The field evidence overturns two opposite myths
One myth says people without electric light naturally sleep eight or nine uninterrupted hours from sunset to sunrise. Another says short field-study averages prove modern adults have been told to sleep too much. Neither follows from the data. Researchers are observing living communities with their own histories, work, food systems, social lives, sleeping places, climates, and exposure to surrounding economies. The studies broaden the range of known human sleep. They do not reveal a single untouched baseline.
The three-society study by Yetish and colleagues measured Hadza in Tanzania, San in Namibia, and Tsimane in Bolivia. Estimated sleep duration ranged from 5.7 to 7.1 hours even though the sleep period was longer. People generally fell asleep several hours after sunset, woke around or before sunrise, and slept nearly an hour longer in winter in the groups with enough seasonal contrast. Extended middle-of-the-night waking was uncommon in those samples.
| Evidence | Population and measure | Main observation | Boundary |
|---|---|---|---|
| Yetish et al., 2015 | Hadza, San, and Tsimane; wrist actigraphy | 5.7–7.1 hours asleep; later than sunset; seasonal timing linked to light and temperature | Three contemporary tropical or subtropical societies cannot represent all human history |
| Samson et al., 2017 | 33 Hadza volunteers; 393 total recording days | About 6.25 hours, low measured efficiency, flexible timing, and opportunistic naps | Actigraphy estimates sleep from movement and one community cannot define a species norm |
| Samson and McKinnon, 2025 | 54 industrial and non-industrial population studies; more than 5,000 participants | Industrial samples slept longer and more efficiently; non-industrial samples showed stronger circadian function | A study-level synthesis combines countries, devices, years, and protocols rather than tracking matched individuals |
| Toba/Qom electrification studies | Community comparisons and longitudinal actigraphy in northern Argentina | Electricity accompanied later, shorter, and less regular sleep, especially through later onset | Electrification changes technology, activity, work, and social life together, so light is not the only possible cause |
Hadza sleep was short, fragmented, and flexible
The dedicated Hadza sleep-biology study collected a total of 393 days of actigraphy recordings from 33 volunteers. Estimated night sleep averaged about 6.25 hours, and measured sleep efficiency was low. Activity, age, light, moon phase, day length, and nighttime temperature were associated with parts of the pattern. Naps appeared on many days, but not as a rigid planned schedule. The authors described flexible sleep-wake behavior embedded in ecology rather than one fixed block.
That finding matters because sleep efficiency is often treated as a universal score. A noisy, socially active camp may produce movement and awakenings while sleepers still judge their sleep in the context of ordinary life. This does not make fragmentation harmless. It shows that a laboratory-derived metric and a person's appraisal can describe different parts of the same night.
Industrial sleep may be longer but less strongly timed
A broader comparison found longer sleep in industrial societies. The 2025 analysis of 54 population studies estimated 7.1 hours for industrial samples and 6.4 hours for non-industrial samples. Industrial sleep was also more efficient. A separate analysis found stronger circadian function in non-industrial populations. Comfortable, protected bedrooms may support longer consolidated sleep while electric light, indoor days, fixed schedules, and evening activity weaken alignment with environmental time cues.
This is a population-level tradeoff hypothesis, not proof that any particular city resident has a weak body clock. The source studies span decades and use different equipment and sampling frames. Country, age, sex, climate, and culture remain intertwined. The result is most useful as a warning against treating duration as the whole of sleep health.
Electrification shows change in real time
Two studies of Toba/Qom communities make the transition more concrete. A 2015 field comparison associated access to electric light with later sleep onset and roughly 40 to 60 minutes less sleep, depending on season. The samples were small and compared communities at one stage of change.
A later longitudinal preprint covering 2012 to 2024 assembled more than 12,000 nightly records and followed electrification as it spread. Sleep shifted later, became less regular, and shortened in the rural community. The record is unusually valuable, but it remains observational and used several actigraph devices and algorithms over time. Electricity arrived with phones, television, social changes, and new options after dark. The study supports a transition; it cannot assign every minute to light alone.
What evolutionary sleep research can and cannot say
Comparative primate work suggests that humans evolved unusually short, consolidated, REM-rich sleep relative to predictions based on other primates. The evolutionary review by Samson and Nunn connects changes in sleeping sites, predation risk, social learning, and cognition with that pattern. These are testable evolutionary explanations, not a clinical protocol. Fossils do not preserve sleep stages, and contemporary behavior cannot be projected directly backward tens of thousands of years.
Lessons worth bringing into a modern bedroom
- Strengthen daytime and morning light. The field studies repeatedly place sleep inside a strong natural light-dark cycle, even when total duration is short.
- Let nights become darker and cooler. Light and falling temperature are recurring timing signals, though neither has one universal dose or thermostat setting.
- Allow some flexibility without engineering deprivation. An occasional nap or variable bedtime is different from cutting nightly opportunity to imitate a population average.
- Judge sleep in context. Duration, timing, regularity, awakenings, safety, and daytime function describe different dimensions.
Human sleep remains responsive to environment and culture. Modern bedrooms improve safety, temperature control, and continuity, while modern schedules can detach sleep from natural light and temperature cycles. These comparisons help explain that variation; they do not prescribe a six-hour sleep schedule.