The Anatomy of Violence

by Adrian Raine

  In fact, to get something as strong as the heart rate–antisocial relationship, you have to turn to the effect of nicotine patches in reducing smoking, or the ability of SAT scores to predict later college GPAs. If we now turn to resting heart rate during a stressor, this seemingly innocuous biomarker suddenly explains 12 percent of the variation that exists among us in antisocial behavior. This is as strong as the ability of mammograms to detect breast cancer, the accuracy of home pregnancy test kits, and success of sleeping pills in improving chronic insomnia. It’s hard to ignore these medical relationships. It’s equally hard to ignore the relationship between heart rate and antisocial behavior. It’s clinically meaningful and significant.

  It is not that low heart rate characterizes only one subgroup of antisocial kids. It applies to young as well as older children, and to girls as well as boys. So boys with low heart rates are more antisocial than boys with high heart rates. Girls with low heart rates are more antisocial than girls with high heart rates.13

  However, heart rate may partly explain the gender differences in antisocial behavior. If you take your pulse using your watch, count the number of beats in one minute, and compare it with your opposite-sex sibling or partner’s pulse, you will likely find that if you are female, your heart rate is several beats a minute higher than your male counterpart’s. Males in general have lower heart rates than females; it’s a robust finding.14 There is the same sex difference in antisocial behavior. The sex difference in heart rate is in place as early as age three, with boys having a heart rate that is 6.1 beats a minute lower than girls.15 This sex difference in heart rate starts just before sex differences in antisocial behavior begin to emerge.16 The strong and replicated sex difference in heart rate provides one intriguing clue as to why men commit more crime than women—they have lower heart rates.

  Let’s shift from comparing genders to comparing generations. Twin studies have repeatedly found substantial heritability for resting heart rate.17 They have also found that the offspring of criminal parents have low resting heart rates.18 Given the fact that there is significant heritability for childhood aggression and adult antisocial behavior, and given that there is transmission of antisocial behavior from parent to child, low heart rate may be one of the heritable mechanisms that account for the transmission of antisocial behavior from one generation to the next.

  A lot of studies have measured heart rate and antisocial behavior concurrently, at the same point in time. But a stronger design would be to assess heart rate early in life—and then show that it’s related to antisocial behavior at a later age. That’s called a prospective longitudinal design. Five such longitudinal studies from England, New Zealand, and Mauritius have indeed confirmed that low heart rate in childhood—as early as age three—is a predictor of later delinquent, criminal, and violent behavior.

  Now, it’s important to note that these studies do not demonstrate causality, and nobody is arguing that we can tell who exactly in a classroom of kids is going to become antisocial on the basis of heart rate alone. But it’s a factor, and by teasing out the temporal ordering of the variables in question through research that follows young children into adulthood, we move one step further in support of the causal model that low heart rate early in life raises the odds of someone becoming a future offender.

  Could it be that social factors cause both crime and low heart rates—giving the false impression that low heart rates cause crime? David Farrington, of Cambridge University, one of the world’s leading criminologists, examined this issue in establishing the best independent early predictors of convictions for violence. He found that out of forty-eight predictors (family, socioeconomic position, academic attainment, and personality—everything from low social class to low IQ to impulsivity), only two were related to violence independent of all other risk factors: low resting heart rate and poor concentration.19 Indeed, low heart rate was even more strongly related to measures of violence than having a criminal parent—one of the best social predictors of later crime.20 These findings led Farrington to conclude that “low heart rate may be one of the most important explanatory factors for violence.”21

  Let’s look at this relationship coming from the other direction. While a low heart rate raises the odds that someone will become antisocial, a high heart rate actually reduces the odds of later crime. I conducted a study of English schoolboys who were antisocial at age fifteen but who desisted from adult crime at age twenty-nine. I then matched them against seventeen antisocial adolescents who had become criminal by age twenty-nine and also with seventeen non-antisocial, noncriminal controls. The ones who desisted from crime had significantly higher resting heart rates relative to both criminal and control groups, indicating that a high heart rate protects against adult crime.22

  On the treatment side, medications like stimulants that raise heart rate reduce antisocial behavior.23 Studies are also showing that heart rate may help predict which children will benefit from therapy—and which won’t. One study from Germany found that children who before treatment had low heart rates were less responsive to behavior therapy.24 Interventions may be more effective in antisocial children with normal or high heart rates in whom the causes of their antisocial behavior may be more environmental than genetic. Knowledge of resting heart rate may not just help predict which children are more at risk for later criminal behavior, it may also provide invaluable knowledge in treatment programs.

  Again on the medical side, one of the big problems is that it’s nearly impossible to find a biomarker that is diagnostically specific to just one psychiatric disorder. For example, there are many biological correlates of depression, but they are also found in patients with anxiety and other mental illnesses. An unusual and important feature of the low heart rate–antisocial relationship is its diagnostic specificity. While other psychiatric conditions, including alcoholism, depression, schizophrenia, and anxiety disorders, have if anything been linked to a higher resting heart rate, no psychiatric condition other than conduct disorder—i.e., antisocial and aggressive behavior—has been associated with a low resting heart rate.25

  The above studies have largely focused on violent criminals, psychopaths, and conduct-disordered children. But how much of a transgressor do you have to be to have a slower heart rate?

  I was pondering this issue on the sabbatical I took with my family at Hong Kong University. In Hong Kong it is rare for pedestrians to cross the road on a red light, even when the coast is clear. But there are always a few who do. Whenever I would take my boys out to the park, we’d inevitably come upon a crossing and they would see some adults breaking the rule. They would point at them and call them “naughty penguins”—after Pingu, a cartoon they watched about an adventurous but mischievous little penguin. So it occurred to me—do naughty penguins also have low resting heart rates?

  With the help of eight undergraduates, I collected heart-rate data on 622 Hong Kong students and asked them about their habits, including how many times they ever crossed the road on a red light. We found a difference. It was not big, just two beats a minute, but it was statistically significant and in the right direction. Naughty penguins really do have lower resting heart rates! Of course, this minor infraction is just the tip of the antisocial iceberg, but it indicates that low heart rate covers the whole spectrum of antisocial acts down to the smallest transgression.

  Taking all these points together, it’s hard to deny that a true, replicable relationship exists between low cardiovascular arousal and violence. When one line of scientific evidence supports a hypothesis, it is persuasive. But when many separate lines of evidence from different perspectives converge on the same conclusion, the argument becomes truly compelling.

  Indeed, this body of evidence has raised the intriguing possibility that low heart rate could be considered a biomarker for the diagnosis of conduct disorder.26 Currently, conduct disorder and almost all clinical disorders like schizophrenia are defined not in terms of biology, but in terms of symptoms that are
obtained in an interview with a clinical practitioner. So the clinical symptoms of conduct disorder are things like lying, stealing, fighting, and cruelty to animals. These are all behavioral in nature and rely on subjective verbal reports from caregivers of the children themselves. There are two good reasons that biomarkers are not included in psychological diagnoses. First, they are not found to be diagnostically specific—they apply also to other disorders. Second, in everyday practice it’s not that easy for a doctor to scan a patient to assess brain functioning—to say nothing of the extra financial burden scanning would present.

  Heart rate is different on both counts. It is diagnostically specific, and it is extremely cheap and quick to assess. Think of it yourself. What happens first when you go to your doctor’s office? You have your blood pressure and heart rate taken. Adding an objective biomarker to a subjective diagnosis is the holy grail that psychiatry and clinical psychology are searching for in all mental illnesses. Of course, not everyone with a low heart rate becomes a violent offender. My heart rate in my mid-twenties was 48 beats per minute, and the same will be true for a number of you. Yet at an admittedly imperfect level, low heart rate is a telltale sign of transgressors.

  GETTING THAT AROUSAL BOOST IN LIFE

  So low resting heart rate represents one of the best replicated, most easily measured, and most promising biological correlates of antisocial and aggressive behavior. But why does it predispose someone to antisocial behavior? Even with simple biological measures, unfolding the “mechanism of action”—how low heart rate produces antisocial and aggressive behavior—is highly complex. Let’s examine a few of the prevailing explanations.

  One is fearlessness theory.27 A low heart rate is thought to reflect a lack of fear.28 Although we talk about “resting” heart rate, the term is misleading. In research studies, subjects are brought into a novel environment, met by strangers, and have electrodes slapped on them. This is less like “resting” and more like experiencing a mild stressor. Timid, anxious children will have higher heart rates. Those lacking fear will have lower heart rates.

  As outlined above, there are some particularly fearless individuals such as bomb-disposal experts who function perfectly well in society and also have particularly low heart rates.29 After all, it takes nerves of steel to defuse a bomb. By the same token, antisocial and violent behavior requires a degree of fearlessness. If a boy lacks fear, he is more likely to get into a fight because he is not afraid of getting hurt. Similarly, punishments like prison do not motivate many offenders to desist from violence because this punishment does not hold fear for them.

  Fearlessness theory receives support from research showing that low heart rate provides the underpinning for a fearless, uninhibited temperament in infancy and childhood,30 and that the more uninhibited a preschooler is, the more aggressive he or she will be later in life.31 Adolescents with low heart rates are also better able to stand stress, indicating that such individuals are more insensitive to social stressors, including socializing punishments.32

  Another theoretical explanation of the low heart rate–antisocial behavior connection lies in empathy. Children with low heart rates are less empathic than children with high heart rates.33 Children who lack empathy are less able to put themselves into another person’s shoes and to imagine what it must feel like to be bullied and hit. Those with low empathy may be more aggressive because they have no concern for the feelings of others. Certainly children lacking empathy are more antisocial and aggressive.34

  Another explanation for how low heart rate produces antisocial and aggressive behavior is stimulation-seeking theory. This theory argues that low arousal represents an unpleasant physiological state, and that those who display antisocial behavior seek stimulation to increase their arousal levels to an optimal level.35 We all have an optimal level of arousal at which we can operate effectively and comfortably.36 Think of times you come back home and really need some stimulation—you turn on the TV, brew some coffee, turn up the music, get on your cell phone, or go out and party. You are bored and need a buzz. Yet there are other times you instead come home and leave the TV off, turn off your cell phone, and retreat into your own quiet space. The day has been too much and you’re over-aroused.

  The same need you have applies to kids with chronically low levels of arousal. Preschool boys with low heart rates not only are more antisocial and hyperactive, but they also choose to watch videotapes depicting intense anger more often than kids with more normal heart rates.37 In my own research, resting heart rate at the age of three characterizes both stimulation-seeking behavior at that same age38 and aggressive behavior at the age of eleven. Kids with chronically low levels of arousal may get an arousal boost in life by beating someone up, shoplifting, joining a gang, or getting involved in drugs. The harsh reality is that breaking any rule is fun for most kids—just think back to the days when you were a teenager. Living on the edge may not be what parents want for their teenagers, but for the kids themselves it’s exciting and gives meaning to life. Perhaps it’s not too surprising, then, that resting heart rate is at its lowest in life during adolescence, when stimulation-seeking39 and antisocial behavior are at their highest.40 And that craving for an arousal boost in adolescence may be part of the reason violence peaks in the late teenage years.

  If you have ever experienced this craving for stimulation, as I did when I was a kid, you get into a state of really just not knowing where to put yourself. You experience an intense feeling of restlessness and emptiness that can peak in a sense of agitation, and a real need to release some type of hard-to-describe, built-up tension. I have that feeling right now. You want to move around. Once you can find something to do in order to “shift gears,” you feel better.

  These feelings are exactly what a significant number of serial killers report experiencing prior to their homicides. The intense tension and restlessness. The need to go out in search of a victim. The consequent excitement of the abduction, torture, rape, and killing. And then the sense of relief and release from tension.

  Why would that be? I suspect it’s explained partly by having physiological under-arousal and a stimulation-seeking personality. The important message I really want to convey is a simple medical fact: low heart rate is a significant risk factor for antisocial behavior. Of course, it is not the only process within the autonomic nervous system that has gone awry in antisocial and violent individuals. To put that statement into a societal context, we need to take a trip to Mauritius.

  SHARED EARLY TEMPERAMENTS, DIVERGENT ADULT OUTCOMES

  Mauritius is one of the most beautiful tropical islands in the world and a destination for those seeking a luxury holiday with its consequent peace, quiet, and harmony. It’s also not a bad place to do research. In the past twenty-five years I have had to drop in on the island thirty-nine times. One could, of course, research violence in Detroit, but on balance I slightly prefer Mauritius. “It’s so delicious,” as the advertisements say along the road going from the airport to La Pirogue beach hotel, where I always stay. The sun, the palm trees, the beaches, the volcanic mountains, and some of the warmest and most gracious people I have ever met make for an exotic mix.

  Mauritius is a small island in the Indian Ocean near the Tropic of Capricorn, lying to the east of Madagascar. It extends thirty-eight miles from north to south, and twenty-nine miles from east to west. Part of the African continent, it is a multiracial democratic nation that gained independence from British rule in 1968 and became a republic within the Commonwealth of Nations in 1992. With a population of 1.28 million as of July 2009, it is the third most densely populated country in the world. At the initiation of our longitudinal study, in 1972, Mauritius was a developing country, but now it is developed and widely viewed as a model African country.

  Mauritius is also a wonderful melting pot of cultures, and the country is again notable in that ethnic tensions are rare. So where is the malevolence in Mauritius? Let’s put the previous idea of low arousal and stimulation-see
king into a research context that we undertook there.

  Why Mauritius, you might ask. Back in 1967, the WHO—World Health Organization—wanted to learn more about children who were at risk for the development of clinical disorders later in life. It recommended that a study should be conducted in a developing country, that the study should utilize three-year-old children, and that biological methods should be used to identify children at risk for later mental-health problems.41 Initially, the WHO had targeted India as a possible site,42 but a medical director from Mauritius successfully argued for the geographical advantage of his country. Mauritius was a small island with low emigration—factors that would permit subjects to be contacted more readily over time than in India.

  The Mauritius study was set up in 1972 by Peter Venables, from York University in England, and Sarnoff Mednick, from the University of Southern California. Peter was to become my PhD supervisor five years later, while Sarnoff would eventually lure me to the United States eleven years after that. I became the director of the study in 1987 when Peter retired. The sample was a birth cohort consisting of 1,795 three-year-old children all born in one of two towns—Vacoas and Quatre Bornes, both in the middle of the island and conveniently situated. The research laboratories were in Quatre Bornes.

  The study began like this. Families came to the research unit. Mothers sat down with their three-year-old children, and new toys were placed around them. Would the child leave the secure home base of his or her mother and explore the toys? At one extreme, some children would not leave and sat clinging to their mother—they were stimulation-avoiders. Some would come and go from their mother, using her as a “safety net” for exploration. Yet others would freely explore the toys and the new physical environment—the stimulation-seekers or explorers. Children were also placed in a sandbox and rated on their engagement in social play with other children. Their friendliness to the experimenter and their willingness to chitchat was also assessed. These four separate behavioral indicators formed a measure of stimulation-seeking.43