Evolutionary psychiatry: foundations, progress and challenges

WHAT EVOLUTION CAN AND CAN'T EXPLAIN

Finding evolutionary explanations for traits that make a species vulnerable to disease is an onerous task. A brief overview of natural selection can encourage critical assessment of proposals that are unlikely to be correct, especially those that suggest that a disease is somehow useful or that traits that harm individual fitness can persist because they give benefits to a group.

Textbook examples describe natural selection adapting a species to a changed environment. In the classic example, as Victorian soot darkened tree trunks, lighter colored moths became easier prey for birds, so darker moths had more offspring and became more common over the generations ¹⁷ . Such examples correctly emphasize that traits are adaptive or maladaptive only in relation to a specific environment, but they give the misimpression that natural selection is mostly about change. Far more often, instead, natural selection keeps things the same. Birds with wings too long or too short are more likely to die in storms, so selection stabilizes the average length at an intermediate value ¹⁸ .

Natural selection also shapes physiological and behavioral systems that adapt organisms to cope with changing environments ¹⁹ , ²⁰ , ²¹ . These range from simple reflexes like sweating to mechanisms that mediate decisions of all kinds, from what to have for lunch to whether to continue a marriage. How many control systems are shaped by natural selection? Tens of thousands; they control the expression of every gene, the processes that regulate metabolism and replication in 200+ different kinds of cells, the development of tissues and organs, and, of course, every physiological parameter. Perhaps most important of all, they control behavior.

Behaviors themselves are not shaped by natural selection, but genetic variations cause brain variations that interact with environments to give rise to behavior variations that influence fitness. This process shapes brains that induce behavior maximizing transmission of genes to future generations. This simple principle is the foundation for behavioral science. It does not mean that all behavior by all individuals maximizes genetic fitness in all environments; it applies only on average, in the natural environment, if the mechanisms are intact. However, recognizing that normal behavior has evolved to maximize the number of offspring who survive and reproduce is an essential foundation for evolutionary psychiatry.

Maximizing the number of surviving offspring requires subtle allocation of effort among several tasks: getting food and shelter, staying alive, finding mates and social partners, and mating and investing in offspring. The field of behavioral ecology studies how organisms allocate effort in ways that maximize reproductive success ³ , ⁴ . Diseases of aging are an example. Genes that cause aging and death are selected for if they increase reproduction ²² , ²³ .

EVOLUTIONARY MEDICINE: EXPLAINING DISEASE VULNERABILITY

My early efforts to use evolutionary principles to explain vulnerability to mental disorders ¹³⁰ quickly made it clear that explaining vulnerability to disease in general had to come first. The crucial advance came from working with G.C. Williams to frame the core question: Why does natural selection leave a species with traits that make it vulnerable to disease? ¹³¹ , ¹³² . Table 1 summarizes some categories of explanations for disease vulnerability. Each deserves brief comment.

Individual variations are the predominant explanation for disease vulnerability. They result mainly from mutations and developmental stochasticity that natural selection cannot eliminate. They are akin to limitations on quality control in a manufacturing process.

Multiple deleterious mutations arise in each individual in every generation. They are selected out with a speed proportional to how much they decrease fitness. The result is a few rare variations with large effect sizes, fewer with moderate effects, and thousands with tiny effects. This is exactly what genome wide association studies (GWAS) are showing for major mental disorders ¹³³ . Most alleles that increase the risk of mental disorders persist because their rate of elimination by natural selection is balanced by the rate of new mutations ⁸⁷ , ¹³⁴ .

Developmental variation is also unavoidable, and it increases the risk of disorders such as schizophrenia and autism. Could natural selection maintain a low level of developmental instability because that creates phenotype variations that increase reproduction for a few individuals in specialized niches despite lowering the average fitness for individuals? The possibility should not be accepted uncritically, but it is theoretically intriguing and potentially relevant ¹³⁵ , ¹³⁶ , ¹³⁷ .

Species‐wide vulnerabilities make many traits suboptimal. Genetic drift and path dependence are both important. Genetic drift can leave a whole species vulnerable, as illustrated by our inability to synthesize vitamin C. Mildly deleterious mutations can become more common, especially in a small population, simply from the stochasticity of evolution ⁸³ . Path dependence leaves some traits suboptimal because natural selection cannot redesign a trait from scratch. An automotive engineer can reroute a fuel line that is prone to cause fires, but the path of the urethra through the prostate gland cannot be changed, despite all the trouble it causes. The constraints on brain design are vastly larger, so an allele that gives an advantage by slightly altering one circuit will likely create problems for others.

Parasites that evolve faster than hosts are more important for the rest of medicine, but they are also relevant for psychiatry. Antibodies against streptococci with antigenic coats similar to human proteins can attack heart valves, causing rheumatic fever, as well as cells in the caudate nucleus, causing some cases of obsessive‐compulsive disorder (OCD).

Mismatch with modern environments explains many woes ¹³¹ , ¹³⁸ , ¹³⁹ , ¹⁴⁰ . Natural selection is too slow to keep up with rapid social and environmental changes. Fat, salt and sugar were in short supply in the African savannah, so we have preferences for them and little protection against the diseases that result when they become readily available. Sanitation, immunizations and antibiotics have decreased the burden of infectious diseases, but rates of autoimmune diseases are escalating ¹⁴¹ , ¹⁴² . Myopia is rare in hunter gatherers, but common and increasing rapidly in modern societies; whether the cause is close work, lack of sun, working in closed spaces, or some combinations of factors remains unknown ¹⁴³ .

The above four factors all result from the limitations of natural selection. It cannot prevent all mutations and developmental variations, and it is too slow to protect against fast‐evolving pathogens and fast‐changing environments. However, some vulnerabilities result from systems optimized by natural selection.

Trade‐offs that benefit individuals leave many traits less robust than they might be. High blood pressure causes atherosclerosis, low pressure causes fainting, so natural selection stabilizes the average at an intermediate level, with control systems that adjust the pressure to the situation. The risks of infections and autoimmune diseases stabilize the aggressiveness of immune responses at an intermediate level that nonetheless results in both infections and autoimmune diseases.

Traits that increase reproduction are selected for even if they reduce health and happiness. Competing for mates requires huge investments in appearance, wealth and social status ¹⁴⁴ . The dieting that sets off eating disorders is usually in the service of competition for mates ¹⁴⁵ . Reproductive competition helps to account for mortality rates three times higher in men than women in early adulthood in modern countries ¹⁴⁶ . The tendency for orgasm to occur sooner for males than females maximizes fitness at the cost of mutual sexual satisfaction ¹³⁰ . Pregnancy has obvious costs, and parturition is risky ¹⁴⁷ . Then there are all the efforts, sacrifices and worries required to raise children. Freud's emphasis on the importance of sex was along the right lines, but no one in his time recognized that selection shapes organisms to maximize gene transmission.

Defenses such as pain, cough, anxiety and low mood are useful responses shaped by natural selection ¹⁴⁸ , ¹⁴⁹ . Their aversiveness is essential to their utility, but the resulting suffering is the bane of our lives. The mild unpleasantness of sweating and shivering motivates moving to someplace where heat generated by the body just matches heat radiated, about 20°C. The greater unpleasantness of nausea and vomiting protects against eating toxic things again. Physical pain is not some abstract signal which gently suggests stopping actions that cause tissue damage; it is an excruciating conscious feeling that motivates escaping the situation and avoiding it in the future. Anxiety and low mood provide similar protection against other dangers. These adaptive responses evolved because they protect against harm. They are aversive for good reasons, and their regulation systems are shaped to benefit our genes, sometimes at a cost to ourselves.

The above summary of evolutionary medicine is brief to the point of being telegraphic, and it does not discuss phylogenetic approaches to infectious diseases and cancer that are proving very useful. Several articles put the current field in a historical context that includes many applications before the 1990s ¹⁵⁰ , ¹⁵¹ , ¹⁵² . Since then, evolutionary medicine has grown into a substantial field, with many textbooks ⁶ , ¹⁰ , ¹² , ¹⁵³ and edited volumes ¹⁴⁷ , ¹⁵⁴ , ¹⁵⁵ , ¹⁵⁶ . The International Society for Evolution Medicine and Public Health has annual meetings, and sponsors an open‐access journal and other online resources. Courses on evolutionary medicine are now offered in most research universities in the US ¹⁵⁷ . However, despite many pleas ¹⁵⁷ , ¹⁵⁸ , ¹⁵⁹ , medical schools still provide little or no coverage of evolutionary biology as a basic science for medicine ¹⁶⁰ .

THE DEVELOPMENT OF EVOLUTIONARY PSYCHIATRY

The many books and papers about evolutionary approaches to mental disorders have had little influence on psychiatry. Historical context provides part of an explanation. C. Darwin said little about mental disorders, despite his connection with psychiatrist J. Crichton‐Browne to get illustrations for his book on emotions ¹⁶¹ . For the rest of the 19th and early 20th century, psychiatry was, along with the rest of medicine, enamored with vague notions about the degeneration of families or the species that had little to do with evolutionary biology ¹⁶² , ¹⁶³ , ¹⁶⁴ . Subsequent evolutionary contributions to psychiatry came in three phases: first ethology, then sociobiology and evolutionary psychology, then evolutionary medicine.

Applications of ethology to psychiatry initially made little use of evolutionary biology. Everything changed in the mid‐1960s. Recognition that capacities for social behaviors are shaped by kin selection and benefits from trading favors provided one foundation. The other was recognition that a full explanation for a trait requires an evolutionary explanation of its origins and functions as well as a proximate explanation of its mechanisms. E. Mayr advocated for this distinction effectively ¹⁶⁵ , ¹⁶⁶ , ¹⁶⁷ , but it became more useful as a part of Tinbergen's four questions ²⁵ , ¹⁶⁸ , ¹⁶⁹ , ¹⁷⁰ , ¹⁷¹ , ¹⁷² : What is the mechanism? How does it develop? What is its phylogeny? What is its adaptive significance? The first two questions are about proximate mechanisms, the other two are about evolution ¹⁷¹ . Recognition that all four questions deserve answers is now an established foundation for behavioral biology that continues to inspire commentary ¹⁶⁸ , ¹⁷⁰ , ¹⁷³ , ¹⁷⁴ , ¹⁷⁵ , ¹⁷⁶ , ¹⁷⁷ .

The significance of these advances was widely recognized only after the publication of Sociobiology by E.O. Wilson ¹⁷⁸ in 1975 and The Selfish Gene by R. Dawkins ¹⁷⁹ in 1976. They inspired the first applications of evolutionary ethology to psychiatry. Especially influential were a series of papers by M.T. McGuire ¹⁸⁰ , ¹⁸¹ , ¹⁸² and related articles that he welcomed as Editor of Ethology and Sociobiology (now Evolution and Human Behavior). Other early explorations of the implications for psychiatry came in books by M. Konner ¹⁸³ , and B. Wenegrat's Sociobiology and Mental Disorder ¹⁸⁴ . The term “evolutionary psychiatry” was first used by P.D. MacLean in an article on how a phylogenetic view of the “triune brain” could counter reductionism ¹⁸⁵ .

Studies by J. Bowlby and M.D. Ainsworth on the adaptive significance of infant attachment, and mental problems resulting from its disruption, were foundational for evolutionary psychiatry and have inspired continuing research ¹⁸⁶ , ¹⁸⁷ , ¹⁸⁸ , ¹⁸⁹ , ¹⁹⁰ , ¹⁹¹ . Recent reassessments have proposed that anxious and ambivalent attachment styles are not necessarily pathological; they can be strategies that infants use to get resources from mothers who might not otherwise be forthcoming ¹⁹² , ¹⁹³ , ¹⁹⁴ , ¹⁹⁵ . Especially clinically relevant is a recent proposal that considers how understanding maternal neglect in the light of its evolutionary origins and functions can be helpful for patients ¹⁹⁶ .

L. Sloman and J.S. Price also conducted early important research, first with chickens, then with vervet monkeys, to test the theory that depression can be understood as “involuntary yielding behavior” that prevents attacks after losing a status battle ¹⁹⁷ , ¹⁹⁸ . An early paper envisioning a wide range of evolutionary applications in psychiatry ¹³⁰ inspired a farsighted but sadly ignored plea to avoid speculating about how diseases might be useful to a species ¹⁹⁹ .

Evolutionary medicine emphasized mental disorders from its origins ¹³¹ , ¹³² . In 1998, psychiatrists M.T. McGuire and A. Troisi published Darwinian Psychiatry, the first book using evolutionary medicine principles to understand mental disorders ²⁰⁰ . Since then, a steady stream of books and papers have further developed the field, now called “evolutionary psychiatry” to expand acceptance by those put off by anything “Darwinian” ²⁰¹ , ²⁰² , ²⁰³ , ²⁰⁴ , ²⁰⁵ , ²⁰⁶ , ²⁰⁷ , ²⁰⁸ , ²⁰⁹ , ²¹⁰ , ²¹¹ , ²¹² , ²¹³ , ²¹⁴ , ²¹⁵ . Progress has been especially fast in the UK, where the Special Interest Group on Evolutionary Psychiatry of the Royal College of Psychiatrists has now over 2,000 members ²¹⁶ . Two leaders of that group, R.T. Abed and P. St. John‐Smith, recently edited the first multiauthor overview of evolutionary psychiatry ¹¹ .

The field of evolutionary psychology has grown in parallel, making major contributions for understanding psychopathology. This field initially emphasized research on mating strategies, because they influence reproduction so directly ⁷⁶ , ¹⁴⁴ , ²¹⁷ , ²¹⁸ , ²¹⁹ , ²²⁰ , ²²¹ . A 1988 meeting established the Human Behavior and Evolution Society ²²² , and encouraged development of evolutionary psychology as a broad field that inspired many papers, books ²²³ , ²²⁴ , ²²⁵ , ²²⁶ , ²²⁷ , ²²⁸ , ²²⁹ , ²³⁰ , courses and controversies, many of which had a political flavor ²³¹ , ²³² , ²³³ , ²³⁴ . Several major publications from evolutionary psychology focus on mental disorders ²⁰⁸ , ²²⁷ , ²³⁵ , ²³⁶ , ²³⁷ , ²³⁸ , ²³⁹ .

The fifty years of advances in understanding behavior and mental disorders summarized above have had little influence on psychiatric research and practice. Identifying the obstacles that have slowed adoption of evolutionary biology as a basic science for psychiatry may help to overcome them. The education gap is a major impediment. Few psychiatrists get a chance to learn how evolutionary principles explain behavior. Many do not even know that evolutionary explanations are needed, and few know how to frame and test evolutionary hypotheses ⁸ . Elementary errors result, even in books on the topic. For instance, many who are curious about evolutionary psychiatry are likely to turn to a book by that title ²⁴⁰ , which presented some intriguing Jungian ideas in a mishmash of speculation about diseases as if they were adaptations evolving because they benefit groups. The book aroused justified skepticism ²⁴¹ .

Wariness about evolution in general is also an obstacle, especially in the US, where some religious groups deny that evolution has had any role in shaping humans. Hesitation among scientists arises from perceptions that evolutionary psychology is somehow controversial ²⁴² , ²⁴³ . While work in all fields deserves critique, sensible scientists all recognize that natural selection shaped the brain, and the importance of understanding the adaptive significance of behavior is increasingly acknowledged across the breadth of psychology ²³² , ²⁴⁴ , ²⁴⁵ .

The largest obstacle, however, is uncertainty about what evolutionary biology has to offer. Mental health clinicians need better ways to help their patients now. So, advances in basic behavioral biology can seem abstract. However, evolutionary psychiatry can improve clinical care now by providing sensible explanations that support all kinds of therapy, as well as new ways to frame disorders that patients can understand and appreciate. Each section below emphasizes such practical applications.

EMOTIONS AND THEIR DISORDERS

Negative emotions are, like pain and cough, symptoms that exist because they have given selective advantages. They have evolved in conjunction with control systems that express them in the situations where they are useful. Those systems express false alarms even when functioning normally, and they are prone to malfunctions that cause disorders. Usually, however, anxiety and low mood are symptoms that indicate a problem, not disorders produced by malfunctioning control systems.

Psychiatry textbooks have long chapters about emotional disorders, but little or nothing about normal emotions. How we can regulate our emotions is the topic of many books and papers, but how our emotions regulate us gets little attention. Controversies about the nature of mood and anxiety disorders persist despite the efforts of expert committees ²⁴⁶ , ²⁴⁷ , ²⁴⁸ , ²⁴⁹ , ²⁵⁰ , ²⁵¹ , ²⁵² . Decades of research have not found the expected specific brain or genetic abnormalities. Medication treatments are somewhat effective, but when to use them is the topic of vigorous public debate ²⁵³ , ²⁵⁴ . News articles suggest that tsunamis of emotional problems are sweeping over whole populations. And, while mental health clinicians and researchers do what they can to stem the tide, it will never be enough. The situation arouses appropriate emotions: confusion and frustration.

The standard approach asks why some people have emotional problems and others do not. The responsible factors have been studied in exhaustive detail: individual differences in genetic make‐up, early experiences, drug use, cognitive biases, relationships, family dynamics, and larger social factors. Thousands of papers and textbooks describe why some individuals experience emotional disorders and others do not.

An evolutionary medicine approach asks different questions ¹² , ¹⁶ , ²⁰⁰ , ²⁵⁵ , ²⁵⁶ : Why do we all have capacities for negative emotions? How are they useful? How is their expression regulated? Why are their control systems vulnerable to malfunction? Answers to these questions provide a biological foundation for understanding and treating emotional disorders in the context of normal emotions.

Recent research progress has led to a consensus that emotions are adaptive states shaped by natural selection ²²⁹ , ²⁵⁷ , ²⁵⁸ , ²⁵⁹ , ²⁶⁰ , ²⁶¹ , ²⁶² , ²⁶³ , ²⁶⁴ , ²⁶⁵ , ²⁶⁶ . However, this progress is obstructed by a tendency to tacit creationism that describes emotions as if they were distinct products of a designer's vision, each with a specific mechanism that carries out a specific function ²⁶⁷ , ²⁶⁸ . For instance, anger is said to serve the function of signaling an imminent attack. Or threatening the end of a relationship. Or expressing dominance. Emotions do serve functions, but one emotion can serve many functions, and one function is advanced by many emotions. So, trying to map specific emotions to specific functions generates complexity and controversies.

The obstacle can be overcome by a definition of emotions based on the situations that shaped them. Emotions are special states that adjust physiology, arousal, cognition, facial expression, motivation, memory, behavior, and subjective experience in ways that gave selective advantages when expressed in situations that recurred and influenced reproductive success over the evolutionary history of a species ²²⁹ , ²⁵⁶ , ²⁶⁴ . Control systems process information from multiple internal and external sources to express emotions in the form and to the degree that maximizes fitness in the current situation. On average. In the natural environment. In response to always insufficient information. If the control system is intact. With variations induced by cultural and individual experiences ²⁶⁹ .

Mapping emotions to situations instead of functions helps to quell some persistent controversies: How many emotions are basic and how many are secondary? Which aspects of an emotion are primary, and which are secondary? Does subjective feeling initiate physiological changes, or does perception of bodily changes give rise to the feeling? An evolutionary perspective suggests that these questions do not have specific answers. Instead, multiple aspects of an emotion are expressed somewhat concordantly, influenced by details of the situation, by each other, by expectations, by cultural learning, and by recursive feedback loops.

The driving modes of modern cars offer a useful but imperfect analogy. Setting a car for sport, normal, eco or snow mode adjusts engine timing, gear ratios, suspension firmness, torque distribution, and dashboard appearance in ways that increase the ability to cope with different situations. The analogy is imperfect, because cars come off the assembly line as identical as quality control can make them, while minds are products of slightly different genomes interacting with varying environments. Furthermore, and more importantly, organic control systems are both more jury‐rigged and more adaptable, so variations in a situation can arouse different aspects of an emotion to different degrees. For instance, different kinds of unpropitious situations arouse different symptoms of low mood ²⁷⁰ .

The algorithms that detect the presence of situations in which emotions would be useful are nothing like an engineer's decision tree. They are products of a process that is like machine learning, steadily improving fitness by successively changing different parameters at different levels, and keeping whatever works ²⁷¹ . The resulting organic complexity makes reverse engineering extremely difficult. Further complexity arises because multiple overlapping situations may be present; conflicting goals may be pursued simultaneously; and individuals have different values, goals, resources, strategies, relationships, and prior experience. Emotion control systems generally work, but describing the brain mechanisms that mediate them is an onerous task.

It is easier to describe the situations that an organism encounters. Situations that influence fitness can be categorized on three dimensions: kind of resource (physical or social), valence (opportunity or threat), and the situations that arise routinely during goal pursuit. Table 2 shows 24 situations that arise in goal pursuit and their corresponding emotions. The need to deal with opportunities and threats specific to more specific kinds of resources further differentiates the states. For instance, situations that arise in getting and keeping or losing a mate have shaped capacities for romantic love, sexual arousal, caring, commitment, guilt, jealousy, and grief.

Any approach to emotional symptoms which assumes that all individuals are the same loses the most important information. General checklists of life events and levels of stress do not measure the situations that arouse emotions. Information about the individual's life situation provides a starting point for distinguishing four categories of emotions: useful for the individual; harmful for the individual, but useful for increasing gene transmission; harmful for the individual and for fitness, but arising from normal mechanisms; and harmful products of an abnormal regulation mechanism.

Appraisal theories of emotions are especially helpful in understanding individuals. Emotions are usually aroused not directly, but by an individual's appraisal of what new information means for his/her ability to make progress towards personal goals ²⁷² , ²⁷³ , ²⁷⁴ . Those goals can differ dramatically between individuals and even within an individual at different times. The emotional impact of a positive pregnancy test depends on whether the woman was eager to conceive or considering a divorce. One patient insisted that she should not be depressed now because she had just started a job at a brokerage firm that tripled her previous salary; she was reluctant to talk about having to give up her previous career as a struggling artist.

The implications for psychiatry are profound. General measures of stress and checklists of life events ignore many factors that influence an individual's emotions. Diagnostic criteria based only on the number, severity and duration of symptoms are necessary to get the reliability required for epidemiology, but they are not grounded in biology. Determining if an emotion is normal requires assessing the presence or absence of a situation in the context of an individual's values, goals, strategies, expectations and psychodynamics. Determining if an emotion is useful for the individual is a separate question. An evolutionary framework for understanding the origins, functions and regulation of normal emotions provides a foundation for understanding abnormal emotions, starting with anxiety and mood disorders.

ANXIETY AND ITS DISORDERS

Almost all research on anxiety has focused on why some people have too much of it. An evolutionary perspective brings in the other half of biology to ask how natural selection shaped subtypes of anxiety, why normal control systems may sometimes express so much excess anxiety, and why some people have too little anxiety ²⁷⁵ , ²⁷⁶ , ²⁷⁷ , ²⁷⁸ , ²⁷⁹ , ²⁸⁰ . This reframing of anxiety disorders can improve clinical outcomes.

DEPRESSION AND LOW MOOD

It is hard to see anything useful about depression. Pessimism, hopelessness, lethargy, low self‐esteem, and ruminating about death or suicide are worse than useless, so depression is usually assumed to be abnormal. But, like physical pain, ordinary low mood is a potentially useful response to a bad situation. Both can be expressed excessively or when they are not needed, resulting in the vast suffering. Treatments are somewhat effective, but, as is the case for anxiety, the search for causes of depression has come up short: plenty of statistically significant results, but no specific common genes, neurotransmitters or brain abnormalities have been found. Controversies and calls for new directions abound.

An evolutionary perspective suggests taking a medical approach. Not the crude “medical model” which assumes that symptoms are products of a specific abnormal mechanism, but an approach like that in the rest of medicine, where some symptoms are recognized as useful responses aroused by disease or disadvantageous situations. Progress in understanding mood disorders will come from discerning the origins, functions and regulation of normal mood. That means identifying the situations in which low mood is useful, how it is useful, how it is normally controlled, and why mood control systems are so vulnerable to malfunction.

Scores of papers propose evolutionary explanations for depression, but reading them can be frustrating. Sadness, low mood, depression symptoms, and depression syndromes are not always clearly delineated. Some articles aim to explain the capacity for ordinary mood variations, others for the symptoms of depression, some more for the syndromes of major depression, melancholia or bipolar disorder. Many proposed explanations are framed as “the function of depression”, often arguing for the importance of that function over alternatives proposed by other authors. Table 3 lists some examples. All deserve consideration, but, in full evolutionary context, they are not competitors. They are varying ways that a group of related states can be useful if expressed in a cluster of overlapping untoward situations that have recurred over the course of evolutionary history.

Considering the evolutionary origins of the capacity for mood frames a different question. In what kinds of situations would the characteristics of low and high mood increase inclusive fitness? Low mood can be useful in unpropitious situations in which efforts are likely to be wasted or cause losses. The intense effort and risk‐taking characteristic of high mood can likewise be useful in propitious situations that offer big payoffs for small investments. It is interesting to consider that time‐limited situations are likely to arouse more intense emotions; if good times are likely to end soon, intense activity and risk taking will be worth it; if bad times are likely to end soon, it is best to just wait.

Natural selection has differentiated the global states of low mood into overlapping subtypes whose utility depends on the resource involved and why effort is likely to pay off ²⁷⁰ , ³³⁷ . Table 4 lists some kinds of situations in which aspects of low mood can be useful and how they can have increased fitness in past generations.

This approach frames depression as extreme versions of overlapping states shaped to cope with different unpropitious situations. It provides a framework for considering them together instead of emphasizing one explanation or viewing subtypes as distinctly separate. As is the case for anxiety disorders, the wish for simplicity is undermined by the messiness of organic complexity. Notions that all situational causes for depression can be collapsed into “stress”, whose effects are mediated by the hypothalamic‐pituitary‐adrenal (HPA) axis, are inconsistent with an evolutionary view. The mood system is not nearly that crude. Mood symptoms are differentiated to deal with different kinds of unpropitious situations ³³⁷ , ³³⁸ , ³³⁹ , ³⁴⁰ , ³⁴¹ . However, that does not mean that different patterns of low mood are distinct modules; they are overlapping suites of responses whose structure is very different from anything an engineer would design.

The wish for simplicity helps to explain the prevalence of black or white opinions that depression is usually a product of brain abnormalities or usually a normal response. A more encompassing evolutionary view encourages recognition that some episodes of depression are aroused by current situations, while others are excessive or distorted responses, and others are unrelated to any current situation. In his case series from 1934, A. Lewis concluded that each group comprised about a third of his patients ³¹⁶ . My experience has been similar, but controlled studies of population samples would be valuable. Clinicians in different settings see different proportions of patients in each group, thus explaining some differences of opinion about the causes of depression.

The above summary of current thinking about evolution and depression is telegraphic in its brevity. Many reviews and books are also available for those interested ²¹⁰ , ³²⁹ , ³⁴² , ³⁴³ , ³⁴⁴ , ³⁴⁵ , ³⁴⁶ , ³⁴⁷ , ³⁴⁸ , ³⁴⁹ , ³⁵⁰ , ³⁵¹ , ³⁵² , ³⁵³ , ³⁵⁴ , ³⁵⁵ . Research to test specific proposals is needed, but difficult to carry out. It will be some years until work in this area settles down to the extent that a summary can become a routine part of psychiatric education. In the meanwhile, work in the clinic and the laboratory continues, where some applications of evolutionary thinking can be useful even now.

Mood disorders are manifestations of failed control systems, so control systems theory is needed for a full explanation ³⁵⁶ , ³⁵⁷ , ³⁵⁸ . Several observers have noted the tendency for depression to induce behaviors that perpetuate and escalate symptoms in a positive feedback cycle ³⁵⁹ , ³⁶⁰ , and new research on metacognitive therapy provides ways to disrupt the cycles ³⁶¹ . Pessimism decreases initiative that could lead to success. Low self‐worth inhibits social contacts. Lethargy decreases exercise. In modern societies, people can retreat to a room alone and shut off contact with others, a perfect recipe for making depression worse. Activation therapy can break the cycle ³⁶² .

As noted already, the self‐adjusting nature of defense control systems makes them especially prone to positive feedback cycles that cause disorders. Much depression is like chronic pain that persists because repeated arousal adjusts the response threshold to a lower level ²⁹⁰ , ²⁹¹ . Depression is recognized to be vulnerable to “kindling”, in which repeated episodes make further episodes more likely ³⁶³ , ³⁶⁴ . Brain mechanisms are being explored, but the question of whether kindling is a defect, an epiphenomenon, or an adaptation to an unfavorable environment remains to be fully addressed.

An evolutionary control system perspective helps to make sense of the repeated finding that many people with depression have the tendency to mood fluctuations, even if they do not meet criteria for bipolar or cyclothymic disorder ³⁶⁵ , ³⁶⁶ . All control systems are compromised by the trade‐off between high gain and stability. Bipolar disorder has the characteristics of a control system with excessively high gain ³⁶⁷ . The early stages of mania often ramp up as enthusiasm generates success, making the environment appear extremely propitious, so that even higher energy and investments seem justified, in a runaway vicious cycle. Most people, after a great success, experience a letdown that often seems mysterious. That tendency, a reflection of what psychologists call opponent process ³⁶⁸ , is just the ticket for preventing mood from escalating out of control, and it seems to be missing in people with bipolar disorder.

Why did natural selection leave mood control systems vulnerable to dysregulation? The standard explanations from evolutionary medicine all apply.

Individual genetic and developmental variations leave some individuals more vulnerable than others. The set point for mood seems to be as stable as that for body mass. An individual's mood responsiveness also tends to stay consistent over time. Both are influenced by genetic variations. There is no reason to assume that the responsible alleles are deleterious mutations. Most are probably just variations whose tiny effects would not influence fitness much, at least in ancestral environments.

Mismatch with modern environments may or may not be important; we lack adequate data to be sure and drawing firm conclusions will be difficult. Some evidence suggests that mood disorders are more common now than they were in the past ³⁶⁹ . However, the belief that the incidence of mood disorders has been increasing in recent decades is distorted by the salience of current problems and the tendency to forget bad times in the past ³⁷⁰ . Epidemiological studies that followed the same population over time have not consistently found substantial increases ³⁷¹ . Even the presumed increased incidence of anxiety and depression related to the COVID‐19 pandemic is not confirmed by systematic epidemiological assessment ³⁷² .

Nonetheless, substantially different rates of mood disorders in different countries ³⁷³ imply strong socio‐cultural influences, perhaps partially mediated by physical factors as well. Concern about the influence of social media is certainly justified, even if hard to confirm ³⁷⁴ . We know that mood is influenced by social comparisons, and that people display especially positive views of their lives on social media, making observers feel inadequate by comparison ³⁷⁵ . However, some anthropological reports found depression in populations far removed from such modern influences ³⁷⁶ , ³⁷⁷ . While small group sizes and rapid cultural changes make definitive studies difficult, the question is important enough to justify a major investment.

Trade‐offs are important in shaping mood regulation systems. The inherent trade‐off between high gain and stability has already been mentioned. The smoke detector principle may also be relevant. Large costs are likely to be incurred by foraging when no food is available or by engaging in status competitions with more powerful others. The costs of waiting, conserving resources and avoiding initiative are likely to be low. So, like other defensive responses, low mood is expressed more readily and intensely than seems sensible. This has important treatment implications. Most of medicine consists of relieving painful normal responses. Recognizing that low mood can be useful does not suggest that it should not be treated. However, it does encourage careful assessment of the situation to try to identify unpropitious situations that can be changed.

Other trade‐offs arise because natural selection shapes organisms to maximize reproduction at the expense of individual health and welfare. A night spent in any busy psychiatric emergency room includes plenty of cases precipitated by infidelity, divorce, abandonment, or the dilemma of whether to leave a relationship. Such relationships are so close to the center of human lives that it can seem heartless to observe that these intense feelings are in the service of reproduction at the expense of individual welfare.

How are these principles useful in the clinic? Changes in the approach to four aspects of care are useful: the clinical evaluation, how to talk with patients about depression, how to describe treatment, and how to look for new treatments.

SUBSTANCE USE AND ABUSE

Most of our research and knowledge about substance abuse is about why some people succumb and others do not, and about what treatments are most effective. Those are the standard questions. However, an evolutionary perspective calls attention to several others: Why do plants make psychotropic drugs? Are human motivations to use substances an epiphenomenon of motivations shaped for other reasons, or are they adaptations shaped because taking drugs increases fitness? Why are some people much more vulnerable to addiction than others? Why are most people so confident that they can use drugs and stop whenever they choose?

Plants make psychoactive substances to discourage herbivores ⁴⁰⁶ , ⁴⁰⁷ . Chemicals that disrupt herbivores’ nervous system are especially common, because small doses can have large effects. A mouse that eats a coffee bean will likely die; an ungulate that browses on tobacco will likely get sick and not do it again. However, an arms race ensues: selection shapes herbivores that can deal with toxins, creating selection for new toxins that can better deter herbivores and perhaps give advantages to specialists who can tolerate them. The monarch butterfly caterpillar has evolved the ability to feed on milkweed and store its toxins, making the caterpillars and the butterflies distasteful to birds. Humans are omnivores who cope with diverse toxins by routing products from the digestive tract to the liver, where enzymes destroy most toxins before nutrition is forwarded to the general circulation.

Explaining vulnerability to substance abuse starts with the observation that humans have used psychoactive drugs for thousands of years – alcohol for 5,000 ⁴⁰⁸ , tobacco for 2,000 ⁴⁰⁹ ; opiates, caffeine and coca for nearly as long. Different drugs induce different states with different benefits. For caffeine it is alertness; for nicotine, calm and alertness; for alcohol, disinhibition and social connection; for cannabis, pleasure and calm; for opiates, euphoria and pain relief; for cocaine and amphetamines, pleasure and energetic concentration; for psychedelics, intense experiences of diverse types. Given this diversity of drug actions, many factors will be relevant, all within the general explanation that humans are smart and learn quickly to repeat behaviors that bring benefits; those behaviors include making and selling drugs as well as using them.

Other animals also use substances ⁴¹⁰ , but only humans have discovered ways to concentrate and administer chemicals in ways that increase positive emotions, decrease negative emotions, and provide experiences otherwise not possible. Some drug use is planned and instrumental; for instance, taking caffeine to stay awake and complete a task. Some is planned for pleasure, such as drinking with friends. But many psychoactive substances act on motivation and learning mechanisms to increase intake steadily in the positive feedback pattern we call addiction. Most induce positive feelings, at least at first, but the liking systems that mediate subjective pleasure are only partially congruent with the wanting systems that motivate behavior ⁴¹¹ . So, what starts out as a simple pursuit of pleasure often shifts to compulsive drug use that brings little pleasure.

The reinforcement mechanisms that maintain drug taking behavior are there for good reasons: learning is an adaptation that induces repeating behaviors that increase fitness ⁴¹² . However, the system cannot tell the difference between a real orgasm and the rush from drug stimulation of dopamine receptors.

The reward system is sufficient to maintain drug use, but withdrawal symptoms make it harder to quit. Continued stimulation by a drug induces adaptive desensitization of receptors, so withdrawal of the drug leaves the receptor unable to activate downstream processes, causing distress ranging from the headaches of caffeine withdrawal to epileptic seizures during alcohol withdrawal.

Most people are confident that they can control their behavior. This false belief greatly increases the risk of substance abuse. When they start using drugs or alcohol, people believe that they can stop whenever they choose. Depending on the drug, many people can stop using, making the risk seem abstract. But conscious decisions have a weak influence on behavior. Many people will stop using for a time, to demonstrate their control to themselves, before slipping back into a pattern of escalating use. As for so many behavioral disorders, positive feedback loops are at the root of the problem. Increased use changes the brain in ways that lead to further increased use. On top of that, substance abuse wrecks job, family and other sources of satisfaction, so that pleasure is soon available only from substances.

In summary, the standard evolutionary explanation for substance abuse is that novel substances can hijack learning mechanisms that were never protected from the effects of drugs, because these were not reliably available during most of our evolutionary history ⁴¹³ , ⁴¹⁴ , ⁴¹⁵ , ⁴¹⁶ . From this perspective, drug abuse is a product of mismatch between evolved behavioral control systems and the ready availability of substances and routes of administration that were not present regularly in our evolutionary history. It is not an adaptation; it is an epiphenomenon arising from the effects of drugs on our chemically mediated motivation mechanisms.

The alternative explanation is that natural selection has shaped systems that motivate taking certain drugs because they have given a selective advantage to our ancestors. E.H. Hagen and colleagues have long argued that individuals may have obtained selective advantages from seeking and using drugs, especially nicotine ⁴¹⁷ , ⁴¹⁸ . Indeed, nicotine is an effective anthelminthic that humans may use for deworming ⁴¹⁹ . The interesting question is whether those benefits increased connections between nicotinic receptors and reinforcing pathways in ways that increased survival, reproduction, and the rewards of smoking.

Mutations that increased our ability to metabolize alcohol emerged about 10 million years ago, about the time when our ancestors descended from trees and began eating more overripe fruit on the ground ⁴²⁰ . Those with a preference for alcohol and a better ability to metabolize it would have gained advantages, but were those advantages just extra calories? Several authors have recently suggested that alcohol use may have facilitated the rise of civilization, or at least the cementing of bonds among group members ⁴²¹ , ⁴²² , ⁴²³ , because the release of inhibitions increases bonding. Conversations over drink may be especially useful ⁴²⁴ . These interesting hypotheses are hard to confirm.

How can all of this be useful in the clinic? Teaching people that their behavior is not nearly as much under their conscious control as they think would provide strong protection, but people are loathe to give up that belief. It would be wonderful if we could prevent initial drug use by telling young people that natural selection never shaped mechanisms to protect us against addiction, but youth are notoriously resistant to advice from their elders – possibly for the good evolutionary reason of avoiding manipulation ⁴²⁵ , ⁴²⁶ , ⁴²⁷ . However, an evolutionary perspective on substance abuse can help to relieve stigma and encourage cooperation with treatment. Understanding the vicious cycles that escalate addiction helps in the difficult task of finding ways to stop them.

Much research is about why some individuals are more vulnerable to substance abuse than others ⁴²⁸ , ⁴²⁹ . The risk is heritable in a range from 72% for cocaine, to 50% for alcohol, to 3‐40% for hallucinogens and opioids ⁴³⁰ . But a polygenic risk score predicts only about 3% of the variance, and individuals in the highest decile have risks not significantly different from those in adjacent deciles ⁴³¹ . There is only moderate overlap between the risks for abuse of different drugs, so the notion of a drug seeking personality is only partially supported ⁴³¹ .

Men are more vulnerable than women. It is not clear if this results from the general tendency for men to take more risks, from the risks to the fetus of taking drugs while pregnant, or something else ⁴³² . It has been also suggested that taking drugs and heavy drinking may be displays of vigor to impress potential mates. While this may occasionally be a proximate motive, it seems unlikely to have been a selection force increasing motivation for drug use.

The influence of environmental factors is obvious. Individuals with few sources of pleasure in their lives are more likely to turn to drugs to make up the deficit. Those suffering from social distress or physical pain can get relief from drugs. Both groups are especially likely to get trapped in a positive feedback cycle. A number of life experiences can influence these factors: early abuse, unfair treatment, deprivation, injury, or simply being in an unfortunate life situation ⁴³³ .

Concerning genetic differences that influence vulnerability, an evolutionary perspective suggests that they are not defects, but minor variations that likely had little influence on fitness until recent generations ⁴³⁴ , ⁴³⁵ . Those variations might, however, have influenced normal behavior in the ancestral environment, for instance by affecting foraging strategies. If this is confirmed, there may be implications in terms of development of behavioral tests aiming to predict vulnerability to substance abuse.

EATING DISORDERS

Research on eating disorders illustrates the limitations of looking only for proximate mechanisms, and the opportunities and difficulties associated with seeking evolutionary explanations. The most fatal of all mental disorders, eating disorders have substantial symptom overlap, and their incidence has been increasing in recent decades, especially in developed countries ⁴³⁶ .

Some papers that argue for the primacy of genetic factors ⁴³⁷ suggest that many patients simply lose interest in eating, but this is not consistent with evidence that preoccupation with thinness and dieting precedes eating disorders in most cases ⁴³⁸ . A GWAS on over 70,000 individuals found eight loci with statistically significant influences on the risk of anorexia nervosa, but effect sizes were minuscule ⁴³⁹ and a polygenic score including all available genetic information accounted for only 1.7% of the variation in risk ⁴⁴⁰ . The genetic influences on anorexia nervosa are therefore unlikely to be abnormalities; they probably result from natural variation in psychological traits such as conscientiousness and neuroticism that can mediate risk in modern environments.

The evolutionary explanations proposed for eating disorders are diverse and confusing. For instance, some papers have suggested that restrictive eating might be an evolved strategy for postponing reproduction until more physical or social resources are available ⁴⁴¹ , ⁴⁴² . However, natural selection has shaped a much more efficient and subtle system to turn off reproductive cycling when a pregnancy would be unlikely to result in a surviving offspring. When high levels of energy expenditure are not balanced by sufficient input, the system reverts follicle stimulating hormone (FSH) and luteinizing hormone (LH) to prepubertal levels, stopping reproductive cycling even at normal body weights ⁴⁴³ . That is why cycling stops in some women athletes during times of intense training. This system works fine on its own; it needs no augmentation by food restriction that is likely to be fatal in a time of famine.

The “adapted to flee famine” hypothesis argues that the high exercise levels pursued by some patients with anorexia nervosa might be an evolved strategy that motivates running away from an area of famine to other locations where food is more available ⁴⁴⁴ . However, individuals with anorexia exercise not to find food but to keep body weight low. Exercising while starving is not an adaptation, it is an aspect of a disease. Most studies of starving people report lethargy, not intense activity ⁴⁴⁵ .

Yet another interpretation of anorexia nervosa as an adaptation proposes that it is induced by female‐female competition for status in a group. It suggests that a woman can avoid attacks from other higher‐status women by losing weight and thus demonstrating that she is not competing for mates and thus is not a threat ⁴⁴⁶ . However, there are other more direct and safer ways to signal submission.

R.T. Abed and colleagues ⁴⁴⁷ , ⁴⁴⁸ emphasize the competition among women to have body shapes that will make them desirable. In ancestral societies, the substantial physical effort needed to get limited supplies of food kept body shape variations small. In modern environments, human food preferences have shaped industries that provide ready access to cheap foods with whatever combinations of fat, salt, sugar, protein, taste and texture that people prefer. The resulting epidemic of obesity makes appearance more important than ever in sexual competition. The effect is magnified by mass media portrayals of body shapes that are caricatures of idealized extremes. Restrictive dieting seems like an obvious strategy to get a good mate and to also gain admiration for self‐control. This sexual competition hypothesis provides a convincing explanation for extreme dieting and its excess prevalence in women in modern societies, but it does not fully explain eating disorders initiated by dieting for other motives, bulimia, and why the eating control system is so vulnerable to malfunction.

The eating control system was shaped to ensure protection against starvation in a trade‐off with avoiding risks from being heavy and slow. Starvation is the more potent selection force, so protection against obesity is relatively weak. However, Nettle et al ⁴⁴⁹ note that natural selection has shaped a system that adjusts fat storage to food availability. When sufficient food supplies are reliably available, extra caloric stores are a useless burden. When food supplies are limited or erratic, the system motivates finding food, consuming it fast, and increasing the body weight set point to provide insurance against starvation.

Severe caloric restriction arouses the famine protection response, but attempts to block its effects can initiate a positive feedback cycle that sends the system out of control ⁴⁵⁰ . Intense efforts at caloric restriction inevitably end in out‐of‐control eating episodes that magnify the fear of obesity and motivate redoubled commitments to restrict intake. The increased weight setpoint amplifies the fear. This combines with more hunger experienced at higher weights than previously, to spiral the system out of control.

Most people revert to their usual eating habits and weight after a period of deviation in either direction. Some persist in the patterns of bingeing and purging that characterize bulimia. A few control their eating, or at least their weight, by restriction, purging, extreme exercise, and the preoccupation about eating and body weight that characterize anorexia nervosa.

These ideas should be helpful in preventing eating disorders, as well as in their treatment. Learning that severe caloric restriction may finally result in weight gain should be a potent antidote to behavior patterns that initiate eating disorders. However, as in the case of substance abuse, the belief that conscious resolve can control behavior makes it hard to convince people that just deciding to stop eating is not necessarily a route to persistent thinness. Once established, eating disorders are much harder to control, because they give rise to a sense of identity that is tangled with eating and body weight, a sense of superiority compared to those with less self‐control, and the determination to defy parents who are desperate to get their child to eat.

THE PERSISTENCE OF DELETERIOUS GENETIC VARIATION

The persistence of disease‐causing alleles in the face of natural selection has been recognized as a paradox since the origins of evolutionary genetics ⁴⁵¹ . Proposed explanations have inspired debate for decades, but resolution now seems within reach, thanks to newly available genetic datasets and methods. Schizophrenia is the focus of the discussion below because it is the disorder that has generated most interest and research, but the general principles are also relevant for other disorders.

As recently as the turn of the millennium, there was hope that we would soon find the genes responsible for highly heritable diseases such as schizophrenia and bipolar disorder, but these expectations have been consistently frustrated. Instead of arising from common variants with large effects that code for proteins, the majority of the risk instead arises from thousands of non‐coding alleles with tiny effects ¹³³ . The prevalence of variants is inversely proportional to their effect size, a pattern consistent with purifying selection eliminating mutations with larger effects faster than those with smaller effects ⁴⁵² . Instead of being localized on some chromosomes, the loci associated with schizophrenia are scattered across the genome, with numbers on each chromosome proportional to the chromosome size ⁴⁵³ .

Larger sample sizes and new family methods are now looking for, and finding, rare variants with larger effects, especially copy number variants and de novo mutations ⁴⁵⁴ . These variants are estimated to account for only about 20% of the heritability of schizophrenia, but they could identify relevant neural circuits, and possibly suggest new treatments. As progress is made to identify all variants that increase the risk of schizophrenia, it is worth asking a large question: will finding them all provide a full explanation for schizophrenia? Most probably, it will also require understanding why the mind is vulnerable to the failure mode of schizophrenia. This may simply be the syndrome that results from a certain pattern of disorganized brain development. It is also possible, however, that the vulnerability results from a trait pushed to a performance peak despite associated risks of failure.

For fitness functions with a cliff edge, natural selection will push the trait to the point close to the peak that maximizes gene transmission over multiple generations, despite the low fitness experienced by the few individuals with values over the cliff ⁴⁵⁵ , ⁴⁵⁶ . The situation gets more interesting when you consider that oscillation is expected between the value that maximizes single generation fitness despite costs to offspring and the value that maximizes gene transmission over multiple generations.

Previous hopes that specific genetic constellations would define specific disorders have also been upended. The diagnostic distinction between schizophrenia and bipolar disorder turns out to be far less crisp than had been assumed. Their genetic correlation of 72% arises from the many alleles that increase the risk of both disorders ²⁹⁹ . Genetic correlations are pervasive among all mental disorders, much more so than for neurological disorders ²⁹⁹ , ⁴⁵⁷ . However, this is a fast‐developing area, and current methods neglect the role of assortative mating in overestimating genetic correlations ⁴⁵⁸ .

These findings are consistent with the hypothesis that mutation‐selection balance is responsible for the persistence of most disease‐causing alleles. Human individuals each have about 70 mutations that are not present in their parents, one of which, on average, is in a protein coding region. Beneficial new mutations are exceedingly rare. Deleterious ones are selected out with a speed proportional to their reduction in fitness, but new ones replace them, maintaining the balance between mutation and selection.

However, interesting evolutionary questions remain. Are some alleles maintained by selection which gives advantages in some individuals or situations balancing their costs in others? Are increased intelligence or creativity associated with some alleles for schizophrenia, autism or bipolar disorder? Do alleles with immune functions that help to shape the developing brain also protect from infection?

All these questions are of interest, but the larger evolutionary question is why certain systems are so vulnerable to failing in typical ways. Are some systems intrinsically vulnerable to failure because they have been shaped to a performance peak adjacent to a cliff edge where fitness plummets? Are some control systems shaped to high gain despite the unavoidable risk of instability? Answering these questions is an important long‐term project.

Polygenic risk scores for bipolar disorder and schizophrenia can predict creativity scores ⁴⁵⁹ , but it is very hard to tell if measures of creativity might be confounded by the kinds of jobs open to people who have severe disorders. Cognitive ability can be measured more reliably. Of 75 genomic loci jointly associated with schizophrenia and intelligence, 81% were associated with lower cognitive performance, while of 12 alleles associated with bipolar disorder and intelligence, 75% were associated with higher performance ⁴⁶⁰ . These are intriguing clues to an unsolved mystery.

A different approach considers the possible role of rapid selection for traits that became useful during the major transition to the social cultural niche in the past few hundred thousand years. T.J. Crow wrote extensively about the possibility that psychosis could be the price we pay for the capacity for language ⁴⁶¹ , ⁴⁶² . We now are on the verge of confirming that some health problems can be attributed to wrenching major transitions in which a new niche or strategy selects strongly for traits that make other traits vulnerable because of anatomic, physiologic or pleiotropic constraints. The exemplar is the transition to bipedality and its legacy of vulnerability to hernias, hemorrhoids, back pain, knee pain, plantar fasciitis, varicose veins, and omental torsion ⁴⁶³ . It is painful to imagine how prevalent these problems must have been in the first million years of bipedality.

The wrenching transition to the cognitive social niche may have created even more severe problems, considering the path‐dependent interactions of multiple alleles that influence brain development pathways ⁴⁶⁴ , ⁴⁶⁵ , ⁴⁶⁶ . Imagine a new allele changing the chemical gradients that influence neuronal migration during brain development in ways that give a benefit, perhaps something like more expressive vocalization. If this gives a net selective advantage, the allele will be selected for, despite negative effects that slightly disrupt multiple other adaptations that evolved previously.

A recently proposed model ⁴⁶⁷ suggests that the development of social brain, language and high‐order cognitive functions transformed many neutral alleles into risk alleles for schizophrenia. Around 100,000‐150,000 years ago, there was a “turning point” when the number of those alleles plateaued. A steady decline then began due to natural selection, that also increased the proportion of protective alleles. This hypothesis is supported by the evidence that older alleles are more likely to increase the risk of schizophrenia and newer ones are more likely to decrease it, and by some epidemiological evidence suggesting that the incidence of schizophrenia is declining ³¹⁷ .

CONCLUSIONS

The main conclusion of this overview is simple: evolutionary biology is a basic science that has a lot to contribute to psychiatry. It provides a scientific framework that is missing from psychiatry but foundational for the rest of behavioral science. It is not an alternative to the search for brain mechanisms; it is a complementary framework that can integrate diverse bio‐psycho‐social approaches. It is not a new method of treatment, but it helps all methods of treatment by putting emotional disorders in the context of normal functioning. For patients, this reduces stigma and tendencies to self‐identify as diseased persons. For clinicians, this offers new ways to describe mental disorders and how treatments work. For researchers, it offers new questions whose answers will inspire new approaches to research on brain mechanisms. These practical benefits of evolutionary psychiatry are ready for application, but a larger possible conclusion deserves consideration.