by Adrian Raine
Figure 8.3 Interaction between early biological risk factors and bad home environment in predisposing to teenage aggression in Australia
We again see the seeds of sin conspiring during infancy to create a deadly weapon in later years. Patty Brennan found a fivefold increase in adult violence when nicotine exposure was combined with exposure to delivery complications—but no increase in violence in those who were nicotine-exposed but lacking delivery complications.26 Maternal smoking also interacts with parental absence in predicting early onset of offending in the United States.27
We hear across all these studies a compelling chorus. Social factors interact with biological factors in predisposing someone to violence. As we discussed earlier, Caspi and Moffitt amazed the world in 2002 by establishing that a gene resulting in low levels of MAOA combined with severe early child abuse results in adult antisocial behavior.28 David Farrington, a world-leading criminologist at Cambridge University, found that low resting heart rate combined with parental separation before age ten resulted in voracious violent offending in adulthood.29 In the first-ever functional MRI study of any antisocial group, I found that violent offenders who suffered severe child abuse showed the greatest reduction in right temporal cortical functioning.30 Another study found that if you have high testosterone levels and a deviant peer group you may become conduct disordered—yet if you have that same high testosterone and circulate in a non-deviant peer group you are instead led to become a leader.31 Genes also combine with ghastly parenting to shape adolescent aggressive behavior.32 However you look at it, studies are showing that when biological and social factors interact, they can be far more malignant than any one factor on its own.
THE “SOCIAL-PUSH” PERSPECTIVE
I mentioned earlier that there are two ways of looking at biosocial effects. One is the “interaction” perspective. I’ve given several examples above. The second approach that I describe here I call the “social-push” perspective.
Back in 1977 it was unpopular to posit a biological basis to antisocial behavior in schoolchildren. Even less accepted was the belief that biological factors combined with social factors. So when my first research publication as a young student focused on this biosocial perspective, it was a virtual no-go area. Hans Eysenck, Britain’s best-known and most controversial psychologist, had already lit a fuse with his controversial book Crime and Personality,33 in which he had the audacity to suggest that crime had a biological basis. Despite the controversy, I believed the book contained a fascinating concept that was related but different—an “antisocialization process.” This concept profoundly influenced my work.
The idea was all but lost amid the acerbic criticisms others made of the book. It appears in a section that really resonated with me. Eysenck considered a child whose mother was a prostitute and whose father was a thief—a child in “Fagin’s kitchen.” He suggested that if that child “conditioned” well or learned quickly from his antisocial home role models he would become a good pickpocket—just like the Artful Dodger in Oliver Twist. In contrast, children who do not condition well will paradoxically not be socialized so easily into an antisocial way of life.34
I had my chance to examine this idea when I first learned psychophysiological techniques in the laboratory of my PhD supervisor, Peter Venables. That was at York University in 1977. I learned the fundamentals of the eccrine sweat-gland system. I scoured the classical-conditioning literature to design a fear-conditioning experiment. I studied what types of electrodes should be used, and the chemical content required of the gel that helped the silver/silver chloride electrode to make contact with the fingers. I learned to measure bias potentials on the electrodes and to rechloride them when the bias potential was unacceptable. I worked with our technician Don Spaven on generating the auditory stimuli to play over headphones in the conditioning experiment. I tested out the decibel levels with an artificial ear and a really expensive audiometer, snapping the connector between the two and making Don very upset. But soon after that setback, I was ready to get going on recruitment.
I interviewed school headmasters, met with the teachers, and put up recruitment flyers in schools. I went knocking on the doors of parents to get permission and recruited kids into the study. I chased up those who had not responded to my recruitment letter. I then went to the schools and gave the schoolkids questionnaires to fill out to assess their antisocial personality, and to get home background information. The teachers rated their antisocial behavior. I walked to school to pick up the kids, brought them over to the lab, and walked them back again when we had finished. It was a heck of a grind. But it was my first research study, and I was enormously excited—even in the autumn rain and the winter snow. The kids felt pretty good too because they got fifty pence for taking part in the study, about a week’s pocket money back in 1978.
We discussed fear conditioning earlier, so you’ll recall that it measures anticipatory fear. The task assesses how much a kid sweats when hearing a soft tone that predicts a loud, unpleasant tone. Can they learn—like Pavlov’s dogs—to form an association between two events in time? Can they learn that certain events are followed by punishment? Do they have a “conscience”—a set of classically conditioned emotional responses—that makes them feel uncomfortable even at the thought of doing something antisocial?
I found that the environment mattered. If the schoolkids came from a good home, then those who conditioned poorly were antisocial.35 Yet if they came from a bad home, the reverse was true—those who conditioned well were the antisocial ones, Dickens’s Artful Dodgers. I was really excited because I got these same findings no matter if it was the teachers rating the antisocial behavior or if it was the child self-reporting on his or her own antisocial personality. Findings were replicating across raters who often disagreed with each other, which suggested that the results were robust. The criminologist and historian Nicole Rafter very generously attributes my first finding as a classic study that got biosocial research in criminology under way,36 but the reality is that, like many scientists, I was standing on the shoulders of giants.37
Where does this lead us? I now want to introduce you to the second biosocial theme that I developed in that review in 2002.38 So far we’ve seen that when a biological risk factor interacts with a social risk factor, the outcome is an exponential increase in violence. But “moderation” is another way that social and biological factors can influence each other. A social process can “moderate”—or change—the relationship between biology and violence. That is exactly what the conditioning experiment had demonstrated—that home background moderates the relationship between fear conditioning and antisocial behavior.
Let’s take another example, this time from the PET-scan research on murderers that we discussed earlier. I had shown that murderers in general have poor prefrontal glucose metabolism.39 In another analysis, however, I divided the murderers into those from bad homes and those from relatively normal homes. We assessed eight different forms of home deprivation—factors like child abuse, severe family conflict, and extreme poverty. To get these data we scoured criminal transcript histories, medical reports, newspaper reports, and reports from psychiatrists, psychologists, and social workers. We even interviewed some of the defense attorneys. We then assigned murderers into either a “deprived” home background group or a “non-deprived” group. The question we then asked was, “Which group has the poor prefrontal functioning that predisposes them to violence?”
You can see the answer in Figure 8.4, in the color-plate section. We have here an example of a normal control on the left, who shows good prefrontal functioning—the red and yellow colors at the top. In the middle we have a murderer from a bad home background. And on the right we have a murderer from a good home. You can see that it’s the murderer from the relatively good home background who shows reduced frontal functioning—the cool colors at the top of the image. And that is the result we observed for the groups as a whole.40
The social environment mod
erates—or alters—the link between poor frontal functioning and murder. The bad brain–bad behavior relationship holds true for murderers from one type of home background—but not for those from a different home.
But how do we explain this? One way to think of it is like this: If you are a murderer, and you come from a bad home, what explains why you are violent? Perhaps here we don’t have to look any further than the bad home, which is a well-known social predisposition to violence.
But what if you are a murderer and you come from a good home? What causes violence here? It’s certainly not the home, because in this case it’s pretty good. Instead it has to be something else—a bad brain, perhaps. And that is indeed what we see in Figure 8.4 (in the color-plate section). Murderers from good homes had a 14.2 percent reduction in right orbitofrontal functioning—a brain area of particular relevance to violence. Accidental damage to this brain area in previously well-controlled adults is followed by personality and emotional changes that parallel criminal psychopathic behavior, or what Antonio Damasio has termed “acquired sociopathy.”41
Let’s think back to the case of Jeffrey Landrigan, which we discussed in chapter 2. He had a fabulous home background, with a loving mother, a father who was a geologist, and a sister who was as well educated and straitlaced as her parents. He had all the advantages of life. And yet Jeffrey swiftly spiraled out of control, beginning at age eleven with burglary, and eventually ending in homicide. What was the cause? Here we should suspect genetics and brain dysfunction, given that his biological father—whom he had never seen—was himself on death row for homicide. Great home—yet awful outcome. Gerald Stano was similarly adopted into a loving home six months after birth, but went on to confess to forty-one murders before facing the electric chair. Landrigan and Stano are just two among a number of serial killers reported on by Dr. Michael Stone, a forensic psychiatrist at Columbia University, who were adopted into warm, loving, and supportive home environments.42 Here we should suspect their genetic heritage, rather than bad homes, as a cause of their violence.
This social perspective on biology-violence relationships is not common in research. As we have seen, the “additive” effect of biological plus environmental risk is the prevailing outlook. And yet the alternative social-push perspective makes some sense, and I feel it can help some parents come to terms with the wayward behavior of their children.
Think about it yourself. Think about people who have a bit of the bad seed about them—a friend, a neighbor, or perhaps a family member who went off the deep end even though his or her siblings stayed on the straight and narrow. Sure, some of them come from classic chaotic homes filled with domestic violence and poverty. But don’t some of them have near-normal home backgrounds? Surprisingly loving parents? Two siblings can come from the same family—the same environment, the same upbringing—yet have different outcomes. Here I suggest that you should suspect subtle biological risk factors in nudging your acquaintance into crime, just as we have seen for murderers from good homes.
I often get e-mails from concerned parents desperately trying to help their wayward children. In one such message, a mother described how her seven-year-old son killed a household pet, struck out violently at her, and confessed to his therapist that he enjoyed choking his younger brother. When the mother became pregnant, the child began punching her in the belly and saying that he wanted the baby dead. He showed little remorse and his treatments, including counseling, medication, and hospital stays, did little to help.
Clearly this child is a serious problem, and equally clearly the mother really cares a great deal. Unlike the all-too-common scenario of parental neglect, she is desperately reaching out for help. Yet here it is the son who is callous, uncaring, and lacking remorse. Loving home—unloving child. What can account for such a tragic mismatch?
In this case it might be heritable process. Why? Because what I did not tell you earlier is that this child was adopted.
When children are adopted, it is often because the biological parents do not want their child, or their behavior is such that the child must be taken away from them. We saw earlier how maternal rejection of the child—especially in combination with biological risk factors like birth complications—is a risk factor for later violence. There is a break in the mother-infant bonding process at a critical period in the time before adoption, and it is not easy for a later loving home to mend that break. So here genetic processes may be accounting for the dangerous behavior shown in this child from a good home.
The emergence of genetic and biological factors for antisocial behavior in the midst of a benign home background is something I have termed the “social-push” hypothesis.43 Where an antisocial child lacks social factors that “push,” or predispose him to antisocial behavior, then biological factors may be the more likely explanation.44 In contrast, social causes of criminal behavior may be more important explanations of antisociality in those exposed to adverse early home conditions.45
This is not to say that antisocial children from adverse home backgrounds will never have biological risk factors for antisocial and violent behavior—they clearly will. Instead, the argument is that in such situations the link between antisocial behavior and biology is watered down because the social causes of crime can camouflage the biological contribution. Social causation will be more salient in children from adverse homes. In contrast, when the home is normal, but the child is not, then a bad brain may be the culprit. Here the social spotlight on violence is dimmed—and what now shines through is biology.46
So far I’ve illustrated the social-push hypothesis with respect to poorer frontal functioning in murderers from benign home backgrounds, and low fear conditioning in antisocial kids from poor homes. Yet this pattern of results has been found for a whole host of biological risk factors. As a graduate student I observed the social-moderation effect again soon after seeing the conditioning effect, finding that low resting heart rates particularly predispose schoolchildren from higher social class homes to antisocial behavior.47
More important, a number of other scientists have seen the same thing. Antisocial children from privileged middle-class backgrounds attending private schools in England have low resting heart rates.48 Antisocial English children from intact but not broken homes have lower heart rates.49 Low resting heart rate also characterized English criminals without a childhood history broken by parental absence and disharmony.50 In the Netherlands, Dutch “privileged” offenders—those from high-social-class homes who commit crimes of evasion—show blunted skin-conductance reactivity.51 In Mauritian children, reduced skin conductance responding to neutral tones at age three—a measure of “orienting,” or poor attention—is related to aggressive behavior at age eleven, but only in those from high-social-class backgrounds.52 Similarly in adults, English prisoners who are emotionally blunted and who come from intact homes—but not broken homes—show reduced skin-conductance orienting.53 Catherine Tuvblad, in Sweden, found that the environment moderates the link between genes and environment. As we might expect from what we learned about genetics in chapter 2, she found a genetic contribution to antisocial behavior in boys, but only those from a good home background.54
This same moderation effect has been observed at a molecular genetic level where abnormalities in genes related to the neurotransmitter dopamine55 are associated with early arrests, but only in adolescents from low-risk family environments—those who are socially better off. Again, genetic factors shine forth more in explaining antisocial behavior when social risk factors are less in evidence.
My student graduate Yu Gao also documented a moderating effect with the Iowa gambling task—a neurocognitive indicator of orbitofrontal functioning. Our colleagues Antoine Bechara and Antonio Damasio had demonstrated that patients with lesions to the ventromedial prefrontal cortex did poorly on this task and also showed psychopathic behavior.56 You’ll recall from chapter 5 that the orbitofrontal cortex is critical for generating somatic markers that inform good decision-m
aking and that it also facilitates good fear conditioning. This task was given to schoolchildren alongside assessments of psychopathic-like behavior.57 Gao found that kids who did poorly on the orbitofrontal gambling task were more likely to be psychopaths—but only when they came from normal home backgrounds.58 Just as I’d previously shown that poor fear conditioning predisposes children from good homes to antisocial behavior, so Gao took a measure of this same orbitofrontal cortex and showed the same moderating result.59
Moving from the lab to the real world, we can see the social-push hypothesis in cases of killers. Randy Kraft, the Scorecard Killer, had a very supportive and stable home background. Similarly, Jeffrey Landrigan had the best of home environments, yet went on to become a death-row inmate. Kip Kinkel, a teenager who killed his parents as well as two children at his high school, had a caring home environment in rural Oregon. His parents were devoted professionals, and he had a loving sister. We’ll see later the orbitofrontal dysfunction that contributed to his violence. You cannot pin the blame on poverty, bad neighborhoods, or child abuse all the time—certainly not in these cases. Nor is social deprivation so obvious in many more murderers who, while not exactly having heavenly homes as kids, did have homes not much different from yours and mine.
FROM GENES—TO BRAIN—TO VIOLENCE
Social factors interact with biological factors to increase a propensity for violence. They also moderate the relationship between biology and violence. There’s a third way to view the influence of the environment on biology, but before we peek into that window on the violent soul we need to step back briefly to genes, the brain, and behavior.