Thursday, January 01, 2004

New Insights into Temperament

By: Jerome Kagan, Ph.D.

Introvert-extrovert, shy-bold, loner-outgoing, and similar distinctions we make about people are rooted in differences that can be identified in children four months old. Harvard psychologist Jerome Kagan’s research over many decades has transformed the centuries-old idea of human temperaments into one of the most rigorously defined concepts in the modern science of personality. Now, Kagan and his colleagues have evidence that these early differences correlate with measurable differences in the brain. Does it all add up to determinism? Has nature trumped nurture? Not at all, says Kagan, but to understand how nature and choice interact, we must relinquish our premise that we start life with no personality. Then, child and parent alike can shape the intricate process that is the development of a human being.

—This essay is adapted from The Long Shadow of Temperament (Harvard University Press, fall 2004).

The brain, and its heir the mind, remain continually receptive to any unexpected interruption as they move their host, the body, to places that provide food, shelter, affiliation, achievement, play, enhanced status, protection, or love. Three types of experiences demanding immediate attention disrupt these well-practiced routines: signs of imminent, or actual, damage to the body’s integrity; sudden loss of a desired relationship with another; and unfamiliar events, which can be a familiar experience occurring at an unexpected time or in an unexpected location, or an event that is somewhat different from one the mind knows well. Humans seem to vary more in how they react to unfamiliarity than to illness, attack, criticism, or separation from a significant other, and this category of intrusion occurs far more often than the first two.

To understand the significance of this susceptibility to surprise for how our personality develops, how we face life, and even our vulnerability to anxiety disorders, we must examine its roots in biology as well as psychology. Each time a person encounters something genuinely unexpected—for example, a child with no arms or a black fire engine—a cascade of brain processes begins that results in a psychological reaction less than a second later. The cascade involves many brain areas, but neuroscientists believe that the most relevant areas are located within the temporal lobe, especially the entorhinal cortex, hippocampus, and amygdala. 

Despite its small size, the almond-shaped amygdala consists of many clusters of brain cells, each with a distinct neurochemistry, pattern of connections, and set of functions. The most important anatomical distinction is between the basolateral and medial areas (which receive stimulation from the outside environment and the viscera as well as from the brain’s entorhinal cortex and hippocampus) and the central area, which receives this information and provokes changes in the autonomic nervous system, posture, distress vocalizations, and stress hormone levels.1 A person usually interrupts his ongoing activity, directs his attention to the event, and tries to relate his feelings and thoughts to his existing knowledge. If he does not immediately recognize the event, or senses it is dangerous, he becomes immobile and is likely to experience changes in heart rate, blood pressure, and sweat gland activity. 


Although these general outlines of how the brain reacts to surprise are constant, different individuals of every vertebrate species seem to have different preferences for approaching or withdrawing from unfamiliar objects, animate or inanimate, and exploring or remaining still in unfamiliar places. Animals that are initially reluctant to approach or to explore the unfamiliar are called fearful. More than 75 years ago, the Russian physiologist Ivan Pavlov noted that some dogs in his St. Petersburg laboratory were minimally fearful of humans, whereas others cowered whenever a laboratory assistant approached or made an unexpected movement. Pavlov suggested that this difference was probably influenced by variation in brain activity. 

John Paul Scott, Ph.D., and John Fuller, Ph.D., studied these complementary traits in five breeds of puppies housed in the Jackson Laboratory in Bar Harbor, Maine.2 In one test, puppies that ran to the corner of the common room with a high-pitched yelp as the handler approached were categorized as timid. Three breeds—basenjis, fox terriers, and Shetland sheepdogs—were significantly more timid than beagles and cocker spaniels, even though there are variations within each breed and, indeed, within a single litter. There are timid and bold house cats, as well, and Robert Adamec, Ph.D., has shown that variation in amygdalar excitability contributes to the behavioral reaction of a cat encountering an unfamiliar animal, object, or location. 

So it is perhaps not surprising that, in the presence of an unfamiliar person or setting, a proportion of young children usually approach while a smaller proportion typically avoid doing so. These complementary traits are called temperamental biases when they result, in part, from inherited differences in biology. On a fine-grained level, there are many such biases, each influenced by a distinct neurochemical profile that will most likely be discovered in the future. An infant’s environment soon begins to sculpt these biases into a personality type, however, and by the time the child is six or seven years of age, it is difficult to detect the initial temperament, which, like a drop of colored ink stirred in a vessel of glycerin, is no longer visible. I rephrase a sentence written by Willard van Orman Quine, Ph.D., to suggest that every psychological quality in an adult can be likened to a pale gray fabric woven from thin black threads representing biology and white threads representing experience, neither visible in the homogeneously gray cloth. 

Nonetheless, the heritability of these two temperamental biases in a large sample of identical and fraternal twins observed in a laboratory at the University of Colorado and at home at ages 14, 20, 24, and 36 months approached 0.5—a value close to the heritability of IQ scores among siblings. The heritability estimate was more than 0.9 when the sample was restricted to children who were extremely timid or bold while playing with two other unfamiliar children. Longitudinal studies of children growing up in Berkeley, California, as well as in New Zealand, affirm the modest but definite continuing influence of these two temperaments from early childhood to adolescence and even adulthood.3

Although most parents notice these two temperamental types in the child’s second year, when he or she meets unfamiliar adults or children, their defining feature is not shyness or sociability with strangers, but the initial posture to any unfamiliar event. As one parent put it, “My son isn’t shy, he just doesn’t like newness.” 

A temperamental bias toward reacting to the unfamiliar with caution and avoidance, which my research colleagues and I call inhibition, helps explain why some children —but only a minority—remain anxious after a traumatic experience such as earthquake, hurricane, kidnapping, or exposure to violence. For example, only 10 of 40 school children who had been kidnapped and terrorized for two days developed symptoms of posttraumatic stress disorder. Similarly, a group of children who developed severe anxiety reactions several months after Hurricane Andrew struck South Florida had been evaluated as inhibited before the storm.4


My colleagues and I became interested in temperamental types and their possible basis in the brain for several reasons. First, we were intrigued by the work of Lewis Lipsitt, Ph.D., and his colleagues at Brown University, which showed that newborn infants who increased their rate of sucking when the liquid they were receiving unexpectedly turned sweet became inhibited children two years later.5 Because the unexpected change in taste would have activated the amygdala, it was reasonable to assume that the newborns who showed a large increase in sucking rate and became inhibited toddlers were born with a lower threshold of excitability in the amygdala. 

In addition, more than 40 years ago, Howard Moss, Ph.D., and I discovered that a small group of southwestern Ohio children who were inhibited during the first three years of life retained derivatives of this bias through young adulthood. These adults were introverted, cautious, and overly dependent for support on their parents or others. By contrast, the larger group of bold, sociable children, called uninhibited, developed the opposite set of adult characteristics.6

Fifteen years later, Richard Kearsley, M.D., Philip Zelazo, Ph.D., and I, probing the effects of daycare on Chinese-American and Caucasian infants born in Boston, discovered that the Chinese children, whether reared only at home or attending daycare, were more inhibited in the second year than a matched sample of Caucasian infants. We also learned that more Chinese than Caucasian mothers described their children as apprehensive when they encountered unfamiliar events.7

Those observations motivated me, along with my colleagues Nancy Snidman, Ph.D., and Doreen Arcus, Ph.D., to study a large group of four-month-old, middle-class, healthy Caucasian infants to see whether inhibited or uninhibited behavior in later childhood could be predicted from an infant’s behavior in relation to unfamiliar events. Because the amygdala sends projections to those brain stem structures that mediate vigorous motor behavior and crying, we assumed that the infants who inherited a neurochemistry that rendered the amygdala excitable should show higher levels of limb activity and distress in reaction to unfamiliar visual, auditory, and olfactory stimuli, and that these infants should be biased to become inhibited toddlers. By contrast, infants who showed minimal motor activity and little distress to the same unfamiliar events, who presumably inherited a less excitable amygdala, should become uninhibited children. About 20 percent of the infants showed a combination of vigorous pumping of legs and arms, occasionally accompanied by arching of the back, and frequent crying in reaction to unfamiliar mobiles, taped voices speaking simple sentences, and the smell of dilute butyl alcohol. We called these infants “high reactive.” The 40 percent of the infants who displayed the complementary pattern, minimal motor activity and no distress, were called “low reactive.” 

We observed these two groups again when they were 14 and 21 months old. For example, we noted whether children in the second year cried, withdrew, or resisted when the examiner placed heart rate electrodes on their chest or a blood pressure cuff around their arm, asked them to put their finger into a glass of black fluid, or placed a drop of water on their tongue. In a fourth test, the examiner uncovered an unfamiliar rotating toy, smiled, and spoke a nonsense syllable followed by the child’s name in a friendly voice. On the next trial, though, the examiner frowned and spoke in a stern voice. Some children cried in reaction to the unexpected change in the examiner’s face and voice; others were unperturbed. We also recorded each child’s behavior in response to the appearance of an unfamiliar woman in a white laboratory coat wearing a gas mask, a large novel metal robot that moved unexpectedly, and a person dressed in a clown costume with a red and white mask who invited the child to approach and play. The infants who had been high reactive at four months were more fearful in their reactions to the entire battery of 17 unfamiliar episodes than were the low reactive children.8

At 4 1/2 years of age, the children who had been high reactive were quiet and subdued while being interviewed by an unfamiliar woman and extremely shy while playing with two other unfamiliar children of the same sex and age. But the children who had been low reactive as infants were spontaneous with the examiner and sociable with the other children. At age 71/2 years, 45 percent of the high reactive children, but only 15 percent of the low-reactive children, had developed two or more anxious symptoms, for example, a serious fear of the dark or of large animals, concern about being kidnapped, reluctance to sleep at a friend’s house, or the need of a night-light. 


Our last assessment of these children occurred when they were 11 years old. We used measures of their social behavior but also evaluated them on four biological measures, selected because other research suggested that these measures were indirect indicators of the level of amygdalar excitability. 

One indicator is greater activity in the right hemisphere, compared with the left, especially in the frontal area, a difference that we measured using an electroencephalogram (EEG). Many scientists have established that the left hemisphere is usually more active than the right hemisphere when an individual is in a happy, relaxed state. By contrast, adults who report chronic tension or anxiety are more likely to show greater EEG activation in the right frontal lobe. This asymmetry could be the result of inherent differences in excitability within each hemisphere, differences in input to the cortex from the limbic system or brain stem, or both. Feedback from activity in the cardiovascular system, gut, and muscles is greater in the right than in the left amygdala. Because the left and right amygdala send their projections to the cortex on the same side (that is, right amygdala to right hemisphere and left amygdala to the cortex in the left hemisphere), children with a more reactive viscera should have a more reactive right amygdala and, as a result, greater desynchronization of alpha frequencies in the right hemisphere. That is, these children would show right hemisphere activation. 

A second indicator of amygdalar excitability is the size of the response evoked from a set of neurons in the brain stem (the inferior colliculus) by a series of click sounds. This response, which scientists called Wave 5, is influenced by the excitability of the amygdala because the amygdala indirectly primes the inferior colliculus. For example, stimulating the inferior colliculus of rats elicits behaviors scientists regard as signs of fear, including body immobility and reluctance to enter the brightly lit arms of a maze. If high reactive individuals have a more excitable amygdala, they should show a larger Wave 5 response to a sequence of click sounds than low reactive individuals. 

The amygdala also projects to the sympathetic nervous system, including the cardiovascular system. Therefore greater amygdalar excitability should be correlated with a faster heart rate and a greater influence of cyclical changes in blood pressure and temperature on heart rate. These differences reflect sympathetic nervous system activity. The changes were our third indicator. 

The fourth indicator was the particular EEG waveform called ERP in response to pictures of unfamiliar objects or scenes, such as a picture of a car on a cloud or hearing the sentence: “The street was covered with planets.” Because of the influence of the amygdala on parts of the cortex, children with a more excitable amygdala should show a larger magnitude waveform when confronted with unfamiliar, impossible scenes, which were, in addition, unexpected, because they occurred only 15 percent of the time in a series of 169 pictures. 

Our expectations for all four biological measures were confirmed. More 11-year-olds who had been high, compared with low, reactive at four months showed greater right, rather than left, hemisphere activation in the EEG, a larger Wave 5, greater sympathetic activity in the cardiovascular system, and a larger ERP waveform when reacting to the unfamiliar pictures. 

It was not surprising that more high reactive children were emotionally subdued when they interacted with the unfamiliar examiner; they smiled and talked less often than the low reactive children. The high reactive children also described themselves as not liking new experiences. Each child ranked a printed set of 20 traits from most to least descriptive of self. Some items were: “I like to play with many children at recess,” “I like going on roller coasters at amusement parks,” or “I like going to new places.” Almost one half of the low reactive children, but only 15 percent of the high reactive children, not only described themselves as liking novel experiences but also showed low values on the four biological measures. By contrast, almost one half of the high reactive children, but less than a quarter of the low reactive children, combined a dislike of new experiences with high values on the biological measures. 


Although about one of every four children who had been high or low reactive developed a behavioral and biological profile that was in accord with their infant temperament, very few children—fewer than 1 in 20— developed the profile characteristic of the complementary group. Thus, the prediction that a high reactive infant will not become a sociable, exuberant 11-year-old with left frontal activation, a small Wave 5, low sympathetic tone, and a small magnitude ERP can be made with much greater confidence than the prediction that this child will be a shy, timid adolescent with high levels of cortical and autonomic arousal. And the prediction that a low reactive infant will not become an extremely shy 11-yearold with high biological arousal is much more certain than the prediction that this child will be exuberant and possess low biological arousal. In other words, each temperamental bias limits the acquisition of the features of the opposite group; a temperamental bias eliminates many more future psychological possibilities than it ensures. A temperamental bias can be likened to the basic form of the song of a particular species of bird. The animal’s genome constrains the basic architecture of the song but does not determine all of its features; within broad limits, the adult song depends on exposure to the songs of other birds and the opportunity to hear its own vocalizations. 

A similar principle holds true for the effect of environmental influences on people. Predict the adult psychological profile of a group of children born to economically secure, well-educated, nurturing parents, and your most accurate estimates will be for characteristics that the adult will not develop: criminality, school failure, psychosis, homelessness, drug addiction, and chronic poverty. Your predictions of the more specific components of the adult personality, such as career, quality of marriage, political attitudes, or number of close friends, will be far less reliable. 

Imagine someone trying to predict the final resting place of a stone rolling down a steep mountain over a three-minute interval. You will be able to eliminate a great many final locations after each 10 seconds of descent, but it is not until the final second that you will be able to predict exactly where the stone will rest. When the promises of the Human Genome Project are met, and parents can request the complete genetic analysis of their newborn, experts will be better able to say what the infant will not become— schizophrenic, manic depressive, talented athlete, or brilliant mathematician—than to tell parents the specific characteristics their infant will possess two decades later. 


Members of our two temperamental groups, high reactive and low reactive, almost certainly differ in their usual “feeling tone,” which originates in bodily activity. Some individuals feel relaxed most of the time, whereas others feel chronically tense. It is not surprising that the poet T.S. Eliot, who wrote in The Waste Land, “I will show you fear in a handful of dust,” was an inhibited child and introverted adult. 

People usually assign some kind of meaning to unexpected and unpleasant changes in bodily sensation. In our society, we often interpret those sensations as meaning that we are tense, worried, or guilty. High reactive individuals, with their greater levels of autonomic and cortical arousal, are probably more vulnerable than low reactive individuals to such changes and therefore to more frequent moments of self-doubt. Most 11-yearold children who had been high reactive, compared with low reactive, reported that their greatest fear was criticism from others for not doing what was expected. By contrast, the low reactive individuals, who possess a more relaxed feeling tone, brood less often over a brief fall from grace and carry a surer sense of their essential goodness as long as life is not too harsh. The high reactive 11-year-olds who reported that they felt bad when a parent criticized them for doing something wrong showed more of the biological signs of amygdalar reactivity than high reactive children who did not report frequent guilt, as well as more than the low reactive children who reported feeling bad after parental criticism. Further, the mothers of high reactive girls were more likely than mothers of low reactive girls to describe their daughters as unusually sensitive to punishment. All this suggests that a temperamental bias can contribute to the probability of whether a child will violate a conventional ethical standard promoted by family and community. 

The egalitarian ideal that many Americans hold dear is not completely compatible with the fact that people inherit biases for experiencing very different conscious feelings. Our consciousness, after all, is where will and, therefore, freedom of choice reside. Thus, we can accept that there are inherent differences in, for example, mathematical ability, musical talent, or athletic skill, but many people are less ready to acknowledge variation in qualities of consciousness, because this acceptance carries the uncomfortable implication that all humans are not equally capable of experiencing shame, guilt, joy, and empathy. These emotions are as fundamental to the harmony of a society as the interbreeding of two animal strains is to evolution. If a chronically jovial mood is, in part, the product of an inherited physiology, sanguine adults have an unfair advantage. Should not a sanguine mood have to be earned? Isn’t there injustice in the world if that mood is simply a gift awarded by nature? 

From sentiments like these come a significant resistance to acknowledging the influence of temperament. As scientific evidence accumulates, this resistance is crumbling, but at the same time it is sustained not only by a desire to limit the significance of biology but also by an ethical position defining what each individual must do to feel satisfied at the end of each day. Many Americans believe that happiness has to be earned and are bothered by the possibility that many might gain joy, for example, simply by taking Prozac each morning. However, this blend of puritanism and egalitarianism is inconsistent with temperamental biases that, occasionally, deny joy to the prudent and talented. 

Although the absolute risk is low, children who are high reactive infants are at a slightly greater risk than others of developing extreme shyness with strangers, or even social phobia, in adolescence or early adulthood. But most high reactive children will not become social phobics, and will find an adaptive niche that protects them from dealing frequently with unfamiliarity on an unpredictable schedule. Fortunately, many vocations—writer, bench scientist, computer programmer—permit this protection, while simultaneously awarding dignity, challenge, and financial reward. 

Low-reactive children might have a slightly greater advantage in our current society than high-reactive children, because of their sociability and willingness to take risks. In fact, however, each temperamental bias carries some advantages. In America’s technologic economy, it is desirable to finish college and for some to attain a postgraduate degree. A child with a high grade point average, of course, has a greater probability of gaining admission to an elite college and so to a productive professional career. Adolescents who were high reactive as infants are more concerned with academic failure; therefore, controlling for intellectual ability, they are more likely to study for exams and turn in assigned papers on time. 

In addition, our society needs many professionals who like to work alone rather than in groups. I suspect that after T.S. Eliot attained international fame and a Nobel prize in literature, he would have said, “no,” if asked whether he would have wanted his mother to take him to a therapist to treat his very inhibited childhood personality. Had Eliot conquered his extreme inhibition, he might not have decided to become a writer. 

Inhibited children enjoy another advantage in contemporary society. If they are willing to assume some risks, adolescents confront many temptations that promise pleasure and excitement. Driving at high speeds, experimenting with drugs, enjoying unprotected sex, and cheating on examinations are common temptations that can have unwanted consequences. High-reactive adolescents are more likely than low-reactive adolescents to feel anxious when they consider engaging in any of these activities and, therefore, are more likely to avoid these risky behaviors. It is not a coincidence that introverts live a little longer than extroverts. Lee Dugatkin, Ph.D., who has written about animal personalities for Cerebrum, discovered that guppies with a bold temperament approach large predator fish, whereas timid guppies stay far away from predators. Not surprisingly, the bolder guppies have a shorter life span, although females prefer to mate with them. 

The island of Cayo Santiago, close to mainland Puerto Rico, contains more than 1000 macaque rhesus monkeys and no human residents. Observers visit the small island each day to provide food for the monkeys and make notes on each animal. The consistently timid monkeys are at risk of dying of starvation, because, when food is put out each day, they wait for the other monkeys to eat first. If they wait too long on too many days, they could starve. The bold monkeys get sufficient food but are at a slightly higher risk of dying from wounds, because they are more likely to initiate an impulsive attack on a larger, stronger monkey. Thus, the account is balanced. Each temperamental type enjoys advantages and disadvantages. 

Most humans want to believe that one set of traits is a bit superior to another and are bothered by this balance, but nature does not understand our prejudice. If one strain of shellfish dies because of a hurricane, a related strain that survives in the same ecology gains a larger territory. Humans like overarching principles that permit ranking the desirability of a large number of outcomes so we can collapse long lists of facts into one general rule. Unfortunately, nature usually withholds that prize. 


Our research that followed high reactive and low reactive children has implications for an important controversy about what types of events are most likely to activate the amygdala and produce the psychological states that follow. The debate hinges on whether the amygdala evolved to respond primarily to imminently threatening events, producing a fear state, or to unfamiliar experiences, producing surprise. Resolving this question is not easy, because potentially harmful or threatening events are also usually unexpected; animals do not voluntarily seek danger. But certain neurons in the amygdala respond reliably to unexpected or unfamiliar events, independent of their potential for harm, and these neurons adjust, often rapidly, as an unfamiliar event loses its surprise value. These neuronal ensembles are the brain’s scouts, vigilantly scanning the external and internal environment for any change in the status quo. 

In one experiment, adults lying in a functional magnetic resonance imaging scanner looking at photographs of faces with neutral expressions show immediate amygdalar activation when a new face appears, even though neither that face nor any of the previous ones display a fearful, angry, threatening, or disgusted expression. In another study, adults recalling the circumstances that most often frightened them named unexpected or unfamiliar events. One woman dated the onset of her phobia of birds to an afternoon when, as a seven-year-old, she was watching Hitchcock’s film, The Birds, in which flocks of birds attack humans. The woman remembered being extremely surprised by learning that birds, which she had regarded as benevolent and beautiful, could be so aggressive. The sharp disconfirmation of her childhood assumption, which activated the amygdala and led to a strong bodily reaction, was the basis for the phobia that lasted for 30 years. 

Nonetheless, many eminent scientists continue to believe that the amygdala evolved primarily to create a state of fear in the presence of danger. Snakes are seen as the classic case.9 Consider, though, that most adults who report being afraid of snakes have never been harmed by one; infants rarely show a fearful reaction to snakes. Scientists at the University of Wisconsin reported that most rhesus monkeys raised in the laboratory showed equivalent levels of avoidance of a snake and a roll of blue masking tape.10 Many years ago, Wolfgang Kohler, Ph.D., evoked terror in a chimpanzee by presenting it not with a snake, but with a furry stuffed animal that had dark beads for eyes. Historical records and reports on other cultures suggest that, over the centuries, more adults have been afraid of being harmed by the spirits of dead ancestors, witches, and sorcerers than by snakes. 

I suspect that the snake is a feared animal because its physical features are different from those of most animals. Snakes have an unusual skin covering and an atypical ratio of head to body, and they move in a novel way.  The authors of the tree of knowledge allegory in the Old Testament probably selected the snake as tempter because of these distinctive physical characteristics, rather than the snake’s capacity to cause harm.


Can parents use their understanding of temperamental biases to guide their child’s development? For example, would it be helpful to test all children at four months of age to determine which ones are high reactive so that their parents could take steps to reduce the probability of later shyness or social phobia? I would hesitate to suggest it. Perhaps 10 to 20 percent of high reactive infants will develop social phobia for a period of time in adolescence or adulthood, which means that a prediction at four months of age will turn out to be wrong the majority of times. Why create apprehension in all parents who have a high reactive infant? 

Most high reactive infants will not become extremely shy, timid adolescents. Most will have close friends and a good academic record. Only about one of three will possess characteristics that differentiate him in an obvious way from other children. Yes, these adolescents will take less pleasure from high-risk activity, like sport parachuting or hiking alone in the wilderness; they will initially feel tense at a party of strangers; they will worry a bit more about examinations; and they will be less likely to choose a vocation that requires split-second decisions, continual contact with strangers, and unpredictable challenges. These adolescents will have learned that criticism and possible failure make them feel tense, and, as a result, they will try to avoid placing themselves in such situations. 

Having said this, I do have some suggestions for parents who might have a high reactive infant. First, they should appreciate that the timid, shy behavior is due to a temperamental bias and should not assume that they were the sole cause of their child’s personality. This appreciation should alleviate some of the guilt experienced by parents who misinterpret their child’s timidity as a sign that they could have done something wrong. 

Second, loving parents should not protect their high reactive infants from all challenge or novelty, even though they might be tempted to do so. Parents who gradually expose their high reactive infants to unfamiliar challenges help them overcome their propensity to avoid the unfamiliar. 

Third, parents should have the confidence that every child is malleable and has the capacity for change. There is no fixed determinism between an infant temperament and what that child will become 20 years later. Temperament is not destiny. Many experiences will affect high and low reactive infants as they grow up. Parents are the first influence. Parents who encourage a more sociable, bold persona and discourage timidity will help their high reactive children develop a less-inhibited profile. The parents’ behaviors as role models are also influential. Most children identify with their parents and, as a result, believe that some of the adults’ properties belong to them. The high reactive girl with a highly anxious mother is likely to conclude that her timidity is probably an inherent trait. But a high reactive girl with a bold, sociable parent is likely to come to the opposite conclusion and believe that she, too, has a similar persona. 

Actual success and failure, with people or tasks, is also important. A child who has several satisfying friendships will have less doubt about her sociability than one with no friends. Success in school, athletics, or a hobby can persuade a high reactive child that there is some basis for resisting a judgment of personal inadequacy. 

Finally, parents should temper their desire to change their child’s personality by carefully and sensitively understanding the child’s own wishes. It is possible to find a regimen of rearing that accommodates both the parent’s wishes and the child’s hopes. 

My colleagues and I admit to being surprised by the shadow that the infant temperamental profiles cast. Our initial 45-minute observation of 16-week-old infants many years ago revealed temperamental dispositions that continued to affect some children for more than a decade. We are also aware of the good fortune that led us to observe the infants when they were four months old, rather than younger or older. There is a maturational transition at two to three months that permits four-month-olds to relate events in the environment to their acquired knowledge. We suspect that, had we first observed these infants at one week or at one year, we might not have found as persuasive evidence for the modest preservation of these two temperamental types. Nature occasionally opens her door for a moment to reveal some of the treasures in her mansion. If the scientist happens to be turned the other way, her sanctum remains mysterious.


  1. Davis, M, and Whalen, PJ. “The amygdala.” Molecular Psychiatry 2001; 6: 13-34.
  2. Scott, JP, and Fuller, J. Genetics and the Social Behavior of the Dog. Chicago. University of Chicago Press, 1965.
  3. Caspi, A, and Silva, PA. “Temperamental qualities at age 3 predict personality traits in young adulthood.” Child Development 1995; 66: 486-498.
  4. LaGreca, AM, Silverman, WK, and Wassastein, SB. “Children’s predisaster functioning as a predictor of post-traumatic stress following hurricane Andrew.” Journal of Consulting and Clinical Psychology 1998; 66: 883-892.
  5. LaGasse, L, Gruber, C, and Lipsitt, LP. “The infantile expression of avidity in relation to later assessments.” In J.S. Reznick (Ed.). Perspectives on Behavioral Inhibition. Chicago. University of Chicago Press, 1989: 159-176.
  6. Kagan, J, and Moss, HA. Birth to Maturity. New York. John Wiley, 1962.
  7. Kagan, J, Kearsley, RB, and Zelazo, PR. Infancy: Its Place in Human Development. Cambridge. Harvard University Press, 1978.
  8. Kagan, J. Galen’s Prophecy. New York. Basic Books, 1994.
  9. Ohman, A, and Mineka, S. “Fears, phobias, and preparedness.” Psychological Review 2001; 108: 483-522.
  10. Nelson, EE, Shelton, SE, and Kalin, NH. “Individual differences in the responses of naive rhesus monkeys to snakes.” Emotion 2003; 3: 3-11.

About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Robert Malenka, M.D., Ph.D., Stanford University School of Medicine
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.