91av

It’s late at night. You’re all alone, walking home past the dimly lit park. You feel a sudden chill when you see them. Teenagers. They snigger as you pass, probably about the old lady they mugged that morning while skipping school. Kids today…

From shopping malls banning youths in hoodies to media coverage of the latest teenage stabbing or shooting, there is growing public concern over crime and antisocial behaviour among young people. Politicians and media commentators wrangle endlessly over whether the cause is single-parent families, binge drinking, TV, or simply society going to the dogs, but no one seems to know what to do about it.

Now a new generation of science is illuminating the debate. Aided by advances in brain imaging and molecular genetics, we are starting to unravel what goes wrong when kids go bad – and to developing new methods to nip teenage crime in the bud.

Worry over bad behaviour in teenagers is nothing new. In their 1995 book , child psychiatrist Michael Rutter and criminologist David Smith documented a steady rise in youth crime following the second world war – averaging a fivefold increase from 1950 to 1990 across most developed countries.

More recent trends can be difficult to interpret, thanks to conflicting data sources and changes in the kinds of crimes recorded. The fact that the majority of crime goes unreported is also unhelpful. According to official figures, crime in the UK dropped between 1992 and 1999, but when government researchers asked young men and women about their criminal activity they found an increase in crime during the same period for 14 to 17-year-olds.

One thing that is clear is that crime is a largely teenage pursuit. Official statistics and government surveys from the US and UK suggest that the peak age for criminal activity is 17 to 18 for boys and 14 to 15 for girls (see Graph). But according to Terrie Moffitt at the Institute of Psychiatry in London, these figures only tell half the story. She argues that there are, in fact, two distinct types of teen criminal.

One very common type takes up petty crime in adolescence, drawn by the glamorous allure of badass friends. This group tends to outgrow crime by their early twenties. The other, more problematic, type starts showing signs of antisocial behaviour in kindergarten. These children, Moffitt says, have biological predispositions to behaviour problems which – if paired with the harsh environment of bad parenting, poverty or abuse – puts them at high risk of becoming lifelong violent criminals.

Much of Moffitt’s work in this area has been with the “Dunedin cohort” – 535 males and 502 females born in New Zealand in 1972 and 1973 who were signed up at birth to the Dunedin Multidisciplinary Health and Development Study, run by the University of Otago in Dunedin.

Combining the Dunedin data with other longitudinal studies, Moffitt found that both boys and girls who went on to become “lifelong” criminals were likely to have shown some kind of early-onset neurological impairment or difficult temperament early in life – as young as 3 years of age. These children tended to have low IQ, poor language skills, and especially a diagnosis of attention-deficit hyperactivity disorder. These problems are more common in boys than in girls, as is teen crime in general. This is also true of conduct disorder – a psychiatric disorder characterised by violence, cruelty to people and animals, and a tendency to break the rules.

Kids whose antisocial behaviours do not appear till later are certainly not to be ignored: , and they were more likely to use drugs and suffer mental illness. Generally speaking, though, it is the early-onset group that represents the greater problem. This group accounted for 10 per cent of the Dunedin males, but by age 26 they had accrued almost half of the violent convictions for the entire cohort.

So what’s different in kids who turn nasty early? Many of these cognitive and emotional traits are likely to have a strong genetic factor, says Moffitt. Indeed, decades of twin and adoption studies have shown that antisocial behaviour and criminality tend to run in families. In a 2005 literature review, Moffitt concluded that genetics account for about half of the variation in antisocial behaviour.

In the mix

Cognitive deficits that predispose kids to violent crime may also be acquired in early life through poor nutrition, birth complications or low birth weight. Minor physical anomalies such as low-seated ears or furrowed tongues – which may be connected to poor neural development or damage – may also be a sign.

Thankfully, biology is by no means destiny. In 1994 Patricia Brennan at Emory University in Atlanta, Georgia, and her colleagues studied the interaction of birth complications and parenting in a cohort of Danish men. They found that while complications increase the risk of later criminal behaviour somewhat, maternal rejection in early life greatly increased the probability of violent offending at 18. “It’s not enough to just have the biology – you need the biology and another factor, like a poor social environment or poor parenting,” says Brennan.

Now researchers are beginning to unravel how specific genes and the social environment interact to produce changes in the brain that influence emotional control and violence in kids. One of the most exciting advances came in 2002, when Moffitt’s group focused on the gene for the enzyme monoamine oxidase A. Variants of MAO-A have been linked with aggression in both animals and humans: the enzyme breaks down neurotransmitters including serotonin – a key molecule in the regulation of aggression. The team looked at the interaction between MAO-A, criminality and abuse as a child – a known risk factor for future antisocial behaviour ().

Using blood samples taken from the Dunedin males, Moffitt’s team examined their MAO-A variants, dividing them into high and low enzyme-activity groups. Boys were also grouped according to their experience of maltreatment in childhood – 8 per cent of the boys had been severely abused as children and 28 per cent were judged likely to have suffered some abuse.

They found that the variant of the MAO-A gene the boys carried did not, by itself, predict anything about future behaviour. But when boys had the low-activity variant MAO-A and had also been abused, the result was striking. These boys were three times as likely to be diagnosed with conduct disorder in adolescence and 10 times as likely to have been convicted of a violent crime in adulthood compared with boys of the same genotype who had not been abused. Boys with the high-activity variant seemed all but insulated from the effects of childhood abuse – in terms of antisocial behaviour – as there was little difference between abused boys and those who had not been abused.

One explanation for this interaction is that both early abuse and the gene for low-activity MAO-A seem to heighten emotional arousal and reactivity in children. In psychological tests in 2002, Seth Pollak at the University of Madison-Wisconsin showed that abused 9-year-olds were far quicker to spot anger and hostility in faces compared to kids who had not been maltreated. While this threat-primed outlook may help children adapt to a life filled with danger and deprivation, it may also lead to heightened levels of reactive or impulsive violence, as anxious children lash out to defend themselves against perceived hostility in others.

Similarly, in 2006, Andreas Meyer-Lindenberg of the National Institute of Mental Health (NIMH) in Bethesda, Maryland, used fMRI scans of the brain to show that in healthy people, those carrying the low-activity MAO-A enzyme had higher levels of activity in the amygdala – an area of the brain central to emotional processing, fear and the fight-or-flight response. When matching photographs of angry and fearful faces, and also when recalling negative memories, individuals with the low-activity gene showed both elevated levels of amygdala activity and lower activity in areas of the prefrontal cortex that regulate the amygdala’s response to such stimuli ().

While these results offer support for a combinatory effect of MAO-A variants and child abuse on the potential for kids to show reactive violence, co-author Ahmad Hariri at the University of Pittsburgh, Pennsylvania, warns that it is still a leap from brain imaging to real-world behaviour. “You are not going to be able to measure someone’s amygdala reactivity and then predict whether they are going to punch someone in the face on the playground when they get upset,” he says.

“You can’t measure brain activity and predict whether they will punch someone”

Another problem is that there is a subset of kids with “early-onset” antisocial behaviour who don’t react emotionally at all. In tests, this group scores high for callous and unemotional (CU) traits – a component of the diagnosis for adult psychopathy. These kids lack empathy and guilt, and also score highly for thrill-seeking, fearlessness and narcissism. They have normal IQ, are generally insensitive to punishment yet highly receptive to reward.

Emotional defects lie at the core of their antisocial tendencies, says James Blair, an expert in psychopathy at the NIMH. Without proper processing of threat cues, it’s hard to spot dangerous situations, hence the fearlessness. Without understanding pain in others, it’s hard to avoid being cruel. “During socialisation it’s critical to learn that some things lead to punishment, someone else’s fear or sadness, so you learn to avoid them.” But that type of learning seems impaired in psychopaths, says Blair. “So if they want something and punching someone in the face is the way to extract it, they might be more likely to engage in that kind of behaviour.”

This type of goal-oriented violence is characteristic of CU kids, says Paul Frick at the University of New Orleans in Louisiana. He estimates that about 30 per cent of early-onset antisocial kids show CU traits, and has found that that they are among the most likely to turn violent and stay that way.

Again, the amygdala, plus some areas of the prefrontal cortex, may be to blame. This time, however, the problem is not overactivity but the opposite. Brain imaging studies in adult psychopaths reveal diminished activity in the amygdala and areas of the prefrontal cortex in fearful situations and when asked to recall emotional memories. Research is ongoing to discover if these findings also apply to high-CU children.

Genetics plays an even bigger role in CU kids than in other antisocial behaviours. In a 2005 study of 7-year-old British twins, Essi Viding at the Institute of Psychiatry in London estimated that 67 per cent of the variation in CU traits can be explained genetically. Extreme antisocial behaviour in kids without CU traits had a heritability of just 30 per cent.

In fact, Blair says that CU traits have no known social causes and show a far weaker connection with poor parenting, or even abuse. This is not to say that social forces play no role. Poverty, for example may provide a motive for antisocial crime. “Psychopathy is just an emotional problem – it doesn’t push you to do anything unless there are obvious rewards to antisocial behaviour,” says Blair.

Understanding whether a child is hyperemotional, or callous and unemotional, allows psychologists to identify high-risk children early and to intervene with tailored treatments before the child gets tangled up in crime. “It is a huge difference from 20 years ago when it was just general therapy and counselling, then waiting a few months to see if they got better,” says Stephen Scott at the Institute of Psychiatry.

Improving parenting is among the main tools in this new approach to antisocial behaviour. In kids with a genetic predisposition to bad behaviour or social risk factors like poverty, good parenting can make all the difference. In 2001 Scott tested a three-month parenting class that taught the importance of praising good behaviour, and consistently and calmly punishing bad behaviour. Parents reported significantly fewer conduct problems and, even with no further instruction, the improvements remained a year later.

For CU kids the strategy is slightly different. In 2005, David Hawes and Mark Dadds at the University of New South Wales in Sydney, Australia, tested a similar parenting programme on boys aged 4 to 8, some of whom had CU traits. Punishments proved ineffective for the CU kids but using praise as a reward worked well, although CU boys improved less overall.

No matter what the intervention, the clear message from all researchers is “the earlier the better”. While programmes for teenagers have met with some success, manipulating behaviour in children under 8 years of age is clearly more effective.

Indeed, some of the most successful strategies begin even before kids are born. In 1998 David Olds at the University of Colorado reported on the 15-year follow-up of 315 children born between 1978 and 1980 in New York state. Mothers in the treatment group received a monthly home nurse visit throughout pregnancy and until the child’s second birthday, focusing on positive childcare, nutrition and the mother’s personal development. Among the poor and single mothers, the results were dramatic: at age 15, their kids had racked up only half the number of arrests and one-fifth the number of convictions. The children also smoked and drank less and had fewer sexual partners than those from untreated families.

These kinds of interventions don’t come cheap, but neither do the health effects of a life of crime. The latest report from the Dunedin cohort showed that at age 32, men in the persistent high-crime group were 10 times as likely to be smokers, 22 times as likely to be drug dependent, and 19 times as likely to be hospitalised for mental illness as the low-crime groups. They were also at greater risk of heart disease and stroke. Both men and women in the high-crime group were more likely to abuse their partners and the women were especially likely to hit their own children, potentially setting them along the same path. Now with the tools to spot problem kids early and give the right kind of help, we have the power to break the vicious circle.

Kids today… They’ve got to be worth it.

Teenagers – Learn more about a uniquely human phenomenon in our comprehensive special report.