The Criminal Mind. Advances in genetics and neuroscience are revolutionizing our understanding of violent behavior—as well as ideas about how to prevent and punish crime.
The Wall Street Journal, April 26, 2013
The scientific study of crime got its start on a cold, gray November morning in 1871, on the east coast of Italy. Cesare Lombroso, a psychiatrist and prison doctor at an asylum for the criminally insane, was performing a routine autopsy on an infamous Calabrian brigand named Giuseppe Villella. Lombroso found an unusual indentation at the base of Villella's skull. From this singular observation, he would go on to become the founding father of modern criminology.
Lombroso's controversial theory had two key points: that crime originated in large measure from deformities of the brain and that criminals were an evolutionary throwback to more primitive species. Criminals, he believed, could be identified on the basis of physical characteristics, such as a large jaw and a sloping forehead. Based on his measurements of such traits, Lombroso created an evolutionary hierarchy, with Northern Italians and Jews at the top and Southern Italians (like Villella), along with Bolivians and Peruvians, at the bottom.
These beliefs, based partly on pseudoscientific phrenological theories about the shape and size of the human head, flourished throughout Europe in the late 19th and early 20th centuries. Lombroso was Jewish and a celebrated intellectual in his day, but the theory he spawned turned out to be socially and scientifically disastrous, not least by encouraging early-20th-century ideas about which human beings were and were not fit to reproduce—or to live at all.
The racial side of Lombroso's theory fell into justifiable disrepute after the horrors of World War II, but his emphasis on physiology and brain traits has proved to be prescient. Modern-day scientists have now developed a far more compelling argument for the genetic and neurological components of criminal behavior. They have uncovered, quite literally, the anatomy of violence, at a time when many of us are preoccupied by the persistence of violent outrages in our midst.
The field of neurocriminology—using neuroscience to understand and prevent crime—is revolutionizing our understanding of what drives "bad" behavior. More than 100 studies of twins and adopted children have confirmed that about half of the variance in aggressive and antisocial behavior can be attributed to genetics. Other research has begun to pinpoint which specific genes promote such behavior.
Brain-imaging techniques are identifying physical deformations and functional abnormalities that predispose some individuals to violence. In one recent study, brain scans correctly predicted which inmates in a New Mexico prison were most likely to commit another crime after release. Nor is the story exclusively genetic: A poor environment can change the early brain and make for antisocial behavior later in life.
Most people are still deeply uncomfortable with the implications of neurocriminology. Conservatives worry that acknowledging biological risk factors for violence will result in a society that takes a soft approach to crime, holding no one accountable for his or her actions. Liberals abhor the potential use of biology to stigmatize ostensibly innocent individuals. Both sides fear any seeming effort to erode the idea of human agency and free will.
It is growing harder and harder, however, to avoid the mounting evidence. With each passing year, neurocriminology is winning new adherents, researchers and practitioners who understand its potential to transform our approach to both crime prevention and criminal justice.
The genetic basis of criminal behavior is now well established. Numerous studies have found that identical twins, who have all of their genes in common, are much more similar to each other in terms of crime and aggression than are fraternal twins, who share only 50% of their genes.
In a landmark 1984 study, my colleague Sarnoff Mednick found that children in Denmark who had been adopted from parents with a criminal record were more likely to become criminals in adulthood than were other adopted kids. The more offenses the biological parents had, the more likely it was that their offspring would be convicted of a crime. For biological parents who had no offenses, 13% of their sons had been convicted; for biological parents with three or more offenses, 25% of their sons had been convicted.
As for environmental factors that affect the young brain, lead is neurotoxic and particularly damages the prefrontal region, which regulates behavior. Measured lead levels in our bodies tend to peak at 21 months—an age when toddlers are apt to put their fingers into their mouths. Children generally pick up lead in soil that has been contaminated by air pollution and dumping.
Rising lead levels in the U.S. from 1950 through the 1970s neatly track increases in violence 20 years later, from the '70s through the '90s. (Violence peaks when individuals are in their late teens and early 20s.) As lead in the environment fell in the '70s and '80s—thanks in large part to the regulation of gasoline—violence fell correspondingly. No other single factor can account for both the inexplicable rise in violence in the U.S. until 1993 and the precipitous drop since then.
Lead isn't the only culprit. Other factors linked to higher aggression and violence in adulthood include smoking and drinking by the mother before birth, complications during birth and poor nutrition early in life.
Genetics and environment may work together to encourage violent behavior. One pioneering study in 2002 by Avshalom Caspi and Terrie Moffitt of Duke University genotyped over 1,000 individuals in a community in New Zealand and assessed their levels of antisocial behavior in adulthood. They found that a genotype conferring low levels of the enzyme monoamine oxidase A (MAOA), when combined with early child abuse, predisposed the individual to later antisocial behavior. Low MAOA has been linked to reduced volume in the amygdala—the emotional center of the brain—while physical child abuse can damage the frontal part of the brain, resulting in a double hit.
Brain-imaging studies have also documented impairments in offenders. Murderers, for instance, tend to have poorer functioning in the prefrontal cortex—the "guardian angel" that keeps the brakes on impulsive, disinhibited behavior and volatile emotions.
Of course, not everyone with a particular brain profile is a murderer—and not every offender fits the same mold. Those who plan their homicides, like serial killers, tend to have good prefrontal functioning. That makes sense, since they must be able to regulate their behavior carefully in order to escape detection for a long time.
So what explains coldblooded psychopathic behavior? About 1% of us are psychopaths—fearless antisocials who lack a conscience. In 2009, Yaling Yang, Robert Schug and I conducted structural brain scans on 27 psychopaths whom we had found in temporary-employment agencies in Los Angeles. All got high scores on the Psychopathy Checklist, the "gold standard" in the field, which assesses traits like lack of remorse, callousness and grandiosity. We found that, compared with 32 normal people in a control group, psychopaths had an 18% smaller amygdala, which is critical for emotions like fear and is part of the neural circuitry underlying moral decision-making. In subsequent research, Andrea Glenn and I found this same brain region to be significantly less active in psychopathic individuals when they contemplate moral issues. Psychopaths know at a cognitive level what is right and what is wrong, but they don't feel it.
What are the practical implications of all this evidence for the physical, genetic and environmental roots of violent behavior? What changes should be made in the criminal-justice system?
Let's start with two related questions: If early biological and genetic factors beyond the individual's control make some people more likely to become violent offenders than others, are these individuals fully blameworthy? And if they are not, how should they be punished?
Take the case of Donta Page, who in 1999 robbed a young woman in Denver named Peyton Tuthill, then raped her, slit her throat and killed her by plunging a kitchen knife into her chest. Mr. Page was found guilty of first-degree murder and was a prime candidate for the death penalty.
Working as an expert witness for Mr. Page's defense counsel, I brought him to a lab to assess his brain functioning. Scans revealed a distinct lack of activation in the ventral prefrontal cortex—the brain region that helps to regulate our emotions and control our impulses.
In testifying, I argued for a deep-rooted biosocial explanation for Mr. Page's violence. As his files documented, as a child he suffered from poor nutrition, severe parental neglect, sustained physical and sexual abuse, early head injuries, learning disabilities, poor cognitive functioning and lead exposure. He also had a family history of mental illness. By the age of 18, Mr. Page had been referred for psychological treatment 19 times, but he had never once received treatment. A three-judge panel ultimately decided not to have him executed, accepting our argument that a mix of biological and social factors mitigated Mr. Page's responsibility.
Mr. Page escaped the death penalty partly on the basis of brain pathology—a welcome result for those who believe that risk factors should partially exculpate socially disadvantaged offenders. But the neurocriminologist's sword is double-edged. Neurocriminology also might have told us that Mr. Page should never have been on the street in the first place. At the time he committed the murder, he had been out of prison for only four months. Sentenced to 20 years for robbery, he was released after serving just four years.
What if I had been asked to assess him just before he was released? I would have said exactly what I said in court when defending him. All the biosocial boxes were checked: He was at heightened risk for committing violence for reasons beyond his control. It wasn't exactly destiny, but he was much more likely to be impulsively violent than not.
This brings us to the second major change that may be wrought by neurocriminology: incorporating scientific evidence into decisions about which soon-to-be-released offenders are at the greatest risk for reoffending. Such risk assessment is currently based on factors like age, prior arrests and marital status. If we were to add biological and genetic information to the equation—along with recent statistical advances in forecasting—predictions about reoffending would become significantly more accurate.
In a 2013 study, Kent Kiehl of the University of New Mexico, looking at a population of 96 male offenders in the state's prison system, found that in the four years after their release, those with low activity in the anterior cingulate cortex—a brain area involved in regulating behavior—were twice as likely to commit another offense as those who had high activity in this region. Research soon to be published by Dustin Pardini of the University of Pittsburgh shows that men with a smaller amygdala are three times more likely to commit violence three years later.
Of course, if we can assess criminals for their propensity to reoffend, we can in theory assess any individual in society for his or her criminal propensity—making it possible to get ahead of the problem by stopping crime before it starts. Ultimately, we should try to reach a point where it is possible to deal with repeated acts of violence as a clinical disorder.
Randomized, controlled trials have clearly documented the efficacy of a host of medications—including stimulants, antipsychotics, antidepressants and mood stabilizers—in treating aggression in children and adolescents. Parents are understandably reluctant to have their children medicated for bad behavior, but when all else fails, treating children to stabilize their uncontrollable aggressive acts and to make them more amenable to psychological interventions is an attractive option.
Treatment doesn't have to be invasive. Randomized, controlled trials in England and the Netherlands have shown that a simple fix—omega-3 supplements in the diets of young offenders—reduces serious offending by about 35%. Studies have also found that early environmental enrichment—including better nutrition, physical exercise and cognitive stimulation—enhances later brain functioning in children and reduces adult crime.
Over the course of modern history, increasing scientific knowledge has given us deeper insights into epilepsy, psychosis and substance abuse, and has promoted a more humane perspective. Just as mental disorders were once viewed as a product of evil forces, the "evil" you see in violent offenders today may someday be reformulated as a symptom of a physiological disorder.
There is no question that neurocriminology puts us on difficult terrain, and some wish it didn't exist at all. How do we know that the bad old days of eugenics are truly over? Isn't research on the anatomy of violence a step toward a world where our fundamental human rights are lost?
We can avoid such dire outcomes. A more profound understanding of the early biological causes of violence can help us take a more empathetic, understanding and merciful approach toward both the victims of violence and the prisoners themselves. It would be a step forward in a process that should express the highest values of our civilization.
—Dr. Raine is the Richard Perry University Professor of Criminology, Psychiatry and Psychology at the University of Pennsylvania and author of "The Anatomy of Violence: The Biological Roots of Crime," to be published on April 30 by Pantheon, a division of Random House.