Good Reasons for Bad Feelings

by Randolph M. Nesse

  The problem with abused drugs is not that they arouse pleasure; the problem is that they increase desire. My colleague the psychologist Kent Berridge has shown that this “wanting” system tends to overwhelm and outlast the “liking” system, so that some chronic users desperately want drugs that no longer provide much pleasure.30 “Wanting” hardly begins to describe the tragedy of people in the trap of spending all their available time, effort, thought, and money getting and taking drugs, even though the high is no longer all that pleasurable.

  Why Are Some People Especially Vulnerable?

  Not everyone gets addicted. Some people can even use heroin recreationally, setting it aside when they need to. The variation in vulnerability results, as for so many other traits, mainly from genetic variations.31,32 It might seem that the alleles that make people vulnerable are defects, but they probably didn’t influence fitness much in environments where drugs weren’t available. They probably did, however, influence behavior. Finding out how should be a high priority.

  I suspect that people who are especially vulnerable to addiction may use different foraging strategies from other people. Higher sensitivity to rewards would make them more likely than others to go back to the same places where they found food before. People with brains less vulnerable to addiction would likely wander more widely. It would be well worth watching how children forage for raspberries. Do children from families prone to addiction forage differently from others? If they do, it should be possible to create computer games that would predict, better than any questionnaire, interview, or genetic test, which individuals are especially vulnerable to addiction.

  Different populations vary dramatically in substance use, thanks to cultural variations, especially prohibitions enforced by religious teachings and leaders. Within a population, however, people whose lives are going badly are more vulnerable.33 People who get little pleasure from their daily lives and those plagued by anxiety, low mood, or boredom find the pleasure from drug use more compelling. A huge body of literature describes how personality, experiencing trauma, poverty, and difficult life situations influence vulnerability to addiction.34 They combine with genetic variations to explain why some people are more vulnerable than others.

  Taming the Plague

  An evolutionary perspective is no better at suggesting new quick cures for addiction than any other perspective is. It does not even try to explain how drugs change brain mechanisms. It does, however, correct false ideas and provide suggestions for new studies. For public policy, the implications are discouraging. Criminalization and interdiction have filled prisons and corrupted governments in country after country. However, increasingly potent drugs that can be synthesized in any basement make controlling access increasingly impossible. Legalization seems like a good idea but causes more addiction. Our strongest defense is likely to be education, but scare stories make kids want to try drugs. Every child should learn that drugs take over the brain and turn some people into miserable zombies and that we have no way to tell who will get addicted the fastest. They should also learn that the high fades as addiction takes over.

  New treatments are desperately needed. Nora Volkow, the leader of the National Institute on Drug Abuse, describes how rapid progress in understanding the brain mechanisms that cause addiction should lead to new drugs to block those mechanisms.35 This offers a real way forward. The substance abuse epidemic is created by novel environments, but changing social environments is difficult and changing human nature is impossible. Solutions will more likely come from finding ways to change our brains.

  CHAPTER 14

  MINDS UNBALANCED ON FITNESS CLIFFS

  The superiority of the human brain to others . . . is a reason why mental disorders are certainly most conspicuous and probably most common in man . . . the longest chain of effectively operated neuronic chains, is . . . likely to perform a complicated type of behavior efficiently very close to the edge of an overload, [and] will give way in a serious and catastrophic way . . . very possibly amounting to insanity.

  —Norbert Wiener, Cybernetics; or Control and Communication in the Animal and the Machine1

  Schizophrenia, autism, and bipolar disorder are very different diseases. Schizophrenia is a cognitive collapse in which every event is imbued with excess personal meaning and the inability to separate inner life from outer life gives rise to hallucinations and delusions. Autism is manifested in early childhood by a lack of social connectedness and solitary preoccupation with repeated motions and nonsocial thinking. Bipolar disease is the product of a broken moodostat that causes alternating periods of depression and mania. These are dire diseases.

  Despite their differences, these diseases have overlapping features that make an evolutionary perspective especially useful. Each afflicts approximately 1 percent of populations worldwide. Each has milder forms that affect 2 to 5 percent of people. Vulnerability depends overwhelmingly on what genes a person has, but people with schizophrenia or autism have fewer children than other people. The evolutionary question is obvious: Why hasn’t natural selection eliminated the genetic variations causing these diseases?

  The evidence for genetic causes is strong. Genetic variations explain about 70 percent of the risk of bipolar disorder,2 80 percent of schizophrenia,3 and 50 percent of autism.4 Having a parent or sibling with one of these diseases increases the risk roughly tenfold.5,6,7 Having an identical twin with one of these disorders increases the risk to over 50 percent.8

  Because identical twins do not always have the same diagnosis, some conclude that an environmental factor must also be involved. However, studies of adopted children show that the family in which a child is raised has little influence on the risk. It is more likely that the differences between identical twins result from chance variations influencing brain development, such as which genes are turned on or turned off and when, and the wandering paths of neurons as they grow.

  I wish I had known sooner that these are genetic diseases. I recall trying to console a mother who was bereft because her son’s doctors would not allow her to visit him during a hospital stay of several months for psychosis. Worse, they said, her early relationship with him had contributed to his schizophrenia. Home videotapes of babies who later developed schizophrenia show parents behaving slightly differently toward them compared to their siblings. However, those differences are not because parental behavior causes schizophrenia; it is because babies with a predisposition to schizophrenia are already slightly different.9 My patient felt guilty and distraught, but at the time no one knew enough to tell her and her doctors firmly that parenting has nothing to do with causing schizophrenia.

  Autism was also blamed on parents, especially women viewed as intellectualized “refrigerator mothers.” I saw one such mother, an accomplished academic. She was indeed intellectual, but she was not cold; she was alternately hot with anger about being blamed for her son’s illness, then depressed and guilty because of thinking that it might be true. She had a touch of the social awkwardness that characterizes many relatives of people with autism, not surprising since half of her genes were the same as her son’s. Such incredibly wrong theories about these diseases caused untold harm. Thank goodness, we know more now and can avoid adding unwarranted guilt to the already huge burden borne by the parents of afflicted children.

  Dashed Hopes

  At the turn of the millennium, hope was high that the alleles causing these diseases would soon be found. The human genome had just been sequenced. Cheaper ways to get genetic data were coming online. All signs suggested that the genetic causes would be found soon. Different versions of scores of genes had been investigated as suspected causes of schizophrenia at a cost of about $250 million. However, the first really large studies revealed that those early candidate genes were all innocent.10,11 Whole careers were spent chasing statistical will-o’-the-wisps.

  The next stage was to examine the entire genome instead of specific genes. Researchers looked at
markers spread over all twenty-three chromosomes to see if variations at certain locations were more common in people with these diseases than they were in other people. Every bit of the genome has been scoured. The conclusion is firm: there are no common genetic variations that substantially increase the risk of schizophrenia, autism, or bipolar disease.12,13 Some increase the risk, but almost all by less than 1 percent. All the identified loci that influence the risk of schizophrenia together explain only 5 percent of the variation.14 Furthermore, the alleles that increase the risk of schizophrenia also increase the risk of bipolar disorder.15

  The disappointment has been crushing. Imagine being one of the scientists who spent years in the lab studying genetic variations that cause these diseases and then discovering that they were all just statistical flukes. We thought we would find specific genetic flaws causing specific diseases. What we found is organic complexity beyond our imagining. It is as if archaeologists using radar sensors were sure they saw a new Rosetta stone deep in a pyramid, but when they finally got to it in person, the first flashlight beam revealed that it was just a pile of sand.

  Some genetic disorders are caused by specific genetic mutations with big effects. Huntington’s chorea (aka Woody Guthrie’s disease) is a good example: if you have the allele, you get the disease. For diseases such as cystic fibrosis caused by recessive genes, you get the disease if you have two copies of a defective allele. However, most common genetic diseases are very different. Instead of a few identifiable genetic variations with big effects, they are caused by thousands of variations spread across the genome, each with only a tiny effect. This is the case not only for schizophrenia, autism, and bipolar disease but also for type II diabetes, high blood pressure, coronary artery disease, migraine headaches, and obesity.

  The inability to find specific alleles that cause genetic diseases is called the problem of “missing heritability.”16,17,18 The heritability is not really missing; solid studies document strong effects of genes. What is missing is identification of the specific alleles that account for the heritability. If variations in who gets schizophrenia result mostly from genetic variations, why is it so hard to find the specific responsible alleles?

  One possibility is that the responsible variations are rare and so can’t be found even if they have big effects. Some very rare variations in the number of copies of a gene do increase the risk of major mental disorders by five times or more. However, even they don’t cause disease reliably on their own. For autism, only 5 percent of the heritable variation arises from rare mutations.19 Also, rare variations are unlikely to account for very many cases. One study that looked at both common genetic variations in genes and rare variations in the numbers of copies of genes influencing schizophrenia found that each identified variation explained almost exactly the same proportion of the overall variation in risk: 0.04 percent, that is, 4 parts out of 10,000.20 Hardly any. How curious that they all have the same effect size, albeit tiny.

  Despite the challenges, the search for genetic causes of mental disorders is nonetheless proceeding fast. The current hope is that combinations of genes will be found to have potent effects that identify the brain circuits causing the problems. Such advances could emerge even before this book is published; that would be wonderful. However, it is now seeming likely that what we have been looking for is not there. A leader in this research area, Kenneth Kendler, says, “The most pessimistic prediction that we will observe only a mess is unlikely. But discovering a highly coherent single pathway to illness also seems improbable . . . despite our wishing so, individual gene variants of large effect appear to have a small to non-existent role in the etiology of major psychiatric disorders.”21 In hindsight, this should not have been surprising; natural selection tends to eliminate alleles that cause dire illnesses.

  The mystery of missing heritability is becoming a bit less mysterious. New studies show that even though specific alleles have tiny influences, most effects can be explained by very complex interactions of many alleles.22 However, no combination of three or ten specific alleles causes disease reliably. Instead, disease risk is influenced by variations in thousands of genes with tiny effects interacting with one another and the environment. One recent report discovered that the number of genetic variations on each chromosome that increase the risk for schizophrenia is proportional to the size of the chromosome.23 They are spread randomly over the entire genome like tiny beads on twenty-three strings, with more beads on the longer strings. Another important finding is that most of the alleles influencing the risk of schizophrenia also influence the risk of bipolar disorder.24

  The discovery that the risk of schizophrenia and autism increases with the age of the father but not the mother suggested that new mutations were responsible. This is because a woman’s lifetime supply of eggs is formed about the time of birth but sperm are formed continuously by multiple cell divisions prone to errors.25,26 However, new studies show that the risk comes not from the age a father had a child but from the age at which he had his first child.27,28 Men who start families late in life are different from other men in ways that increase the risk of schizophrenia in their children. Nonetheless, 75 percent of new mutations come from fathers, and this accounts for 10 to 20 percent of cases of schizophrenia in the offspring of older fathers.

  The mountain of new facts is revealing the limits of the standard schema. A mechanic’s model assumes that the brain is constructed of discrete circuits with specific functions. It assumes specific disorders that can be defined by specific identifiable brain pathology with specific genetic causes. It assumes that normal brains are made by normal genomes and abnormal brains are products of abnormal genes. But many alleles that influence risk are not abnormal, and many of them influence multiple disorders. A deeper evolutionary perspective suggests the need to embrace the reality of organic complexity and look for causes not just in mechanisms but in trade-offs that cause intrinsic vulnerabilities.

  Instead of just looking for broken parts and the causes of disease in some individuals, we can also ask why all members of our species are vulnerable. Many US readers will have listened to Car Talk, the wonderful public radio show in which Click and Clack, the Tappet Brothers, diagnosed automotive problems with their thick Boston accents and uproarious laughter.

  Callers describe arcane problems. Sally in Dallas has an MG that won’t start again after being driven on a hot day. Click and Clack first diagnose her car’s problem: vapor lock. They then describe the mechanism that causes the problem: the fuel pump can move only liquid, so when the gasoline in a hot fuel line vaporizes, the car won’t start again until it cools down. Click and Clack then go into engineering mode. They describe the design flaw that makes the problem common for that specific car model: MGs from that year have a fuel line near the hot exhaust manifold, so they are especially prone to vapor lock. Finally they explain why vapor lock is an inherent problem for all cars with carburetors. When I was a teenager, I listened to an automotive engineer neighbor talk about his work trying to prevent vapor lock. I said, “Well, if it is just too much heat, that should be easy to solve.” He said, “Oh, yeah? It is hot above an engine. Where would you put the fuel pump and carburetor?” After spending years creating strategies to minimize the problem, he had little patience for a kid who did not understand that vapor lock is an intrinsic vulnerability with no easy solution.

  Could schizophrenia, autism, and bipolar disorder result from similar intrinsic vulnerabilities in human minds? If so, genetic variations that influence risk are related to the cause only in the same distant way that different auto models differ in their vulnerability to vapor lock. An evolutionary approach suggests looking for inherent constraints in the brain’s information-processing systems.

  The Evolutionary Genetics of Dire Disorders

  People with schizophrenia or autism have far fewer offspring than unaffected siblings, with the reduction stronger for men than for women.29,30 Sisters of affected indivi
duals tend to have very slightly increased numbers of children, perhaps in compensation, but brothers have fewer.31 Selection against these disorders should be strong.

  The most probable evolutionary explanation is that there are limits to what selection can do. In an influential review paper Matthew Keller and Geoffrey Miller discounted the role of environmental mismatch and expressed skepticism about the idea that alleles for mental disorders give advantages.32 They concluded that the most plausible explanation is that new mutations are continually created and only slowly selected out. That is certainly correct and a major cause of mental disorders. They went on to suggest that the brain is especially vulnerable because so many genes are involved in constructing it. This is more questionable. Even more genes influence height, but height abnormalities are uncommon. Machines malfunction if even one part fails, but bodies usually work just fine despite many mutations and minor bits of damage.

  A model based only on mutation implies that some combination of normal genes could prevent disease completely while maximizing fitness. However, this may be incorrect. Vulnerability to dire disorders could persist even if all mutations could be eliminated. Several possibilities deserve consideration.

  One is the intriguing idea, developed by the evolutionary biologist Bernard Crespi and colleagues, that schizophrenia and autism are genetic flip sides of the same coin and that they result from genes that benefit their own transmission despite the cost to the host.33,34 The logic is based on an observation by Robert Trivers, further developed by the Harvard biologist David Haig, that chemical tags placed on chromosomes early in development inhibit the expression of certain genes.35 The relevance of this imprinting process for obesity was described in chapter 13. Imprinting can also turn off genes selectively, depending on whether they come from the mother or from the father.36