Why We Sleep

by Matthew Walker

  The gift of split-brain deep NREM sleep is not entirely unique to aquatic mammals. Birds can do it, too. However, there is a somewhat different, though equally life-preserving, reason: it allows them to keep an eye on things, quite literally. When birds are alone, one half of the brain and its corresponding (opposite-side) eye must stay awake, maintaining vigilance to environmental threats. As it does so, the other eye closes, allowing its corresponding half of the brain to sleep.

  Things get even more interesting when birds group together. In some species, many of the birds in a flock will sleep with both halves of the brain at the same time. How do they remain safe from threat? The answer is truly ingenious. The flock will first line up in a row. With the exception of the birds at each end of the line, the rest of the group will allow both halves of the brain to indulge in sleep. Those at the far left and right ends of the row aren’t so lucky. They will enter deep sleep with just one half of the brain (opposing in each), leaving the corresponding left and right eye of each bird wide open. In doing so, they provide full panoramic threat detection for the entire group, maximizing the total number of brain halves that can sleep within the flock. At some point, the two end-guards will stand up, rotate 180 degrees, and sit back down, allowing the other side of their respective brains to enter deep sleep.

  We mere humans and a select number of other terrestrial mammals appear to be far less skilled than birds and aquatic mammals, unable as we are to take our medicine of NREM sleep in half-brain measure. Or are we?

  Two recently published reports suggest humans have a very mild version of unihemispheric sleep—one that is drawn out for similar reasons. If you compare the electrical depth of the deep NREM slow brainwaves on one half of someone’s head relative to the other when they are sleeping at home, they are about the same. But if you bring that person into a sleep laboratory, or take them to a hotel—both of which are unfamiliar sleep environments—one half of the brain sleeps a little lighter than the other, as if it’s standing guard with just a tad more vigilance due to the potentially less safe context that the conscious brain has registered while awake. The more nights an individual sleeps in the new location, the more similar the sleep is in each half of the brain. It is perhaps the reason why so many of us sleep so poorly the first night in a hotel room.

  This phenomenon, however, doesn’t come close to the complete division between full wakefulness and truly deep NREM sleep achieved by each side of birds’ and dolphins’ brains. Humans always have to sleep with both halves of our brain in some state of NREM sleep. Imagine, though, the possibilities that would become available if only we could rest our brains, one half at a time.

  I should note that REM sleep is strangely immune to being split across sides of the brain, no matter who you are. All birds, irrespective of the environmental situation, always sleep with both halves of the brain during REM sleep. The same is true for every species that experiences dream sleep, humans included. Whatever the functions of REM-sleep dreaming—and there appear to be many—they require participation of both sides of the brain at the same time, and to an equal degree.

  UNDER PRESSURE

  The fourth and final difference in sleep across the animal kingdom is the way in which sleep patterns can be diminished under rare and very special circumstances, something that the US government sees as a matter of national security, and has spent sizable taxpayer dollars investigating.

  The infrequent situation happens only in response to extreme environmental pressures or challenges. Starvation is one example. Place an organism under conditions of severe famine, and foraging for food will supersede sleep. Nourishment will, for a time, push aside the need for sleep, though it cannot be sustained for long. Starve a fly and it will stay awake longer, demonstrating a pattern of food-seeking behavior. The same is true for humans. Individuals who are deliberately fasting will sleep less as the brain is tricked into thinking that food has suddenly become scarce.

  Another rare example is the joint sleep deprivation that occurs in female killer whales and their newborn calves. Female killer whales give birth to a single calf once every three to eight years. Calving normally takes place away from the other members of the pod. This leaves the newborn calf incredibly vulnerable during the initial weeks of life, especially during the return to the pod as it swims beside its mother. Up to 50 percent of all new calves are killed during this journey home. It is so dangerous, in fact, that neither mother nor calf appear to sleep while in transit. No mother-calf pair that scientists have observed shows signs of robust sleep en route. This is especially surprising in the calf, since the highest demand and consumption of sleep in every other living species is in the first days and weeks of life, as any new parent will tell you. Such is the egregious peril of long-range ocean travel that these infant whales will reverse an otherwise universal sleep trend.

  Yet the most incredible feat of deliberate sleep deprivation belongs to that of birds during transoceanic migration. During this climate-driven race across thousands of miles, entire flocks will fly for many more hours than is normal. As a result, they lose much of the stationary opportunity for plentiful sleep. But even here, the brain has found an ingenious way to obtain sleep. In-flight, migrating birds will grab remarkably brief periods of sleep lasting only seconds in duration. These ultra–power naps are just sufficient to avert the ruinous brain and body deficits that would otherwise ensue from prolonged total sleep deprivation. (If you’re wondering, humans have no such similar ability.)

  The white-crowned sparrow is perhaps the most astonishing example of avian sleep deprivation during long-distance flights. This small, quotidian bird is capable of a spectacular feat that the American military has spent millions of research dollars studying. The sparrow has an unparalleled, though time-limited, resilience to total sleep deprivation, one that we humans could never withstand. If you sleep-deprive this sparrow in the laboratory during the migratory period of the year (when it would otherwise be in flight), it suffers virtually no ill effects whatsoever. However, depriving the same sparrow of the same amount of sleep outside this migratory time window inflicts a maelstrom of brain and body dysfunction. This humble passerine bird has evolved an extraordinary biological cloak of resilience to total sleep deprivation: one that it deploys only during a time of great survival necessity. You can now imagine why the US government continues to have a vested interest in discovering exactly what that biological suit of armor is: their hope for developing a twenty-four-hour soldier.

  HOW SHOULD WE SLEEP?

  Humans are not sleeping the way nature intended. The number of sleep bouts, the duration of sleep, and when sleep occurs have all been comprehensively distorted by modernity.

  Throughout developed nations, most adults currently sleep in a monophasic pattern—that is, we try to take a long, single bout of slumber at night, the average duration of which is now less than seven hours. Visit cultures that are untouched by electricity and you often see something rather different. Hunter-gatherer tribes, such as the Gabra in northern Kenya or the San people in the Kalahari Desert, whose way of life has changed little over the past thousands of years, sleep in a biphasic pattern. Both these groups take a similarly longer sleep period at night (seven to eight hours of time in bed, achieving about seven hours of sleep), followed by a thirty- to sixty-minute nap in the afternoon.

  There is also evidence for a mix of the two sleep patterns, determined by time of year. Pre-industrial tribes, such as the Hadza in northern Tanzania or the San of Namibia, sleep in a biphasic pattern in the hotter summer months, incorporating a thirty- to forty-minute nap at high noon. They then switch to a largely monophasic sleep pattern during the cooler winter months.

  Even when sleeping in a monophasic pattern, the timing of slumber observed in pre-industrialized cultures is not that of our own, contorted making. On average, these tribespeople will fall asleep two to three hours after sunset, around nine p.m. Their nighttime sleep bouts will come to an end just prior to, or soon after,
dawn. Have you ever wondered about the meaning of the term “midnight”? It of course means the middle of the night, or, more technically, the middle point of the solar cycle. And so it is for the sleep cycle of hunter-gatherer cultures, and presumably all those that came before. Now consider our cultural sleep norms. Midnight is no longer “mid night.” For many of us, midnight is usually the time when we consider checking our email one last time—and we know what often happens in the protracted thereafter. Compounding the problem, we do not then sleep any longer into the morning hours to accommodate these later sleep-onset times. We cannot. Our circadian biology, and the insatiable early-morning demands of a post-industrial way of life, denies us the sleep we vitally need. At one time we went to bed in the hours after dusk and woke up with the chickens. Now many of us are still waking up with the chickens, but dusk is simply the time we are finishing up at the office, with much of the waking night to go. Moreover, few of us enjoy a full afternoon nap, further contributing to our state of sleep bankruptcy.

  The practice of biphasic sleep is not cultural in origin, however. It is deeply biological. All humans, irrespective of culture or geographical location, have a genetically hardwired dip in alertness that occurs in the midafternoon hours. Observe any post-lunch meeting around a boardroom table and this fact will become evidently clear. Like puppets whose control strings were let loose, then rapidly pulled taut, heads will start dipping then quickly snap back upright. I’m sure you’ve experienced this blanket of drowsiness that seems to take hold of you, midafternoon, as though your brain is heading toward an unusually early bedtime.

  Both you and the meeting attendees are falling prey to an evolutionarily imprinted lull in wakefulness that favors an afternoon nap, called the post-prandial alertness dip (from the Latin prandium, “meal”). This brief descent from high-degree wakefulness to low-level alertness reflects an innate drive to be asleep and napping in the afternoon, and not working. It appears to be a normal part of the daily rhythm of life. Should you ever have to give a presentation at work, for your own sake—and that of the conscious state of your listeners—if you can, avoid the midafternoon slot.

  What becomes clearly apparent when you step back from these details is that modern society has divorced us from what should be a preordained arrangement of biphasic sleep—one that our genetic code nevertheless tries to rekindle every afternoon. The separation from biphasic sleep occurred at, or even before, our shift from an agrarian existence to an industrial one.

  Anthropological studies of pre-industrial hunter-gatherers have also dispelled a popular myth about how humans should sleep.fn3 Around the close of the early modern era (circa late seventeenth and early eighteenth centuries), historical texts suggest that Western Europeans would take two long bouts of sleep at night, separated by several hours of wakefulness. Nestled in-between these twin slabs of sleep—sometimes called first sleep and second sleep, they would read, write, pray, make love, and even socialize.

  This practice may very well have occurred during this moment in human history, in this geographical region. Yet the fact that no pre-industrial cultures studied to date demonstrate a similar nightly split-shift of sleep suggests that it is not the natural, evolutionarily programmed form of human sleep. Rather, it appears to have been a cultural phenomenon that appeared and was popularized with the western European migration. Furthermore, there is no biological rhythm—of brain activity, neurochemical activity, or metabolic activity—that would hint at a human desire to wake up for several hours in the middle of the night. Instead, the true pattern of biphasic sleep—for which there is anthropological, biological, and genetic evidence, and which remains measurable in all human beings to date—is one consisting of a longer bout of continuous sleep at night, followed by a shorter midafternoon nap.

  Accepting that this is our natural pattern of slumber, can we ever know for certain what types of health consequences have been caused by our abandonment of biphasic sleep? Biphasic sleep is still observed in several siesta cultures throughout the world, including regions of South America and Mediterranean Europe. When I was a child in the 1980s, I went on vacation to Greece with my family. As we walked the streets of the major metropolitan Greek cities we visited, there were signs hanging in storefront windows that were very different from those I was used to back in England. They stated: open from nine a.m. to one p.m., closed from one to five p.m., open five to nine p.m.

  Today, few of those signs remain in windows of shops throughout Greece. Prior to the turn of the millennium, there was increasing pressure to abandon the siesta-like practice in Greece. A team of researchers from Harvard University’s School of Public Health decided to quantify the health consequences of this radical change in more than 23,000 Greek adults, which contained men and women ranging in age from twenty to eighty-three years old. The researchers focused on cardiovascular outcomes, tracking the group across a six-year period as the siesta practice came to an end for many of them.

  As with countless Greek tragedies, the end result was heartbreaking, but here in the most serious, literal way. None of the individuals had a history of coronary heart disease or stroke at the start of the study, indicating the absence of cardiovascular ill health. However, those that abandoned regular siestas went on to suffer a 37 percent increased risk of death from heart disease across the six-year period, relative to those who maintained regular daytime naps. The effect was especially strong in workingmen, where the ensuing mortality risk of not napping increased by well over 60 percent.

  Apparent from this remarkable study is this fact: when we are cleaved from the innate practice of biphasic sleep, our lives are shortened. It is perhaps unsurprising that in the small enclaves of Greece where siestas still remain intact, such as the island of Ikaria, men are nearly four times as likely to reach the age of ninety as American males. These napping communities have sometimes been described as “the places where people forget to die.” From a prescription written long ago in our ancestral genetic code, the practice of natural biphasic sleep, and a healthy diet, appear to be the keys to a long-sustained life.

  WE ARE SPECIAL

  Sleep, as you can now appreciate, is a unifying feature across the animal kingdom, yet within and between species there is remarkable diversity in amount (e.g., time), form (e.g., half-brain, whole-brain), and pattern (monophasic, biphasic, polyphasic). But are we humans special in our sleep profile, at least, in its pure form when unmolested by modernity? Much has been written about the uniqueness of Homo sapiens in other domains—our cognition, creativity, culture, and the size and shape of our brains. Is there anything similarly exceptional about our nightly slumber? If so, could this unique sleep be an unrecognized cause of these aforementioned accomplishments that we prize as so distinctly human—the justification of our hominid name (Homo sapiens—Latin derivative, “wise person”)?

  As it turns out, we humans are special when it comes to sleep. Compared to Old- and New-World monkeys, as well as apes, such as chimpanzees, orangutans, and gorillas, human sleep sticks out like the proverbial sore thumb. The total amount of time we spend asleep is markedly shorter than all other primates (eight hours, relative to the ten to fifteen hours of sleep observed in all other primates), yet we have a disproportionate amount of REM sleep, the stage in which we dream. Between 20 and 25 percent of our sleep time is dedicated to REM sleep dreaming, compared to an average of only 9 percent across all other primates! We are the anomalous data point when it comes to sleep time and dream time, relative to all other monkeys and apes. To understand how and why our sleep is so different is to understand the evolution of ape to man, from tree to ground.

  Humans are exclusive terrestrial sleepers—we catch our Zs lying on the ground (or sometimes raised a little off it, on beds). Other primates will sleep arboreally, on branches or in nests. Only occasionally will other primates come out of trees to sleep on the ground. Great apes, for example, will build an entirely new treetop sleep nest, or platform, every single night. (Imagine having t
o set aside several hours each evening after dinner to construct a new IKEA bedframe before you can sleep!)

  Sleeping in trees was an evolutionarily wise idea, up to a point. It provided safe haven from large, ground-hunting predators, such as hyenas, and small blood-sucking arthropods, including lice, fleas, and ticks. But when sleeping twenty to fifty feet up in the air, one has to be careful. Become too relaxed in your sleep depth when slouched on a branch or in a nest, and a dangling limb may be all the invitation gravity needs to bring you hurtling down to Earth in a life-ending fall, removing you from the gene pool. This is especially true for the stage of REM sleep, in which the brain completely paralyzes all voluntary muscles of the body, leaving you utterly limp—a literal bag of bones with no tension in your muscles. I’m sure you have never tried to rest a full bag of groceries on a tree branch, but I can assure you it’s far from easy. Even if you manage the delicate balancing act for a moment, it doesn’t last long. This body-balancing act was the challenge and danger of tree sleeping for our primate forebears, and it markedly constrained their sleep.

  Homo erectus, the predecessor of Homo sapiens, was the first obligate biped, walking freely upright on two legs. We believe that Homo erectus was also the first dedicated ground sleeper. Shorter arms and an upright stance made tree living and sleeping very unlikely. How did Homo erectus (and by inference, Homo sapiens) survive in the predator-rich ground-sleeping environment, when leopards, hyenas, and saber-toothed tigers (all of which can hunt at night) are on the prowl, and terrestrial bloodsuckers abound? Part of the answer is fire. While there remains some debate, many believe that Homo erectus was the first to use fire, and fire was one of the most important catalysts—if not the most important—that enabled us to come out of the trees and live on terra firma. Fire is also one of the best explanations for how we were able to sleep safely on the ground. Fire would deter large carnivores, while the smoke provided an ingenious form of nighttime fumigation, repelling small insects ever keen to bite into our epidermis.