do not predict that the male will be especially aggressive. Aggression is a multidimensional motivational state.

Controlling and Harnessing Aggression
As we will discuss at length in Chapter 7, all mammals have connections between the prefrontal cortex and the subcortical structures to manage self-control. Here we will briefly consider self-control with respect to aggression in social mammals.

During evolution of the mammalian brain, the efficient and reliable circuitry already in place in reptiles for life-sustaining behavior was expanded and modified. Basic life-sustaining circuitry that worked well in reptiles was not dismantled; pain and pleasure, pivotal in learning mechanisms, were not demolished in favor of a wholly new design. Instead, they were modified and refitted. In mammals and birds, subcortical structures for aggression and defense, for fighting and fleeing, for drive and motivations (the “limbic brain”) are powerful and effective in maintaining our lives. The cortex is useless without them.

Pathways between the prefrontal cortex and the hypothalamus add flexibility and greater intelligence to emotional responses; they add foresight and creativity. They allow our behavior to be more considered and intelligent than instinctual.

The prefrontal cortex is richly connected to reward structures as well as to motivational and emotional circuitry. The connection to reward systems allows for shaping of the behavioral expression of emotions and drives and even for habitual suppression of certain kinds of behavior, such as aggression within the group. Social mammals learn what forms of kin competition are acceptable and what are not, and play in the young is a crucial part of that learning. Under the power of disapproval, the reward system restricts the occurrence of certain behaviors, such as aggression, to very specific conditions, such as the need for self-defense. Part of what is learned is how to shift attention elsewhere and how to damp-down the very feeling of anger. Social habits learned early are very deep in the brain’s system, and they do not change easily.

Hostility between groups, as well as within groups, is subject to local conventions. If you are an Inuit living in the far north, you will have grown up in a community that discourages aggressive behavior except under very specific conditions, usually involving territorial transgression or certain highly structured games. The anthropologist Franz Boas reported in 1888 that it was not uncommon for someone in an Intuit camp to kill a hunter who had strayed into the group’s hunting land. Nevertheless, the Inuit did not, so far as Boas could determine, engage in group warfare against each other. They did have summer trade meetings, where many groups would congregate, exchanging tools and allowing courtship among the adolescents.

The Yanomamo in Brazil, by contrast, tended to encourage aggressive behavior among children, teaching them combat skills (at least during recent times when anthropologists studied them). Under pressure of population growth, the adult males frequently engaged in raids against other groups. This seems to have been true also of the Haida, who lived on Haida Gwaii (formerly the Queen Charlotte Islands), off the north coast of British Columbia. In this they contrasted with the neighboring Tlingit and Salish tribes, who were raided, but usually not themselves raiders. (See also Chapter 6 for more discussion of these differences.)

We live in a matrix of social practices, practices that shape our expectations, our beliefs, our emotions, and our behavior—even our gut reactions. Our personalities and temperaments are bent and formed within the scaffolding of social reality. The matrix gives us status and strength, and, above all, predictability. The matrix of social conventions is both a boon and a bind—sort of the way a sailboat is. You need it to move through the waters, but you have to play by its rules.

Different sailboat configurations serve the sailor in different ways, and some of these configurations were not always obvious. The keel, so critical to sailing, was first invented by the Vikings around 800 CE, giving them a huge advantage in sea travel, not to mention conquering and pillaging. Likewise, social institutions, such as having an independent police force paid by taxes on the citizenry or allowing women to vote, were not obvious, at least not until they were put in place. Once tried with success, social institutions tend to stabilize, mature, and spread. They become second nature. Their justness seems so transparently, unmistakably self-evident. We make up myths to depict them as universal among all civilized humans, as having been in force since the dawn of Homo sapiens, or perhaps as having been handed down by a supernatural being.

A major target of exploration of this chapter has been the association of aggression and hate with pleasure. Oddly, there is very little research on the neurobiology of this association and not much psychological research either. At several points in thinking about the link, I have wondered whether I am just plain wrong in perceiving that the link exists. I suspect that I am not wrong. I expect that hostility does not always involve pleasure, but in some conditions, particularly when groups fight groups, the two seem closely linked. Evidently, also important is the hormonal balance and the receptor density and distribution for the various hormones. To a first approximation, human males and females display differences in aggressive behavior that are linked to male and female hormones, though these behavioral dispositions can be modulated by the cultural matrix.

At the same time, aggressive impulses in all mammals are subject to self-control. But before taking a closer look at how the brain regulates self-control, I want to explore a more basic question in the next chapter: Do our genes dispose us to warfare against other humans?