Four. Building a Personal Brain

When I was a fledgling psychiatrist, a colleague gave me a tip on how he gets to know a new patient. Early in the first visit he briefly imagines the patient as a ten-year-old child. The point of this exercise is to look past someone’s current troubles and picture the person as still little. Was she shy or popular? Was he a bully or a wimp?

I’ve found this tip useful because it immediately dials up compassion: The image of anyone as a child warms my heart. But it also creates a hunch to explore. Forming an imaginary picture of someone in grade school stimulates me to learn about the development of their personality.

When I got this tip in the 1960s, my limited knowledge of personality development was based on the ideas of Erik Erikson. A psychoanalyst who worked with children, Erikson thought we become ourselves by going through a series of well-defined stages as we progress from the extreme dependence of infancy to the responsibilities of adult life. The early stages seemed most important to him because he believed that they leave particularly enduring residues. As he explained in Childhood and Society:

Every adult ... was once a child. He was once small. A sense of smallness forms a substratum in his mind, ineradicably. His triumphs will be measured against this smallness, his defeats will substantiate it. The questions as to who is bigger and who can do or not do this or that, and to whom—these questions fill the adult’s inner life far beyond the necessities and the desirabilities which he understands and for which he plans.¹

Erikson’s view of personality is appealing because he reminds us of the lasting influence of childhood events. But two things are missing: genes and the brain. When Erikson wrote about the development of individual differences, he assumed that they were mainly due to upbringing and life experiences because very little was known about the influence of genetic variations. And when he described the transitions from one stage to the next, he thought of them primarily as psychological responses to a succession of challenges because very little was known about what was going on in the maturing brain.

This has changed. We now know a great deal about the way our brains develop under the guidance of our personal gene variants and our personal environments. Instead of just thinking of ourselves as solving the challenges of our youth with the brain we were born with, we have come to realize that each brain—like each face—has its own innate building plans. Furthermore, the brain’s building plan was not drafted by the systematic methods of professional architects. Instead, each brain uses a scheme that would drive contractors crazy, with continuous remodeling due to changes in both genetic and environmental instructions while the project is still underway.

This continuous remodeling has a purpose. By remaining open to the interactions of our unique set of genes and environments during the more than two decades of basic construction, we each come to have a truly personal brain. Within it are the deeply ingrained components of our unique personalities that continue to guide us for the rest of our lives.

The Brain Builds Itself

The adult human brain is built of about 100 billion nerve cells (neurons), most of which were made before we were born. But not all of these neurons were created equal. As the fertilized human egg divides, it generates many types of primitive neurons, each of which is destined to play a particular role in the brain. Having been assigned their approximate fates by a process that turns on and off specific genes, the primitive neurons migrate to their designated places guided by chemical signals that they selectively respond to. When they get there, they start building connections with other neurons to form the neuronal circuits and networks that are the basis of all our behavior.

To build these connections, the neurons make branches called dendrites to receive signals and other branches called axons to send signals. Dendrites are short and studded with spines. Axons can be long enough to reach other neurons anywhere in the brain and to embrace them with clusters of little nerve endings, called boutons. Signaling between boutons of one neuron and dendrites of another occurs at structures called synapses.

A synapse is activated when a bouton releases a chemical neurotransmitter such as serotonin or dopamine onto the spine of a dendrite. The neurotransmitter travels across the synapse and binds to receptors embedded on the spine. This transmits information to the dendrite, a process called synaptic signaling or synaptic transmission.

Many types of synaptic signaling exist between neurons, governed by the dozens of different chemical neurotransmitters that are squirted from boutons onto receptors on the spines. Every neuron manufactures a particular neurotransmitter and displays a particular set of receptors. So every neuron has both a spatial address, defined by its location in a particular brain circuit, and a chemical signature, defined by its neurotransmitter and receptors.

The complicated process of spatial assembly of neurons into circuits and networks is well on its way by the time a person is born. Among the circuits that operate in infancy are some in the amygdala, a brain structure that I mentioned in discussing the SERT gene. The amygdala is a hub for a complex set of circuits that integrate our emotions. Using these infantile circuits, babies experience joy, contentment, fear, anger, and the distress of separation. Neuronal controls of these emotions are gradually put in place over the next two decades, and they have major effects on the developing personality.

Circuit maturation doesn’t depend only on adding new synaptic connections. While useful ones are strengthened, others are eliminated. The same selective remodeling process is also applied to the neurons themselves. Some of them grow and sprout more branches; others are destroyed by a specialized mechanism of cell death called apoptosis, which is an indispensable part of the developmental process. Much of this happens in fetal life and during the first few years after birth, but some goes on through adolescence and into adulthood.

A notable case of remodeling occurs in a group of neurons in the hypothalamus that play an essential role in the establishment of female or male patterns of sexual behavior. In the female fetus, these neurons die off as part of the developmental program that sets up female-specific sexual circuits. But in the male fetus, testosterone from the fetal testicles rescues these neurons from the apoptotic grim reaper and stimulates them to build male-specific brain circuits.² The timing of this effect of testosterone is crucial. If it comes too late in fetal development, the key neurons in the hypothalamus are already dead, and the brain is set on an irreversible female course. Other regulators of neuronal death may also have decisive behavioral effects, but none is as obvious as testosterone.

Brain circuits can also be modified by progressively wrapping axons with a fatty substance called myelin. Myelin acts like the insulation around an electrical cord, which facilitates the speed of conduction of electrical signals. Myelination is often a final and essential step in the genetically controlled development of a circuit.

Although this overall developmental program is at work in all of us, each of our brains is different because their structural details are influenced by thousands of gene variants in our personal genomes. There is also a little sloppiness in the assembly process, due to random variations in the movement of neurons and in the expression of critical genes. This is one reason that even the brains of identical twins are not exactly the same.³

Understanding the step-by step nature of brain construction explains why it is so difficult to go back and make changes in brain circuitry and in the aspects of personality that the circuits control. Once neurons have taken up their positions, they are pretty well settled. Once they have established useful connections, those connections tend to be maintained. Although there is always some residual capacity for change, it takes a lot of work to remodel structures that are built by a developmental program that unfolds over more than two decades. Even our extraordinary human ability to learn new things may not be up to the challenge of modifying patterns that were laid down in this way. This is true not only of patterns that were strongly influenced by genes. It is equally true of those patterns that were shaped by our personal environments during phases of brain development called critical periods.

Critical Periods in Brain Development

A critical period is a window in time when certain brain circuits are open to essential environmental information. Arrival of this information shapes the circuits in a lasting way.⁴ Once this shaping is completed, the window is closed.

The most famous example of a critical period comes from Konrad Lorenz, who studied the behavior of baby geese. Lorenz found that each baby is primed to pay special attention to the first moving creature it sees after hatching—generally, its mother. This information is immediately imprinted in its brain, which leads it to follow its mother in those cute little trails of goslings. But if the mother goose is removed during hatching and replaced by another moving creature—such as Lorenz himself—the babies may imprint on him instead. The result is recorded in pictures of goslings trailing the bearded scientist.

Another well-known example is the development of the vocalizations of male songbirds, and it, too, involves a social interaction. In this case, the critical period of brain development is not confined to the minutes after hatching, but lasts for a few months. During this time, each juvenile male bird shapes its simple innate song by progressively matching it to the complex song of an adult male.⁵ Without such instruction during this critical period, it will never be able to sing like an adult.

These critical periods in goslings and songbirds provide the opportunity to incorporate essential environmental information that is uniquely valuable to each species. For humans, a notable example is learning to speak, which develops during a critical period that lasts for more than a decade.⁶ During this period, children don’t only learn their native language. They also pick up the accent of the people they grow up with, especially their peers.⁷ As this critical period closes, it becomes very difficult to speak like a native. This is why immigrants such as Henry Kissinger, who learned English in his teens, speak with a foreign accent. Even natives who migrate to a different region can be spotted in this way: Four decades in California have not erased the vestiges of my own linguistic imprinting in New York City.

Although researchers have studied these critical periods of brain development for many years, we still have limited information about their number and the ways they are closed. But the main message is clear: Certain brain circuits become established at particular times, and their properties tend to endure. A similar process appears to be at work in the development of many aspects of our personalities.

What Will My Child Be Like?

Although a baby is born with an immature brain, it immediately becomes a player in the world. At first, it can only cry to signal distress or coo to signal contentment. But its behavioral repertoire grows rapidly in its first few years of life as it builds new brain circuits and remodels others.

As brain development continues, parents begin wondering if their child’s early patterns of behavior can provide clues about his or her mature personality. Researchers have tried to answer this question by examining children repeatedly from infancy to adulthood. Because each research group uses its own system for describing behavioral patterns, it’s difficult to compare the results. Nevertheless, there is general agreement that early patterns persist in some children, whereas other children change a lot.

Evidence for some persistence of patterns comes from pioneering studies by Stella Chess and Alexander Thomas,⁸ a wife-and-husband team of child psychiatrists. From their observations of babies they identified three broad patterns of behavior, which they called temperaments. Forty percent of the babies were called “easy” because they approached new situations without difficulty, had high adaptability to change, accepted most frustration with little fuss, and were not very moody. In contrast, the 10% of the babies who were called “difficult” were much more inclined to be irritable, showed intense negative emotions, and had trouble adapting to change. Another 15%, called “slow to warm up,” were initially uncomfortable in new situations but adapted after repeated contact. The remaining 35% showed a mixed picture.

Follow-ups of the children as young adults indicated that there were “only modest levels of consistency in temperament over time for a group of subjects as a whole.” Their conclusion in 1986 fits well with what we know today: “Maturational factors, neurophysiologic changes, and a host of environmental influences—all these serve to produce continuity in some individuals and change in others.”⁹

A series of studies led by Jerome Kagan also found evidence for both continuity and change. Kagan identified subgroups of children that he called inhibited and uninhibited, based on their willingness to engage with unfamiliar people when they were 2 and 7 years old. When he re-examined them in adolescence, he found that the majority of the children in the inhibited group remained quiet and serious, while only 15% were as lively and talkative as the average teen from the uninhibited group. Of the children in the uninhibited group, 40% maintained that style as teens, and only 5% had become subdued and quiet. As Kagan summed it up, “[A]bout one-half the adolescents retained their expectable demeanor, while only 15 percent had changed in a major way.”¹⁰

Some behavioral continuity of members of the two groups was also observed at age 22. In brain imaging studies, the inhibited group showed significantly more activation of the amygdala when shown pictures of unfamiliar faces.¹¹ This sign of a stronger emotional response to new faces is reminiscent of their greater wariness of strangers as toddlers. Other researchers have also found that children retain many of their characteristics as adults.¹²

Evidence of continuity into adulthood is particularly strong for a subgroup of children who show signs of antisocial behavior in grade school. If they are sufficiently aggressive and impulsive to be singled out as having a conduct disorder before the age of 10, they tend to maintain this antisocial pattern when they have grown up.¹³ In contrast, children who don’t show signs of antisocial behavior until their teens are more easily reformed and have a better chance of becoming law-abiding adults.¹⁴

Other kinds of behavior that are prominent in childhood may change dramatically in adolescence, including some behaviors that are known to be heritable. For example, heritable childhood fears of heights, snakes, or blood frequently disappear by the time the children are in their teens.¹⁵ How these and other waxing and waning genetic effects eventually play out also depends, in part, on interactions with the person’s environment.

Gene–Environment Dialogues

Persuasive evidence of the combined effects of environment and genes comes from studies of people with antisocial personalities.¹⁶ Everyone who has watched The Sopranos knows that antisocial behavior runs in families, and studies show that 10% of a community’s families commit most of its crimes.¹⁷ So you won’t be surprised to learn that studies with twins show a 40% to 50% heritability of antisocial traits.¹⁸ But in this case, family environment also has a significant effect. Furthermore, adopted children raised in antisocial families have an increased risk of developing an antisocial personality pattern,¹⁹ even though they are genetically unrelated.

Added support for the importance of family environment comes from a study of a group of children in Dunedin, New Zealand. The researchers enrolled all of the 1,037 children born in this city from April 1972 through March 1973, assessed them at multiple intervals through the age of 26, and stored the data for subsequent analysis. This provided detailed information about child development in the entire community without preconceptions about what might show up.

One notable finding was that many of the children were abused: 8% had “severe” maltreatment, 28% had “probable” maltreatment, and only 64% had no maltreatment.²⁰ But this should not be taken to mean that New Zealanders are particularly nasty. In carefully controlled interviews of 8,667 American adults, 22% reported sexual abuse during childhood, 21% reported physical abuse, and 14% reported witnessing their mother being beaten; many reported all three.²¹ A substantial portion also described repeated emotional abuse.

Having detected considerable child abuse in Dunedin, the researchers wondered whether it was correlated with the development of an antisocial personality pattern. To answer this question, they concentrated on boys because they are more likely than girls to develop this pattern. They found that the degree of maltreatment of the boys was, indeed, correlated with the degree of antisocial behavior. But there was considerable individual variation. Some of the severely maltreated boys developed a troublesome antisocial pattern, whereas others did not.²² Why?

One possibility is that the boys who became antisocial had a genetic predisposition to turn out this way. For example, they might have been innately defiant or aggressive, which might have called forth more abuse. Such interactions between a child’s innate tendencies and parental reactions are one reason children raised in the same family turn out to be so different,²³ and this likely played some role in the Dunedin study. But in this case, the researchers decided to get more specific by looking for a single gene variant that influenced the antisocial outcome.

To get started, they examined a plausible suspect: the MAOA gene. This gene makes monoamine oxidase-A, an enzyme that degrades serotonin, norepinephrine, and dopamine, three neurotransmitters that control brain circuits involved in emotional behaviors. Two characteristics of the MAOA gene made it seem relevant: brain levels of monoamine oxidase-A influence many types of antisocial behavior;²⁴ and variants of the MAOA gene’s promoter control the manufacture of different amounts of the monoamine oxidase-A enzyme in the brain.²⁵ Furthermore, the MAOA gene happens to be located on the X chromosome, which simplifies its study in boys because they have only one copy (girls have two). In the Dunedin study, 63% of the boys had the high-MAOA variant, which makes a lot of the enzyme in the brain, and 37% had the low-MAOA variant, which makes less of it.²⁶

Is having the high- or low-MAOA gene variant correlated with antisocial behavior? The researchers found that, by itself, it is not. Boys who hadn’t been abused had little antisocial behavior, regardless of which variant they had. But among the abused children, there was a significant effect. Those abused children with the low-MAOA variant were more likely to become antisocial.²⁷

Several subsequent studies of antisocial men support these findings.²⁸ So does a study of women from an American Indian tribe who had experienced childhood sexual abuse.²⁹ In this case, too, abuse was correlated with an antisocial pattern of behavior, and those abused women with two copies of the low-MAOA gene (one on each of their X chromosomes) had the highest rate of antisocial behavior. In contrast, those with two high-MAOA genes had the lowest rate of antisocial behavior. Furthermore, as with men, the MAOA gene didn’t matter in the absence of abuse.

This doesn’t mean that being born with the low-MAOA variant is bad news. Having more or less monoamine oxidase-A has multiple effects on brain functions,³⁰ and these may have desirable or undesirable consequences. The outcome depends on individual circumstances, other gene variants, and one’s taste in personalities. The big story from the studies of childhood abuse and MAOA is more general. It illustrates the principle that genetic differences can influence the effects of childhood environments on a personality.

Enduring Effects on Gene Expression

It also works the other way: Environment can have enduring effects on the expression of particular genes that affect behavior. The best example comes from studies in Michael Meaney’s laboratory of the effects of rat mothering on the personalities of their pups. The studies began by comparing the behavior of the offspring of two types of rat mothers: high-lickers who licked and groomed their pups vigorously, and low-lickers who were less enthusiastic.³¹ When these offspring were tested months later, those raised by the high-lickers were less fearful and less reactive to stress than those raised by the low-lickers. Furthermore, their greater emotional stability was apparent not only in behavioral tests, such as open field activity, but also in their blood levels of glucocorticoids, stress-related hormones released from the adrenal gland.

Was the greater emotional stability of the highly licked pups caused by the maternal behavior (nurture)? Or did the high-licking mothers also have genetic differences that were transmitted to their pups via their DNA (nature)? To find the answer, pups born to high-licking mothers were swapped immediately after birth with those born to low-licking mothers, the adoption tactic that Galton had proposed to distinguish nurture from nature. The results of this cross-fostering pointed to nurture, the maternal behavior, rather than the maternal genes. High-licking foster mothers did just as good a job as high-licking biological mothers in producing stress-resistant pups, and vice versa.

Having observed this behavioral result, Meaney and his colleagues looked for differences in the brains of the two groups of pups. They found that the highly licked animals had a more active form of the gene that makes the glucocorticoid receptor (GR), a protein that responds to glucocorticoid hormones. This change, which was observed in neurons in brain circuits that control emotions, was already detectable in the pups’ brains during the first week of nursing and was maintained throughout their lives.

To find out how this came about, the researchers searched for modifications in the promoter part of the GR gene, which regulates the gene’s activity. It is known that promoters can be modified by a natural biochemical reaction, called an epigenetic change (from the Greek epi, which means “over” or “above”), which adds or removes a tiny methyl group at a precise point in their DNA, and that an epigenetic change may modify the promoter’s effectiveness and alter the activity of the gene. The researchers discovered that the promoter of the GR gene was less methylated in the highly licked animals and that this change of their brain DNA, which was caused by their mothering, led to an increase in the manufacture of the gene’s protein product, the glucocorticoid receptor.³²

Furthermore, the behaviorally induced change in the methylation of the gene’s promoter was maintained in the highly licked animals as they grew up. So, too, was the activity of the GR gene. This suggested that the enduring epigenetic change in the DNA of these animals, and the resultant increase in glucocorticoid receptors, had shifted the settings of a brain circuit that controls the stress response. The result was a sustained effect on their personalities.³³

The research with high-licking mothers has attracted a lot of attention because it has something for everyone. Geneticists like it because it demonstrates the importance of an environmentally induced chemical modification of a gene. Psychologists like it because it shows that behavior can affect genes as dramatically as genes can affect behavior. Neuroscientists like it because it adds to their understanding of the ways that experience can produce a sustained change in brain circuits. And, to all of them, a major implication of these studies is that experiences, especially those in early life,³⁴ can produce epigenetic modifications of DNA that have enduring effects on personality.

Such environmentally induced epigenetic changes keep accumulating as we grow up. One way we know this is from studies of identical twins. Derived from a single fertilized egg, these twins start out with identical DNA. Nevertheless, the methylation pattern of their DNA becomes progressively different as the twins grow older.³⁵ These epigenetic differences in the DNA of identical twins are believed to be due, in part, to the many differences in the environments the two twins grew up in. Although the functional significance of these epigenetic differences is not yet known, it is reasonable to assume that they give rise to some of the observable differences between identical twins, including differences in their personalities.³⁶

Adolescent Remodeling

Although a great deal of brain development takes place in fetal life and childhood, extensive remodeling also occurs in our teens. Some of this structural remodeling is initiated by a few thousand specialized neurons in the hypothalamus that trigger the hormonal changes of puberty. These neurons make a small protein, gonadotropin-releasing hormone (GnRH), which signals the pituitary gland to activate the ovaries or testes to secrete estrogen in girls and testosterone in boys.³⁷ Bursts of these hormones then modify, enlarge, and activate the brain circuits for sexual behavior that were first built in the fetus.³⁸

The sex hormones also do much more. By activating neurons that have receptors for estrogen or testosterone, they change the activity and settings of many other brain circuits. This gives rise to behavioral changes that are typical of adolescence, such as increased sexual interest, risk taking, impulsivity, and social awareness.³⁹

But sex hormones are only one factor in the brain remodeling and behavioral changes of adolescence. Many other sex-specific changes in brain gene expression don’t depend on these hormones. Both the hormone-induced and the hormone-independent processes lead to enduring modifications in brain circuits, some of which distinguish male from female brains.⁴⁰

As in other periods of brain development, adolescence provides opportunities for genetic variations to make themselves felt. For example, some gene variants that influence cognitive abilities may not exert their full effects until the mid-teens. We know this, in part, from studies of the IQs of adopted children. These studies show that their IQs become progressively more like those of their biological parents during adolescence, as the influence of gene variants that influence cognitive abilities becomes more apparent.⁴¹ This increasing effect of the gene variants that influence cognitive abilities was confirmed in a study of 11,000 pairs of identical or fraternal twins. The researchers found that the heritability of general cognitive abilities increased from 41% at age 7, to 55% at age 12, and to 66% at age 17.⁴²

Adolescent brain remodeling is not apparent solely from the behavioral changes of the teen years. It has also been observed directly by magnetic resonance imaging (MRI) of brain structures at various stages of development.⁴³ The most extensively studied anatomical changes are those in the front part of the brain, especially the prefrontal cortex, which sits behind the forehead. As adolescence progresses, changes take place in the structure of regions of prefrontal cortex and their connections to brain regions such as the amygdala, which regulates emotional expression.

Changes in the connectivity and organization of brain networks during adolescence and early adulthood has not been observed just by looking at static brain structure. Functional magnetic resonance imaging (fMRI), which measures the activity of brain circuits during the performance of mental tasks, has also been used. These studies of mental activity reveal substantial changes in the functional connectivity of the brain in the progression from adolescence to adulthood.⁴⁴

The long critical period of adolescence is also open to environmental influences. While the brain is actively rewiring, life goes on, and peers play extremely important roles in transmitting values and social skills.⁴⁵ This openness to peer influence is of particular interest to parents, educators, and clinicians, who would like to prevent the many troublesome personality patterns that start showing up at this stage of life.⁴⁶

Closing Some Windows in the Brain and the Environment

When is brain development completed? MRI studies of individuals show that brain structure stabilizes at around age 25.⁴⁷ Although a little more myelination may continue for at least another decade,⁴⁸ changes that show up on brain scans after age 40 are generally signs of wear and tear rather than additional developmental remodeling. Furthermore, studies of the integrated activity of brain regions that is measured by functional MRI show that mature brain networks are also well established by young adulthood.⁴⁹

This doesn’t mean that the adult brain has become fixed and immutable. One of its most important functions is to keep learning and storing new information by making microscopic changes in the structure and function of synapses. Nevertheless, young adulthood marks a milestone in brain development, when we have largely built the personal instrument that will continue to guide us for the rest of our lives.

Development of basic personality traits follows a similar trend but lags behind. As anatomical changes in the brain are winding down in our third decade, changes in the Big Five are winding down too. Repeated testing shows considerable stabilization of a person’s Big Five scores by age 20, significantly more stabilization by age 30, and a little more stabilization until about age 50.⁵⁰

This progressive stabilization is not only due to the closing of windows of brain development. As Roberts and Caspi point out,⁵¹ it is also due to the increasing constancy of the young adult’s social environment. This is the environment that is populated by the friends, partners, and coworkers whom they have selected—and who have selected them.

The result of selecting a fairly constant social environment during young adulthood is that we subsequently spend most of our time with a limited cast of familiar people. These people provide stability because they keep behaving in ways that we have come to expect. They also elicit stability because they keep us behaving in ways that they have come to expect. This mutual stabilization of our social environment plays a big part in the creation and maintenance of the two overarching aspects of personality that I now turn to: character and sense of identity.