The Fear Factor Read online

  As I would come to learn, the NIMH is a byzantine kind of place, especially for newcomers. Paperwork confirming my postdoc was probably wending its way through a labyrinth of cubicles somewhere on campus, but it hadn’t yet gotten a final stamp of approval. No matter. I had a computer, I had a desk, I eventually located James, and I was ready to get to work.

  My first goal was to publish my dissertation. In a series of five studies, I had found that sensitivity to fearful facial expressions is a reliable predictor of empathic concern in response to others’ distress. The first study found that participants who were best at recognizing fearful expressions offered more money and time to help Katie Banks, the distressed woman I had portrayed in the radio interview. The remaining studies found that volunteers who were best at recognizing fear also evaluated strangers’ physical appearances more kindly if they thought those strangers would receive the evaluations, and they expressed a greater desire to help distressed strangers described in short vignettes.

  James Blair’s research could be the key to understanding these odd findings. In 1995, James had published a novel hypothesis of psychopathy. A hallmark of psychopaths is their frequent perpetration of proactive aggression—cold, purposeful aggression, be it physical, verbal, or social—aimed at achieving some goal. Criminals who kill people after robbing them so they can’t be identified later, or who use threats of violence to extort money, are often psychopaths. James proposed that the mechanism that prevents most of us from engaging in these sorts of behaviors—and that appears to be malfunctioning in psychopaths—is a system he termed the Violence Inhibition Mechanism, or VIM (later updated and renamed the Integrated Emotion Systems model).

  James developed the idea of the VIM based in part on the work of animal behavior experts like Konrad Lorenz and Irenäus Eibl-Eibesfeldt, who observed that, among group-living animals in the wild, conflicts over resources or status can be quelled before actual aggression erupts through the use of body postures and vocalizations that send specific signals. Take wolves, which are good analogs to humans for two reasons. First, the organization of their packs is not unlike small bands of prehistoric humans or modern hunter-gatherers. Both consist of smallish, mutually dependent, and interrelated groups of adults and their young who work cooperatively to defend their territory, care for juveniles, and hunt for food. Second, many behaviors that wolves use to communicate are familiar to us because they have been retained in wolves’ domesticated descendants—our dogs.

  If, during a walk in the woods, you were to encounter a wolf that approached you with its fur standing on end, its body stiff, and its tail and head held high, growling in a low tone, you would need no translator to tell you how much trouble you were in. The wolf would be telling you that loud and clear. Not because it saw you as prey, mind you—wolves don’t bother communicating like this with animals they are planning to eat. This wolf’s behavior is an intimidation display—one that is meant to be seen and that signals it views you as a competitor or a threat, perhaps because you wandered too near a kill or its pups. So what would you do?

  If you were you (a human), you’d be in a bad spot. Wolves are usually fearful of humans, but when they’re not, there is little that a lone, unarmed human can do to fend them off. Outrunning a wolf is impossible, as is, probably, winning a physical fight with a creature whose jaws can crush a moose femur. Your best bet would be to walk backward slowly, avoid eye contact, and pray.

  If you were another wolf, however, you’d have much better options. One would be to exploit the Violence Inhibition Mechanism. Through it, you might be able to make the approaching wolf not want to attack you anymore. You’d need to start by avoiding eye contact and crouching down low. No, even lower. Your goal would be to try to look a fraction of your actual size. Even better, you might roll on your back, fold your legs in close to your body, flatten your ears, and emit a few whimpers—the higher-pitched and more helpless sounding the better. If you really wanted to go for the gold, and if the wolf came close enough, you could try licking the bristly underside of its jaws or peeing on yourself. If you have owned a very submissive dog before, you have probably seen these behaviors play out. If not, it all might seem distasteful and counterintuitive to you. Why signal how weak and pitiful you are to an animal that is threatening you?

  Of course, you wouldn’t do any of these things if you were threatened by a nonsocial creature like a rattlesnake or a shark. Go ahead and try licking a rattlesnake and see where it gets you. But the social wolf is acutely attuned to signs that another member of its species is raising the equivalent of a white flag. In adopting postures and vocalizations that make itself look smaller and weaker, a wolf under attack can signal that it will not—cannot—contest its would-be attacker’s dominance and power, rendering an actual physical battle unnecessary. Weak and subordinate wolves display these cues all the time during conflicts with stronger, more dominant wolves, who largely inhibit their aggressive attacks in response. Such a system makes pack life better for everyone. Because they use body language and vocalizations to communicate their relative power and aggressive or non-aggressive intentions so effectively, wild wolves rarely resort to actual violence.

  We humans may not pee on ourselves or roll onto our backs to signal fear and submission to one another, but we do have other cues that serve similar purposes. Body postures, vocal cues, and facial expressions that communicate dominance and subordination are built into our communicative repertoires just as surely as they are in wolves. And as is true for wolves, fearful and submissive cues tend to make a person appear physically smaller and weaker. Fearful body movements are cringing and crouching—think a hunched posture with shoulders huddled and hands and arms drawn in close to the body to shrink the visual silhouette. Fearful vocalizations are high-pitched, like the cry of a small creature with a small and minimally resonant larynx. And fearful facial expressions convey vulnerability and powerlessness through a combination of round, widened eyes, raised and upwardly angled brows, and a grimace. These cues are designed to appease—to literally disarm a potential attacker. Imagine hitting someone who was whimpering and cringing and looking terrified in front of you. It’s hard to contemplate without feeling like a terrible person. Consider your potential for Violence Inhibited.

  James laid out how the Violence Inhibition Mechanism causes typically developing children to acquire an aversion to hurting others. Young children are almost always at least a little aggressive, with age two being the most violent year of a person’s life, statistically speaking. This, by the way, is another good argument against the idea that all aggression is learned. Most toddlers occasionally engage in reactive aggression, including punching, scratching, even biting, regardless of whether they have ever seen these behaviors modeled and without ever having been rewarded for them. Aggression is such a deeply rooted, primitive behavior that it doesn’t need to be learned. So, thanks to their violent propensities, young children eventually learn what happens when they hurt someone: that someone gets upset. When people get hurt, they cry or cringe or act otherwise distressed. And just as is true for wolves, these sorts of behaviors are fairly effective at terminating aggression in normal children. In one study conducted in the 1970s, researchers examined the behavior of young children forced to share a box full of gerbils—the elementary school equivalent of heroin—with another child. The researchers set the box down between each pair of children, said, “Go!” then hustled from the room. They were right to hustle, as 441 separate conflicts over the gerbils erupted during the study, which involved only 72 pairs of children.

  The outcome of each conflict was recorded by the researchers, who discovered that one of the best ways for a child in possession of the gerbils to keep another child from grabbing them away was to adopt what are called oblique brows—the raised and upwardly angled brows that are a key component of both fearful and sad expressions. They look like this: / . Oblique brows were a more effective gerbil retention strategy than any attempts to use logic (“
My turn!” “I have to feed him!”) or physical force. Just as is true for wolves, the best way to resist attacks by gerbil-crazed six-year-olds is to use appeasement—to make them not want to attack.

  The VIM has deterrent effects as well. As they develop and gain experience in social conflicts, children become able to predict in advance what sorts of behaviors will cause others to exhibit distress, and eventually they refrain from engaging in these behaviors with their peers. This is an essential part of the process of turning uncivilized toddlers into trustworthy members of their social group. The mechanism continues working throughout life. A recent study found that even during negotiations between adults, up to 12 percent more value can be accrued by a negotiator who expresses sadness as compared to anger or no emotion.

  At least this is how things are supposed to proceed, but unfortunately, in a small percentage of children, the VIM doesn’t work. Approximately 7 percent of children, or about one in fifteen, will qualify for a diagnosis of conduct disorder at some point during childhood. Children receive this diagnosis when they persistently engage in behaviors that are violent or cruel or otherwise violate the rights of others. The occasional schoolyard fight or squabble over gerbils doesn’t count as conduct disorder. Children with this disorder threaten, bully, steal, and vandalize. They may set fires or engage in forcible sex. They are genuinely problematic.

  Here are the full criteria for a conduct disorder diagnosis, according to the most recent Diagnostic and Statistical Manual of Mental Disorders (also known as DSM-5), published in 2013 by the American Psychiatric Association. Children with conduct disorder must have exhibited at least three of these fifteen criteria in the past year, with at least one criterion present in the past six months:

  Aggression to People and Animals

  1. Often bullies, threatens, or intimidates others.

  2. Often initiates physical fights.

  3. Has used a weapon that can cause serious physical harm to others (e.g., a bat, brick, broken bottle, knife, gun).

  4. Has been physically cruel to people.

  5. Has been physically cruel to animals.

  6. Has stolen while confronting a victim (e.g., mugging, purse snatching, extortion, armed robbery).

  7. Has forced someone into sexual activity.

  Destruction of Property

  8. Has deliberately engaged in fire setting with the intention of causing serious damage.

  9. Has deliberately destroyed others’ property (other than by fire setting).

  Deceitfulness or Theft

  10. Has broken into someone else’s house, building, or car.

  11. Often lies to obtain goods or favors or to avoid obligations (i.e., “cons” others).

  12. Has stolen items of nontrivial value without confronting a victim (e.g., shoplifting, but without breaking and entering; forgery).

  Serious Violations of Rules

  13. Often stays out at night despite parental prohibitions, beginning before age 13 years.

  14. Has run away from home overnight at least twice while living in the parental or parental surrogate home, or once without returning for a lengthy period.

  15. Is often truant from school, beginning before age 13 years.

  Obviously, any child who exhibits three or more of these behaviors is seriously troubled. But not all children with conduct disorder are troubled in the same way, or for the same reasons. Somewhere between half and two-thirds of children with conduct disorder exhibit primarily the reactive form of aggression. They are usually not deliberately cruel; rather, their fights and threats and destructiveness seem to be driven by fear or frustration. Importantly, they are also emotionally reactive after they act out. If they hurt someone or lose control, they may cry, wonder aloud what is wrong with them, or express unprompted remorse for what they have done. They seem genuinely sorry that their behavior may have hurt siblings or parents or friends they care about. These are the children for whom conduct disorder has most likely resulted from experiences of trauma or abuse, or perhaps an innately reactive or dysregulated temperament crossed with a moderately stressful environment. In these children, the Violence Inhibition Mechanism itself is probably intact, but its effects are sometimes overwhelmed by stronger forces. Focusing on these other forces, by either ameliorating sources of stress or trauma in the child’s environment or treating symptoms of depression, anxiety, or lack of impulse control with medication or psychotherapy often causes conduct disorder symptoms to abate as well. This outcome suggests that these children’s conduct disorder is a secondary diagnosis that is itself being caused by something else.

  What about the other one-third to one-half of children with conduct disorder? For these 2 or 3 percent of all children, conduct disorder is not secondary to depression or anxiety. These children are, if anything, emotionally underreactive. Their aggression often isn’t accompanied by anger or upset—sometimes it seems to come out of nowhere, and to be purposefully cruel. Worse, it isn’t followed by any display of appropriate emotions like guilt or remorse in response to others’ distress. Being unruffled by signs of the distress that their cruel or violent behavior has caused others makes these children especially worrisome because it suggests that, for them, the Violence Inhibition Mechanism itself is impaired.

  One reason these children may not respond appropriately to others’ distress is that they have trouble recognizing others’ distress. Specifically, as James discovered, these children show a deficit that is the mirror image of what I had found in my own work: the least empathic children are also the worst at recognizing facial expressions of fear. When shown images of frightened faces like the ones I used in my graduate work, or when played recordings of frightened voices, they fail to recognize the faces and voices as fearful, and they fail to show the same empathic responses to them that healthy children show, such as an increase in sweat on the palms of the hands.

  And these are the children who are most at risk for becoming psychopaths.

  I vividly remember the day I met my first such child. In 2005, another research group at the NIH called to tell us that a boy enrolled in their protocol might be a better fit for ours. The other group initially thought that simple mood dysregulation might be his main problem because he had frequent temper tantrums. But mood dysregulation alone doesn’t produce tantrums like Dylan’s.*

  First, Dylan was twelve at the time, an age when, for most children, actual tantrums are in the distant past. Normally the preschool years are the ones that brew the most tantrums. And while a two-year-old’s tantrums may be annoying or frustrating, they are rarely significant problems. Now imagine an equally deranged tantrum produced by a boy who was five-foot-three and 120 pounds and able to reach every potential weapon in the house—knives, matches, baseball bats—and imagine it lasting an hour or more. Terrifying, right? That was Dylan. His bouts of rage would usually start over little nothings—irritation over not getting something he wanted or being punished for some misbehavior—and then quickly escalate, sometimes to red-faced screaming, other times to threatened or actual violence and destruction. During some of his worst episodes, he had threatened his parents with violence, punched or kicked holes in walls and doors, and, in one instance, smeared the walls of a room he’d been locked in with his own excrement. Once he threatened his mother with a knife. On that occasion and others, his mother brought his younger sisters to a relative’s house to spend the night for fear Dylan might actually follow through on his threats.

  The details of Dylan’s tantrums provided our first clue that something beyond simple mood dysregulation was afoot. When a person has a tantrum, it sometimes appears that he has completely lost control and might literally be capable of anything, so helpless is he to counter the forces of the emotional maelstrom. But this is an illusion. Tantrums can actually be experimentally induced in laboratory animals like cats and monkeys by stimulating a region of the brain called the medial hypothalamus. The old-school technique was to insert a tiny electrode into this small, evolut
ionarily ancient structure and turn on the current. When electricity is sent surging through the medial hypothalamus of a cat, it can send the cat into a snarling, spitting, clawing rage that looks for all the world like a toddler having a tantrum—but only if there is another living creature in the cat’s environment. Rage, even electrically generated rage, needs a target. The same phenomenon has been demonstrated more recently using a much more precise technique, called optogenetics, in which the DNA of neurons is manipulated to make them fire in response to pulses of light. Then a tiny light-emitting optical fiber is embedded within the brain. When it is turned on, it will cause nearby genetically altered neurons to fire. Optogenetic triggering of neurons within the medial hypothalamus of a mouse will similarly cause it to launch coordinated, rageful attacks against other mice, or even against a wiggly rubber glove. But if the cage is empty when the light pulses—no aggression.

  So even though a tree that falls in an empty forest will still make a noise, a child who trips and falls in an empty room will probably not throw a tantrum. What’s the point? Rage is an emotion designed to make things happen, usually by cowing someone else into submission (as the angry wolf did). This is probably why, among animals like monkeys that observe clear social hierarchies, electrically induced rage will not be directed at just any living target, but mainly at targets who are lower-ranking than the raging monkey. Higher-ranking monkeys, who are likely to be unimpressed by a subordinate’s rage attack, are usually spared. What does this mean? It means that even when an external machine is generating the rage attack—which would seem to be as uncontrollable as rage can get—the resulting behavior can still be modulated. The organism remains capable of maintaining some level of conscious or unconscious control over its behavior in accordance with basic biological rules—attack only living things, but not if they outrank you. This means that a child whose only problem is mood regulation will be very unlikely to go to extremes like threatening to stab his parents with a knife or smearing his own excrement on the walls, even in the grip of the most towering rage.