Scientists are leaning more and more toward the view that a gene specifically "for" a disorder or condition such as antisocial behavior is very unlikely (Rutter, Moffitt, and Caspi, 2006). Dean Hamer, director of the Gene Structure and Regulation Unit at the National Cancer Institute, writes, "Human behaviors, and the brain circuits that produce them, are undoubtedly the product of intricate networks involving hundreds to thousands of genes working in concert with multiple developmental and environmental events" (2002, p. 72).

Because there are so many elements involved and they interact with one another in such complex ways, it is extremely difficult to disentangle the influence of genes from the influence of the environment. To tease out these different strands and estimate their relative power, behavioral geneticists use twin studies, comparing identical twins (who share all their genes) and fraternal twins (who share about half their genes). The genetic influence on a characteristic such as intelligence, temperament, personality, cognitive style, or psychophysiology is greater when the trait is more similar in identical twins than in fraternal twins. Researchers also study adopted children to see if they are more like their biological parents (with whom they share genes) or their adoptive parents (with whom they share the environment). These studies, which are becoming increasingly analytical and sophisticated, show that antisocial behavior is moderately heritable (Moffitt, 2005), especially antisocial behavior that begins early in life (Arseneault et al., 2003; Rhee and Waldman, 2002).

In addition, scientists have discovered that some genes interact with a particular environment to actually produce a disorder (Rutter et al., 2006); some genes are expressed or turned on (or not) because of physical, social, and cultural factors in the environment; and some genes—for example, those that influence difficult temperament, impulsivity, novelty seeking, and lack of empathy—predispose people to be exposed to environmental risks. Genes even help shape the environment. Genes influence how parents bring up their children; genes affect the responses that children evoke from their families and the others around them; and, as children grow older, genes sway their choice of companions and surroundings (Caspi and Silva, 1995; Plomin, Owen, and McGuffin, 1994).

It's important to remember that heredity is not destiny. With the right environmental interventions at the right time, even a trait with a strong genetic foundation (such as antisocial behavior) can be altered.


Almost all experts agree that boys are at greater risk for physical aggression than girls (Underwood, 2003), and several large longitudinal studies bear this out (Broidy et al., 2003). Boys seem more susceptible to many of the risk factors for aggressive behavior—difficult temperament, ADHD, learning disabilities, and nervous system dysfunction, for example (Moffitt and Caspi, 2001; Rutter, Giller, and Hagell, 1998) —and the prevalence rate for aggression in boys is three to four times as high as the rate for girls (American Psychiatric Association, 1994). The social context has a strong influence on how and whether aggressive behavior appears (Maccoby, 2004). Parents treat anger and aggression differently in their sons and daughters and use more physical punishment and power-assertive discipline with their boys (Zahn-Waxler and Polanichka, 2004). Boys hit, push, kick, tease, and insult each other more; spend more time in rough-and-tumble play; and accept aggressive behavior more readily than girls. Physical aggression starts to taper off as boys get older and develop more impulse control (Maccoby, 2004).

Like boys who don't renounce physical aggression before they enter school, girls who continue to act aggressively face the prospect of school failure and rejection by their peers and are also more likely to be depressed (Underwood, 2003). They often join groups of boys, fight with boys, and eventually date—and marry—boys who act aggressively. Without the social and problem-solving skills they need to sustain an intimate relationship, they may find themselves in increasing danger as the boys grow bigger and stronger (Pepler and Craig, 1999). They are also more likely to become teenage mothers (Pepler and Craig, 1999). Although arrest statistics indicate that the rate of physical violence among girls is rising, other studies show that this is a myth: It is not the level of violence that has changed but the labeling of offenses (Chesney-Lind and Belknap, 2004).

As researchers turn their attention to aggressive behavior in girls, they are looking hard at indirect and relational aggression, where the goal is to damage another's self-esteem, social status, or both (Underwood, 2003). Covert tactics—exclusion, back-stabbing, gossiping, belittling, and the like—become more sophisticated and prevalent in middle childhood and are fairly widespread among girls during adolescence. Relational aggression provides a way for them to act on their angry feelings, seek revenge, and assure themselves that they're accepted by the group (Underwood, 2003).


In 1956, New York University psychiatrists Alexander Thomas and Stella Chess and their colleagues (Thomas, Chess, and Birch, 1968) began a pioneering longitudinal study of temperament. By collecting data on a sample of 133 children from infancy through young adulthood, they discovered that each child is born with a distinct temperament—an observable, constitutionally based pattern of behavior and emotions, a characteristic way of experiencing and interacting with the world.

Thomas and Chess identified nine traits that appear in different people in different combinations and quantities. These traits emerge early, often become stable by 4 months of age (Shonkoff and Phillips, 2000), and remain into adulthood, although by that time they may look entirely different from the outside (Caspi and Silva, 1995). This explains why a self-assured young woman who talks easily to strangers at a party may still regard herself as a shy person.

Thomas and Chess (Thomas et al., 1968) made another interesting observation: They found three distinct types of children, whom they classified as easy, slow to warm up, and difficult. The children in the easy group had a sunny outlook on life and adapted easily to change. Over the years, a mere 7.5 percent of them developed behavior problems (Chess and Thomas, 1984). The children who were slow to warm up took a long time to get used to new things, but with patient care they became interested and involved. About half of them had some problems with behavior (Chess and Thomas, 1984). The difficult group of children cried loudly and often, had tantrums, resisted new things and changes in routine, and always seemed to be in a bad mood. Although they made up just 10 percent of the study sample, about 70 percent of them developed problem behaviors (Chess and Thomas, 1984).

In trying to figure out why 30 percent of the children with difficult temperaments managed not to develop problems, Thomas and Chess (Thomas et al., 1968) evolved the concept of goodness of fit. Serious disturbances are more likely to arise, they found, when the temperament of the child and the expectations of the family or teacher are out of sync. Traits of temperament are neither good nor bad in themselves; what matters is how the environment responds to them.

The theorists who've followed Thomas and Chess—Jerome Kagan, Mary Rothbart, John Bates, and others—have started to pinpoint more precisely the temperamental dimensions associated with problem behavior. Perhaps the most important traits they've identified have to do with the way children experience and regulate their emotions (Frick, 2004).

Some children are easily unsettled and feel their negative emotions, such as anger and frustration, intensely and frequently (Lahey and Waldman, 2003). Untamed, this negative emotional reactivity can put them at risk of defiance, tantrums, and other problem behaviors; lead to peer rejection; and impair the development of cognitive skills such as social information processing. However, when negative emotional reactivity is balanced by a good dose of another temperamental quality, effortful control, or the ability to inhibit feelings and behavior and focus attention, the risk is attenuated (Rothbart and Jones, 1998).

At the opposite pole of this prickly emotional reactivity is a temperamental trait that also carries a high risk for behavior problems (Frick and Morris, 2004). When their reactions are low key and difficult to arouse, children seem to feel no fear and may actively seek out excitement, novelty, and danger. Those with an extreme version of this trait—variously called uninhibited (Kagan, 1998), daring (Lahey and Waldman, 2003), callous-unemotional (Frick and Morris, 2004), or low in harm avoidance (Tremblay, Pihl, Vitaro, and Dobkin, 1994)—aren't deterred by the threat of punishment or moved by others' distress; and they have trouble developing empathy, guilt, and moral reasoning (Frick and Morris, 2004). Their aggressive behavior is likely to be covert and instrumental.

However, like Thomas and Chess before them, the new temperament researchers do not believe that temperament is destiny. Families and teachers who understand and accommodate temperamental traits can gradually extend a child's capacity to cope—to learn to regulate her emotions, maintain relationships, develop empathy and guilt, and follow societal norms (Frick and Morris, 2004). Jerome Kagan (1998), a developmental psychologist who's studied inhibited and uninhibited temperamental traits in hundreds of babies, has found that only about one-quarter of them still show the same behavioral and biological profile at the age of 11 years (Kagan and Snidman, 2004). Daily experience helped the inhibited children learn to control their fear and irritability and the uninhibited to manage their disobedience, aggressive behavior, and impermeability to adult criticism—or not. Children with more extreme temperamental traits find such learning more difficult, which makes it harder to teach and care for them.

Although the environment can influence temperament, it is important to remember that biology comes first. On average, traits of temperament are about 50 percent heritable (Plomin et al., 1994). Kagan (1994; Kagan and Snidman, 2004) suspects that differences in neurochemical inheritance provide the basis for temperament traits, and he has found several physiological differences between inhibited and uninhibited children. For example, when uninhibited children are stressed, their heart rate is less likely to accelerate—a characteristic that has also been seen in older children with aggressive conduct disorders (Reiss and Roth, 1993).

Influenced by both biology and child-rearing, temperament varies with culture and geography (Kagan and Snidman, 2004). In 1974, medical student Marten deVries (1989) went to Kenya and Tanzania to collect information about the temperament of the children of the Masai, a semi-nomadic tribe on the Serengeti Plain. Using temperament scales based on Thomas and Chess's criteria, he identified 10 infants with easy temperaments and 10 with difficult temperaments. The area was in the midst of a severe drought, and when deVries returned five months later, most of the Masai's cattle and many of their people had died. Although deVries couldn't locate all of the babies, he found seven with easy temperaments and six with difficult temperaments. Of the seven with easy temperaments, five had died; all but one of those with difficult temperaments had survived.

What accounted for the unexpected survival of the children with difficult temperaments? DeVries credited several factors. First, the Masai admire their warriors and encourage aggressiveness and assertiveness in their children. Second, shared caregiving in the Masai's large extended families makes it easier to deal with children who are difficult. Third, Masai mothers breast-feed on demand, and children who are fussy ask for—and receive—more nourishment. The qualities that European American middle-class families regard as difficult—loud and frequent crying, for instance—are an advantage in an environment of scarcity (Chess and Thomas, 1989; DeVries, 1989).