Behave: The Biology of Humans at Our Best and Worst

by Robert M. Sapolsky

  In the 1950s the British psychiatrist John Bowlby challenged the view of infants as simple organisms with few emotional needs; his “attachment theory” birthed our modern view of the mother-infant bond.*24 In his trilogy Attachment and Loss, Bowlby summarized the no-brainer answers we’d give today to the question “What do children need from their mothers?”: love, warmth, affection, responsiveness, stimulation, consistency, reliability. What is produced in their absence? Anxious, depressed, and/or poorly attached adults.*

  Bowlby inspired one of the most iconic experiments in psychology’s history, by Harry Harlow of the University of Wisconsin; it destroyed Freudian and behaviorist dogma about mother-infant bonding.25 Harlow would raise an infant rhesus monkey without a mother but with two “surrogates” instead. Both were made of a chicken-wire tube approximating a torso, with a monkey-ish plastic head on top. One surrogate had a bottle of milk coming from its “torso.” The other had terry cloth wrapped around the torso. In other words, one gave calories, the other a poignant approximation of a mother monkey’s fur. Freud and B. F. Skinner would have wrestled over access to chicken-wire mom. But infant monkeys chose the terry-cloth mom.* “Man cannot live by milk alone. Love is an emotion that does not need to be bottle- or spoon-fed,” wrote Harlow.

  Evidence for the most basic need provided by a mother comes from a controversial quarter. Starting in the 1990s, crime rates plummeted across the United States. Why? For liberals the answer was the thriving economy. For conservatives it was the larger budgets for policing, expanded prisons, and three-strikes sentencing laws. Meanwhile, a partial explanation was provided by legal scholar John Donohue of Stanford and economist Steven Levitt of the University of Chicago—it was the legalization of abortions. The authors’ state-by-state analysis of the liberalization of abortion laws and the demographics of the crime drop showed that when abortions become readily available in an area, rates of crime by young adults decline about twenty years later. Surprise—this was highly controversial, but it makes perfect, depressing sense to me. What majorly predicts a life of crime? Being born to a mother who, if she could, would have chosen that you not be. What’s the most basic thing provided by a mother? Knowing that she is happy that you exist.*26

  Harlow also helped demonstrate a cornerstone of this book, namely what mothers (and later peers) provide as children grow. To do so, he performed some of the most inflammatory research in psychology’s history. This involved raising infant monkeys in isolation, absent mother or peers; they spent the first months, even years, of their lives without contact with another living being, before being placed in a social group.*

  Predictably, they’d be wrecks. Some would sit alone, clutching themselves, rocking “autistically.” Others would be markedly inappropriate in their hierarchical or sexual behaviors.

  There was something interesting. It wasn’t that these ex-isolates did behaviors wrong—they didn’t aggressively display like an ostrich, make the sexually solicitive gestures of a gecko. Behaviors were normal but occurred at the wrong time and place—say, giving subordination gestures to pipsqueaks half their size, threatening alphas they should cower before. Mothers and peers don’t teach the motoric features of fixed action patterns; those are hardwired. They teach when, where, and to whom—the appropriate context for those behaviors. They give the first lessons about when touching someone’s arm or pulling a trigger can be among the best or worst of our behaviors.

  I saw a striking example of this among the baboons that I study in Kenya, when both a high-ranking and a low-ranking female gave birth to daughters the same week. The former’s kid hit every developmental landmark earlier than the other, the playing field already unlevel. When the infants were a few weeks old, they nearly had their first interaction. Daughter of subordinate mom spotted daughter of dominant one, toddled over to say hello. And as she got near, her low-ranking mother grabbed her by the tail and pulled her back.

  This was her first lesson about her place in that world. “You see her? She’s much higher ranking than you, so you don’t just go and hang with her. If she’s around, you sit still and avoid eye contact and hope she doesn’t take whatever you’re eating.” Amazingly, in twenty years those two infants would be old ladies, sitting in the savanna, still displaying the rank asymmetries they learned that morning.

  ANY KIND OF MOTHER IN A STORM

  Harlow provided another important lesson, thanks to another study painful to contemplate. Infant monkeys were raised with chicken-wire surrogates with air jets in the middle of their torsos. When an infant clung, she’d receive an aversive blast of air. What would a behaviorist predict that the monkey would do when faced with such punishment? Flee. But, as in the world of abused children and battered partners, infants held harder.

  Why do we often become attached to a source of negative reinforcement, seek solace when distressed from the cause of that distress? Why do we ever love the wrong person, get abused, and return for more?

  Psychological insights abound. Because of poor self-esteem, believing you’ll never do better. Or a codependent conviction that it’s your calling to change the person. Maybe you identify with your oppressor, or have decided it’s your fault and the abuser is justified, so they seem less irrational and terrifying. These are valid and can have huge explanatory and therapeutic power. But work by Regina Sullivan of NYU demonstrates bits of this phenomenon miles from the human psyche.

  Sullivan would condition rat pups to associate a neutral odor with a shock.27 If a pup that had been conditioned at ten days of age or older (“older pups”) was exposed to that odor, logical things happened—amygdala activation, glucocorticoid secretion, and avoidance of the odor. But do the same to a younger pup and none of that would occur; remarkably, the pup would be attracted to the odor.

  Why? There is an interesting wrinkle related to stress in newborns. Rodent fetuses are perfectly capable of secreting glucocorticoids. But within hours of birth, the adrenal glands atrophy dramatically, becoming barely able to secrete glucocorticoids. This “stress hyporesponsive period” (SHRP) wanes over the coming weeks.28

  What is the SHRP about? Glucocorticoids have so many adverse effects on brain development (stay tuned) that the SHRP represents a gamble—“I won’t secrete glucocorticoids in response to stress, so that I develop optimally; if something stressful happens, Mom will handle it for me.” Accordingly, deprive infant rats of their mothers, and within hours their adrenals expand and regain the ability to secrete plenty of glucocorticoids.

  During the SHRP infants seem to use a further rule: “If Mom is around (and I thus don’t secrete glucocorticoids), I should get attached to any strong stimulus. It couldn’t be bad for me; Mom wouldn’t allow that.” As evidence, inject glucocorticoids into the amygdalae of young pups during the conditioning, and the amygdalae would activate and the pups would develop an aversion to the odor. Conversely, block glucocorticoid secretion in older pups during conditioning, and they’d become attracted to the odor. Or condition them with their mother present, and they wouldn’t secrete glucocorticoids and would develop an attraction. In other words, in young rats even aversive things are reinforcing in Mom’s presence, even if Mom is the source of the aversive stimuli. As Sullivan and colleagues wrote, “attachment [by such an infant] to the caretaker has evolved to ensure that the infant forms a bond to that caregiver regardless of the quality of care received.” Any kind of mother in a storm.

  If this applies to humans, it helps explain why individuals abused as kids are as adults prone toward relationships in which they are abused by their partner.29 But what about the flip side? Why is it that about 33 percent of adults who were abused as children become abusers themselves?

  Again, useful psychological insights abound, built around identification with the abuser and rationalizing away the terror: “I love my kids, but I smack them around when they need it. My father did that to me, so he could have loved me too.” But once again something b
iologically deeper also occurs—infant monkeys abused by their mothers are more likely to become abusive mothers.30

  DIFFERENT ROUTES TO THE SAME PLACE

  I anticipated that, with mothers now covered, we’d next examine the adult consequences of, say, paternal deprivation, or childhood poverty, or exposure to violence or natural disasters. And there’d be the same question—what specific biological changes did each cause in children that increased the odds of specific adult behaviors?

  But this plan didn’t work—the similarities of effects of these varied traumas are greater than the differences. Sure, there are specific links (e.g., childhood exposure to domestic violence makes adult antisocial violence more likely than does childhood exposure to hurricanes). But they all converge sufficiently that I will group them together, as is done in the field, as examples of “childhood adversity.”

  Basically, childhood adversity increases the odds of an adult having (a) depression, anxiety, and/or substance abuse; (b) impaired cognitive capabilities, particularly related to frontocortical function; (c) impaired impulse control and emotion regulation; (d) antisocial behavior, including violence; and (e) relationships that replicate the adversities of childhood (e.g., staying with an abusive partner).31 And despite that, some individuals endure miserable childhoods just fine. More on this to come.

  We’ll now examine the biological links between childhood adversity and increased risk of these adult outcomes.

  THE BIOLOGICAL PROFILE

  All these forms of adversity are obviously stressful and cause abnormalities in stress physiology. Across numerous species, major early-life stressors produce both kids and adults with elevated levels of glucocorticoids (along with CRH and ACTH, the hypothalamic and pituitary hormones that regulate glucocorticoid release) and hyperactivity of the sympathetic nervous system.32 Basal glucocorticoid levels are elevated—the stress response is always somewhat activated—and there is delayed recovery back to baseline after a stressor. Michael Meaney of McGill University has shown how early-life stress permanently blunts the ability of the brain to rein in glucocorticoid secretion.

  As covered in chapter 4, marinating the brain in excess glucocorticoids, particularly during development, adversely effects cognition, impulse control, empathy, and so on.33 There is impaired hippocampal-dependent learning in adulthood. For example, abused children who develop PTSD have decreased volume of the hippocampus in adulthood. Stanford psychiatrist Victor Carrion has shown decreased hippocampal growth within months of the abuse. As a likely cause, glucocorticoids decrease hippocampal production of the growth factor BDNF (brain-derived neurotrophic factor).

  So childhood adversity impairs learning and memory. Crucially, it also impairs maturation and function of the frontal cortex; again, glucocorticoids, via inhibiting BDNF, are likely culprits.

  The connection between childhood adversity and frontocortical maturation pertains to childhood poverty. Work by Martha Farah of the University of Pennsylvania, Tom Boyce of UCSF, and others demonstrates something outrageous: By age five, the lower a child’s socioeconomic status, on the average, the (a) higher the basal glucocorticoid levels and/or the more reactive the glucocorticoid stress response, (b) the thinner the frontal cortex and the lower its metabolism, and (c) the poorer the frontal function concerning working memory, emotion regulation, impulse control, and executive decision making; moreover, to achieve equivalent frontal regulation, lower-SES kids must activate more frontal cortex than do higher-SES kids. In addition, childhood poverty impairs maturation of the corpus callosum, a bundle of axonal fibers connecting the two hemispheres and integrating their function. This is so wrong—foolishly pick a poor family to be born into, and by kindergarten, the odds of your succeeding at life’s marshmallow tests are already stacked against you.34

  Considerable research focuses on how poverty “gets under the skin.” Some mechanisms are human specific—if you’re poor, you’re more likely to grow up near environmental toxins,*35 in a dangerous neighborhood with more liquor stores than markets selling produce; you’re less likely to attend a good school or have parents with time to read to you. Your community is likely to have poor social capital, and you, poor self-esteem. But part of the link reflects the corrosive effects of subordination in all hierarchical species. For example, having a low-ranking mother predicts elevated glucocorticoids in adulthood in baboons.36

  Thus, childhood adversity can atrophy and blunt the functioning of the hippocampus and frontal cortex. But it’s the opposite in the amygdala—lots of adversity and the amygdala becomes larger and hyperreactive. One consequence is increased risk of anxiety disorders; when coupled with the poor frontocortical development, it explains problems with emotion and behavior regulation, especially impulse control.37

  Childhood adversity accelerates amygdaloid maturation in a particular way. Normally, around adolescence the frontal cortex gains the ability to inhibit the amygdala, saying, “I wouldn’t do this if I were you.” But after childhood adversity, the amygdala develops the ability to inhibit the frontal cortex, saying, “I’m doing this and just try to stop me.”

  Childhood adversity also damages the dopamine system (with its role in reward, anticipation, and goal-directed behavior) in two ways.

  First, early adversity produces an adult organism more vulnerable to drug and alcohol addiction. The pathway to this vulnerability is probably threefold: (a) effects on the developing dopamine system; (b) the excessive adult exposure to glucocorticoids, which increases drug craving; (c) that poorly developed frontal cortex.38

  Childhood adversity also substantially increases the risk of adult depression. Depression’s defining symptom is anhedonia, the inability to feel, anticipate, or pursue pleasure. Chronic stress depletes the mesolimbic system of dopamine, generating anhedonia.* The link between childhood adversity and adult depression involves both organizational effects on the developing mesolimbic system and elevated adult glucocorticoid levels, which can deplete dopamine.39

  Childhood adversity increases depression risk via “second hit” scenarios—lowering thresholds so that adult stressors that people typically manage instead trigger depressive episodes. This vulnerability makes sense. Depression is fundamentally a pathological sense of loss of control (explaining the classic description of depression as “learned helplessness”). If a child experiences severe, uncontrollable adversity, the most fortunate conclusion in adulthood is “Those were terrible circumstances over which I had no control.” But when childhood traumas produce depression, there is cognitively distorted overgeneralization: “And life will always be uncontrollably awful.”

  TWO SIDE TOPICS

  So varied types of childhood adversity converge in producing similar adult problems. Nonetheless, two types of adversity should be considered separately.

  Observing Violence

  What happens when children observe domestic violence, warfare, a gang murder, a school massacre? For weeks afterward there is impaired concentration and impulse control. Witnessing gun violence doubles a child’s likelihood of serious violence within the succeeding two years. And adulthood brings the usual increased risks of depression, anxiety, and aggression. Consistent with that, violent criminals are more likely than nonviolent ones to have witnessed violence as kids.*40

  This fits our general picture of childhood adversity. A separate topic is the effects of media violence on kids.

  Endless studies have analyzed the effects of kids witnessing violence on TV, in movies, in the news, and in music videos, and both witnessing and participating in violent video games. A summary:

  Exposing children to a violent TV or film clip increases their odds of aggression soon after.41 Interestingly, the effect is stronger in girls (amid their having lower overall levels of aggression). Effects are stronger when kids are younger or when the violence is more realistic and/or is presented as heroic. Such exposure can make kids more accepting of aggression—in one study, watc
hing violent music videos increased adolescent girls’ acceptance of dating violence. The violence is key—aggression isn’t boosted by material that’s merely exciting, arousing, or frustrating.

  Heavy childhood exposure to media violence predicts higher levels of aggression in young adults of both sexes (“aggression” ranging from behavior in an experimental setting to violent criminality). The effect typically remains after controlling for total media-watching time, maltreatment or neglect, socioeconomic status, levels of neighborhood violence, parental education, psychiatric illness, and IQ. This is a reliable finding of large magnitude. The link between exposure to childhood media violence and increased adult aggression is stronger than the link between lead exposure and IQ, calcium intake and bone mass, or asbestos and laryngeal cancer.

  Two caveats: (a) there is no evidence that catastrophically violent individuals (e.g., mass shooters) are that way because of childhood exposure to violent media; (b) exposure does not remotely guarantee increased aggression—instead, effects are strongest on kids already prone toward violence. For them, exposure desensitizes and normalizes their own aggression.*

  Bullying

  Being bullied is mostly another garden-variety childhood adversity, with adult consequences on par with childhood maltreatment at home.42