Randolph M. Nesse
Mental health research is in a crisis (Brüne et al., 2012). The predominant framework has assumed that mental disorders have specific brain causes of the sort that explain multiple sclerosis. However, thousands of reductionist researchers supported by billions of dollars have not found a specific cause, or even a reliable biomarker, for any of the major mental disorders. The evolutionary perspective that revolutionized understanding of animal behavior over a generation ago (Alcock, 2001) can help explain why the search has not succeeded, and it can suggest new approaches.
At the core of that revolution is recognition that a full explanation of any biological trait requires a description of its evolutionary history as well as its mechanism (Nesse, 2013; Tinbergen, 1963). For a disease, a full understanding requires knowing not only why some individuals get sick, but also why all members of the species have traits that are vulnerable to failure (Nesse, 2005b; Nesse & Williams, 1994; Williams & Nesse, 1991). Seeking such explanations has led to rapid developments in evolutionary medicine (Nesse et al., 2010; Stearns, 2012). They are now ready for application in psychiatry.
The core insight is that evolution can explain not only why most traits work so well, but also why many are vulnerable to failure. Attempts to explain why the mind is vulnerable have a long history. What is new is taking such questions seriously and posing and testing alternative evolutionary hypotheses. This chapter summarizes recent progress and the substantial challenges associated with seeking evolutionary explanations for mental disorders.
Early applications of ethology to mental disorders (McGuire & Fairbanks, 1977) gave rise to more specific and comprehensive evolutionary approaches (McGuire & Troisi, 1998; Nesse, 1984; Wenegrat, 1990). Several books cover specific conditions (Baron-Cohen, 1997; Gilbert, 1992; Horwitz & Wakefield, 2007, 2012; Wenegrat, 1995). The field finally has a textbook (Brüne, 2008) to augment the previous major treatment (McGuire & Troisi, 1998). Many articles address specific mental disorders, and some provide a new foundation for defining the categories that describe disorders (Cosmides & Tooby, 1999; Nesse & Stein, 2012; Wakefield, 1992). Several chapters in this Handbook and many general books about evolutionary psychology tackle specific disorders.
The diverse ideas in these sources can be summarized in eight fundamental contributions that an evolutionary perspective offers to psychiatry and clinical psychology (see Table 43.1). A brief summary of each sets the stage for considering specific disorders.
Table 43.1 Eight Contributions of an Evolutionary Perspective on Mental Disorders
|
The task of explaining why all humans are vulnerable to mental disorders is no different from that of explaining why we are vulnerable to other diseases. The tendency in both cases has been to attribute vulnerability to the limited powers of natural selection. These limits are important explanations for some diseases, but there are five other possible reasons why the body and mind are not better designed (Nesse, 2005b; Nesse & Williams, 1994; Williams & Nesse, 1991) (see Table 43.2). Brief summaries of each kind of explanation are an important prelude to their application to mental disorders.
Table 43.2 Six Evolutionary Explanations for Vulnerability
|
Mutations happen, and deleterious ones can persist despite natural selection. Also, there is no starting fresh with bodily designs, so substandard aspects abound. But this is not the main explanation for vulnerability to disease.
Most common chronic diseases are caused by novel environmental factors (Gluckman & Hanson, 2006). For instance, atherosclerosis and breast cancer are prevalent now because our bodies are not well-suited for life in a modern environment. Whether rates of mental disorders are increasing remains uncertain. An international effort to gather prevalence data on mental disorders from 72,000 interviews in 14 countries (Kessler & Ustun, 2000) uses urban or rural agricultural sites. No comparable effort has estimated disorder rates in hunter-gatherer populations. Technical difficulties make this understandable, but it is nonetheless unfortunate, because such studies may not be possible in the next generation.
Mental disorders are often blamed on the modern environment (Stevens & Price, 1996), and that is certainly correct to some extent, but it is hard to tell which disorders are really more common. Retrospective data suggested that depression rates were increasing rapidly with each generation (Cross-National Collaborative Group, 1992). However, data gathered using consistent questions in the same population over recent decades showed no such increase (J. M. Murphy, Laird, Monson, Sobol, & Leighton, 2000).
Natural selection cannot remove vulnerability to infection because pathogens evolve faster than we can (Ewald, 1994). Furthermore, the defenses that protect us, especially immune responses, tend to cause problems themselves. Some mental disorders may result from arms races with pathogens and their autoimmune sequelae. For instance, some cases of obsessive compulsive disorder may result from streptococcal-induced autoimmune damage to the caudate nucleus (Swedo, Leonard, & Kiessling, 1994). Prenatal exposure to infection may predispose to schizophrenia (Ledgerwood, Ewald, & Cochran, 2003). Infectious causes have been proposed for a wide range of mental disorders, especially affective disorders (Ewald, 2000), and the role of inflammation in depression is increasingly recognized (Raison & Miller, 2013).
Design trade-offs make perfection impossible for any trait, natural or human-made. We could have less anxiety, but only at the cost of being more likely to be injured or killed. We could be more trusting, but at the cost of being exploited.
A gene that decreases health, longevity, or happiness will nonetheless be selected for if it increases reproductive success (RS). Such genes are likely responsible for many of our least valued characteristics, such as bitter competition, envy, greed, and unquenchable sexual desire and jealousy (Buss, 2000). The differences between the sexes arise largely because different reproductive strategies shape different physical and mental traits, even at the expense of longevity and individual well-being (Cronin, 1991; Daly & Wilson, 1983; Geary, 1998; Kruger & Nesse, 2004).
As noted above, pain, cough, fever, and other protective responses are unpleasant but useful responses that protect us from danger and loss. The prevalent tendency to confuse these defenses with diseases and defects has been called “The Clinician's Illusion” (Nesse & Williams, 1994). Most physicians know that cough and inflammation are adaptations, but the utility of fever, diarrhea, and anxiety is less widely recognized. A naïve view sees our vulnerability to negative emotions as examples of poor design. But natural selection does not care a fig for our happiness; it just mindlessly shapes whatever emotional tendencies increase RS (Nesse, 1991a; Tooby & Cosmides, 1990). The smoke detector principle (discussed later) explains why many instances of negative emotions are excessive or unnecessary (Nesse, 2005c).
Some evolutionary approaches to mental disorders emphasize one of these six possible explanations, to the exclusion of others. For instance, some authors emphasize the effects of living in a modern environment (Glantz & Pearce, 1989; Stevens & Price, 1996); others emphasize infection (Raison & Miller, 2013) or constraints, tradeoffs, or path dependencies (Crow, 1997; Horrobin, 1998). Others propose that mental disorders persist because of fitness benefits, even for conditions such as schizophrenia (J. S. Allen & Sarich, 1988; Shaner, Miller, & Mintz, 2004), bipolar disorder (Wilson, 1998), and suicide (deCatanzaro, 1980). Emphasizing one cause to the exclusion of others causes substantial confusion.
This is only one of 10 mistakes that are common in attempts to find evolutionary explanations for diseases. See the text box for a list of others. They can be prevented by addressing 10 questions systematically when pursuing projects in evolutionary medicine (Nesse, 2011).
When a patient comes to the general medical clinic with a cough or kidney failure, the physician knows that a cough is a protective response and that the kidney regulates salt and water balance. By contrast, when a patient comes to a mental health clinic with a phobia, the utility of anxiety may never be considered. When someone comes with jealousy, consideration of its normal functions is unlikely. Mental health professionals lack knowledge about normal emotional functions comparable to the understanding physiology offers to general medicine. Evolutionary psychology is beginning to provide this missing body of knowledge, as shown by the chapters in this Handbook, and by evolutionary perspectives on motivation (French, Kamil, & Leger, 2000), emotion (Nesse, 1990a; Plutchik, 2003), and specific topics such as grief (Archer, 1999; Nesse, 2005a).
EP can bring information about an individual's idiosyncratic values, goals, and life situations into a scientific framework. Consider John, a depressed 20-year-old man who works two jobs in local stores to support his disabled mother. When he was 14, his dying father made him promise to take care of his mother always. He has been doing that ever since, but with increasing resentment and depression. These three sentences give more insight into his depression than a dozen demographic variables and a brain scan. An evolutionary understanding of motivation can bring such information into a nomothetic framework based on the behavioral ecology categories of life history effort: somatic, reproductive, and social.
EP's greatest contribution may be a deeper understanding of relationships. For instance, Bowlby's insights about the evolutionary functions of attachment (Bowlby, 1969) have been extended by suggestions that apparently “abnormal” kinds of attachment may represent alternative strategies for infants to get resources from their mothers in difficult circumstances (Belsky, 1999; Chisholm, 1996) and a deeper understanding of women's reproductive strategies in general (Hrdy, 1999). Analysis of mutually beneficial reciprocal exchanges has led to extensive studies of economic games (Fehr & Fischbacher, 2003) that illuminates the origins of the social emotions (Fessler & Haley, 2003; Fiske, 1992). However, interpreting all human relationships as calculated exchanges ignores aspects of human life that are essential to understanding mental disorders, such as our capacities for moral action, and the emotions of pride and guilt (Katz, 2000). Selection has shaped capacities for commitment that are sometimes superior to rational calculation (Frank, 1988; Gintis, 2000; Nesse, 2001a). People choose the best partners available. These selfish choices create powerful selection forces that shape prosocial capacities, including altruism and moral sensitivities (Nesse, 2010; Noë & Hammerstein, 1995).
Freud's theories are ridiculed because some are wrong and psychoanalytic treatment is not reliably effective. However, the reality of repression is a profound fact of human nature that needs an evolutionary explanation (Badcock, 1988; Nesse, 1990b; Sulloway, 1985). Trivers and Alexander separately suggested that self-deception is a strategy for deceiving others (Alexander, 1975; Trivers, 1976, 2011), but people also may repress the sins of others to preserve valuable relationships (Nesse, 1990b).
Closely related is Trivers's insight that regression may be a strategy used by offspring to manipulate their parents into providing resources that would be appropriate only if they were younger or sick (Trivers, 1974). His more general theory of parent-offspring conflict is the neglected foundation for understanding many childhood disorders (Trivers, 1974). Attempts to provide an evolutionary foundation for psychodynamics are developing (Badcock, 1988; Nesse & Lloyd, 1992; Slavin & Kriegman, 1992; Sulloway, 1985), but remain relatively unappreciated by psychoanalysts. This may be because an evolutionary view fosters skepticism that undermining repression will be helpful routinely (Slavin & Kriegman, 1990).
Developmental psychology now offers sophisticated assessments of extensive data about what children do at different stages of life, and how these phenomena vary across cultures. It increasingly considers evolution (P. P. G. Bateson & Martin, 2000; Geary & Bjorklund, 2000; Rutter & Rutter, 1993). In the midst of a burst of interest in facultative developmental mechanisms and their role in evolution (Hall, 1998; West-Eberhard, 2003), evolutionary psychologists have begun looking for mechanisms that use environment inputs to adjust developmental pathways (Del Giudice, Ellis, & Cicchetti, in press).
A possible adaptation with particular relevance for mental disorders is the adjustment of the gain in the hypothalamic pituitary axis system in response to early stress, and the transmission of this sensitivity across the generations by maternal influences on fetal brain development (Ellis & Del Giudice, 2014; Essex, Klein, Eunsuk, & Kalin, 2002; Teicher et al., 2003).
Advances have been rapid in applying evolutionary thinking to aspects of child development that contribute to psychopathology (Bjorklund & Pellegrini, 2002; Del Giudice et al., in press; Frankenhuis & Del Giudice, 2012; Narvaez, Panksepp, Schore, & Gleason, 2012). In particular, intensive work focuses on early effects, even in utero, that influence stress reactivity via genomic imprinting (Bateson et al., 2004; B. J. Ellis, Jackson, & Boyce, 2006; Gluckman et al., 2009; Meaney, 2010). An important debate is underway about how to distinguish epiphenomena from facultative adaptations that detect and transmit information that increases fitness for future generations.
Most mental disorders are emotional disorders. People come for treatment because they experience anxiety, depression, anger, or jealousy. Many assume that such negative emotions are abnormal, but they are useful, at least for our genes. People with depression and anxiety are so obviously impaired that it is hard to see how such emotions could be useful. However, the principles of signal detection theory explain why selection shapes emotion regulation mechanisms that often give rise to normal but useless suffering—the smoke detector principle (Nesse, 2005c). Emotions are now routinely recognized as special states shaped by selection to give advantages in fitness-significant situations that have recurred over evolutionary time (Ekman, 1992; Nesse, 1990a; Nesse & Ellsworth, 2009; Plutchik, 2003; Tooby & Cosmides, 1990), but this fundamental principle has yet to be incorporated into research on emotional disorders.
When is an emotion abnormal? The criteria for psychiatric diagnoses are based on symptom intensity, duration, and associated disability, with no consideration of context (American Psychiatric Association, 2013). The extremes are reliably abnormal, but without knowing the functions of emotions, the line between normal and abnormal remains subjective (D. Murphy & Stich, 2000; Nesse, 2001b; Nesse & Stein, 2012; Troisi & McGuire, 2002; Wakefield, 1992). The lack of an evolutionary foundation fosters serious errors, including describing continuous emotions as categories and neglecting abnormal conditions characterized by excess positive or deficient negative emotions. If general medicine ignored context the way psychiatric diagnosis does, it would diagnose “abnormal cough disorder” based on cough frequency and severity without considering possible causes. Far from genuinely atheoretical, the DSM system fosters a crude view of mental disorders that considers only mechanisms, while neglecting the other half of biology (Horwitz, 2002).
Wakefield (1992) offers a strong critique of the DSM, using the concept of “harmful dysfunction” to clarify what is and is not a disorder. This sophisticated evolutionary analysis of psychiatric diagnosis argues that it is essential for mental as well as physical disorders to separate normal from abnormal phenomena based whether they are harmful and whether they arise from a dysfunction (Wakefield, Chapter 42, this volume).
Most psychiatric research attempts to explain individual differences—why some people get sick while others do not. Despite growing agreement on the importance of gene-environment interactions (Kendler, Kuhn, & Prescott, 2004; Ridley, 2003; Rutter & Rutter, 1993), different authorities emphasize different causal factors (genetic, developmental, situational, etc.). Far from emphasizing genetic differences, an evolutionary view provides a framework that highlights the relationships among all factors and levels. It also contributes strategies for avoiding some simple mistakes.
Much misunderstanding arises from confusing attempts to explain the existence of a trait with attempts to explain variations in a trait. As Richard Lewontin has emphasized, it is senseless to ask whether the area of a rectangle is caused more by its length or its width. However, variation among rectangles can only arise from differences in width or length. Likewise, all traits result from genes interacting with environments, but variations among individuals can result only from differences in genes, differences in environments, and GxE interactions. The proportion of variance attributable to each component is, of course, not fixed, but varies depending on the particular environment and the range of genotypes. Preoccupation with nature versus nurture has distracted attention from the many different routes to a disorder. An evolutionary approach fosters simultaneous consideration of the many factors that may explain individual variation in a trait (see also Arslan & Penke, Chapter 45, this volume).
The task of accounting for individual differences should not be reduced to arguing about the relative importance of one factor compared to another. It is, instead, the challenge of explicating how each contributes to individual differences in a particular trait, and how their contributions to a particular trait may be different not only among families, populations, or cultures, but even among individuals. The responsible factors may be mainly genetic in one individual and mainly environmental in another. An individual can have one depression episode triggered by relationship failure and another triggered by inflammation. This has practical implications for mental health research. For instance, research that looks for “the cause” of depression is unlikely to succeed.
The fundamental principles of evolutionary biology prove their utility when applied to specific mental disorders. Following traditional categories, these principles apply to emotional, behavioral, and cognitive disorders.
Most mental disorders are emotional disorders, but diagnosis and treatment are not yet based on knowledge about the evolutionary origins and functions of emotions. Instead, intense or prolonged negative emotions are viewed as abnormal, irrespective of the situation, while deficits in negative emotions and excesses of positive emotions are rarely recognized as disorders. Progress will require correcting this fundamental error.
Although anxiety can be useful, a dry mouth and tremor when standing before a large group is worse than useless. Likewise, the symptoms of panic may help escape from a lion, but they are unhelpful in a grocery store. We now have vast knowledge about the responsible brain mechanisms, but no comparable body of knowledge about the evolutionary origins and utility of social anxiety or panic (Nesse, 1987). Similarly, hundreds of studies document every aspect of excessive anxiety states, but only a handful look for states of deficient anxiety, the hypophobias (Marks & Nesse, 1994). One study tried to confirm that fear of heights often results from severe falls early in life. It found adult fear of heights in 18% of the control group, but only 3% of the group that had experienced a fall early in life. Those with hypophobia early in life still had deficient anxiety decades later (Poulton, Davies, Menzies, Langley, & Silva, 1998).
Anxiety illustrates the diversity of the body's regulation mechanisms. For instance, rigid defensive responses to fixed cues, such as chicks hiding from hawk-shaped shadows, are useful when a correct response to the first encounter is essential, but they result in many false alarms and do not protect against novel dangers. Flexible learning mechanisms protect against novel dangers but may fail during a crucial initial exposure, and they are prone to result in phobias. Social learning is another solution. Infant rhesus monkeys show no innate fear of snakes; however, a single observation of another monkey displaying fear of snakes induces long-lasting avoidance. Watching another monkey display fear of a flower induces no such fear (Mineka, Keir, & Price, 1980).
Exposure treatment is effective for phobias, but it does not undo the learned fear response. Instead, new cortical processes suppress the fear response (Quirk, 2002). Exposure to danger disrupts this suppression. Thus, the great flood in Moscow caused the reemergence of previously extinguished fears in Pavlov's dogs. This may reflect a constraint in a path-dependent mechanism; simple unlearning may not be possible.
Other fears cannot be extinguished. For instance, post-traumatic stress disorder (PTSD) illustrates one-time learning of the strongest sort. Proximate science is steadily homing in on the mechanisms that account for this syndrome, but evolutionary hypotheses about the potential utility of PTSD are just being considered (Cantor, 2005). Nearly dying is so important to fitness that a learning mechanism that responds to a single experience to cause severe symptoms could nonetheless give an advantage.
Fear has distinct subtypes that seem to have been partially differentiated from generic anxiety to cope with domain-specific challenges (Marks & Nesse, 1994). For instance, panic flight is just the ticket to escape from a predator, but frozen immobility is superior when teetering on a cliff. Social anxiety is present in most people (Gilbert, 2001; Leary & Kowalski, 1995), and people who lack it are often insufferable, even if they do not qualify for a psychiatric diagnosis. One wonders how they would have fared in small hunter-gatherer groups.
The smoke detector principle helps to explain some apparent peculiarities of the mechanisms that regulate anxiety and other defenses (Nesse, 2005c). Because most anxiety responses are inexpensive and protect against huge potential harms, an optimal system will express many alarms that are unnecessary in the particular instance, but nonetheless perfectly normal. This suggests that using drugs to block defenses may be safe in most instances, but that in some situations blocking a defense may be fatal. A sophisticated formal mathematical treatment should help to advance research in this area (M. Bateson, Brilot, & Nettle, 2011). A major book takes an evolutionary view of when anxiety is a harmful dysfunction (Horwitz & Wakefield, 2012), and why it is so difficult for psychiatric researchers to recognize the utility of anxiety.
The utility of sadness and depression is less obvious than for anxiety, but Bibring long ago suggested that depression signaled the need to detach when libido persists in a connection to an unrewarding object (Bibring, 1953). Hamburg (Hamburg, Hamburg, & Barchas, 1975) and Klinger (Klinger, 1975) described how emotions regulate goal pursuit more generally, with inability to reach a goal first arousing aggressive attempts to overcome a frustrating obstacle, then low mood motivating disengagement. If the person does not give up, the ordinary low mood escalates into depression. This principle, now confirmed by much research (Brickman, 1987; Carver & Scheier, 1990; Carver & Scheier, 1998; Emmons & King, 1988; Janoff-Bulman & Brickman, 1982; Little, 1999; Wrosch, Scheier, & Miller, 2003), provides the foundation for a more general approach to mood as a mechanism that allocates effort proportional to propitiousness (Nesse, 1991b, 2000, 2009). When payoffs are high, positive mood increases initiative and risk-taking. When risks are substantial or effort is likely to be wasted, low mood blocks investments. In this perspective, ordinary episodes of sadness and low mood motivate changing behavioral strategies (Watson & Andrews, 2002). If no alternative is found and the goal is essential, persistence may result in depression (Klinger, 1975; Wrosch, Scheier, & Carver, 2003).
Observations of chickens and monkeys who lost their positions in the hierarchy has suggested a view of depression as “involuntary yielding” that protects against continuing attack (Gilbert, 1992; Sloman, Price, Gilbert, & Gardner, 1994). This is consistent with data showing that stressful events cause depression mainly if they are characterized by humiliation and/or being trapped in an impossible quest (Brown, Harris, & Hepworth, 1995; Kendler, Hettema, Butera, Gardner, & Prescott, 2003). Also related is the suggestion that sex differences in depression may arise from the male tendency to strive for position and resources, leaving many women vulnerable to depression because they have fewer options (Gilbert, 1992; Wenegrat, 1995). The role of defeat and entrapment observed in animal studies (Blanchard, Litvin, Pentkowski, & Blanchard, 2009), in human studies by Brown (Brown et al., 1995), inspired by Price (Price, Sloman, Gardner, Gilbert, & Rohde, 1994), has been extended by Gilbert (2006), and now is the focus of an extensive literature confirming that generic stress is not nearly as potently depressogenic as defeat and entrapment (Taylor, Gooding, Wood, & Tarrier, 2011).
Depression has also been viewed as a social manipulation (Hagen, 2002; Watson & Andrews, 2002). Hagen sees postpartum depression as a “blackmail threat” to abandon the infant, but other theories could also explain the association of postpartum depression with poor resources and relationships. In a related but more general view, Watson and Andrews suggest that depression facilitates “social navigation” by signaling that current strategies are failing and new directions are needed (Watson & Andrews, 2002). This approach echoes psychoanalyst Emmy Gut's work on productive and unproductive depression (Gut, 1989). Nettle (2004) notes inadequacies of the social navigation hypothesis, and emphasizes the possible adaptive value of neuroticism.
The role of rumination in solving social dilemmas is highlighted in a recent article that gained wide attention (Andrews & Thomson, 2009), but also criticism from those who view rumination as an unhelpful symptom with little evidence for utility and some evidence that tendencies to rumination increase the risk of depression (Nolen-Hoeksema & Morrow, 1993). Earlier work by Gut considers the possibility that such rumination can be productive or unproductive, depending on the circumstances (Gut, 1989).
DeCatanzarro proposed that suicide can be adaptive if an individual has no chance for reproduction but can increase future reproduction of kin by ceasing to use resources that they could use instead (deCatanzaro, 1980). Data showing that suicides are more common in old and sick people are consistent; however, alternative explanations are available, separation from kin does not protect against suicide, and there are no animal examples. The role of the group is also central to Allen and Badcock's model, in which people carefully monitor what they can contribute to a group (N. Allen & Badcock, 2003). People who realize they can contribute little retreat into depression that is hypothesized to prevent active expulsion from the group.
These approaches are quite different from the prevalent view that depression is a brain disorder. The brain mechanisms that mediate mood certainly can go awry, but two questions need consideration. First, is low mood a useful response, like a cough, or an abnormality unrelated to defenses, like epileptic seizures? Second, do individual differences in vulnerability to depression arise mainly from primary brain differences, or from brain changes mediated by social experience? These are not mutually exclusive alternatives, and most depression is best understood as the outcome of gene × environment interactions (Caspi et al., 2003). Also, there are different routes to depression, some of which progress irrespective of environment, others of which arise mainly from life circumstances, perhaps especially those involving pursuit of unreachable goals. Then, of course, there is the reality that depression can have different causes in different individuals, or even in the same individual at different times.
Anxiety and depression get the attention, but every emotion is subject to at least two kinds of disorder: excesses or deficits. For instance, pathological jealousy is common, but few clinicians know why jealousy exists (Buss, Larsen, Westen, & Semmelroth, 1992). Jealousy may arise for good reasons (Buss et al., 1999) or from delusions. Feelings of inadequacy make some men think that their partners might prefer someone else, and then that they do prefer someone else. Depression treatment often relieves pathological jealousy (Stein, Hollander, & Josephson, 1994). The syndrome of pathological lack of jealousy has yet to be described.
Other disorders involve inability to control behavior. Most obvious are the addictions and other habits, but other problems of behavioral control range from eating disorders to violence.
The human toll taken by addictions is magnified because their effects harm others as well as the addict. A special issue of Addiction was devoted to evolutionary approaches (Hill & Newlin, 2002), with suggestions about the adaptive significance of addiction (Sullivan & Hagen, 2002), life history theory (Hill & Chow, 2002), and the significance of fermentation (Dudley, 2002), among others. One evolutionary insight is simple; learning is chemically mediated, so exogenous substances that directly stimulate reward mechanisms can cause addiction (Nesse, 1994; Nesse & Berridge, 1997). The subjective sensations are pleasurable and the associated reinforcement increases the frequency of drug taking. Withdrawal symptoms stimulate further drug taking. Over time, subjective “liking” wanes, withdrawal effects become more severe, and the habit strength of “wanting” increases, trapping the addict in a vicious cycle that may offer little pleasure, even as it consumes most of what is valuable in life.
Vulnerability to substance abuse results from our novel environment. The reliable availability of pure chemicals and clever new routes of administration increase the rate of drug taking. Tobacco administered via the technological advance of cigarettes is the most widespread and harmful addiction, with alcohol a close second. The so-called hard drugs of abuse, such as amphetamines and cocaine, act even more directly on ascending dopamine tracts to establish addiction. Substance abuse is a universal human vulnerability to drugs that hijack reward mechanisms. Several recent papers challenge this model, arguing that long coevolution with drug-producing plants has shaped humans to pursue and use drugs that give advantages, via medical benefits or by augmenting energy and abilities (Hagen et al., 2009; Sullivan, Hagen, & Hammerstein, 2008). This interesting possibility seems to complement the possibility that drug abuse has become common because novel environments make pure drugs readily available via novel routes.
Half of Americans are now overweight and a third are clinically obese. They spend billions on books and treatment, but nothing works very well. Vast amounts of research have tried to understand what is wrong with the heavy half. An evolutionary approach suggests a different question: Why are we all vulnerable to obesity? A simple answer is that our behavior-regulation mechanisms were shaped in the very different environment of the African savannah, where the penalty for eating too little was swift and fatal. Even when food was plentiful, obesity remained rare because choices were limited and getting food involved burning as many calories per day as a modern aerobics instructor (Eaton, Shostak, & Konner, 1988).
Attempts to control weight by willpower leads to the other eating disorders, anorexia nervosa and bulimia. Explanations for anorexia as an adaptive variant mating strategy have been suggested (Surbey, 1987; Voland & Voland, 1989). However, a simpler starting place is the observation that these disorders usually begin with strenuous diets motivated by attempts to be attractive and thus preferred as a mate, a strategy that has been more important for women than men. Such diets cause episodes of gorging, a hallmark of bulimia, but life-saving during famine. Gorging precipitates shame, feelings of lack of control, more intense fear of obesity, and new resolutions in a vicious cycle of escalating anorexia and bulimia. Eating disorders are also fostered by the intense mating competition in large social groups, augmented by media images that make real bodies seem inadequate. In light of the pervasiveness of mating competing, this makes perfect sense (Buss, 1988, 1994).
Given its importance, you might think selection would have made sex foolproof. Instead, it exemplifies the vulnerabilities of a trait shaped by multiple strong forces of selection (Troisi, 2003). For instance, men complain about premature orgasm while women complain about lack of orgasm. Why? Sex differences in brain mechanisms and differences in anatomic proximity to stimulation, yes, but these are proximate explanations. Why is the system so poorly designed for mutual satisfaction? Because selection does not shape mechanisms for mutual satisfaction. Women who had orgasms very quickly might well have had fewer children, as might men who dallied too long when interruption is likely (Nesse & Williams, 1994). This is consistent with the observation that premature ejaculation is a problem mainly for men who are young or fearful.
Not every disorder fits neatly into a traditional category, especially some of the more serious disorders such as schizophrenia and autism. For these disorders the focus is less on any possible adaptive significance and more on evolutionary reasons why systems are vulnerable to failure.
Schizophrenia is the most serious common mental disorder of adults, with a prevalence rate consistent at about 1% across cultures (Jablensky, Satorius, & Ernberg, 1992). This undermines the idea that novel environment explains psychosis, although new reports suggest some variations of rates between cultures (van Dongen & Boomsma, 2013). There is also strong evidence that schizophrenics have lower than average RS: 0.3 of average for males and 0.5 for females (Pulver et al., 2004). The same data show no increased fitness of their close relatives, arguing against any selective benefits via kin selection. As mentioned already, infection has also been implicated. Low fitness should pose strong selection against alleles that increase vulnerability to schizophrenia, but they persist, posing the paradox of genes for schizophrenia. A comprehensive review of possible explanations was published recently (van Dongen & Boomsma, 2013).
Selection might not be powerful enough to purge deleterious mutations if they occur often enough to balance selection. This seems increasingly likely, as we have been unable to find any common alleles that account for even 1% of the vulnerability to schizophrenia and increasing evidence of the influence of rare alleles and copy number variations. Further support for the role of new mutations comes from recent confirmation that the rates of schizophrenia increase dramatically for children who were conceived when their fathers were over 40 (Malaspina et al., 2002). Genes transmitted by the mother have divided only 24 times per generation, compared to 800 cell divisions for the DNA in sperm of older fathers, suggesting that many cases of schizophrenia arise from new mutations.
Another possibility is that so many genes are involved that selection can act on them only weakly, so normalizing selection can never shape a design parameter to an extremely narrow zone, and some individuals will have parameters beyond the range of good functioning (Keller & Miller, 2006). This general idea is an important antidote to theories positing benefits from schizophrenia, even though an enormous number of genes contributing to a trait may make it less vulnerable, not more.
Schizophrenia genes might also spread if they are linked to strongly beneficial genes (Burns, 2005), but pleiotropic effects are more important. Cliff-edge effects offer a related possibility. For instance, racehorse breeding has resulted in longer and thinner leg bones that increase speed but are prone to fracture. If some mental characteristic gives increasing fitness up to a point where some mental trait suffers catastrophic failure, such cliff-edge effects could account for the genetic patterns seen in schizophrenia and manic-depressive illness (Burns, 2007; Nesse, 2004).
The same lines of reasoning also apply to other severe mental diseases that also have an incidence of about one in 100, autism in particular. Baron-Cohen has suggested that the manifestations of autism are examples of a pathological extreme of cognitive styles that are typically male (Baron-Cohen, 2002). This would help to explain the 4 to 1 predominance of males for autism with higher IQ scores.
One of the more significant lines of research in evolutionary psychiatry, by Crespi and colleagues, investigates the possibility that schizophrenia and autism are flip sides of a coin that is influenced by genetic imprinting that benefits paternal versus maternal genomes (Badcock & Crespi, 2006; Crespi, Summers, & Dorus, 2007). The idea is derived from Trivers's recognition that alleles derived from paternal and maternal genomes can have conflicting interests (Burt & Trivers, 1998) and Haig's work showing differential imprinting of genes that regulate the resources a fetus extracts from a mother in patterns consistent with the hypothesis that paternal alleles gain an advantage by extracting more and maternal alleles gain an advantage by controlling this manipulation (Haig, 2010). Emerging evidence supporting Crespi's theory includes an axis from schizophrenia to autism in a principle components analysis of symptoms (Dinsdale, Hurd, Wakabayashi, Elliot, & Crespi, 2013), and genetic evidence that autism and schizophrenia are best described as diametric disorders (Crespi, Stead, & Elliot, 2010).
The evolutionary origins of attention-deficit/hyperactivity disorder (ADHD) have been the focus for several articles suggesting possible functions (Baird, Stevenson, & Williams, 2000; Brody, 2001; Jensen et al., 1997; Shelley-Tremblay & Rosen, 1996), or that it is a facultative adaptation to certain environments (Jensen et al., 1997). The striking male bias of the sex ratio, over 5 to 1, gives hints that ADHD may simply be the extreme end of a continuum on which males tend to be higher than females, much akin to a recent suggestion about autism (Baron-Cohen, 2002). In the ancestral environment, a tendency to move quickly to a new activity when current efforts are unproductive is a foraging strategy that may pay off more for hunting males than gathering females. As for the capacity to sit in one place indoors for hours under enforced contact with a boring book, that is so far from anything the natural environment ever required, it is astounding that any of us can do it!
Child abuse has been a major focus for mental health prevention and treatment. Understanding the evolutionary origins and functions of attachment has helped to explain why most parents do not abuse their children despite provocations (Bowlby, 1984). An evolutionary perspective motivated two behavioral ecologists to ask the now-obvious question: Is child abuse more common in families with a stepparent? Their astounding result, not recognized by decades of previous work, is that death at the hands of parents is 80 times more common if there is a stepparent in the house (Daly & Wilson, 1988). This finding is commonly presented in a context framed by the tendency of males in many species to kill all unweaned infants shortly after they take over a female mating group (Hrdy, 1977). However, the mating pattern of humans does not routinely involve males fighting to take over a harem with multiple females who are nursing infants, so the analogy is incorrect. Instead, the mechanisms that protect babies from violence in families with two related parents seem more prone to fail in reconstituted families (Gelles & Lancaster, 1987).
The question one usually gets about evolutionary psychiatry is, “So, what can you cure that others cannot?” This is a fair question if evolutionary psychiatry is just another approach to therapy along with a hundred others. However, what evolution offers psychiatry is two steps back from proposing new cures, but far more fundamental. Instead of new treatments, it seeks explanations for mental disorders. Instead of explanations for why some individuals get sick and others do not, it seeks to explain why we all have minds that leave us vulnerable to mental disorders, and why natural selection has not eliminated genetic variations that result in disease for some.
Its most profound contribution is a solid framework for understanding how behaviors are regulated to accomplish the many conflicting tasks of life, from getting food and surviving, to competing for status, to finding mates and protecting children. Instead of viewing one kind of life as normal and others as deviations, it sees the inherent conflicts in relationships, the struggles that go on in groups, and the dilemmas every person faces to allocate efforts among a host of competing needs. Far from providing a rigid and cold perspective, an evolutionary view fosters deeper empathy for the challenges we all face, and deeper amazement that so many people are able to find loving relationships, meaningful work, and a way to juggle a bevy of responsibilities with good humor, and even joy.
Does this presage a new kind of psychotherapy? There certainly are major implications for how to do psychotherapy (Gilbert & Bailey, 2000) and psychoanalysis (Slavin & Kriegman, 1992), and some have called for applications now (Gilbert & Bailey, 2000; McGuire & Troisi, 2006; Troisi, 2012), but it is essential not to reduce evolutionary approaches to just another theory in the crowded clinical marketplace. It is broader and deeper. Every clinician should understand how selection shaped behavior, and all kinds of therapy should make use of evolutionary principles.
Finally, an evolutionary view of mental disorders in no way suggests accepting the pains and difficulties of the human condition; it instead suggests that much suffering is useless, even when it is normal. It supports using medications or other methods when they can relieve suffering safely, and it provides knowledge about the functions of negative emotions and the smoke detector principle that are crucial for making wise, individualized decisions. Most of all, it fosters a deeper sympathy for the human plight.