Show Summary Details

Substantive editing and updates to citations.

Updated on 5 April 2016. The previous version of this content can be found here.
Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 11 April 2021

Temperamental Contributions to Inhibited and Uninhibited Profiles

Abstract and Keywords

A temperamental bias is currently defined as a behavioral profile with a partial origin in the child’s biology that varies among individuals. These biases, which appear early in development, are sculpted by experience into a variety of personality profiles. This chapter first describes possible genetic and nongenetic bases for temperamental categories, followed by a detailed presentation of the research on high- and low-reactive infants who are biased to become inhibited or uninhibited children. The chapter concludes with a discussion of the implications of these two temperaments to psychopathology and speculations on the temperamental variation among reproductively isolated human groups. A large number of questions remain unanswered. Perhaps the most critical is discovering the genes and resulting neurochemical or neuroanatomical features that contribute to the high- and low-reactive profiles.

Keywords: temperament, genes, neurobiology, reactivity, inhibition, amygdala, psychopathology

Key Points

  1. 1. Temperaments are biases favoring certain psychological dispositions that originate in biological processes.

  2. 2. The biological processes can be inherited neurochemical or neuroanatomical conditions or the consequence of prenatal or early postnatal events.

  3. 3. The potential number of biological conditions is far greater than the number of temperaments detected thus far. Many of these conditions make no contribution to temperaments and different biological conditions can result in similar temperamental phenotypes.

  4. 4. Several neurochemistries appear to be bases for temperaments; these include the products of polymorphisms of the genes for the serotonin transporter, one or more of the dopamine receptors, COMT, and variations in oxytocin, vasopressin, and the sex hormones.

  5. 5. Variation in the initial psychological reaction to unfamiliar or unexpected events is present in all vertebrates and appears to be an important temperamental bias in humans.

  6. 6. Kagan and colleagues and Fox and colleagues have studied 4-month-old infants who display either a profile of distress and vigorous motor activity or the absence of both features to unfamiliar stimuli. This research reveals that the former, infants called high reactive, are biased to become shy, timid children and adolescents who are anxious over unrealistic future events. The relaxed infants, called low reactive, become bold children and extraverted adolescents.

  7. 7. The childhood and adolescent physiologies of the high-reactive infants imply excitable limbic circuits and a bias favoring greater activation of the right rather than the left prefrontal cortex.

  8. 8. Although most high reactives lose their earlier shyness, they remain vulnerable to a state of uncertainty when challenges occur.

  9. 9. The main effect of a temperamental bias is to restrict the development of certain profiles rather than lead to a particular profile.

  10. 10. The different genomes of reproductively isolated human groups imply differences in temperamental biases.

The Idea of Temperament

Each object or life form possesses properties that determine how it will react to varied events. Candles and the bacterium E. colirespond differently to a lit match and penicillium mold. All humans vary in the frequency and intensity of many behaviors and feelings. When the person’s inherent biology, rather than experience alone, is the source of the variation the resulting profiles are called temperamental biases. Most appear early in development, but not always during the opening weeks, and are shaped by experience into a large, but limited, envelope of psychological profiles. Although most cultures acknowledge these biases, there have been brief intervals during which they were denied. One example was the interval between 1900 and 1960, when a large number of Americans regarded biologically based human variation as politically incorrect because it was inconsistent with the nation’s hope that the illiterate European immigrants and their children would assimilate successfully into American society. It is also relevant that acknowledging the importance of temperaments would affirm Hitler’s rhetoric regarding the inherent superiority of Aryans.

A temperamental bias can originate in at least five different biological processes: an inherited neurobiological profile, season of conception, maternal stress or illness during pregnancy, extreme prematurity, or events during the opening months of life that alter the growth of the still-immature brain (Kagan & Snidman, 2004; Schmidt, Miskovic, Boyle, Saigal, 2008). Although the first mechanism is assumed to be the most common source of temperamental biases, the others are of potential influence.

For example, embryos conceived during the spring or fall months in the Northern Hemisphere, when the hours of daylight are either increasing or decreasing at a faster-than-usual rate, are exposed to decreasing or increasing levels of maternally secreted melatonin. In addition, as newborns they will experience variation in the hours of daylight during the days immediately following birth. The first condition can affect fetal development; the second entrains the newborn’s circadian clock to either a long or short light period.

The pregnant mother’s secretion of high levels of melatonin influence brain development in diverse ways because this molecule binds to receptors in many brain sites, contributes to cell death, and suppresses both dopamine release and cortisol production (Ciesla, 2001; Torres-Farton et al., 2004; Zisapel, 2001). Hence, a fetus with a particular genome could develop a temperamental bias if conceived during the fall or spring months. Early fall conceptions in the Northern Hemisphere (as well as conceptions in February through April in the Southern Hemisphere) are associated with more extreme levels of shyness in children (Gortmaker, Kagan, Caspi, & Silva, 1997), affective disorder in adults (Joiner, Pfaff, Acres, & Johnson, 2002; Pjrek et al., 2004; Torrey et al., 1997), and variation in dopamine turnover in the brain (Chotai & Adolfsson, 2002; Chotai, Serretti, Lattuada, Lorenzi, & Lilli, 2003). By contrast, adults conceived during the spring months in the Northern Hemisphere are at higher risk for illicit drug use (Goldberg & Newlin, 2000) and usually feel maximally alert during the morning rather than the evening hours (Caci, Robert, Dossio, & Boyer, 2005).

Maternal infection or chronic stress during pregnancy provokes the mother’s immune system to produce cytokines and antibodies that can influence brain development and lead to temperamental biases. For example, mothers who contracted the flu during the second trimester had a slightly higher probability of giving birth to a child who later developed schizophrenia (Patterson, 2006). Even the risk of developing the symptoms of autism was a little higher in children born to mothers who experienced a severe hurricane or tropical storm during the fifth or sixth week of their pregnancy. This is the time when the s layers of the cortex are differentiating, myelin is being laid down, and many neurotransmitters are emerging (Kinney, Miller, Crowley, Huang, & Gerber, 2008).


Inherited variation in the brain’s chemistry is likely to be the most frequent basis for a temperamental bias. The human brain contains over 150 different molecules that vary in concentration, as well as in the density and location of the molecule’s multiple receptors. Receptor density can be independent of the molecule’s concentration. These conditions influence the excitability of neuronal clusters and the multiple circuits in which they participate. Because the number of possible combinations of molecule concentrations and receptor patterns is large, the number of possible brain states that could influence the behaviors defining a temperament exceeds by a substantial amount the number of different categories of observable behavior. If each of the 150 molecules has a low, moderate, or high concentration, and each molecule has, on average, three different receptors, each of which can be low, moderate, or high in density there will be over 4,000 neurochemical profiles that could represent the foundations of human temperaments.

Some Examples

The molecules that appear to be the most relevant for temperamental biases are serotonin, GABA, glutamate, dopamine, norepinephrine, opioids, corticotropin-releasing hormone (CRH), oxytocin, vasopressin, acetylcholine, prolactin, gastrin-releasing peptide, and, of course, the sex hormones. For example, two polymorphisms of the gene for the serotonin transporter (5-HTTLPR), which absorbs serotonin from the synapse, assumes the form of a short (s) or a long (l) allele in the gene’s promoter region. (Hariri & Brown, 2006). The short allele is associated with less effective transcription of the gene; therefore, serotonin will remain in the synapse for a slightly longer time. Many scientists believe that this phenomenon is followed by inhibitory feedback on the source of serotonin (the raphe nucleus), resulting in a lower tonic level of serotonin. Monkeys who inherit the short allele are more likely to avoid unfamiliar events or settings; young children with the short allele are more irritable than most infants (Auerbach et al., 1999; Battaglia et al., 2005; Bethea et al., 2004). Reduced levels of serotonin transporter can alter the thalamic input to the amygdala and cortex leading to greater excitability (Bigos et al., 2008; Reimold et al., 2008; Young, Bonkale, Holcomb, & Hicks, 2008). Hungarian women with one or two short alleles reported more labile moods and more frequent bouts of sadness than those with two long alleles (Gonda et al., 2008). However, the brain state that accompanies inheritance of the short allele does not automatically lead to any specific idea, emotion, or action. This brain state occasionally evokes a conscious feeling that the person interprets. It is the interpretation that determines the behavior or mood that develops. Two adults who possess the short allele often impose different interpretations on the same feelings.

Another genetic polymorphism refers to the presence of the base adenine or guanine in codon 158 of the gene for the enzyme catechol-o-methyltransferase (COMT), which is accompanied by either the amino acid valine or methionine in the COMT molecule. Those who inherit the allele that leads to methionine have slightly better working memories because of slower degradation of dopamine in the neurons of the frontal cortex (Barnett et al., 2007).

Variation in dopamine function is third source of temperanents. The expectation of a desirable event, the unexpected receipt of a desired event, and most violations of expectations unrelated to rewards are accompanied by increases in dopamine and altered excitability in select circuits. The two major classes of dopamine receptors (D1 and D5 are excitatory and D2, 3 and 4 are inhibitory) vary in their density in different brain locations. The moment a rat places its paws in a novel space or receives an unexpected reward there is an immediate release of dopamine in the nucleus accumbens (Rebec, Christianson, Guevra, & Bardo, 1997).

Tonically high levels of dopamine in the frontal lobe or striatum are associated with proportionately smaller transient or phasic rises in dopamine and, as a result, a muted feeling of pleasure. By contrast, individuals who possess lower tonic levels experience larger phasic increases in dopamine and, presumably, a slightly more intense moment of pleasure to an unexpected event. The difference in feeling between someone sated on chocolate and one who has not had chocolate for 12 months provides an analogy. The former derives less pleasure from a chocolate bar than the latter.

These facts and hypotheses imply that children or adults who inherit genes for tonically higher dopamine activity in the striatum should have a smaller phasic increase in dopamine activity and experience less pleasure following a new or rarely experienced desirable event compared with those who possess tonically lower levels of dopamine activity. Most females have higher tonic levels of dopamine activity than males, partly because estrogen interferes with the dopamine transporter, resulting in dopamine remaining in the synapse for a longer time (Kaasinen, Nagren, Hietala, Farde, & Rinne, 2001). Men, who have a lower tonic level of dopamine, show a larger phasic release of dopamine following the administration of amphetamine than women (Munro et al., 2006). These facts may explain why women seem to extract less pleasure from high-risk novel activities, such as sport parachuting, drag racing, and high-stakes gambling (Hyde, 2005).

Finally, the gene for the D4 receptor, called the DRD4 allele, varies in the number of repeats of a 48 base pair sequence. The longer, 7-repeat variant was more common among 1-year-old Israeli infants with compromised attention to objects (Auerbach, Benjamin, Faroy, Geller, & Ebstein, 2001) and Italian 3-year-olds with an intense reaction to novelty (de Luca et al., 2003).

The gene for the molecule MAO A, which degrades serotonin, norepinephrine, and epinephrine in the synapse, varies in the number of repeats (3 or 4) in the promoter region (Hartl & Jones, 2005). Polish women with panic disorder were more likely than others to have four repeats (Samochowiec et al., 2004), whereas European men who chose a violent means to commit suicide were more likely to have the shorter number of repeats (Courtet et al., 2005).

An allele of the gene for the gastrin-releasing peptide could affect temperament because it acts on the receptors of interneurons, which release GABA, leading to an inhibition of neural activity. This allele is associated with larger concentrations of the molecule, less GABA activity in the amygdala, and preservation of a fear state for a longer period of time (Maren, Yap, & Goosens, 2001; Sanders, 2001).

Variation in the concentration of norepinephrine or in the density of one of its receptors affects the preferred reaction to novelty, level of alertness, the ability to sustain attention in the face of distractors, and the threshold for detecting very subtle changes in sensory input. Rats from the Wistar strain that explored unfamiliar areas had greater norepinephrine activity in the nucleus accumbens than other animals from the same strain (Roozendaal & Cools, 1994).

Variation in the genes responsible for the production of opioids is relevant to human temperaments. Visceral afferent feedback from the body arrives at the nucleus tractus solitarius in the medulla, where opioids modulate the information sent from this structure to the cortex. An individual with less effective opioid activity in the medulla would be vulnerable to greater activation of limbic sites and more frequent or more intense states of worry, tension, or dysphoria (McNally & Westbrook, 2003). Individuals with greater opioid activity in the medulla should enjoy more moments of serenity (Miyawaki, Goodchild, & Pilowsky, 2002; Wang & Wessendorf, 2002).

CRH, secreted by the hypothalamus, affects many systems but especially the hypothalamic-pituitary-adrenal (HPA) axis. The human stress hormone cortisol (corticosterone in rats) is one product of activity in this axis. Capuchin monkeys with high cortisol levels were more avoidant than animals with lower levels (Byrne & Suomi, 2002), and children born to a parent suffering from panic disorder who, in addition, had a particular allele for the CRH gene were more avoidant than others (Smoller et al., 2005). However, there is no simple relation between cortisol and any psychological profile because cortisol level, like any biological variable, is subject to a broad variety of temporary and chronic psychological states (Gunnar, 1994). For example, high levels of cortisol are more common among economically disadvantaged adults with varied personalities (Li, Power, Kelly, Kirschbaum, & Hertzman, 2007).

Men who normally have high levels of circulating testosterone show a rise in cortisol when they lose a competition; men who have low levels of the male hormone do not display this reaction (Mehta, Jones, & Josephs, 2008). Adults administered either 20 or 40 milligrams of cortisol, compared with a placebo, showed the expected rise in circulating hormone, but there was no relation between their cortisol level and their subjective feelings or their ratings of unpleasant words or pictures (Abercrombie et al., 2003). One reason for the loose relation between cortisol level and psychological qualities is that the time of day when cortisol is gathered is important. Individuals with high cortisol levels in the morning report sad or melancholic moods, but the mean cortisol level across the entire day is associated with a happier mood (Simpson et al., 2008).

Variations in the sex hormones or their receptors are associated with temperamental biases. A female embryo developing next to her male sibling in a twin pair is subject to the masculinizing effects of the testosterone secreted by her unborn brother between weeks 8 and 24 of gestation. It is likely that this exposure contributes to the fact that, at age 4, these girls were more active than the average female and their play resembled that of boys (Collaer & Hines, 1995). The ratio of the length of the index to the ring finger, called the 2D:4D ratio, is a rough index of the amount of male hormone secreted during the fetal period. Although only male fetuses secrete testosterone, the adrenal glands of both sexes secrete a molecule that is chemically similar to testosterone. Most males have a slightly smaller index than ring finger (average ratio of .97 to 0.98), whereas most females have more similar lengths for both fingers (average ratio of 0.99 to 1.00). About two thirds of adult men, but only one third of adult women, have a smaller index than ring finger. Variation in the 2D:4D ratio is moderately heritable and is correlated with a variety of physical and psychological qualities in both children and adults (Bailey & Hurd, 2005; Fink et al., 2007; Hall & Schaeff, 2008).

For example, men with a very masculine ratio (shorter index than ring finger) report being more dominant in social interactions, engage in more frequent sexual activity, are less likely to be depressed, and have a square rather than a round or narrow face. Boys with a masculine ratio are a little more likely to be hyperactive and aggressive; girls with a feminine ratio are friendlier and more sociable (Fink et al., 2005; Manning & Fink, 2008). Surprisingly, the finger ratio even predicted young children’s style when asked to draw a desirable object. Second- and third-grade girls with a feminine finger ratio were more likely to draw flowers with a lot of pink and purple, whereas girls with a less feminine ratio more often drew people or artifacts with darker colors (Turgeon, 2008).

Some Caveats

Most psychological profiles that have a temperamental foundation are the result of combinations of a large number of alleles for many molecules (Jabbi et al., 2007). For example, the peptide oxytocin potentiates the secretion of norepinephrine when a novel event occurs (Pfister & Muir, 1989), and androgen levels modulate norepinephrine release following encounter with novelty (Handa et al., 1994; Hanson, Jones, & Watson, 2004). One-year-olds who showed extremely avoidant behavior to a stranger possessed both the two short alleles of the dopamine transporter gene, as well as the 7-repeat polymorphism of the DRD4 receptor. The least avoidant children combined the two long forms of the dopamine transporter molecule with the 7-repeat polymorphism (Lakatos et al., 2003). Japanese women who reported avoiding potentially dangerous situations combined the short allele of the serotonin transporter along with an allele that reduced the functioning of the norepinephrine transporter (Suzuki et al., 2008). Investigators had to combine the effects of 19 molecules in order to account for 60% of the variation in avoidant behavior in a sample of rats (Ray, Hansen, & Waters, 2006). The many inconsistencies in the reported relations between genes and psychological outcomes are due to the fact that most investigators usually measure only one allele rather than a combination (Arbelle et al., 2003). Specific patterns of genes, molecules, and experience are the bases of psychological outcomes, not the additive consequences of single processes.

Laboratory assays exist for only a small proportion of the molecules that affect psychological states. When investigators are able to measure more of the alleles and associated molecules that contribute to feelings and behaviors, they will discover that patterns of genes and rearing environments predict the psychological variations defining human temperaments (D’Souza & Craig, 2006). The fact that no gene has been unequivocally identified as the cause of any major psychiatric disorder is likely to remain true, because different environments, like potters in different cities working with clays of varying malleability, sculpt each child’s pattern of temperamental biases into distinctive personality profiles (Abdolmaleky, Thiagalingam, & Wilcox, 2005).

For example, some adults with the short allele for the serotonin transporter gene were exposed to the eye of several hurricanes that struck Florida in 2004. But the only persons who developed posttraumatic stress disorder (PTSD) were those who had been deprived of social support; those possessing the short allele who enjoyed social support did not develop PTSD symptoms (Kilpatrick et al., 2007). Possession of the short allele is associated with more aggressive behavior in female monkeys who are normally dominant, but with submissive behavior in females who are normally submissive (Jarrell et al., 2008).

The person’s social class can influence the outcome of a temperamental bias. Adults with the short form of the serotonin transporter who grew up in economically compromised families showed a blunted release of serotonin to a drug that normally releases this molecule; whereas college-educated adults with the same allele showed a normal level of release (Manuck et al., 2005). The level of distress in German adults were best predicted by a combination of gender (women higher than men) and being unemployed (Grabe et al., 2005).

Social class also affects the consequences of the alleles of the DRD4 gene. Upper-middle-class adults with fewer repeats of the gene for the DRD4 receptor were attracted to novel experiences; whereas those from economically disadvantaged backgrounds with the same number of repeats had low values on measures of novelty seeking (Lahti et al., 2006). The combination of an economically disadvantaged background and childhood maltreatment was a more significant predictor of adult criminality than an allele of the MAO molecule. Scientists trying to predict criminality, depression, or level of anxiety would do far better if they relied on a person’s social class, gender, and genetic markers. Because historical events alter the prevalence of poverty and parental treatment of children, the relation between genes and psychological outcomes will vary across history and culture. Much of the inconsistency in the results from different laboratories is due to interactions among the diverse products of genes, experience, ethnicity, social class, gender, and the cultural background of the participants (Manuck, Flory, Ferrell, & Muldoon, 2004).

What Temperaments to Study

The absence of robust relations between any neurobiological profile and any temperament leaves psychologists free to speculate on the important biases that are amenable to study. Mary Rothbart’s two temperamental dimensions of reactivity, meaning ease of arousal, and self-regulation dominate discussions of infant and child temperaments (Rothbart et al., 2001). Her original definition of reactivity referred to high levels of motor activity, crying, smiling, or autonomic responsiveness. Rothbart’s recent writings have concentrated on the behavioral signs of distress or fear, which she classifies as negative reactivity (Rothbart & Bates, 2006).

Self-regulation referred to the processes that modulate reactivity, including attention, approach, withdrawal, attack, and self-soothing. Rothbart’s recent concerns center on infants and children who are unable to regulate strong negative emotions, whom she calls “dysregulated.” This temperamental bias affects how adults will behave toward the child and the probability that the child will expose himself or herself to risky situations. Research by Rothbart and others reveals interactions between dysregulation and experience. Dysregulated infants raised in harsh or neglecting environments are at a higher risk for problem behaviors than others (Rothbart & Bates, 2006). But dysregulated children more often grow up in disadvantaged homes. The experiences associated with disadvantage predict problem behaviors in children who do not possess a dysregulated temperament. Interested readers should consult Rothbart and Hwang (2005) and Rothbart, Ahadi, and Evans (2000).

Most of the current conclusions about temperament are based on parental descriptions of their children, rather than direct observations. These two sources of evidence are not highly correlated, in part because parental descriptions are influenced by the parent’s personality, education, ethnic group, and nationality. Mothers who did not graduate high school more often describe their infants as “difficult” (Jansen et al., 2008). Russian parents describe their infants as more irritable and less exuberant than American parents (Gartstein et al., 2005). In addition, some signs of a temperamental bias are so subtle parents do not notice them and parents have no knowledge of their child’s biological properties.

A few investigators favor strategy that measures the child’s behaviors directly and withholds premature judgment on the name of the temperament until the evidence has been gathered. My colleagues and I chose this strategy. We began with the reliable observation, which parents affirm, that young children vary in their reaction to unfamiliar people, events, and situations.

Inhibited and Uninhibited Children

My colleagues and I have studied two infant temperaments that create a bias favoring either avoidance of or approach to unfamiliar events in one and two-year olds. Unfamiliar events can be familiar objects encountered in unfamiliar settings (e.g., seeing a robin on a dining-room table), events that share some but not all features with a familiar one (a dog with one eye ), or objects or situations never experienced before (a photograph of a distant galaxy sent by a space telescope). Unfamiliar events can be alterations of the immediate perceptual surround (a clap of thunder) or alterations of a person’s long-term store of knowledge (seeing a table move spontaneously). These different origins of unfamiliarity might create different brain states; hence, psychologists should not assume that novelty is a unitary concept.

Unfamiliar or unexpected events usually provoke a burst of firing in the thalamus, followed by excitation of varied cortical sites and the release of dopamine and norepinephrine (Bunzeck & Duzel, 2006; Rolls et al., 2005). The psychological state that English calls uncertainty often follows encounters with unfamiliar events. The state called event uncertainty recruits activity in sensory areas, the amygdala, the parahippocampal region, and often the prefrontal cortex. A second class of uncertainty is generated when a person is unsure about what decision or behavior should be implemented when alternatives are available. This state, called response uncertainty, recruits brain activity in the cingulate cortex, frontal lobe, and basal ganglia. Infants younger than 8 or 9 months are not mature enough to recognize they have a choice; hence, response uncertainty is rare during the first 7 or 8 months.

The brain of every animal is exquisitely sensitive to unfamiliar and unexpected events. The simplest stimulus, for example an unexpected sound piercing the quiet, perturbs the chemistry of the genes within neurons to provoke production of the protein c-fos in the rat’s cochlear nucleus (Kandiel, Chen, & Hillman, 1999). Unfamiliarity can, on occasion, compete with an event’s hedonic quality. The increase in CRH in the central nucleus of the rat amygdala is equally large when rats are unexpectedly restrained, which they do not like, and when they are unexpectedly fed, which they do like (Merali et al., 1998). Of course, the degree of uncertainty created by an unfamiliar event always depends on the context in which it appears. The approach of an adult stranger is more likely to produce a cry in 8-month-olds if they are in an unfamiliar laboratory than in a familiar room at home. This principle holds for most incentives. There are few predictable consequences of a stimulus, only consequences of a stimulus in a particular context.

Despite this caveat, a small proportion of infants and children are hyperresponsive to events that are either unexpected or discrepant from their past experience. We believe that extreme responses to unfamiliar events are the product of a temperamental bias due, in part, to an inherited neurochemistry that affects the excitability of the amygdala. If the unfamiliar event can be assimilated easily, the child shows minimal signs of uncertainty. Most 1-year-olds reach for a new toy after having played with a different one for several minutes because the new object poses no threat, is assimilated at once, and a relevant behavior is available. However, not all 1-year-olds reach toward a stranger who has extended her hand because that event is assimilated less easily and they may not know what to do. Hence, a stranger can provoke both event and response uncertainty in children older than 9 or 10 months. Children, like adults, live in a corridor bordered on one side by the appeal of events that are comprehensible transformations of what they know and, on the other, by an avoidance of novelty they do not understand and for which they do not have a coping behavior.

Children older than 12 months who consistently display a vigilant, avoidant, or affectively subdued reaction to most unfamiliar events are called inhibited; those who show minimal avoidance and vigilance are called uninhibited. The heritability of these two categories, based on behavioral observations of monozygotic and dizygotic twins observed four times in the first 3 years of life, approached 0.5 (Kagan & Saudino, 2001). Although some inhibited children who are shy with strangers preserve this quality, albeit modestly, through 18 years of age (Caspi & Silva, 1995), most inhibited children do not become excessively shy adolescents because parents, teachers, and peers encourage them to be bold, and because they want to develop a more sociable profile. Inhibited 2- and 3-year-olds were most likely to preserve their persona if they had intrusive-hypercritical mothers, and likely to lose their initial avoidant style if their mothers consistently discouraged withdrawal and timidity (Rubin, Burgess, & Hastings, 2002). Because an outgoing, sociable, exuberant profile is regarded as desirable in North America and Europe, it is more likely to be preserved over the childhood years than an inhibited profile (Stifter, Putnam, & Jahromi, 2008).

There is some disagreement over whether the concepts inhibited and uninhibited should be treated as categories or a continuum. Thomas and Chess (1977) treated the variation in each of their nine temperamental dimensions as continua and wrote about an approach–withdrawal dimension, as if each infant and child could be placed on a scale reflecting a tendency to withdraw or to approach unfamiliar events.

My colleagues and I believe that extremely inhibited children are qualitatively different from those who are only moderately shy, and extremely uninhibited children are qualitatively different from those who are moderately sociable. I base this suggestion on the premise that extreme phenotypes are mediated by distinct neurochemical profiles. Support for this claim is the observation that moderately inhibited (or moderately uninhibited) 1-year-olds develop different psychological profiles as adolescents than extremely inhibited (or extremely uninhibited) children (Kagan & Snidman, 2004). (See Fox et al., 2005, for a similar result.)

The statistical techniques psychologists use are one reason for the favorable attitude toward continua. Because the correlation coefficient, t-test, and analysis of variance and covariance are most powerful when computed on continuous variables, it is inconvenient to assume qualitatively distinct categories of individuals. The problem with positing qualitative categories is that there is no consensus on the criteria investigators can use to classify a small number of subjects as belonging to a distinct group. An investigator unfamiliar with Down syndrome who was studying the relation between maternal age and children’s intelligence in a large, middle-class sample would find no correlation between the two variables for the whole sample. However, examination of a scatterplot of these variables might reveal that the two children with the lowest IQ scores happened to have the two oldest mothers in the sample. Under these conditions, the investigator might consider the possibility that these two children were qualitatively different from the rest of the sample.

The majority of significant correlations between pairs of psychological variables, or between a psychological and a biological variable, are less than 0.40. More important, when the correlation is statistically significant it is usually because of the scores of the participants in the top or bottom 15% to 20% of the distributions. The correlation for the remaining 60% to 70% is close to zero. This fact implies that these samples consist of two qualitatively discrete groups. Height and weight are perfect examples of continuous variables. But adults who are taller than seven feet and weigh more than 300 pounds, as well as those who are less than four feet and weigh less than 100 pounds, are qualitatively distinctive groups with special biologies.

History of the Concept

My interest in inhibited and uninhibited children began in 1957 when Howard Moss and I were studying a sample of Caucasian adults, born between 1929 and 1939, who were participants in a longitudinal study at the Fels Research Institute in Ohio. I interviewed and tested the adults who were in their third decade while Moss analyzed the rich corpus of childhood information on them. Our seminal discovery was that a small group of children who avoided unfamiliar people and situations during the first 3 years preserved some derivatives of that early disposition as young adults: they were introverted, cautious, and overly dependent on others for emotional support. By contrast, an equally small group of bold, sociable 2- and 3-year-olds were extraverted, liked risk, and preferred competitive, entrepreneurial lifestyles (Kagan & Moss, 1962).

About 15 years later, Richard Kearsley, Philip R. Zelazo, and I were reflecting on longitudinal observations gathered on 53 Chinese-American and 63 Caucasian infants living in Boston who were participants in a study of the effects of infant daycare. Some infants attended our experimental daycare center from 3 to 29 months; matched controls were reared only at home. The most significant result, based on five assessments, was that the Chinese-American infants, whether attending the daycare center or raised at home, were more inhibited than the Caucasian children. The former stayed closer to their mothers in unfamiliar settings, cried more intensely when temporarily separated from the mother, and were wary when playing with an unfamiliar child (Kagan, Kearsley, & Zelazo, 1978). These results motivated study of the temperamental biases my colleagues and I called inhibited and uninhibited.

The first investigation by Cynthia Garcia-Coll involved 117 Caucasian infants, 21 months old, as they encountered unfamiliar people, objects, and situations in laboratory settings (Garcia-Coll, Kagan, & Reznick, 1984). The category inhibited was defined by crying, withdrawal, and the absence of spontaneous affect to most of the unfamiliar events. The classification uninhibited was defined by minimal crying and a spontaneous approach to these events. The remaining children were inconsistent and were not seen again. When Garcia-Coll observed the same inhibited and uninhibited children several weeks later, most retained the behavioral style they had displayed earlier.

The inhibited children at age 4 were more subdued when playing with an unfamiliar child and glanced repeatedly at the examiner. A small number were observed during their first day at kindergarten. Those who had been classified as inhibited at 21 months were more solitary and quieter than the uninhibited children (Rimm-Kaufman, 1996).

Nancy Snidman categorized a different group of 31-month-old children as inhibited or uninhibited based on their behavior while playing with one unfamiliar child of the same age and sex and their reaction to a woman wearing a plastic cover over her head who entered the playroom and invited each child to approach. The children who remained close to their mother, were quiet during most of the play session, and did not approach the stranger were called inhibited; those with the opposite traits were classified as uninhibited. The Garcia-Coll and Snidman samples were assessed again at 4, 5, and 7 years of age. About one third of the inhibited and one third of the uninhibited children retained their behavioral style, whereas less than 10% changed from one category to the other.

Because the corpus of evidence implied that these two temperamental biases had a biological foundation, we tried to discover early signs of these two categories in 16-week-old infants who are too young to display obvious avoidance of or approach to unfamiliar events. Reliable signs of avoidance to unfamiliarity first appear between 7 and 10 months, usually in the form of a fear reaction to strangers. The fact that the amygdala is usually provoked by unfamiliarity provided the critical clue to the infant variables we quantified (Fitzgerald et al., 2006; Rolls et al., 2005).

The Amygdala

The amygdala consists of three major neuronal clusters—basolateral, corticomedial, and central areas—each with distinct profiles of connectivity, neurochemistry, and function. Each cluster projects to many sites and receives input from many regions, resulting in about 600 known amygdalar connections (Stefanacci & Amaral, 2002). The basolateral area receives information from the thalamus, sensory cortices, and parahippocampal region. If the event deviates from the immediate past or long-term knowledge, the basolateral nucleus activates the central nucleus, which sends projections to targets that lead to body immobility, defensive behavior, sympathetic activity, and/or activation of the HPA axis (Stark et al., 2005).

The hypothesis that infants with an excitable amygdala are likely to display inhibited behavior in the second year is supported by the fact that newborn infants whose rate of sucking increased dramatically following an unexpected change in taste sensation, from water to a sweet liquid, were more inhibited in the second year than newborns who showed a minimal increase in sucking following the same change in taste (LaGasse, Gruber, & Lipsitt, 1989). The unexpected change in taste, from water to sweet, which activated in sequence the corticomedial and then the central nucleus, led to activation of the motor centers that control sucking (Barot & Bernstein, 2005). Hence, newborn infants with a more excitable central nucleus should display a greater increase in sucking rate than those with a less excitable central nucleus.

About one in seven house cats resemble inhibited children because they fail to explore unfamiliar places, withdraw from unfamiliar objects, and are reluctant to attack rats. This profile, which first appears at about 30 days of age (comparable to 9 months in humans), becomes a stable trait by 60 days (comparable to 14 months in children) when the kitten’s amygdala gains control of the circuits that mediate avoidant behavior. The inhibited kittens show a larger increase in amygdalar activation than bold kittens when they hear sounds that resemble the threat howl of another cat (Adamec, 1991).

In addition, activation of the amygdala of many mammalian species is followed by limb movements, mediated by projections from the basolateral nucleus to the ventral striatum and ventral pallidum, and arching of the back and crying, mediated by projections of the central nucleus to the central gray and the nucleus ambiguus (Pitkanen, 2000).

These facts imply that 4-month-old infants who inherited a neurochemistry that rendered the amygdala, or one of its nuclei, unusually excitable should display more vigorous motor activity, including arching of the back, and more frequent crying to unfamiliar events than infants who inherited a neurochemistry that rendered the amygdala less excitable. Although the amygdala might be the proximal cause of the increased motor activity and crying, other structures are likely to be implicated. The hippocampus, parahippocampal gyrus, perirhinal cortex, and prefrontal cortex, which are reciprocally connected to the amygdala, can excite or modulate the amygdala (Strange et al., 2005; Fried, MacDonald, & Wilson, 1997; Witter et al., 2000).

The 4-Month Evaluation

We selected 4 months as an optimal time for the infant assessment because the amygdala and its projections were mature enough to provoke vigorous motor activity and distress cries to unfamiliar events, but not mature enough for projections from the prefrontal cortex to modulate the amygdala. Small clusters of GABA-ergic neurons, called the intercalated cells, located between the basolateral and central nuclei which receive input from the prefrontal cortex, inhibit activity in the basolateral and central regions, resulting in reduced activation of motor and autonomic targets.

We evaluated over 500 healthy Caucasian infants born to middle-class women, most with college degrees who gave birth to healthy babies at term. The 4-month evaluation, which lasted a little less than an hour, involved the presentation of mobiles composed of one, four, and seven unfamiliar colorful toys that moved back and forth in front of the infant’s face, a cotton swab dipped in dilute butyl alcohol applied to the infant’s nostrils, and a tape recording of eight short sentences read by a blend of human voices and a female voice speaking three nonsense syllables at three different loudness levels without any visible human as the source of the sounds.

The 20% of this sample who displayed frequent crying and vigorous limb movement, accompanied by arching of the back, were qualitatively different from the remaining infants. We assumed that these infants, whom we called high reactive, had inherited an excitable amygdala. The 40% of the sample that showed minimal motor activity and very little distress were called low reactive. These two groups, along with some children who belonged to neither the high- nor low-reactive categories, were studied for the next 18 years.

Later Assessments

When these children were 14 and 21 months they encountered a large number of unfamiliar events, including people, objects, procedures, and rooms in laboratory settings with the mother always present. One third of the sample was obviously fearful or avoidant on one or both occasions (they cried and/or retreated to 4 or more of the 17 unfamiliar episodes). These children were called inhibited. A complementary group of uninhibited children showed minimal signs of fear on both occasions. As expected, one third of the 1- to 2-year-olds who had been high-reactive infants were clearly inhibited and very few were uninhibited; one third of the low reactives were uninhibited and very few were inhibited.

At 4.5 years the children were administered a variety of cognitive tests and observed in a play session with two other unfamiliar children of the same age and sex while the mothers sat on a couch in the playroom. Twice as many low as high reactives were extremely sociable and talkative with the two unfamiliar peers; about half of the high reactives were shy, quiet, and timid.

We interviewed the mother and the child’s teacher for signs of anxiety to unfamiliarity when the children were 7.5 years old. One fourth of the sample had two or more anxious symptoms (e.g., they were very shy with unfamiliar children, avoided novel foods, had phobias of animals or lightning, or were reluctant to sleep at a friend’s house). The 7-year-olds who had been high reactive were more likely than low reactives to meet criteria for two or more anxious symptoms. About one of every five high-reactive infants, but not one low reactive, was inhibited at all four assessments ( 14 and 21 months, as well as at 4.5 and 7.5 years) and no high reactive was consistently uninhibited across all four evaluations.

When these children were 11 years old we coded their behavior with an examiner and obtained both self-reports and maternal descriptions for each child. Frequency of spontaneous smiles while interacting with the examiner was a sensitive sign of the earlier infant temperaments. High reactives rarely smiled; whereas low reactives smiled and laughed frequently. However, the mother’s descriptions, as well as the children’s descriptions of their shyness or sociability, bore a minimal relation to their early temperaments.

We were not surprised by the latter finding because the validity of every inference is affected by its source of evidence. I noted earlier that most of the published reports on children’s temperaments are based on verbal reports provided by parents, less often by teachers or peers (Rothbart et al., 2001). However, parent or teacher verbal reports have a unique structure that differs from the structure of behavior. Hence, one is not a proxy for the other. This statement does not mean that parental descriptions have no value, for they can reflect the parents’ perceptions of the child. However, these perceptions need not be correlated with actual observations of the child in ecologically natural situations.

However, most social scientists prefer measures that are easy to obtain and have the appearance of being objective. Questionnaires are easy to administer, and although a score on a questionnaire index of a temperamental trait seems to be free of ambiguity, its interpretation is not. Although adult descriptions of children seem to have the advantage of sampling behaviors across many occasions and in settings not possible in the laboratory, this source of evidence has special problems. Psychologists can ask parents only about qualities that the latter understand with words that are part of a familiar vocabulary. Questionnaires cannot ask about a child’s biological properties. A small proportion of infants are minimally irritable, smile often, have a low heart rate, and show less alpha power in the resting electroencephalogram (EEG; and therefore greater activation) in the left compared with the right frontal area (Kagan & Snidman, 2004). Psychologists who invented a temperamental concept for this combination of qualities could not ask parents about it because the parents have no access to the child’s biology.

More important, parents vary in the accuracy of their descriptions. Mothers and fathers did not agree in their ratings of the fearfulness or sociability of their infants, in part because fathers interpreted high activity level as reflecting a positive emotional mood but mothers regarded the same behavior as a sign of anger (Goldsmith & Campos, 1990). Human languages are not rich enough to describe all the behaviors that can be components of a temperamental category. That may be why parents, teachers, and peers usually do not agree in their ratings of a variety of behavioral and emotional properties in another person (Achenbach, 1985). There was not even a robust correlation between the replies of college students to questions asking about their worries and the daily logs they kept for 6 days in which they wrote down each time they felt worried or anxious (Vercuil, Brosschot, & Thayer, 2007). Although questionnaire data should not be ignored, investigators who rely only on questionnaires must recognize that the validity of their inferences is restricted to that class of information. Of course, the same caveat applies to those who rely only on behavioral observations or biology. Every answer to a query on a questionnaire is the product of a judgment, often constructed at the moment, and is always affected by how the question is worded and the context. That is why Schwarz wrote, “Unfortunately, self-reports are a fallible source of data” (Schwarz, 1999, p. 93). Georg von Bekesy, who received a Nobel Prize for his research on hearing, once told a young scientist, “The method is everything.”

The Biological Variables

EEG Asymmetry

We assessed a number of biological variables that are indirect indexes of amygdalar excitability when the children were 11 years old. One measure was hemispheric asymmetry in alpha power in the EEG. The EEG represents the synchronized activity of large numbers of cortical pyramidal neurons, which, at any moment, have a dominant frequency of oscillation at varied sites. When children or adults are mentally and physically relaxed, maximal power is usually in the frequency range of 8 to 13 Hz, called alpha. About 50% to 60% of most samples display less alpha power in the left than the right frontal area, implying greater cortical activation in the left frontal area because during periods of arousal or thought alpha frequencies are replaced with higher-frequency bands. About 25% of most samples show less alpha in the right frontal area. Individuals in a happy, relaxed state are more likely to show greater activation in the left rather than the right frontal lobe (Atchley, Ilardi, & Enloe, 2003; Calkins et al., 1996; Davidson, 2003; Davidson, Ekman, & Saron, 1990; Fox et al., 2005; Piefke et al., 2003; Schmidt, Fox, Schulkin, & Gold, 1999). Depressed patients who showed improvement following drug therapy were more likely to display left hemisphere activation; those who were not helped by the drug were more likely to be right hemisphere active (Bruder et al., 2008). Because amygdalar activity is transmitted to the frontal lobe via the nucleus basalis, it is possible that greater activation of the right frontal lobe reflects greater activity in the right amygdala (Kapp, Supple, & Whalen, 1994). It is possible that the distribution of receptors for CRH, or level of CRH, affects the asymmetry of EEG activation, for monkeys with extreme right frontal activation across a 4-year interval had higher CRH levels (Kalin, Shelton, & Davidson, 2000).

In light of this evidence it is not surprising that the 11-year-olds in our sample who had been high-reactive infants and were inhibited in the second year were more likely than the low-reactive infants to show greater activation over the right than the left frontal area (Kagan & Snidman, 2004). High-reactive infants from a different sample who, at 4 and 7 years, were reticent while playing with unfamiliar children were more likely than others to show right frontal activation when they were 9 months old (Polak, Fox, Henderson, & Rubin, 2005).

Brainstem Auditory Evoked Potential

A second biological measure that differentiated high and low reactives was the fifth wave in the brainstem auditory evoked potential (BAER) evoked by a series of click sounds. This waveform reflects the neuronal activity generated by termination of the fibers of the lateral lemniscus on the inferior colliculus and occurs between 5.5 and 6 msec following the onset of a sound (Chiappa, 1983). The critical fact is that the amygdala projects to the inferior colliculus. As a result, an excitable amygdala can enhance the magnitude of the fifth waveform. Adults display an enhanced Wave 5 when they thought they might receive an electric shock, a warning that would activate the amygdala, compared with times when they were certain that no shock would be delivered (Baas, Milstein, Donlevy, & Grillon, 2006). Our expectation that high reactives, who are supposed to possess a more excitable amygdala, would have a larger Wave 5 than low reactives at 11 years was confirmed.

Sympathetic Tone

High reactives should show a greater sympathetic response in the cardiovascular system because the amygdala projects to relevant sympathetic targets. A spectral analysis of supine heart rate separates power in the lower-frequency band of the spectrum, reflecting sympathetic and parasympathetic activity, from power in the higher-frequency band, reflecting parasympathetic or vagal tone. This measure was combined with the adolescent’s resting heart rate. More high than low reactives combined greater power in the lower-frequency band of the cardiac spectrum with a higher resting heart rate. The 11-year-olds who showed the opposite pattern of more power in the higher frequency band and a lower heart rate, called high vagal tone, smiled frequently during the second year and described themselves as happy most of the time. Children who were reticent while playing with peers had higher sympathetic tone in the cardiovascular system compared with those who played alone but showed no sign of anxiety or those who were very sociable (Henderson, Marshall, Fox, & Rubin, 2002). Fetuses with high heart rates, implying greater sympathetic tone, displayed less smiling and laughter when they were 6 months old (Di Pietro et al., 1996). A very variable heart rate, which is a sign of high vagal tone, is associated with the tendency to approach unfamiliar events (Richards & Cameron, 1989) and people (Fox, 1989).

Event-Related Potentials

The magnitude of the event-related potential in the electroencephalogram between 250 and 500 msec to unfamiliar or unexpected scenes also separated high and low reactives at 11 years. Children with a more excitable amygdala should show larger waveforms to unfamiliar events because the amygdala sends projections to the locus coeruleus, ventral tegmentum, and basal nucleus of Meynert, which, in turn, project to pyramidal neurons that mediate the magnitude of the event-related potential to unfamiliarity. Dopamine neurons in the ventral tegmentum show increased activity to most unfamiliar or unexpected events. The high reactives displayed larger waveforms to ecologically invalid scenes, such as a child’s head on an animal’s body, than low reactives.

Other Biological Measures

The 11-year-old high reactives were a little more likely than low reactives to possess the combination of light blue eyes, a smaller body size, and a narrower face. This result would not surprise scientists who study the changes in physical characteristics that accompany the domestication of fox, mink, and cattle. Untamed fox have hairs that are black at the base and silver white at the outer edge, stiff erect ears, and a tail that turns down. Russian scientists selectively bred the small proportion of male fox who were tame with equally tame females for over 40 generations. The offspring of the 40 generations of selective breeding displayed white spots in their coat that were free of melanin pigmentation, floppy ears, an upturned tail, a different shape face, as well as lower levels of cortisol and higher levels of serotonin (Trut, 1999). A large region on fox chromosome 12, which corresponds to a region on chromosome 5 in dogs, contains genes that differentiate tame from wild foxes. Rats bred for tameness also show low cortisol levels (Albert et al., 2008), and barn owls that have large melanin spots on their breast feathers are more resistant to the usual consequences of increases in corticosterone (Almasi et al., 2008). Owners of cocker spaniels who have a red coat color say they are more nervous than the more common, darker colored spaniels (Keeler, 1947). Eighteenth-century manuals written for European mothers advised them to avoid hiring red-headed wet nurses because their irascible personalities rendered their breast milk dangerous for infants (Kagan, 2014).

If tame foxes inherit a combination of an inhibited temperament and select physical features, it is not surprising that high and low reactive Caucasian children differ in eye color and facial shape. These features are derivatives of neural crest cells in the developing embryo which migrate to become the melanocytes, which are the sources of melanin in the skin, hair, and iris, as well as facial bone and cartilage. Many genes contribute to the final fate of the neural crest cells (Wakamatsu, Nomura, Osumi, & Suzuki, 2014). Alleles of the genes that mediate the time of migration of neural crest cells and their molecular features may contribute to the cluster of physical, physiological, and behavioral features found in tame foxes, as well as inhibited children who had been high-reactive infants.

Not all of the biological measures we gathered differentiated inhibited from uninhibited children or high- from low-reactive infants. Although many investigators have suggested that potentiated startle to a cue that signals an aversive event is a measure of anxiety, we did not affirm that hypothesis when the threatening incentive was a light that warned of an unpleasant air puff to the throat or pictures of aversive scenes. Barker and colleagues (2015) reported a similar result with seven-year-old inhibited children from Fox’s sample.

The Evaluation at Age 15

The assessment at age 15 included a long interview in the youth’s home, the administration of two Q-sorts, and a laboratory evaluation of EEG asymmetry, event-related potentials, heart rate, and Wave 5 from the BAER.

The biological variables measured 4 years earlier were still differentiating, but to a lesser degree. Left or right frontal activation in the EEG was modestly stable from 11 to 15 years. More relevant is the fact that more high than low reactives showed right frontal activation at both ages. The magnitude of Wave 5 was also stable, and more high reactives had larger Wave 5 magnitudes at both 11 and 15 years of age. Finally, the high reactives showed shallower habituation of the waveform at 400 msec. to alternating blocks of different sets of valid and invalid pictures (Kagan, Snidman, Kahn, & Towsley, 2007).

We confirmed what other investigators have reported—namely, low correlations among the different biological variables. Left or right frontal activation is only one component of a brain state that must be integrated with other aspects of the person’s biology, his or her past experience, and, of course, the immediate setting. Left frontal activation does not predict an approach tendency in most individuals, independent of their temperament, history, and gender (Coan & Allen, 2004).

The behavioral consequences of most events cannot be predicted from recorded patterns of brain activity because the brain’s response is influenced not only by the person’s subjective interpretation of the incentive, but also by its physical features, familiarity, the event the brain expects to occur in the next moment, and the setting. The pattern of brain activity in the anterior cingulate to a painful heat stimulus depended on whether the self or an experimenter administered the painful event (Mohr et al., 2005). The brain state evoked by any event is the foundation of an envelope of possible psychological states. The one that is actualized depends on the person’s history and interpretation of the immediate context (Balaban, Alper, & Kasamon, 1996). The same caveat applies to psychological measures. Affirming that self is happy most of the time has an ambiguous meaning, for this self-description had different correlates in low and high reactives. A single measure, whether behavioral or biological, is ambiguous in meaning because it can be the product of more than one set of conditions. That is why patterns of measures are preferred.

Psychological Measures at Age 15

Variation in restless motoricity during the 3-hour home interview clearly separated the two temperaments. Almost one of every two high reactives, but only one of six low reactives, showed frequent small movements of the fingers or legs suggestive of tension and uncertainty. Although the high-reactive boys talked less often than the low-reactive boys, the high-reactive girls were as talkative as the low reactives when conversing with the female interviewer.

The answers to questions asking about sources of worry separated the two temperaments. We distinguished between realistic worries over performance in school or in extracurricular activities that did not meet the adolescent’s standards, on the one hand, and unrealistic worries, such as meeting strangers, encountering crowds, traveling to new places, the health of a pet, or the future, on the other. One high-reactive adolescent girl told the interviewer she did not like spring because of the unpredictability of the weather. A high-reactive boy said he felt anxious when he had to interact with someone because he was aware of the alternative things he could say and was always uncertain as to which statements would be interpreted as undesirable. Two thirds of high reactives, but only 20% of low reactives, nominated one or more unrealistic fears as a frequent source of worry, and these high reactives described themselves (on a Q-sort) as serious, thinking too much before deciding what to do, and wishing they were more relaxed and easygoing.

The distinction between realistic worries over inadequate school or athletic performance and less realistic worries over social interactions or new places resembles Freud’s contrast between realistic and neurotic anxieties. Unlike realistic worries, over which adolescents have some control, youths have a compromised ability to control the outcomes of encounters with strangers, new places, and novel challenges. The number of unrealistic worries of monozygotic twins and their spouses was more heritable than realistic fears of illness, an automobile accident, or being criticized for a mistake (Sundet et al., 2003). A survey of the fears of college students from seven different cultures revealed no gender or cultural differences in the number of realistic fears but significant cultural and gender differences in the frequency of unrealistic fears (Davey et al., 1998). It is likely that unrealistic worries have a different neurobiological foundation than concern over a final examination or poor performance on the soccer field. Fox odor and a conditioned stimulus signaling electric shock activate different genes in rats, and the behaviors that accompany these two incentives are uncorrelated across or within varied rat strains (Rosen, Adamec, & Thompson, 2005).

Because adolescents meet new people, visit unfamiliar places, and cannot know the future, we have to ask why high reactives more often named these events as primary sources of worry. One likely answer is that they are more susceptible to spontaneous, visceral feedback, especially from the autonomic nervous system and gut, that pierces consciousness and creates uncertainty because it is unpredictable, has an ambiguous origin, and the person is unsure of its meaning. This feeling shares features with the state evoked when, as children, they encountered unfamiliar people and situations. Because the amygdala participates in both states, it is possible that the psychological state created by unexpected visceral feedback functions as a conditioned stimulus to provoke the earlier state of uncertainty and, on occasion, an anxious or melancholic mood.

The 15-year-old high reactives were more likely than low reactives to affirm the statement, “I feel bad when one of my parents said I did something wrong.” Furthermore, the mothers of these high reactives reported that their children seemed to be unusually sensitive to punishment. It is possible that the greater religiousity of high-reactive adolescents is attributable to the fact that their spiritual commitment muted their guilt.

Low reactives, by contrast, experience fewer bouts of unpredictable bodily sensations. As a result, their attention is directed outward to the social environment, whereas high reactives, who are susceptible to these unpredictable bodily feelings, direct their attention inward. Over time, this habit can create a tendency to be preoccupied with one’s feelings and thoughts. Jung had suggested in his 1928 classic book, Psychological Types that this difference in focus of attention separates introverts from extraverts.

However, the frequent perception of change in bodily tone, which is often interpreted as anxiety or depression, could also result from failure of prefrontal modulation of the amygdala and/or a more reactive autonomic system. There is some evidence to support this claim. The anatomy and blood flow profiles of 135 high or low reactives were evaluated when they were 18 to 19 years of age by Carl Schwartz of Massachusetts General Hospital (Schwartz et al., 2010). One significant finding was that high reactives were more likely than low reactives to possess a thicker cortex in a small region in the anterior part of the ventromedial prefrontal cortex in the right hemisphere. The low-reactive boys with extremely thin cortices in this region (about 20% of this group) had displayed left frontal activation, high vagal tone, and low cortical arousal in the EEG at age 11 years. This site projects to the amygdala and to sites in the central gray that are responsible for the excessive arching of the back displayed by high reactives at 4 months. It is also relevant that cortical thickness in this area in 11-year-olds has a heritability of about 40% (Schmitt et al., 2008).

The ventromedial prefrontal cortex receives input from the hippocampus and parahippocampal area, sites that respond to discrepant events. This fact implies that a thicker cortex in this area might be associated with a lower threshold for the detection of novelty. (See Price, 2007, for a review of the connections of the orbitofrontal and medial prefrontal cortex.) Examination of the behaviors displayed at 4 and 14 months suggest that the greater cortical thickness of high reactives may have been present during infancy. The most anxious high-reactive adolescents who also possessed thicker values were among the most distressed and aroused 4-month-old infants in our sample. One sign was their frequent arching of the back to the unfamiliar stimuli.

In addition, the high–reactive infants displayed a larger surge of blood flow (BOLD signal) to the amygdala when they were presented with pictures they did not expect to see, as well as a shallower slope of habituation of the BOLD signal to the amygdala to alternating blocks of familiar and unfamiliar pictures during the 18-year evaluation (Schwartz et al., 2012). The shallower habituation implies a failure of amygdalar neurons to adapt to continued novelty (Kagan, 2014).

High-reactive youth in American and European societies are biased to interpret the feeling evoked by unexpected visceral feedback as implying they are worried or guilty. Because strangers, new places, and new challenges are the most frequent novelties in their lives, and the folk theory they were taught leads them to assume that these feelings are due to a compromise in their psychological characteristics, they often describe themselves as anxious. Members of other cultures might impose different interpretations on the same visceral feedback. Cambodian refugees living in Massachusetts interpret an unexpected bout of tachycardia as implying a weak heart caused by a loss of energy following a lack of sleep or a diminished appetite (Hinton et al., 2005a, 2005b).

The Saulteaux Indians of Manitoba worry about contracting a serious disease because illness implies they violated an ethical norm on sexual, aggressive, or sharing behavior (Hallowell, 1941). Social anxiety is a natural phenomenon, but it is especially salient in cultures where encounters with strangers and crowds are frequent and social acceptance is an important motive. A temperamental bias renders individuals vulnerable to select members of a family of emotions that psychiatrists call anxiety. History and culture select the specific member of the family that will be prevalent. Occupational and social failure in contemporary America have replaced the seven ancient sins of pride, anger, envy, avarice, sloth, gluttony, and lust as a basis for anxiety, shame, or guilt.


About one of five high reactives and one of four low reactives preserved, to age 15, some of the theoretically expected psychological and biological features of their temperament. About one third of an adult sample who remembered being extremely inhibited as children also said they were shy and introverted (Gladstone & Parker, 2005). However, less than 5% of high or low reactives developed a combination of behavior and biology characteristic of the other temperamental category. Therefore, the power of the early temperamental bias was to constrain the development of a behavioral profile rather than determine the development of a particular personality. The prediction that high reactives would not become consistently extraverted and sociable was affirmed more often than the prediction that a high reactive would become extremely shy and introverted.

Seven high reactives (four boys and three girls, representing about 10% of all the high reactives) were prototypic examples of their temperamental category. These seven were difficult to soothe when they were 4 months old, were extremely inhibited in the second year, and displayed either right frontal activation in the EEG, a large Wave 5, high sympathetic tone, or a large event-related potential to discrepant scenes at 11 or 15 years.

A small group of low-reactive boys was equally distinctive. Twelve boys, about 25% of all low-reactive boys, earned the sobriquet “Strong, silent, Clint Eastwood types.” These boys were uninhibited at 14 and 21 months and as adolescents were relaxed, regarded themselves as happy, and showed high vagal tone, left frontal activation, a small Wave 5, and small-magnitude waveforms to the discrepant pictures.

Fortunately, Fox and colleagues have reported similar results based on their longitudinal studies of infants who were administered a battery at 4 months that was similar to the one we used to classify the high- and low-reactive infants. Fox classified his large sample into three groups. Two resembled our high and low reactives and a third group was low in motoricity and crying but babbled and smiled frequently. The high reactives were most likely to be inhibited in the first and second years and most likely to show right frontal activation in the EEG (Fox et al., 2001, 2005). Furthermore, the high reactives who had displayed right frontal activation at 9 months of age were the most reticent when playing with three unfamiliar peers at age 4 and 7 years.


A number of implications follow from this evidence. First, high reactives are more vulnerable than most youths to develop social anxiety, even though most will not become social phobics. Less than 50% of children who were chronically shy developed social anxiety disorder as adults (Biederman et al., 2001), and most shy adults do not meet the criteria for social phobia. Although more high-reactive girls than high-reactive boys develop social anxiety disorder, one high-reactive boy developed an extreme degree of social anxiety. This boy displayed frequent arches of the back and a chronically unhappy facial expression at 4 months and was extremely fearful in the second year. He screamed when a stranger entered the playroom, when a blood pressure cuff was applied to his arm, and when a clown unexpectedly entered the room where he was playing with his mother on the couch. He missed many days during his senior year of high school because of social anxiety and feelings of panic when he was in large crowds. However, rather than presenting a timid persona, he was an angry adolescent who peppered his interviews with obscenities and denied any happiness or hope for a gratifying future. Very few high reactives develop this unusual profile. Among a sample of college students who reported being socially anxious, almost half also said that they liked high-risk activities and were not timid conformists (Kashdan, Elhai, & Breen, 2008). Whether they actually behaved this way or were describing their ego ideal cannot be known because no observations of their actual behavior were gathered. This is, as I noted, a serious problem when self-report represents the only evidence.

A diagnosis of depression or anxiety was most common among high-reactive girls. This result is in accord with a study of 500 Canadian children (Brendgen et al., 2005). In addition, college students who reported a melancholic mood across many days in a 3-week period were likely to have social anxiety (Kashdan & Steger, 2006). It is possible that the neurochemistry of the limbic areas of high reactives interferes with their ability to enjoy the pleasure that accompanies the receipt or anticipation of an unexpected, or a larger-than-anticipated, desirable event (Schultz, 2006). However, CRH neurons, which are activated by the anticipation of threat and are present in dopamine-producing brain sites, can suppress dopamine release when a reward is imminent (Austin, Rhodes, & Lewis, 1997). Perhaps one reason high-reactive adolescents reported not liking risky activities, despite the fact they promised the pleasure of excitement, is that they do not experience intense pleasure when they think about visiting a new city, meeting a new person, or engaging in a new activity. As a result, they avoid rather than seek these events (Netter, 2006).

The most important conclusion is that a temperamental bias does not guarantee any particular adult profile. It only limits the range of profiles that are likely to develop. The temperaments I inherited made it unlikely that I would become a test pilot or portfolio manager because the level of uncertainty that is an essential element in these careers makes me feel uncomfortable. But those biases could not determine the particular combination of traits I possess. Each temperament can be likened to a drop of black ink that disappears after being stirred in a vessel of glycerin. No observer could discern my pattern of temperaments from my current behavior, even though these biases exert an influence on my actions (Kagan, 2014). A pattern of temperaments, like the menu at a restaurant, proposes a variety of possibilities. A person’s life history selects one outcome from the larger set.

The two temperaments of high and low-reactivity have implications for variation in moral development. High reactives are especially vulnerable to feelings of guilt following an ethical violation. Kochanska (1991, 1993) has reported that shy, timid children raised by mothers who used reasoning in their socialization developed the strictest consciences (based on a projective measure). Most children experience anxiety, shame, or guilt following a violation of a moral standard, but high reactives are vulnerable to more intense forms of these emotions, whereas low reactives are biased to experience less intense feelings. Although the vast majority of low reactives will not become delinquents, if they are raised in environments that are overly permissive of aggression, lying, and stealing, the odds of such an outcome are a bit greater.

Ethnicity and Temperament

Human populations that have been reproductively isolated for many hundreds of generations differ in alleles that have implications for temperament. North and South Americans, Europeans, Africans, and Asians had been reproductively isolated for about 2,000 generations. Only 1,000 generations of isolation were needed to evolve the domestic cat from the aggressive wild cat or the pet dog from the wolf (Driscoll et al., 2007; Saetre et al., 2004). Thus, it is not surprising that Africans, Caucasians, and South and East Asians differ in many alleles (Li et al., 2008). For example, Caucasians differ from East Asians in at least 25% of the alleles in promoter regions for genes that could influence mood and behavior (Esau et al., 2008; Gardner, Bertranpetit, & Comas, 2008; Guthery et al., 2007; Spielman et al., 2007). Even within Caucasian subpopulations of Europeans there are small, but significant, genetic differences between those living in northwestern regions, for example Finland, and those in southeastern regions, such as Italy (Lao et al., 2008).

I noted earlier that the short and long alleles in the promoter region of the gene for the serotonin transporter molecule (the s and l alleles) are differentially linked to variation in moods. Japanese and Chinese adults are more likely than European-Caucasians and Africans to inherit the short allele. It is possible that the different genetic profiles of Asians and Caucasians help explain why 4-month-old Chinese infants are calmer than Caucasians. The former show less crying when presented with unfamiliar visual and auditory stimulation (Kagan et al., 1994). Japanese infants, too, are less easily aroused (Caudill & Weinstein, 1969) and less reactive to strangers than Caucasians (Lewis, Ramsay, & Kawakami, 1993). The parents of Thai children were more concerned with their children’s low energy and somatic problems, whereas the parents of Caucasian-American children reported more concern with disobedience and hyperactivity (Weisz et al., 1995, 2003). Although these results appear to be inconsistent with the fact that the Chinese-American 1-year-olds in the daycare study cited earlier were more likely to cry following separation from the mother, these infants were less aroused as younger infants and were described by their parents as less active and less exuberant than the Caucasian infants (Kagan, Kearsley, & Zelazo, 1978).

These early differences might be related to the fact that more Chinese than Caucasian adults prefer a calm to an aroused state. The Chinese who use illicit drugs usually abuse opiates, which induce relaxation; in contrast, most Caucasians prefer cocaine and amphetamines, which generate higher states of arousal (Tsai, 2007).

It is worth noting, as I did 20 years ago (Kagan, 1994), that East Asians, Caucasians, and Africans possess different facial skeletons. The Asians have the flattest faces; Africans have the most prominent nose, jaw, and chin. This variation is mediated by genes controlling the final fate of the neural crest cells that formed during the first trimester. It is possible that these genes also contribute to temperamental biases.

These scattered facts invite reflection on the two religious philosophies that have been differentially appealing to Asians and Europeans. The Reformation theologians emphasized dysphoric moods: both Martin Luther and John Calvin were convinced that anxiety and guilt were endemic to the human condition. Buddhist philosophy, which was more attractive to Asians than Christianity, made a feeling of serenity, not reduced levels of anxiety or guilt, the primary goal of life. The Buddhist imperative urges the elimination of all desire for material and sensory pleasures because a frustrated wish is the primary cause of a suffering that makes serenity impossible. It is tempting to speculate that the temperamental differences between Asians and Europeans made a small but nonetheless real contribution to the attractiveness of these two philosophies. If many Europeans were experiencing high levels of cortical and autonomic arousal and interpreted these sensations as implying guilt, fear, or anxiety, a philosophy that urged serenity would have met resistance because that state seemed unattainable. By contrast, a philosophy that accepted anxiety and guilt as definitive of the human condition would seem less valid to Asian adults whose temperaments permitted them more moments of serenity; achieving a chronically serene state might seem a real possibility for this group. Perhaps biology and culture came together in a symbiotic alliance to influence the preferred philosophical orientations of these two populations (Kagan, 1994).

The Need for Two Vocabularies

Ideally, temperamental biases should be defined by combinations of behavior and biology. That would require a special vocabulary because the words that describe the activity of neurons and circuits are not easily translated into the vocabulary that describes behaviors, thoughts, or feelings. It remains possible that some sentences describing brain patterns can never replace the sentences that describe the psychological features of temperaments.

The problem surrounding the selection of the best vocabulary for a phenomenon is captured by a frustrating nineteenth-century discussion between Heinrich von Helmholtz, who studied the physics of sound, and Johannes Brahms. The two men could not understand each other because Brahms spoke about form and counterpoint, whereas the physicist talked about sine waves and sound spectra. Although the latter are the physical foundations of all musical compositions, Brahms could not conceive of using Helmholtz’s vocabulary to describe his symphonies. A description of the pattern of neuronal excitation that occurs when a rat enters a cage containing a novel object cannot replace the statement, “Rats prefer to enter places where they have encountered an unfamiliar object in the past.” This conclusion does not mean that psychologists should ignore the circuits activated and the chemicals released when a rat places his paws in a cage containing a novel object. The psychological profile psychologists study can be likened to a gray cloth woven from thin black threads, representing biology, and thin white ones, representative of experience, that are invisible to an observer.

The plea for distinctive vocabularies for biological and psychological concepts and mechanisms has an analog in Bohr’s insistence that physicists must use the concepts of classical physics, not those of quantum theory, to describe an experiment, even though the latter is the foundation of the experimental procedures and results. When Edward Teller, father of the H-bomb, challenged Bohr’s skeptical position, the older man replied that if the experiments were summarized in the language of quantum mechanics, the two of them would not be sitting together drinking tea, but imagining their interaction. It is difficult to imagine a time when a mother’s guilt over giving birth to a damaged infant because she abused cocaine during her pregnancy will be explained or predicted by a particular pattern of neural activity.

Neuroscientists who use the concept fear rely on a semantic network that contains the terms amygdala, conditioned freezing, and electric shock as central terms. However, the central words in psychologists’ semantic network for fear are criticism, failure, illness, rejection, and uncertainty. Thomas Kuhn used the example of the French word douce and the English word sweet to make this point. Although sugar is called douce by the French and sweet by Americans, only the French would use the word douce to describe a bland-tasting soup, and only Americans would apply the word sweet to ingenuous girls. Thus, douce and sweet have related, but not synonymous, meanings. The behaviors, thoughts, feelings, and brain states that define a temperamental bias are the products of a series of phases that comprise a cascade. The forms and processes that define the phases of most biological cascades have to be described with different vocabularies. Chromosomes separate; neurons synapse; animals mate; and species evolve. High and low reactive, and inhibited and uninhibited, name collections of behaviors that, at the moment, cannot be translated into the sentences that describe genes, neurons, and circuits.

Final Questions

A large number of puzzles remain unresolved. Perhaps the most critical is discovery of the genes and the resulting neurochemical or neuroanatomical features that contribute to the high- and low-reactive profiles. A shy posture with strangers or timidity in unfamiliar settings can be the result of mechanisms other than those that create a high-reactive temperament. Many adolescents who were high-reactive infants and inhibited toddlers were not exceptionally shy at 15 years, and some who were not high reactive were shy adolescents. Thus, cross-sectional studies that use shyness in older children as an index of a high-reactive temperament will exclude many valid cases.

Two related issues are (1) determining the most sensitive set of biological measures that, when combined with behavior, discriminate high from low reactives and (2) discovering whether these measures change when a high-reactive infant who was an inhibited child becomes less fearful as an adult or when a low-reactive adolescent who was a fearless child becomes shy. Is the threshold of amygdalar excitability malleable to experience, or are these thresholds relatively independent of the individual’s persona? In addition, it would be important to determine whether belonging to a disadvantaged social class or ethnic minority affects the development of high and low reactives. It is reassuring that about the same proportion of very shy, socially anxious 2-year-old boys from a working-class population that included African-American and Hispanic families maintained an anxious profile across the first decade (Feng, Shaw, & Silk, 2008).

Finally, it would be important to determine whether the strategy we adopted in the study of high and low reactives will prove fruitful for probing other temperamental biases that are the origins of forms of hyperactivity, impulsivity, emotional callousness, extreme levels of exuberance, or bouts of depression.


The inclusion of biological evidence in studies of temperament is a welcome development. The history of the sciences is rich with examples of the progress that occurs when previously separate domains probe common phenomena. The union of two domains often provides new information that refines existing terms and eliminates concepts that have outlived their usefulness. Psychologists may eventually replace the currently popular constructs that describe children and their environments as separate entities with synthetic terms that classify a particular temperamental type growing up in a particular environment. Instead of writing about high-reactive infants on the one hand, and permissive families on the other, psychologists might invent a new construct that describes the envelope of profiles possible when high-reactive children grow up in permissive homes. This suggestion is an instance of the more general rule that the construct chosen depends on the investigator’s purpose. Light can be described as a wave or a particle. Depending on one’s theory, one can describe a psychological profile in terms of its current features, biological origins, or history. Each child’s temperament influences the probability that an event will evoke a change in feeling; the child’s history determines the meaning he or she will impose on that feeling.

My colleagues and I were surprised by the length of the shadow cast by the infant profiles we called high and low reactive and believe we were fortunate to have initiated the assessments at 4 months old rather than at an earlier or later age. Nature occasionally opens her gate to reveal some of the exotic flowers in her courtyard. If the scientist happens to be turned away at the critical moment, her sanctum remains a mystery.

Questions for Future Research

  1. 1. What biological mechanisms lead to each temperamental bias?

  2. 2. Specifically, what are the neurochemistries of high- and low-reactive infants, and what genes contribute to these distinct neurochemistries?

  3. 3. What are the postnatal experiences that allow some high reactives to conquer their extreme avoidant tendencies and some low reactives to remain relaxed, exuberant extraverts?

  4. 4. How can clinicians differentiate between adult social phobics who inherited a high-reactive biology and those who did not?

  5. 5. What combination of initial temperaments and later experiences contributes to the psychological differences between adults with an Asian pedigree and those with a European-Caucasian pedigree?


Abdolmaleky, H. M., Thiagalingam, S., & Wilcox, M. (2005). Genetics and epigenetics in major psychiatric disorders: Dilemmas, achievements, applications, and future scope. American Journal of Pharmacogenomics, 5, 149–160.Find this resource:

Abercrombie, H. C., Kalin, N. H., Thurow, M. E., Rosenkranz, M. A., & Davidson, R. J. (2003). Cortisol variation in humans affects memory for emotionally laden and neutral information. Behavioral Neuroscience, 117, 505–516.Find this resource:

Achenbach, T. M. (1985). Assessment and taxonomy of child and adolescent psychopathology. Newbury Park, CA: Sage.Find this resource:

Adamec, R. E. (1991). Anxious personality in the cat. In B. J. Carroll & J. E. Barrett (Eds.), Psychopathology and the brain (pp. 153–168). New York, NY: Raven Press.Find this resource:

Albert, F. W., Shchepina, O., Winter, C., Rompler, H., Teupser, D., Palme, R., … Paabo, S. (2008). Phenotypic differences in behavior, physiology, and neurochemistry between rats selected for tameness and for defensive aggression towards humans. Hormones and Behavior, 53, 413–421.Find this resource:

Almasi, B., Roulin, A., Jenni-Eiermann, S., & Jenni, L. (2008). Parental investment and its sensitivity to corticosterone is linked to melanin-based coloration in barn owls. Hormones and Behavior, 54, 217–223.Find this resource:

Arbelle, S., Benjamin, J., Galin, M., Kremer, P., Belmaker, R. H., & Ebstein, R. P. (2003). Relation of shyness in grade school children to the genotype for the long form of the serotonin transporter promoter region polymorphism. American Journal of Psychiatry, 160, 671–676.Find this resource:

Atchley, R. A., Ilardi, S. S., & Enloe, A. (2003). Hemispheric asymmetry in the processing of emotional content in word meanings. Brain and Language, 84, 105–119.Find this resource:

Auerbach, J. G., Benjamin, J., Faroy, M., Geller, V., & Ebstein, R. (2001). DRD4 related to infant attention and information processing. A developmental link to ADHD? Psychiatric Genetics, 11, 31–35.Find this resource:

Auerbach, J., Geller, V., Lezer, S., Shinwell, E., Belmaker, R. H., & Levin, J. (1999). Dopamine D4 receptor (D4DR) and serotonin transporter promoter (5-HTTLPR) polymorphisms in the determination of temperament in two-month-old infants. Molecular Psychiatry, 4, 369–373.Find this resource:

Austin, M. C., Rhodes, J. L., & Lewis, D. A. (1997). Differential distribution of corticotropin-releasing hormone immunoreactive axons in monoaminergic nuclei of the human brainstem. Neuropsychopharmacology, 17, 326–341.Find this resource:

Baas, J. M., Milstein, J., Donlevy, M., & Grillon, C. (2006). Brainstem correlates of defensive states in humans. Biological Psychiatry, 59, 588–593.Find this resource:

Bailey, A. A., & Hurd, P. L. (2005). Depression in men is associated with more feminine finger length ratios. Personality and Individual Differences, 39, 829–836.Find this resource:

Balaban, E., Alper, J. S., & Kasamon, Y. L. (1996). Mean genes and the biology of aggression: A critical review of recent animal and human research. Journal of Neurogenetics, 11, 1–43.Find this resource:

Barker, T. V., Reeb-Sutherland, B., Degnan, K. A., Walker, O. L., Chronis-Tuscano, A., Henderson, H. A., Pine, D. S. & Fox, N. A. (2015). Contextual startle responses moderate the relation between behavioral inhibition and anxiety in middle childhood. Psychophysiology, Sep. 2 doi: 10.1111/psyp 12517.Find this resource:

Barnett, J. H., Heron, J., Ring, S. M., Golding, J., Goldman, D., Xu, K., & Jones, P. B. (2007). Gender-specific effects of the Catechol-O-Methyltransferase Val108/158 Met polymorphism on cognitive function in children. American Journal of Psychiatry, 164, 142–149.Find this resource:

Barot, S. K., & Bernstein, I. L. (2005). Polycose taste pre-exposure fails to influence behavioral and neural indices of taste novelty. Behavioral Neuroscience, 119, 1640–1647.Find this resource:

Battaglia, M., Ogliari, A., Zanoni, A., Cittterio, A., Pozzoli, U., Giorda, R., Maffei C., & Marino C. (2005). Influence of the serotonin transporter promoter gene and shyness on children’s cerebral responses to facial expressions. Archives of General Psychiatry, 62, 85–94.Find this resource:

Bethea, C. L., Streicher, J. M., Coleman, K., Pau, F. K., Moessner, R., & Cameron, J. L. (2004). Anxious behavior and fenfluramine-induced prolactin secretion in young rhesus macaques with different alleles of the serotonin reuptake transporter polymorphism (5HTTLPR). Behavior Genetics, 34, 295–307.Find this resource:

Biederman, J., Hirshfeld-Becker, D. R., Rosenbaum, J. F., Herot, C., Friedman, D., Snidman, N., Kagan, J., & Faraone, S. V. (2001). Further evidence of association between behavioral inhibition and social anxiety in children. American Journal of Psychiatry, 158, 1673–1679.Find this resource:

Bigos, K. L., Pollock, B. G., Aizenstein, H. J., Fisher, P. M., Bies, R. R., & Hariri, A. R. (2008). Acute 5-HT reuptake blockade potentiates human amygdala reactivity. Neuropsychopharmacology, 33, 3221–3225.Find this resource:

Brendgen, M., Wanner, B., Morin, A. J. S., & Vitaro, F. (2005). Relations with parents and with peers, temperament, and trajectories of depressed mood during early adolescence. Journal of Abnormal Child Psychology, 33, 579–594.Find this resource:

Bruder, G. E., Sedoruk, J. P., Stewart, J. W., McGrath, P. J., Quitkin, F. M., & Tenke, C. E. (2008). Electroencephalographic alpha measures predict therapeutic response to a selective serotonin reuptake inhibitor antidepressant: Pre-and post-treatment findings. Biological Psychiatry, 63, 1171–1177.Find this resource:

Bunzeck, N., & Duzel, F. (2006). Absolute coding of stimulus novelty in the human substantia nigra/VTA. Neuron, 51, 369–379.Find this resource:

Byrne, J., & Suomi, S. J. (2002). Cortisol reactivity and its relation to home cage behavior and personality ratings in tufted capuchin (Cebux apella) juveniles from birth to 6 years of age. Psychoendocrinology, 27, 139–154.Find this resource:

Caci, H., Robert, P., Dossio, C., & Boyer, P. (2005). Morningness—Eveningness for Children Scale: Psychometric properties and month of birth effect. Encephale, 31, 56–64.Find this resource:

Calkins, S. D., Fox, N. A., & Marshall, T. R. (1996). Behavioral and physiological antecedents of inhibited and uninhibited behavior. Child Development, 67, 523–540.Find this resource:

Caspi, A., & Silva, P. A. (1995). Temperamental qualities at age 3 predict personality traits in young adulthood. Child Development, 66, 486–498.Find this resource:

Caudill, W., & Weinstein, H. (1969). Maternal care and infant behavior in Japan and America. Psychiatry, 32, 12–43.Find this resource:

Chiappa, K. H. (1983). Evoked potentials in clinical medicine. New York, NY: Raven Press.Find this resource:

Chotai, J., & Adolfsson, R. (2002). Converging evidence suggests that monoamine neurotransmitter turnover in human adults is associated with their season of birth. European Archives of Psychiatry and Clinical Neuroscience, 252, 130–134.Find this resource:

Chotai, J., Serretti, J., Lattuada, E., Lorenzi, C., & Lilli, R. (2003). Gene-environment interaction in psychiatric disorders as indicated by season of birth variations in tryptophan hydroxylase (TPH), serotonin transporter (5HTTLPR) and dopamine receptor (DRD4) gene polymorphisms. Psychiatry Research, 119, 99–111.Find this resource:

Ciesla, W. (2001). Can melatonin regulate the expression of prohormone convertase 1 and 2 gene via monomeric and dimeric forms of RZR/ROR nuclear receptor, and can melatonin influence the processes of embryogenesis or carcinogenesis by disturbing the proportion of cAMP and cGMP concentrations? Medical Hypothesis, 56, 181–193.Find this resource:

Coan, J. A., & Allen, J. J. B. (2004). Frontal EEG asymmetry as a moderator and mediator of emotion. Biological Psychology, 67, 47–49.Find this resource:

Collaer, M. L., & Hines, A. (1995). Human behavioral differences. Psychological Bulletin, 118, 55–107.Find this resource:

Courtet, P., Jollant, F., Buresi, C., Castelnau, D., Mouthon, D., & Malafosse, A. (2005). The monoamine oxidase A gene may influence the means used in suicide attempts. Psychiatric Genetics, 15, 189–193.Find this resource:

Davey, G. C., McDonald, A. S., Hirisave, U., Prabhu, G. G., Iwawaki, S., Jim, C. I., … Reimann B. C. (1998). A cross-cultural study of animal fears. Behaviour Research and Therapy, 36, 735–750.Find this resource:

Davidson, R. J. (2003). Affective neuroscience and psychophysiology. Psychophysiology, 40, 655–665.Find this resource:

Davidson, R. J., Ekman, P., & Saron C. D. (1990). Approach withdrawal and cerebral asymmetry: Emotional expression and brain physiology. Journal of Personality and Social Psychology, 58, 330–341.Find this resource:

De Luca, A., Rizzardi, M., Buccino, A., Alessandroni, R., Salvioli, G. P., Filograsso, N., Novelli, G., & Dallapiccola, B. (2003). Association of dopamine D4 receptor (DRD4) exon III repeat polymorphism with temperament in 3-year-old infants. Neurogenetics, 4, 207–212.Find this resource:

Di Pietro, J. A., Hodgson, D. M., Costigan, K. A., & Johnson, T. R. B. (1996). Fetal antecedents of infant temperament. Child Development, 67, 2568–2583.Find this resource:

Driscoll, C. A., Menotti-Raymond, M., Roca, A. L., Hupe, K., Johnson, W. E., Geffen, E., … MacDonald, D. W. (2007). The near Eastern origin of cat domestication. Science, 317, 519–523.Find this resource:

D’Souza, U. M., & Craig, I. W. (2006). Functional polymorphisms in dopamine and serotonin pathway genes. Human Mutation, 27, 1–13.Find this resource:

Esau, L., Kaur, M., Adonis, L., & Arieff, Z. (2008). The 5-HTTLPR polymorphism in South African healthy populations: A global comparison. Journal of Neural Transmission, 115, 755–760.Find this resource:

Feng, X., Shaw, D. S., & Silk, J. S. (2008). Developmental trajectories of anxiety symptoms among boys across early and middle childhood. Journal of Abnormal Psychology, 117, 32–47.Find this resource:

Fink, B., Manning, J. T., Williams, J. H. G., & Podmore-Nappin C. (2007). The 2nd to 4th digit ratio and developmental psychopathology in school-aged children. Personality and Individual Differences, 42, 369–379.Find this resource:

Fink, B., Grammer, K., Mitteroecker P., Gunz P., Schaefer K., Bookstein F. L., & Manning J. T. (2005). Second to fourth digit ratio and face shape. Proceedings of the Royal Society B, 272, 1995–2001.Find this resource:

Fitzgerald, D. A., Angstadt, M., Jelsone, L. M., Nathan, P. J., & Phan, K. L. (2006). Beyond threat. NeuroImage, 30, 1441–1448.Find this resource:

Fox, N. A., Henderson, H. A., Marshall, T. J., Nichols, K. E., & Ghera, M. N. (2005). Behavioral inhibition: Linking biology and behavior within a developmental framework. In S. T. Fiske, A. E. Kazdin, & D. L. Schacter (Eds.), Annual review of psychology (Vol. 56, pp. 235–262). Palo Alto, CA: Annual Reviews.Find this resource:

Fox, N. A., Henderson, H. A., Rubin, K. H., Calkins, S. D., & Schmidt, L. A. (2001). Continuity and discontinuity of behavioral inhibition and exuberance. Child Development, 72, 1–21.Find this resource:

Fox, N. A. (1989). Psychophysiological correlates of emotional reactivity during the first year of life. Developmental Psychology, 25, 364–372.Find this resource:

Fried, I., MacDonald, K. A., & Wilson, C. L. (1997). Single neuron activity in human hippocampus and amygdala during recognition of faces and objects. Neuron, 18, 753–765.Find this resource:

Garcia-Coll, C., Kagan, J., & Reznick, J. S. (1984). Behavioral inhibition in children. Child Development, 55, 1005–1009.Find this resource:

Gardner, M., Bertranpetit, J., & Comas, D. (2008). Worldwide genetic variation in dopamine and serotonin pathway genes. American Journal of Medical Genetics Part B, 82, 1–6.Find this resource:

Gartstein, M. A., Knyazev, G. G., & Slobodskaya, H. R. (2005). Cross-cultural differences in the structure of infant temperament. Infant Behavior and Development, 28, 54–61.Find this resource:

Gladstone, G., & Parker, G. (2005). Measuring a behaviorally inhibited temperament style: Development and initial validation of new self-report measures. Psychiatry Research, 135, 133–143.Find this resource:

Goldberg, A. E., & Newlin D. B. (2000). Season of birth and substance abuse: Findings from a large national sample. Alcohol Clinical and Experimental Research, 24, 774–780.Find this resource:

Goldsmith, H. H., & Campos, J. J. (1990). The structure of temperamental fear and pleasure in infants. Child Development, 61, 1944–1964.Find this resource:

Gonda, X., Fountoulkis, K. N., Rihmer, Z., Lazary, J., Laszik, A., Akiskal, K. K., Akiskal, H. S., & Bagdy G. (2008). Towards a genetically validated new affective temperamental scale: A delineation of the temperament “phenotype” of 5-HTTLPR using the TEMPS-A. Journal of Affective Disorders, 112, 19–29.Find this resource:

Gortmaker, S. L., Kagan, J., Caspi, A., & Silva, P. A. (1997). Daylength during pregnancy and shyness in children. Developmental Psychobiology, 41, 107–114.Find this resource:

Grabe, H. J., Lange, M., Wolff, B., Volzke, H., Lucht, M., Freyberger, H. J., John, U., & Cascorbi I. (2005). Mental and physical distress is modulated in the 5-HT transporter gene interacting with social stressors and chronic disease burden. Molecular Psychiatry, 10, 220–224.Find this resource:

Gunnar, M. R. (1994). Psychoendocrine studies of temperament and stress in early childhood. In J. Bates & T. Wachs (Eds.), Temperament: Individual differences at the interface of biology and behavior (pp. 175–198). Washington, DC: American Psychological Association.Find this resource:

Guthery, S. L., Salibury, B. A., Pungliya, M. S., Stephens, J. C., & Banshad, M. (2007). The structure of common genetic variation in United States populations. American Journal of Human Genetics, 81, 1221–1231.Find this resource:

Hall, P. A., & Schaeff, C. M. (2008). Sexual orientation and fluctuating asymmetry in men and women. Archives of Sexual Behavior, 37, 158–165.Find this resource:

Hallowell, A. I. (1941). The social function of anxiety in a primitive society. American Sociological Review, 6, 869–891.Find this resource:

Handa, R. J., Nunley, K. M., Lorens, S. A., Louie, J. P., McGivern, R. F., & Bullnow, M. R. (1994). Androgen regulation of adrenocorticotropin and corticosterone secretion in the male rat following novelty and foot shock stressors. Physiology and Behavior, 55, 117–124.Find this resource:

Hanson, D. K., Jones, B. A., & Watson, N. V. (2004). Distribution of androgen receptor immunoreactivity in the brainstem of male rats. Neuroscience, 127, 797–803.Find this resource:

Hariri, H. R., & Brown, S. M. (2006). Serotonin. American Journal of Physiology, 163, 12.Find this resource:

Hartl, D., & Jones, E. W. (2005). Genetics (6th ed.). Boston: Jones & Bartlett.Find this resource:

Henderson, H. A., Marshall, T. J., Fox, N. A., & Rubin, K. H. (2002). Psychophysiological and behavioral evidence for varying forms and functions of non-social behavior in preschoolers. Unpublished manuscript, Department of Human Development, University of Maryland.Find this resource:

Hinton, D. E., Pich, V., Chean, D., Pollack, M. H., & McNally, R. J. (2005a). Sleep paralysis among Cambodian refugees: Association with PTSD diagnosis and severity. Depression and Anxiety, 22, 47–51.Find this resource:

Hinton, D. E., Pich, V., Safren, S. A., Pollack, M. H., & McNally, R. J. (2005b). Anxiety sensitivity in traumatized Cambodian refugees. Behavioral Research and Therapy, 43, 1631–1643.Find this resource:

Hyde, J. S. (2005). The gender similarities hypothesis. American Psychologist, 60, 581–592.Find this resource:

Jabbi, M., Korf, J., Kema, I. P., Hartman, C., van der Pompe, G., Minderaa, R. B., Ormel, J., & den Boer J. A. (2007). Convergent genetic modulation of the endocrine stress response involves polymorphic variations of 5-HTT, COMT and MAOA. Molecular Psychiatry, 12, 483–490.Find this resource:

Jansen, P. W., Raat, H., Mackenbach, J. P., Jaddoe, V. W., Hofman, A., Verhulst, F. C., & Tiemeier, H. (2008). Socioeconomic inequalities in infant temperament: The Generation R Study. Social Psychiatry Psychiatric Epidemiology, 44, 87–95.Find this resource:

Jarrell, H., Hoffman, J. B., Kaplan, J. R., Berga, S., Kinkead, B., & Wilson, M. E. (2008). Polymorphisms in the serotonin reuptake transporter gene modify the consequences of social status on metabolic health in female rhesus monkeys. Physiology and Behavior, 93, 807–819.Find this resource:

Joiner, T. E., Pfaff, J. J., Acres, J. G., & Johnson, F. (2002). Birthmonth and suicidal and depressive symptoms in Australians born in the Southern vs. the Northern Hemisphere. Psychiatry Research, 112, 89–92.Find this resource:

Kaasinen, V., Nagren, K., Hietala, J., Farde, L., & Rinne, J. O. (2001). Sex differences in extrastriatal dopamine d(2)-like receptors in the human brain. American Journal of Psychiatry, 158, 308–311.Find this resource:

Kagan, J. (1994). Galen’s Prophecy. New York, NY: Basic Books.Find this resource:

Kagan, J. (2014). Temperamental contributions to the development of psychological profiles. I and II. In S. G. Hofmann & P. DiBartolo, (Eds.), Social Anxiety, 3rd ed. (pp. 378–450). New York, NY: Elsevier.Find this resource:

Kagan, J., Arcus, D., Snidman, N., Yufeng, W., Hendler, J., & Greene, S. (1994). Reactivity in infants: A cross-national comparison. Developmental Psychology, 30, 342–345.Find this resource:

Kagan, J., Kearsley, R., & Zelazo, P. R. (1978). Infancy. Cambridge, MA: Harvard University Press.Find this resource:

Kagan, J., & Moss, H. A. (1962). Birth to maturity: A study in psychological development. New York, NY: Wiley (reprinted by Yale University Press, 1982).Find this resource:

Kagan, J., & Saudino, N. (2001). Behavioral inhibition and related temperaments. In R. N. Emde & J. K. Hewitt (Eds.), Infancy to early childhood (pp. 111–122). New York, NY: Oxford University Press.Find this resource:

Kagan, J., & Snidman, N. (2004). The long shadow of temperament. Cambridge, MA: Harvard University Press.Find this resource:

Kagan, J., Snidman, N., Kahn, V., & Towsley, S. (2007). The preservation of two infant temperaments into adolescence. Monographs of the Society for Research in Child Development, 72, 1–75.Find this resource:

Kalin, N. H., Shelton, S. E., & Davidson, R. J. (2000). Cerebrospinal fluid corticotrophin—releasing hormone levels are elevated in monkeys with patterns of brain activity associated with fearful temperament. Biological Psychiatry, 47, 579–585.Find this resource:

Kandiel, A., Chen, S., & Hillman, D. E. (1999). c-fos gene expression parallels auditory adaptation in the adult rat. Brain Research, 839, 292–297.Find this resource:

Kapp, B. S., Supple, W. F., & Whalen, R. (1994). Effects of electrical stimulation of the amygdaloid central nucleus on neocrotical arousal in the rabbit. Behavioral Neuroscience, 108, 81–93.Find this resource:

Kashdan, T. B., Elhai, J. D., & Breen, W. E. (2008). Social anxiety and disinhibition. Journal of Anxiety Disorders, 22, 925–935.Find this resource:

Kashdan, T. B., & Steger, M. F. (2006). Expanding the topography of social anxiety. Psychological Science, 17, 120–128.Find this resource:

Keeler, C. E. (1947). Coat color, physique, and temperament. The Journal of Heredity, 38, 271–277.Find this resource:

Kilpatrick, D. G., Koenen, K. C., Ruggiero, K. J., Acierno, R., Galea S., Resnick, H. S., … & Gelernter, J. (2007). The serotonin transporter-genotype and social support and moderation of posttraumatic stress disorder and depression in hurricane-exposed adults. American Journal of Psychiatry, 164, 1693–1699.Find this resource:

Kinney, D. K., Miller, A. M., Crowley, D. J., Huang, E., & Gerber, E. (2008). Autism prevalence follows prenatal exposure to hurricanes and tropical storms in Louisiana. Journal of Autism & Developmental Disorders, 38, 481–488.Find this resource:

Kochanska, G. (1991). Socialization and temperament in the development of guilt and conscience. Child Development, 62, 1379–1392.Find this resource:

Kochanska G. (1993). Toward a synthesis of parental socialization and child temperament in early development of conscience. Child Development, 64, 325–347.Find this resource:

La Gasse, L., Gruber, C., & Lipsitt, L. P. (1989). The infantile expression of avidity in relation to later assessments. In J. S. Reznick (Ed.), Perspectives on behavioral inhibition (pp. 159–176). Chicago, IL: University of Chicago Press.Find this resource:

Lahti, A., Raikkonen, K., Ekelund, J., Peltonen, L., Raitakari, O. T., & Keltikangas-Jarvinen, L. (2006). Socio-demographic characteristics moderate the association between DRD4 and novelty seeking. Personality and Individual Differences, 40, 533–543.Find this resource:

Lakatos, K., Nemoda, Z., Birkas, E., Ronai, Z., Kovacs, E., Ney, K., … Gervai, J. (2003). Association of D4 dopamine receptor gene and serotonin transporter polymorphism with infants’ response to novelty. Molecular Psychiatry, 8, 90–98.Find this resource:

Lao, O., Lu, T. T., Nothnagel, M., Junge, O., Freitag-Wolf, S., Callebe, A., … Holmlund G., et al. (2008). Correlation between genetic and geographic structure in Europe. Current Biology, 26, 1241–1248.Find this resource:

Lewis, M., Ramsay, D. S., & Kawakami, K. (1993). Differences between Japanese infants and Caucasian-American infants in behavioral and cortisol response to inoculation. Child Development, 64, 1722–1731.Find this resource:

Li, J. Z., Absher, D. M., Tang, H., Southwick, A. M., Casto, A. M., Ramachandran, S., …Myers, R. M. (2008). Worldwide human relationships inferred from genome-wide patterns of variation. Science, 319, 1100–1104.Find this resource:

Li, L., Power, C., Kelly, S., Kirschbaum, C., & Hertzman, C. (2007). Life-time socio-economic position and cortisol patterns in mid-life. Psychoneuroendocrinology, 32, 824–833.Find this resource:

Manning, J. T., & Fink, B. (2008). Digit ratio (2D:4D), dominance, reproductive success, asymmetry, and sociosexuality in the BBC internet study. American Journal of Human Biology, 23, 527–533.Find this resource:

Manuck, S. B., Bleil, M. E., Petersen, K. L., Flor, J. D., Mann, J. J., Ferrell, R. E., & Muldooon, M. F. (2005). The socio-economic status of communities predicts variation in brain serotonergic responsivity. Psychological Medicine, 35, 519–528.Find this resource:

Manuck, S. B., Flory, J. D., Ferrell, R. E., & Muldoon, M. F. (2004). Socio-economic status variation in the serotonin transporter gene-linked polymorphic region. Psychoneuroendocrinology, 29, 651–668.Find this resource:

Maren, S., Yap, S. A., & Goosens, K. A. (2001). The amygdala is essential for the development of neuronal plasticity in the medial geniculate nucleus during auditory fear conditioning in rats. Journal of Neuroscience, 21, RC135.Find this resource:

McNally, G. P., & Westbrook, R. F. (2003). Opioid receptors regulate the extinction of Pavlovian fear conditioning. Behavioral Neuroscience, 117, 1292–1301.Find this resource:

Mehta, P. H., Jones, A. C., & Josephs, R. A. (2008). The social endocrinology of dominance. Journal of Personality and Social Psychology, 94, 1078–1093.Find this resource:

Merali, Z., McIntosh, J., Kent, T., Michaud, D., & Anisman, H. (1998). Aversive and appetitive events evoke the release of corticotropin-releasing hormone and Bombesin-like peptides of the central nucleus of the amygdala. Journal of Neuroscience, 18, 4758–4766.Find this resource:

Miyawaki, T., Goodchild, A. K., & Pilowsky, P. M. (2002). Activation of mu-opioid receptors in rat ventrolateral medulla selectively blocks baroreceptor reflexes while activation of delta opioid receptors blocks somato-sympathetic reflexes. Neuroscience, 109, 133–144.Find this resource:

Mohr, C., Binkofski, F., Erdmann, C., Buchel, C., & Helmchen, C. (2005). The anterior cingulate cortex contains distinct areas dissociating external from self-administered painful stimulation: A parametric fMRI study. Pain, 114, 347–357.Find this resource:

Munro, C. A., McCaul, M. E., Wong, D. F., Oswald, L. M., Zhou, Y., Brasic, J., … Wand G. S. (2006). Sex differences in striatal dopamine-release in healthy adults. Biological Psychiatry, 59, 966–974.Find this resource:

Netter P. (2006). Dopamine challenge tests as an indicator of psychological traits. Human Psychopharmacology, 21, 95–99.Find this resource:

Patterson, P. H. (2006). Pregnancy, immunity, schizophrenia, and autism. Engineering & Science, 3, 11–21.Find this resource:

Pfister, H. P., & Muir, J. L. (1989). Influence of exogenously administered oxytocin on central noradrenaline dopamine and serotonin levels following psychological stress in nulliparous female rats (Rattus norvegicus). International Journal of Neuroscience, 45, 221–229.Find this resource:

Pitkanen, A. (2000). Connectivity of the rat amygdaloid complex. In J. P. Aggleton (Ed.), The amygdala (2nd ed., pp. 31–116). New York, NY: Oxford University Press.Find this resource:

Piefke, M., Weiss, P. H., Zilles, K., Markowitsch, H. J., & Fink G. R. (2003). Differential remoteness and emotional tone modulate the neural correlates of autobiographical memory. Brain, 126, 650–668.Find this resource:

Pjrek, E., Winkler, D., Heiden, A., Praschak-Rieder, N., Willeit, M., Konstantinidis, A., Stastny, J., & Kasper, S. (2004). Seasonality of birth in seasonal affective disorder. Journal of Clinical Psychiatry, 65, 1389–1393.Find this resource:

Polak, C., Fox, N. A., Henderson, H. A., & Rubin, K. H. (2005). Behavioral and physiological correlates of socially wary behavior in middle childhood: Does social wary behavior mediate the relation between infant temperament and later childhood maladjustment. Unpublished manuscript.Find this resource:

Price, J. L. (2007). Definition of the orbital cortex in relation to specific connections with limbic and visceral structures and other cortical regions. Annals of the New York Academy of Sciences, 1121, 54–71.Find this resource:

Ray, J., Hansen S., & Waters, N. (2006). Links between temperamental dimensions and brain monoamines in the rat. Behavioral Neuroscience, 120, 85–92.Find this resource:

Rebec, G. V., Christianson, J. R., Guevra, C., & Bardo, M. T. (1997). Regional and temporal differences in real time dopamine efflux in the nucleus accumbens during food choice novelty. Brain Research, 776, 61–67.Find this resource:

Reimold, M., Batra, A., Knobel, A., Smolka, M. N., Zimmer, A., Mann, K., … Heinz A. (2008). Anxiety is associated with reduced central serotonin transporter availability in unmedicated patients with unipolar major depression: a [11C] PET study. Molecular Psychiatry, 13, 606–613.Find this resource:

Richards, J. E., & Cameron D. (1989). Infant heart rate variability and behavioral developmental status. Infant Behavior and Development, 12, 45–58.Find this resource:

Rimm-Kaufman, S. E. (1996). The contribution of behavioral inhibition in reaction to transition to kindergarten. Unpublished doctoral dissertation, Harvard University, Cambridge, MA.Find this resource:

Rolls, E. T., Browning, A. S., Inoue, K., & Hernadi, I. (2005). Novel visual stimuli activate a population of neurons in the primate orbitofrontal cortex. Neurobiology of Learning and Memory, 84, 111–123.Find this resource:

Roozendaal, B., & Cools, A. R. (1994). Influence of the noradrenergic state of the nucleus accumbens in basolateral amygdala mediated changes in neophobia of rats. Behavioral Neuroscience, 108, 1107–1118.Find this resource:

Rosen, J. B, Adamec, R. E., & Thompson, B. C. (2005). Expression of egr-1 (zif268) RNA in select fear-related brain regions following exposure to a predator. Behavioral Brain Research, 162, 279–288.Find this resource:

Rothbart, M. K., Ahadi, S. A., & Evans, D. E. (2000). Temperament and personality: Origins and outcomes. Journal of Personality and Social Psychology, 78, 122–135.Find this resource:

Rothbart, M. K., Ahadi, B. A., Hershey, K. L., & Fisher, D. (2001). Investigation of temperament at three to seven years. Child Development, 72, 1394–1408.Find this resource:

Rothbart, M. K., & Bates J. E. (2006). Temperament. In N. Eisenberg (Vol. 3 Ed.) and W. Damon and R. M. Lerner (Eds.), Handbook of child psychology (pp. 99–166). New York, NY: John Wiley.Find this resource:

Rothbart, M. K., & Hwang, J. (2005). Temperament and the development of competence and motivation. In A. J. Eliot & C. S. Dweck (Eds.), Handbook of competence and motivation (pp. 167–184). New York, NY: Guilford Press.Find this resource:

Rubin, K. H., Burgess, K. B., & Hastings, D. D. (2002). Stability and social behavioral consequences of toddlers’ inhibited temperament and parenting behaviors. Child Development, 73, 483–495.Find this resource:

Saetre, P., Lindberg, J., Leonard, J. A., Olsson, K., Pettersson, U., Ellegren, H., … Jazin E. (2004). From wild wolf to domestic dog: Gene expression changes in the brain. Brain Research Molecular Brain Research, 126, 198–206.Find this resource:

Samochowiec, J., Hajduk, A., Samochowiec, A., Horodnicki, J., Stepie, G., Grzywacz, A., & Kucharska-Mazur, J. (2004). Association studies of MAO-A, COMT, and 5-HTT gene polymorphisms in patients with anxiety disorders of the phobic spectrum. Psychiatry Research, 128, 21–26.Find this resource:

Sanders, S. K. (2001). Cardiovascular and behavioral effects of GABA manipulation in the region of the anterior basolateral amygdala of rats. Dissertation Abstracts International, Section B, Sciences and Engineering, 62, 1060.Find this resource:

Schmidt, L. A., Fox, N. A., Schulkin, J., & Gold, P. W. (1999). Behavioral and psychophysiological correlates of self-presentation in temperamentally shy children. Developmental Psychobiology, 35, 119–135.Find this resource:

Schmidt, L. A., Miskovic, V., Boyle, M. H., & Saigal, S. (2008). Shyness and timidity in young adults who were born at extremely low birth weight. Pediatrics, 122, 181–187.Find this resource:

Schmitt, J. E., Lenroot, R. K., Wallace, G. L., Ordaz, S., Taylor, K. N., Kabani, N., … Giedd, J. N. (2008). Identification of genetically mediated cortical networks. Cerebral Cortex, 18, 1737–1747.Find this resource:

Schultz, W. (2006). Reward and addiction. In S. T. Fiske, A. E. Kazdin, & D. A. Schacter (Eds.), Annual Review of Psychology (Vol. 57, pp. 87–116). Palo Alto, CA: Annual Reviews.Find this resource:

Schwartz, C. E., Kunwar, P. S., Greve, D. N., Moran, L. R., Viner, J. C., Covino, J. M., … Wallace, S. R. (2010). Differences in adult orbital and ventromedial prefrontal cortex predicted by infant temperament at 4 months of age. Archives of General Psychiatry, 67, 1–7.Find this resource:

Schwartz, C. E., Kunwar, P. S., Greve, D. N., Kagan, J., Snidman, N. C., & Bloch, R. B. (2012). A phenotype of early infancy predicts reactivity of the amygdala in male adults. Molecular Psychiatry, 17, 1042–1050.Find this resource:

Schwarz, N. (1999). Self-reports. American Psychologist, 54, 93–105.Find this resource:

Simpson, E. E. A., McConville, C., Rae, G., O’Connor, J. M., Stewart-Knox, B., Coudray, C., & Strain, J. J. (2008). Salivary cortisol, stress and mood in healthy older adults: The Zenith study. Biological Psychology, 78, 1–28.Find this resource:

Smoller, J. W., Yamaki, L. H., Fagerness, J. A., Biederman, J., Racette, S., Laird, N. M., … Sklar P. B. (2005). The corticotropin-releasing hormone gene and behavioral inhibition in children at risk for panic disorder. Biological Psychiatry, 15, 1485–1492.Find this resource:

Spielman, R. S., Bastone, L. A., Burdick, J. T., Morley, M., Ewens, W. J., & Cheung, V. G. (2007). Common genetic variants account for differences in gene expression among ethnic groups. Nature Genetics, 39, 226–231.Find this resource:

Stark, R., Schienle, A., Girod, C., Walter, B., Kirsch, P., Blecker, C., Ott, U., Schafer, A., Zimmerman M., & Vital D. (2005). Erotic and disgust inducing pictures—differences in the hemodynamic responses of the brain. Biological Psychology, 70, 19–29.Find this resource:

Stefanacci, L., & Amaral, D. G. (2002). Some observations on cortical input for the macaque amygdala. Journal of Comparative Neurology, 451, 301–323.Find this resource:

Stifter, C. A., Putnam, S., & Jahromi, L. (2008). Exuberant and inhibited toddlers: Stability of temperament and risk for problem behavior. Development and Psychopathology, 20, 401–421.Find this resource:

Strange, B. A., Hurlemann, R., Duggins, A., Heinz, H. J., & Dolan, R. J. (2005). Dissociating intentional learning from relative novelty responses in the medial temporal lobe. NeuroImage, 25, 51–62.Find this resource:

Sundet, J. M., Skre, I., Okkenhaug, J. J., & Tambs, K. (2003). Genetic and environmental causes of the interrelationships between self-reported fears. A study of a non-clinical sample of Norwegian identical twins and their families. Scandinavian Journal of Psychology, 44, 97–106.Find this resource:

Suzuki, A., Matsumoto, Y., Oshino, S., Kamata, M., Goto, K., & Otani K. (2008). Combination of the serotonin transporter and norepinephrine transporter gene promoter polymorphisms might influence harm avoidance and novelty seeking in healthy females. Neuroscience Letters, 439, 52–55.Find this resource:

Thomas, A., & Chess, S. (1977). Temperament and development. New York, NY: Brunner/Mazel.Find this resource:

Torres-Farton, C., Richter, H. G., Germain, A. M., Valenzuela, G. J., Campino, C., Rojas-Garcia, P., … Seron-Ferre, M. (2004). Maternal melatonin selectively inhibits cortisol production in the primate fetal adrenal gland. Journal of Physiology, 554, 841–856.Find this resource:

Torrey, E. F., Miller, J., Rawlings, R., & Yolken, R. H. (1997). Seasonality of births in schizophrenia and bipolar disorder: A review of the literature. Schizophrenia Research, 28, 1–38.Find this resource:

Trut, L. N. (1999). Early canid domestication. American Scientist, 87, 160–169.Find this resource:

Turgeon, S. M. (2008). Sex differences in children’s free drawings and their relationship to 2D:4D ratio. Personality and Individual Differences, 45, 527–532.Find this resource:

Tsai, J. C. (2007). Ideal affect: Cultural causes and behavioral consequences. Perspectives on Psychological Science, 2, 242–253.Find this resource:

Vercuil, B., Brosschot, J. F., & Thayer, J. F. (2007). Capturing worries in daily life. Behaviour Research and Therapy, 45, 1835–1844.Find this resource:

Wang, H., & Wessendorf, M. W. (2002). Mu- and delta-opioid receptor mRNAs are expressed in periaqueductral gray neurons projecting to the rostral ventromedial medulla. Neuroscience, 109, 619–634.Find this resource:

Wakamatsu, Y., Nomura, T., Osumi, T., & Suzuki, K. (2014). Comparative gene expression analyses reveal heterochrony for Sox9 expression in the cranial neural crest during marsupial development. Evolution & Development, 16, 197–206.Find this resource:

Weisz, J. R., Chaiyasit, W., Wiss, B., Eastman, K. L., & Jackson, E. W. (1995). A multidimensional study of problem behavior among Thai and American children in school. Child Development, 66, 402–415.Find this resource:

Weisz, J. R., Weiss, B., Suwanlert, S., & Chaiyasit, W. (2003). Syndromal structure of psychopathology in children of Thailand and the United States. Journal of Consulting and Clinical Psychology, 71, 375–385.Find this resource:

Witter, M. P., Wouterlood, F. G., Naber, P. A., & Van Haeften, T. (2000). Anatomical organization of the parahippocampal-hippocampal network. Annals of the New York Academy of Sciences, 911, 1–24.Find this resource:

Young, K. A., Bonkale, W. L., Holcomb, L. A., & Hicks, P. B. (2008). Major depression, 5HTTLPR gentotype, suicide and antidepressant influences on thalamic volume. British Journal of Psychiatry, 192, 285–289.Find this resource:

Zisapel, N. (2001). Melatonin-dopamine interactions: From basic neurochemistry to a clinical setting. Cell Molecular Neurobiology, 21, 605–616.Find this resource: