Is sleep a trait that has been shaped by natural selection? There are several reasons to think so. First, the trait is widespread among animals and perhaps universal among vertebrates. In some animals that seem not to sleep, such as dolphins, one half of the brain in fact sleeps while the other stays awake, possibly because they must repeatedly swim to the surface to breathe. Second, all vertebrates seem to share the same sleep regulation mechanisms, with the center that controls dreaming sleep consistently located in the ancient parts of the brain. Third, the patterns of mammalian sleep, with its periods of rapid eye movement and rapid brain waves, are also shared with birds, whose evolution diverged from ours before the time of the dinosaurs. Fourth, the wide variation in the actual patterns of sleep, even in closely related mammals, suggests that whatever kind of sleeping was done by our most recent common ancestor could evolve rapidly to match the species’ particular ecological niche. Finally, if deprived of sleep, all animals function poorly.
In order to better understand sleep difficulties, we would like to understand how the capacity and necessity for sleep increase fitness. One major contribution to the problem came in 1975 from British biologist Ray Meddis, who proposed that the amount and timing of our sleep are set by our potential for productive activity in different phases of the day-night cycle. As one reviewer of Meddis’s book put it, our motivation to sleep at night arises from the desirability of staying off the streets. If there are special dangers in being abroad in the dark and little likelihood of positive accomplishment then, we are better off resting. This explains why humans and other animals benefit from a daily cycle of activity, but it does not explain why we sleep instead of just spending the night quietly awake, ready for any opportunities or dangers that may arise. It also does not explain why we have become so dependent on sleep that its lack makes us barely able to function.
Here is one possible perspective on the evolutionary origins of sleep. Imagine that some distant ancestor needed no sleep. If one line of its descendents had experienced greater dangers at one part of the day-night cycle (let’s assume for simplicity that it was night) and greater opportunities during the day, then individuals who were inactive at night would have had a fitness advantage. As the species gradually came to confine its activity to the daylight hours, its nocturnal quiescence grew ever more prolonged and profound until it reliably spent many hours of every night inactive.
Given such a reliable daily period of inactivity, other evolutionary factors would be expected to act. It is unlikely that all needed cellular maintenance activities would proceed equally well whether an animal were awake or asleep. If some needed processes worked more efficiently when the brain was disengaged from its usual tasks, selection would act to delay them during the wakeful day and catch up during the night, thus favoring development of the state we recognize as sleep. In this way, as suggested in 1969 by Ian Oswald of Edinburgh University, some brain maintenance processes would be confined more and more to sleep and we would become more and more dependent on sleep. During this period, of course, it would be necessary for sleeping individuals to be quite safe, otherwise sleep would quickly have been selected against. Just as we became dependent on getting vitamin C from foods only because we could reliably get plenty of it, the steady availability of a period of safe rest was necessary before certain bodily maintenance mechanisms could be carried out only during sleep. One implication is that a search for metabolic processes confined to sleep, or taking place at a much greater rate during sleep, will provide insights on why we need to sleep. Indeed, brain scans have shown that protein synthesis is greatest during dreamless sleep and that mechanisms for synthesizing certain neurotransmitters can’t keep up with daytime utilization and therefore must catch up at night. Furthermore, cell division is fastest in all tissues during sleep.
Once sleep was established for physiological repair, natural selection might well have relegated other functions to this period. Those most often suggested have been the memory-regulation functions. Researchers Allan Hobson and Robert McCarley have argued that dreaming sleep supports the physiology that consolidates learning. Francis Crick and Graeme Mitchison have evidence that dreaming sleep functions to purge unnecessary memories, much as we periodically discard unnecessary files from our computers. We won’t consider these suggestions in detail but will only point out only that these are not necessarily mutually exclusive alternatives, nor are they at odds with Oswald’s idea that sleep evolved as a period of tissue repair. None of this contradicts Meddis’s observation that sleep regulates activity periods depending on the animal’s ecology. Like other traits, sleep undoubtedly has many important functions. While each hypothesized function needs to be tested, support for one alternative provides evidence against another only if the functions are incompatible. Studies of sleep patterns in many different animal groups in relation to their ways of life and evolutionary relationship to one another could provide helpful evidence.
Now that we are seldom threatened by nocturnal predators such as tigers, and now that artificial light makes productive activity possible throughout the night, the need for regular sleep has become a great bother, especially when we fly across the world and our bodies insist on living according to our original time zone. Looking for the functions of sleep may well provide the knowledge we need to adapt it better to our present needs—or, at the very least, to make it possible to read in the evening without falling asleep and then to sleep soundly through the night despite our worries about the crises tomorrow might bring.
DREAMING
Dreaming has interested people since the dawn of history and no doubt through much of prehistory. In recent years, many theories have been proposed about the functions of dreams, from Freud’s theory of dreams fulfilling forbidden wishes to Francis Crick’s theory that dreams erase and reorganize memories. But the debate has been so inconclusive that some current major authorities, like Harvard’s Allan Hobson, can still argue that dreams may have no specific function but are mainly epiphenomena of brain activities. This seems unlikely to us, given the simple observation that deprivation of dreaming sleep causes severe psychopathology. For instance, cats kept on tiny islands in a pool were able to sleep, but the loss of muscle tone that accompanies dreaming sleep slipped them into the water and woke them. Such deprivation of dreaming sleep made these unfortunate cats wild and hyper sexual and shortened their lives.
Even without delineating the function of dreams, an evolutionary approach can contribute to their understanding. Donald Symons, an evolutionary anthropologist at the University of California (Santa Barbara), recently proposed that there are, for evolutionary reasons, serious constraints on the stimuli we experience in dreams. While individual sleep behavior varies enormously, we tend, in dreams, to experience a wealth of our own actions and of sights but very little sound, smell, or mechanical stimulation. We can dream about doing things without actually moving because our motor nerves are paralyzed when we are in the kind of sleep that permits dreaming. We remember what people in dreams look like and what they tell us, but we do not remember as easily what their voices sounded like. We may remember enjoying a dreamworld glass of wine, but we often cannot recall its bouquet. We can dream that someone strikes us but may not remember what it felt like.
The reason for these constraints, Symons suggests, is that they were required by Stone Age realities. We could afford visual hallucinations, because closed eyes made sight useless; it was too dark for effective vision anyhow. By contrast, a cry of alarm, the smell of a tiger, or the panicky grasp of a child were important cues that required unimpaired vigilance of our senses of hearing, smell, and touch. Some species sleep with their eyes open, but we sleep with our ears open: we cannot let our dreams distract us from important sounds. Symon’s theory explains some of the peculiarities of dreaming (and predicts some not yet noticed), and it will stand or fall according to how well its expectations conform to actual findings on the sensory composition of dreams. So far it seems to account for most
of the available evidence.
THE FUTURE OF PSYCHIATRY
Psychiatry has recently emulated the rest of medicine by devising clear (if somewhat arbitrary) diagnostic categories, reliable methods of measuring symptoms, and standard requirements for experimental design and data analysis. Psychiatric research is now just as quantitative as that in the rest of medicine. Has all this apparent rigor brought psychiatry acceptance as just another medical specialty like neurology, cardiology, or endocrinology? Hardly. The research findings are solid, but they are not connected in any coherent theory. In its attempt to emulate other medical research by searching for the molecular mechanisms of disease, psychiatry has ironically deprived itself of precisely the concepts that provide the tacit foundation for the rest of medical research. By trying to find the flaws that cause disease without understanding normal functions of the mechanisms, psychiatry puts the cart before the horse.
Research on the anxiety disorders exemplifies the problem. Psychiatrists now divide anxiety disorders into nine subtypes, and many researchers treat each as a separate disease, investigating its epidemiology, genetics, brain chemistry, and response to treatments. The difficulty is, of course, that anxiety is not itself a disease but a defense. To appreciate the problems this creates, imagine what would happen if doctors of internal medicine studied cough the way modern psychiatrists study anxiety. First, internists would define “cough disorder” and create objective criteria for diagnosis. Perhaps the criteria would say you have cough disorder if you cough more than twice per hour over a two-day period or have a coughing bout that lasts more than two minutes. Then researchers would look for subtypes of cough disorder based on factor-analytic studies of clinical characteristics, genetics, epidemiology, and response to treatment. They might discover specific subtypes of cough disorders such as mild cough associated with runny nose and fever, cough associated with allergies and pollen exposure, cough associated with smoking, and cough that usually leads to death. Next, they would investigate the causes of these subtypes of cough disorder by studying abnormalities of neural mechanisms in people with cough disorders. The discovery that cough is associated with increased activity in the nerves that cause the chest muscles to contract would stimulate much speculation about what neurophysiological mechanisms could make these nerves overly active. The discovery of a cough-control center in the brain would give rise to another set of ideas as to how abnormalities in this center might cause cough. The knowledge that codeine stops cough would lead other scientists to investigate the possibility that cough results from deficiencies in the body’s codeinelike substances.
Such a plan of research is obviously ludicrous, but we recognize its folly only because we know that cough is useful. Because we know that cough is a defense, we look for the causes of cough not in the nerves and muscles that generate a cough, or even in the brain mechanisms that regulate cough, but instead in the situations and stimuli that normally arouse the protective cough reflex. While some rare cases of cough may be caused by abnormalities of the cough-regulation mechanisms, the vast majority are adaptive responses that expel foreign matter from the respiratory tract. Only after searching for such a natural stimulus does a physician consider the possibility that the cough-regulation mechanism itself might be awry.
Many psychiatrists have studied individual differences in susceptibility to anxiety with the worthy goal of helping the many people who experience panic, tension, fear, and sleeplessness throughout their lives. Nonetheless, this approach fosters much confusion. What if research on cough were to focus on those individuals who have a lifelong tendency to cough in response to the least stimulus? Such people would be told they have a cough disorder. Soon there would be campaigns to identify people predisposed to cough disorder in order to find the genes that cause this abnormality in the cough-regulation mechanism. There undoubtedly are people with a genetic susceptibility to ready coughing, but studying them would tell us little about the cause of most coughs.
There are limits to this analogy. Anxiety is much more complicated than cough, its functions are less obvious, and it varies much more from individual to individual. More important, the cues that arouse anxiety are far less tangible than those that arouse cough. Cough is caused by foreign material in the respiratory tract, while anxiety is aroused by diverse cues processed by the mind in mysterious ways. The most obvious anxiety cues are images of dangerous objects or stimuli that have been paired with pain or some other noxious stimulus. Most clinical anxiety is aroused, however, by complex cues that require subtle interpretation. If, for example, the boss doesn’t greet you, you are not invited to a meeting, and a friend avoids you on a day when layoff notices are to be distributed, you may feel serious apprehension. If it is your birthday, however, and you suspect a surprise party may be in the works, the same stimuli will arouse a very different reaction. This example only begins to tap the complexity of the mental systems that regulate anxiety. Many wishes and feelings never make it to consciousness but nonetheless cause anxiety. The woman whose panic attacks started when she began an affair insisted that the two were unrelated. Just because many of the cues that cause anxiety are hard to identify does not mean that they are not there, and it certainly does not mean that the anxiety they cause is useless or a product of abnormal brain mechanisms.
Conversely, just because much anxiety is normal, that does not mean it is all useful. Furthermore, many anxiety disorders are caused by genetic predispositions. We don’t yet know whether these are best understood as genetic defects or normal variations. Certainly, the kinds and dangerousness of various threats vary considerably from one generation to the next, and this should maintain considerable genetic variation in the anxiety-regulation mechanisms.
If psychiatry stays on its current course, it will be left treating only those disorders caused by demonstrable brain defects, while the pains and suffering of everyday life will be left to other clinicians. This would be unfortunate for patients as well as psychiatrists. The rest of medicine treats normal defensive reactions; why shouldn’t psychiatry do the same? In this as well as other ways, an evolutionary view is psychiatry’s route to genuine integration with the rest of medicine. An intensive effort to understand the functions of the emotions and how they are normally regulated would provide, for psychiatry, something comparable to what physiology provides for the rest of medicine. It would provide a framework in which pathopsychology could be studied like pathophysiology, so that we can understand what has gone wrong with the normal functioning of bodily systems. There is every expectation that an evolutionary approach will bring the study of mental disorders back to the fold of medicine, relying not on a crude “medical model” of emotional problems but on the same Darwinian approach that is so useful in the rest of medicine.
15
THE EVOLUTION OF MEDICINE
Nothing in biology makes sense except in the light of evolution.
—Theodosius Dobzhansky, 1973
You are crossing a heath on a well-worn path when a flash of early sunlight reflects from something lying over by an older trail. You follow the gleam to its source, brush away some dirt, and discover an old-fashioned gold pocket watch. Perhaps it is the same old watch that people have been finding for two centuries, but some details have been overlooked.
Its perfection still elicits wonder. The seam around the case is all but invisible; the crystal is symmetrical and gleaming; the chain is made of exquisitely miniature gold links. The face has numerals sharply etched around the logo of the Lifetime Watch Company. But even as you admire the watchmaker’s skill, the light reveals some surprising imperfections. The crystal is laced with slight distortions. And the chain, though beautiful and flexible, is thin and broken, thus explaining why the watch is here and not in a pocket. A notch in the seam is perfectly shaped for a thumbnail but large enough for dirt and water to enter easily. Odd, these flaws. You open the back, and the exquisite mechanism again inspires awe. How could anyone have designed, much less constructed, so
many perfectly cut gears of rustproof brass, the hairlike spring of steel, the balance wheel suspended by minuscule jewels? But when you try to set the watch, the knob is so tiny you can barely grasp it and a dozen twirls advance the hands only a single hour. You shake the watch. It ticks for five seconds, then is stopped by flakes of rust from that steel spring. What an odd device this is! So perfect in many respects, in others makeshift at best. How could the creator of such a masterpiece have allowed so many obvious flaws? Inside the case is an inscription in tiny letters. You take out your magnifying glass and read:
OVERVIEW OF CAUSES OF DISEASE
We now return to where we began, to a seeming incongruity at the core of medicine. Despite their exquisite design, our bodies have crude flaws. Despite our multiple defenses, we have a thousand vulnerabilities. Despite their capabilities for rapid and precise repairs, our bodies inevitably deteriorate and eventually fail. Before Darwin, physicians could only wonder at the incongruity of it all, perhaps with the hope that our bodies are part of an unfathomable divine plan, or with the suspicion that they are some cosmic prank. Ever since Darwin, the incongruity has often mistakenly been attributed to the supposed weakness or capriciousness of natural selection. In the light of modern Darwinism, however, the incongruity unfolds into a sharply blocked tapestry with a place for each of several distinct causes of disease.
Why isn’t the body more reliable? Why is there disease at all? As we have seen, the reasons are remarkably few. First, there are genes that make us vulnerable to disease. Some—though fewer than has been thought—are defectives continually arising from new mutations but kept scarce by natural selection. Other genes cannot be eliminated because they cause no disadvantages until it is too late in life for them to affect fitness. Most deleterious genetic effects, however, are actively maintained by selection because they have unappreciated benefits that outweigh their costs. Some of these are maintained because of heterozygote advantage; some are selected because they increase their own frequency, despite creating a disadvantage for the individual who bears them; some are genetic quirks that have adverse effects only when they interact with a novel environmental factor.
Why We Get Sick Page 28
