Splitting your sleep into two chunks is not only possible, it appears to be something the human body can adapt to under the right conditions. Laboratory experiments show that when people are placed in long, dark nights mimicking winter, their sleep naturally breaks into two bouts separated by one to three hours of quiet wakefulness. That finding, combined with historical records of a “first sleep” and “second sleep” in pre-industrial Europe, has fueled widespread interest in segmented sleep as an alternative to the single consolidated block most of us aim for. But the story is more complicated than “our ancestors did it, so it must be natural.” Whether a split schedule helps or hurts depends on how much total sleep you get, when the segments fall, and what you are trying to optimize.
What the Long-Night Experiments Actually Showed
The strongest evidence that biphasic sleep is built into human biology comes from a series of studies in the early 1990s led by sleep researcher Thomas Wehr at the National Institute of Mental Health. When healthy volunteers were moved from a standard schedule of 16 hours of light and 8 hours of darkness to a simulated winter schedule with only 10 hours of light and 14 hours of darkness, their sleep expanded and split into two distinct bouts. The sleep episodes were roughly symmetrical, each lasting several hours, with a waking gap of about one to three hours between them.1PubMed. In short photoperiods, human sleep is biphasic Their total sleep time stretched to around 11 hours on the long-night schedule, compared to roughly 7.7 hours under the summer-like schedule.2PubMed. Conservation of photoperiod-responsive mechanisms in humans
The body’s internal chemistry shifted along with the sleep pattern. Melatonin secretion lasted longer under long nights, and so did periods of elevated prolactin. In a related experiment, melatonin duration averaged about 12.5 hours under the winter photoperiod versus about 10.3 hours under summer conditions.3The Journal of Clinical Endocrinology & Metabolism. The Durations of Human Melatonin Secretion and Sleep Respond to Changes in Daylength (Photoperiod) These changes mirror what happens in other mammals exposed to seasonal shifts in day length. The takeaway is that consolidated eight-hour sleep is not the only mode the human clock can run in. Give people enough darkness, and their sleep architecture reorganizes itself into two phases.
The Historical Record Versus Hunter-Gatherer Data
The historian Roger Ekirch famously documented hundreds of references to “first sleep” and “second sleep” in European literature, court records, and medical texts from before the Industrial Revolution. This two-bout pattern was common enough that it had its own vocabulary in multiple languages. People would wake after their first sleep, spend an hour or two praying, reading, chatting, or having sex, and then return to bed for a second round.4PubMed Central. Segmented Sleep in Preindustrial Societies
But when researchers went looking for segmented sleep in modern-day hunter-gatherer societies living without electricity, the pattern did not show up as expected. A landmark study tracked sleep in three pre-industrial groups (the Hadza of Tanzania, the San of Namibia, and the Tsimané of Bolivia) using wrist-worn activity monitors and found that their sleep was not typically interrupted by long waking intervals. Sleep durations ranged from about 5.7 to 7.1 hours, and the sleep period was generally continuous. None of these groups began sleeping near sunset; on average, they fell asleep about 3.3 hours after the sun went down and usually woke before sunrise.5PubMed Central. Natural sleep and its seasonal variations in three pre-industrial societies There was roughly an hour’s difference in sleep between summer and winter, but not the dramatic biphasic split seen in Wehr’s laboratory.
The Hadza, studied separately with more detailed actigraphy, averaged about 6.25 hours of sleep per night with relatively low sleep efficiency. They napped on roughly half of all days, with naps lasting an average of about 48 minutes. Their sleep duration was influenced by activity levels, light exposure, temperature, moon phase, and day length.6PubMed. Hadza sleep biology: Evidence for flexible sleep-wake patterns in hunter-gatherers So the pattern that emerged was not the medieval European two-sleep system but something more flexible: a shortish night of sleep supplemented by opportunistic daytime naps.
How do you reconcile these findings? The laboratory experiments used 14 hours of enforced darkness with nothing to do, which is longer than a typical tropical night. Pre-industrial Europeans at higher latitudes experienced genuinely long winter nights. Equatorial hunter-gatherers do not. The segmented pattern seems to emerge when people have far more darkness than they need for sleep, which depends heavily on geography and season. It is less a universal human default and more a mode the body can slip into when conditions call for it.
How Split Sleep Affects Your Thinking
If you are considering a segmented schedule for practical reasons (shift work, caregiving, military deployment, studying), the question that probably matters most is whether you will be able to think clearly. Research on this has been surprisingly encouraging, with one important caveat: total sleep time matters enormously.
A well-designed study in adolescents compared split sleep (a main nighttime block plus an afternoon nap) against a single consolidated block, at both 8 hours and 6.5 hours of total time in bed. When participants got a full 8 hours, vigilance performance was statistically indistinguishable between the split and continuous groups. When total time was cut to 6.5 hours, the split sleepers actually performed better on sustained-attention tasks than the continuous sleepers, accumulating fewer lapses over the course of the protocol.7Sleep. Cognitive effects of split and continuous sleep schedules in adolescents differ according to total sleep opportunity That result suggests that when you are slightly sleep-restricted, splitting the deficit across two opportunities may be less damaging to alertness than taking the full hit in one shortened block.
Memory encoding showed a similar benefit. In a study that tested factual learning after split versus continuous sleep, participants in the split groups performed significantly better on afternoon memory tasks regardless of whether total sleep was 8 hours or 6.5 hours. The researchers attributed this to the midday nap reducing homeostatic sleep pressure before the afternoon learning session, essentially giving the brain a partial reset.8Scientific Reports. Splitting sleep between the night and a daytime nap reduces homeostatic sleep pressure and enhances long-term memory
These are controlled laboratory studies, mostly in young people on regimented schedules. Real-world split sleeping tends to be messier. You might not have a quiet, dark room for your daytime nap. Your circadian rhythm might fight you if you try to nap at the wrong time of day. Still, the laboratory evidence offers a useful principle: if your schedule forces you to cut total sleep, splitting what you have may protect cognitive function more than you would expect.
The Glucose Problem
Not every outcome tilts in favor of split sleep. One area where researchers have found a genuine downside is blood sugar regulation. In a study of adolescents on sleep-restricted schedules (6.5 hours of total time in bed), those on the split schedule showed significantly larger spikes in blood glucose during an oral glucose tolerance test than those who slept in a single block. This difference appeared during both cycles of sleep restriction and vanished when participants returned to a full 9 hours of sleep.9PubMed Central. Differential effects of split and continuous sleep on neurobehavioral function and glucose tolerance in sleep-restricted adolescents
This matters because impaired glucose tolerance is an early marker on the path toward metabolic problems. It does not mean that splitting your sleep will give you diabetes, but it does mean that the cognitive benefits of split sleep come with a metabolic trade-off under conditions of sleep restriction. When total sleep was adequate, the glucose differences disappeared. The pattern keeps repeating: total sleep time is the variable that matters most. Splitting is a scheduling tool, not a way to cheat on the amount of sleep your body needs.
Oxidative Stress and Longer-Term Health Signals
A separate line of research has looked at biomarkers of oxidative stress in people who habitually sleep in segmented patterns. In one population-based study, individuals with biphasic-napping patterns (the largest group at about 45% of participants) had higher levels of malondialdehyde, a marker of cell damage from oxidative stress, compared to monophasic sleepers. People who slept in a polyphasic pattern (three or more episodes per day) had higher basal oxidizability. Meanwhile, monophasic sleepers had significantly higher total glutathione, which is one of the body’s main antioxidant defenses.10PubMed Central. Elevated oxidative stress biomarkers in adults with segmented sleep patterns
These are cross-sectional findings, meaning they show an association, not a cause. People who split their sleep may differ from consolidated sleepers in other ways that drive the oxidative stress signal. And it is worth noting that in another study of Omani adults, the biphasic-siesta pattern was the most common sleep type (about 35% of participants) and was associated with younger age, while roughly 90% of the sample napped in the afternoon with an average duration of about 45 minutes.11PubMed Central. Sleep Patterns and Quality in Omani Adults In cultures where afternoon napping is the norm, billions of people practice a mild form of split sleep without obvious population-level health consequences. The oxidative stress findings are worth flagging as an open question, not treating as a settled verdict.
Why Extreme Polyphasic Schedules Are a Different Animal
Online communities have long promoted polyphasic schedules like the “Uberman” (six 20-minute naps spread across 24 hours, totaling about 2 hours of sleep) or the “Everyman” (a short core sleep of 3-4 hours plus several naps). These are fundamentally different from the biphasic split being discussed here. They involve radical reductions in total sleep, and the evidence against them is clear.
A National Sleep Foundation consensus panel reviewed the available data and concluded that polyphasic schedules involving significant reductions in total 24-hour sleep and fragmentation into multiple short episodes are associated with a range of adverse outcomes for physical health, mental health, and performance. The panel explicitly recommended against striving to adopt such schedules.12PubMed. Adverse impact of polyphasic sleep patterns in humans: Report of the National Sleep Foundation sleep timing and variability consensus panel
A laboratory study that directly compared polyphasic short sleep (multiple naps totaling about 4-5 hours) to monophasic short sleep of the same duration found that the polyphasic group fared worse. They showed greater impairment in vigilance, particularly in the morning, along with more time spent in light sleep stages and less deep sleep.13PubMed Central. Neurobehavioral functions and sleep architecture during polyphasic and monophasic short sleep schedules Fragmenting already-insufficient sleep into tiny pieces makes things worse, not better. The distinction matters: splitting 7-8 hours into two chunks is very different from splitting 3 hours into six chunks.
The Body Clock Can Learn to Anticipate a Split
One of the more interesting questions about segmented sleep is whether the body truly adapts or just tolerates the disruption. A study of cloistered monks and nuns who have practiced a split-sleep schedule for years provides a striking answer. These religious communities follow a centuries-old tradition of waking for Matins (midnight prayer services), sleeping before and after. When researchers measured their core body temperature rhythms, they found a genuinely biphasic pattern: temperature dropped in the evening, then plateaued or rose during the first sleep bout, dropped again after the midnight waking, and rose normally in the morning. Remarkably, the temperature rise began anticipating the nocturnal awakening by about 85 minutes, even though the monks and nuns still relied on alarm clocks to wake up.14PubMed. Ring the bell for Matins: circadian adaptation to split sleep by cloistered monks and nuns
This suggests that with consistent practice over months and years, the circadian system does not merely cope with a split schedule but actively reorganizes around it, pre-adjusting physiology to match the expected pattern. The monks and nuns did not report excessive daytime sleepiness or health complaints related to their schedule. This is a small, specific population, but it provides real evidence that long-term adaptation to segmented sleep is possible when the schedule is stable and total sleep time is adequate.
Shift Work and Split Sleep Strategies
For people who work rotating or night shifts, split sleep is often less a lifestyle choice and more a survival strategy. A laboratory simulation compared three conditions: consolidated nighttime sleep (10 hours in bed from 10 PM to 8 AM), consolidated daytime sleep (10 hours in bed from 10 AM to 8 PM), and split sleep (two 5-hour opportunities at 3 AM-8 AM and 3 PM-8 PM). The nighttime sleepers got the most total sleep at about 8.4 hours per day. The split sleepers averaged about 7.2 hours, and the daytime-only sleepers got the least at about 6.4 hours. Despite sleeping less than the nighttime group, the split sleepers did not show significantly worse cognitive performance, and they were less sleepy than the daytime-only group.15PubMed. Investigation of the effectiveness of a split sleep schedule in sustaining sleep and maintaining performance
The practical implication is that if you cannot get a full night of sleep because of your work schedule, a split approach that anchors part of your sleep in the nighttime hours and supplements with a second block later can be more effective than trying to cram all your sleep into daytime hours, when circadian drive for wakefulness makes sustained sleep difficult. Biomathematical models of fatigue risk in aviation have explored this principle, simulating how best to distribute limited sleep opportunities during 24-hour layovers, and consistently find that splitting sleep across two well-timed blocks outperforms a single poorly timed block.16SLEEPJ. Fatigue Risk Management in Aviation: Optimal Planning of Sleep Opportunities During 24-Hour Layovers Through Biomathematical Modeling
How Sleep Changes With Age
Your ability to maintain a single consolidated sleep block changes across your lifetime, which has implications for whether a split schedule might actually suit you. Slow-wave (deep) sleep declines markedly during adolescence and continues to drop with age. Meanwhile, sleep consolidation weakens in older adults: nighttime awakenings become more frequent and longer, and the circadian clock tends to shift earlier.17PubMed. Modelling changes in sleep timing and duration across the lifespan: Changes in circadian rhythmicity or sleep homeostasis? Older people’s sleep is particularly vulnerable to disruption during the rising phase of the body temperature rhythm, meaning that the early morning hours become a weak point for maintaining sleep.18PubMed. Contribution of circadian physiology and sleep homeostasis to age-related changes in human sleep
For many older adults, the fragmented nighttime sleep they experience is not a disorder to be fought but a natural drift in sleep architecture. Rather than lying in bed frustrated from 4 AM to 6 AM, some sleep researchers have suggested that accepting a shorter nighttime block and supplementing with an afternoon nap may be a more realistic and less distressing approach. The variability of the relationship between the sleep-wake cycle and circadian rhythms increases with age, and sleep timing naturally shifts earlier.19PubMed. Circadian regulation of human sleep and age-related changes in its timing, consolidation and EEG characteristics A rigid insistence on a single 8-hour block can create anxiety around nighttime wakefulness that makes the problem worse.
Caffeine, Alcohol, and the Quality of Each Bout
If you are going to split your sleep, the substances you use during waking hours matter more, not less, because you have two sleep onsets to protect instead of one. A study of financial traders wearing sleep trackers found that caffeine consumption significantly reduced sleep quantity, while alcohol significantly degraded sleep quality. When both substances were consumed on the same day, the effects partially canceled out in terms of measurable sleep metrics, but not in a way that actually produced good sleep. The sedating effect of alcohol offset some of caffeine’s impact on sleep duration, and caffeine’s stimulation masked some of alcohol’s impact on sleep quality.20PubMed Central. Sleep, alcohol, and caffeine in financial traders
For someone on a split schedule, the timing of caffeine becomes critical. A cup of coffee that clears your system before your nighttime sleep might still be circulating when your afternoon nap window opens. Alcohol consumed in the evening might help you fall asleep for your first bout but fragment the second half of nighttime sleep, which is exactly the interval you cannot afford to lose when your total sleep opportunity is already divided. Managing substance timing on a split schedule requires more forethought than on a standard monophasic one, because each bout is shorter and more vulnerable to disruption.
Individual Variation in Sleep Architecture
Not everyone’s circadian system is wired the same way. Researchers have identified a subgroup of people who display a “bimodal chronotype,” meaning their alertness and body temperature patterns show two peaks rather than the usual single afternoon peak. A population-based study attempted to characterize this phenotype and its relationship to sleep, though the association with a specific genetic variant in the PER3 clock gene did not reach statistical significance.21PubMed. Characterization of bimodal chronotype and its association with sleep: A population-based study The genetics here remain murky, but the existence of the bimodal phenotype suggests that some people may be pre-disposed to function well on a two-bout schedule while others would struggle with it.
Sleep researchers increasingly talk about the two-process model: a homeostatic drive that builds the longer you are awake (Process S) and a circadian rhythm that promotes wakefulness and sleepiness on a roughly 24-hour cycle (Process C).22PubMed Central. The two-process model of sleep regulation: Beginnings and outlook How these two processes interact in any given person determines whether split sleep feels restorative or miserable. If your circadian dip aligns well with your planned nap window and your homeostatic pressure has built up enough to let you fall asleep quickly, a split schedule can feel natural. If your nap window falls during a circadian peak of alertness, you will lie there awake and waste the opportunity. The success of split sleep is, in this sense, a scheduling problem as much as a biological one.