Five crew members sleeping aboard Mir gave researchers something space medicine rarely gets: a long run of repeated sleep records before launch, during flight and after return. Across 256 usable nights tied to missions that lasted about six months, the study found that sleep in orbit grew more fragmented, with about an extra hour of wakefulness during the scheduled sleep episode compared with life on Earth.
The paper, published in the Journal of Sleep Research, tracked sleep with the Nightcap monitor before, during and after Mir missions flown between 1996 and 1998. The researchers found shorter total sleep time, lower sleep efficiency, longer time to fall asleep and more wake after sleep onset during flight. They also found a subtler change inside the night itself: REM sleep fell early in the mission, then climbed back toward preflight levels over time while non-REM sleep gave up more of its share.
Sleep architecture can sound technical, but it describes how a night is divided between wakefulness and major sleep stages. Those divisions matter because both REM sleep and non-REM sleep support memory, learning, emotional balance and next-day performance. NASA’s Sleep on Station overview notes that poor sleep quality in orbit can affect attention, concentration, problem-solving, decision-making and judgment.
What the Mir study actually measured
The data set came from five crew members, all men with an average launch age of 43.5 years, who lived aboard Mir for an average of 180 days. Eight people originally agreed to participate, but three only served as backup crew and did not record any inflight sleep. The final analysis kept 112 usable preflight nights, 83 inflight nights and 61 postflight nights, which added up to 256 nights that met the study’s inclusion rules.
Researchers used the Nightcap sleep monitor, a lighter system than full laboratory polysomnography. That mattered because payload limits and operational constraints make space sleep studies difficult. The device could sort each night into wakefulness, REM sleep and non-REM sleep, which gave the team a way to compare the broad architecture of sleep across Earth and orbit even though it could not split non-REM into finer sub-stages.
The authors did more than compare one average night on the ground with one average night in space. They also looked at how sleep changed over time during the mission. That second question is important for long expeditions, because crews might adapt to microgravity, light schedules, noise, workload, or stress in uneven ways rather than following a simple steady pattern.
Wakefulness rose even though sleep opportunity did not shrink
The sharpest result was the growth of wakefulness in orbit. Average total sleep time fell to 5.7 hours inflight from 6.7 hours preflight, even though the time set aside for sleep stayed almost the same at about 7.6 to 7.7 hours. In other words, the problem was not only a shorter slot on the schedule. Crew members spent more of that slot awake.
Sleep efficiency dropped to 73 percent inflight from 89 percent preflight, which the paper described as a decline of roughly 16 percentage points. Sleep onset latency nearly doubled, rising to 54.5 minutes from 28.8 minutes. Wake after sleep onset climbed from about 20 minutes, or 4.5 percent of the sleep opportunity, to about 67 minutes, or 14.5 percent. The team’s summary of NASA sleep countermeasures helps explain why that is a persistent concern in orbit, where lighting, airflow, noise, carbon dioxide, schedules and mission demands can all interfere with rest.
The paper does not pin the blame on one single cause. The discussion points instead to a cluster of likely pressures on Mir, including environmental conditions, stress, circadian misalignment and the realities of living in microgravity for months. That caution matters, because the data show what changed in sleep, while the exact mix of causes still needs follow-up work.
REM and non-REM both fell at first, then the balance shifted
The extra wakefulness took time away from both major sleep states early in flight. REM sleep fell to 19.6 percent of the sleep opportunity from 26.4 percent preflight, while non-REM sleep fell to 53.5 percent from 62.4 percent. In minutes, REM dropped to about 90 minutes from 120 minutes and non-REM dropped to about 250 minutes from 283 minutes.
Mission time then changed that pattern. The paper found that REM sleep was initially reduced and later recovered toward preflight levels by about 180 days in space. That recovery did not happen because the crew started sleeping much longer overall. Total sleep time did not increase significantly over the mission. Instead, REM gained back time while non-REM sleep lost more of its share.
That tradeoff is one of the paper’s most interesting findings because it suggests sleep in orbit may reorganize internally even when the night’s total duration stays short. The authors connect that possibility to long-term sleep regulation, while also noting that the Nightcap device could not show which part of non-REM was being squeezed hardest. A related npj Microgravity study on short shuttle missions found altered sleep spindles and slow waves, which gives one clue about the kinds of deeper non-REM changes future space studies may need to resolve.
Why non-REM losses could be a real operational concern
Non-REM sleep is a broad category and the Mir paper stops short of claiming exactly which substage was lost. That restraint is important. Some Earth studies suggest the brain often tries to preserve the deepest slow-wave sleep under chronic restriction, while stage 2 sleep gives way first. The Mir data cannot prove whether the same pattern happened in space, because the monitor lacked EEG channels detailed enough to split non-REM into subtypes.
Even so, the concern is easy to understand. Stage 2 sleep contains many sleep spindles and spindles have been linked with learning. Slow-wave sleep has its own importance for recovery and brain health. The Mir authors note that both possible losses would matter for people who have to work safely in a demanding spacecraft. Another spaceflight study, reported in Sleep, linked sleep deficiency on six-month International Space Station missions with degraded neurobehavioral function and higher stress.
The Mir findings therefore widen the question from “How long did the crew sleep?” to “What kind of sleep did the crew lose?” A shorter night is already a problem for fatigue. A night that also redistributes time away from one sleep state into another could carry extra consequences for cognition, mood and long-term health, even if the exact mechanism still has to be mapped more precisely.
What returned to normal and what still needs better answers
One encouraging result is that the postflight sleep architecture returned to its preflight distribution in the metrics the team measured. The researchers did not find differences between preflight and postflight phases, which suggests the Mir missions did not leave irreversible changes in these broad sleep measures. That is reassuring, but it should be read alongside the study’s limits rather than as proof that long missions leave no lingering sleep cost.
The limitations are straightforward. The sample was small, the missions were flown on Mir rather than the modern International Space Station and the Nightcap monitor could not separate non-REM into stage 2 and slow-wave sleep. The authors also note that sleep in space may involve unusual eye or head movement patterns, which could affect how a portable monitor classifies REM, non-REM and wakefulness. A classic NASA’s current overview of sleep and irregular-schedule risks and later work on circadian misalignment both show that space sleep research has long had to balance useful data against hardware limits.
The practical message stays clear even with those caveats. Long missions in orbit cut into sleep by increasing wakefulness during the sleep episode and the internal pattern of sleep may keep shifting across the months aloft. For future crews headed toward longer expeditions, that means countermeasures should aim at more than just extending the sleep window. Better schedules, better lighting, better sleep environments and better stage-level monitoring may all be needed if explorers are going to keep both quantity and quality of sleep intact.
