REM is ~20-25% of a healthy adult night, back-loaded toward morning, with awake-like EEG, vivid dreams, and muscle atonia. What it does for memory and emotion, and why short sleep costs REM first.
What REM sleep is (atonia, awake-like EEG, eye movements)
REM (rapid eye movement) sleep is the stage Aserinsky and Kleitman first described in 1953 (Science): bursts of fast eye movement under closed lids, a desynchronized low-voltage EEG that resembles wakefulness, and near-complete skeletal-muscle atonia that prevents you from acting out dreams. It is sometimes called 'paradoxical sleep' because the brain looks awake while the body is paralyzed. Vivid, narrative dreaming concentrates here, though dreaming also occurs in NREM. REM alternates with NREM stages across the night in the ~90-minute cycle.
How much REM you need (~20-25% of the night)
The most-cited normative dataset is Ohayon et al. 2004, a meta-analysis of 65 polysomnography studies, which puts REM at roughly 20-25% of total sleep time in healthy adults. On a 7-8 hour night that is about 90-120 minutes. The fraction is relatively stable across adulthood compared with slow-wave (N3) sleep, which falls steeply with age. There is no single 'target number' to chase on a tracker: total sleep time and an unbroken late-night window are the variables that determine your REM total.
Why REM is back-loaded toward morning
REM is not evenly distributed. The first REM period of the night may last only ~10 minutes; the last, near morning, can run 30-60 minutes. Two systems stack to produce this. Slow-wave (N3) pressure is highest at sleep onset and dissipates with each cycle, freeing later cycles for REM. And REM propensity is gated by the circadian clock, peaking near the core-body-temperature minimum in the last hours before habitual wake (Carskadon & Dement). This is why cutting sleep short by waking 1-2 hours early removes disproportionately more REM than deep sleep.
What REM does (emotional and procedural memory)
REM is implicated in emotional-memory processing, procedural and motor learning, and overnight recalibration of emotional reactivity (Rasch & Born 2013). The complementary picture from the two-process memory model is that slow-wave sleep favors declarative consolidation while REM favors procedural and emotional integration, though the division is not absolute. Selective REM deprivation produces a REM rebound on recovery nights, an increase in REM beyond baseline, which is the clearest evidence that the brain homeostatically defends a REM quota rather than treating it as optional.
What suppresses REM (alcohol, antidepressants, deprivation)
Alcohol suppresses REM in the first half of the night and then fragments the second half with rebound REM and arousals (Ebrahim et al. 2013). Many antidepressants, particularly SSRIs and SNRIs, robustly suppress REM and delay its onset; this is a well-documented pharmacological effect, not a sign the drug is failing. Sleep deprivation and fragmented 'catch-up' sleep preferentially cost the late-night REM that is back-loaded toward dawn. THC blunts REM acutely with a rebound on withdrawal. The practical lever for most people is simply protecting the last two hours of sleep.
Can you increase REM, and do trackers measure it
Mostly you protect REM rather than 'boost' it: keep total sleep adequate, do not wake early, and avoid alcohol before bed. Because REM rides the final cycles, an extra hour of morning sleep yields a large REM gain. As for measurement, consumer wearables (Oura, Whoop, Apple Watch) infer stages from heart rate, heart-rate variability, movement, and temperature, not EEG, so the REM number is an estimate. Against polysomnography, stage-level classification is the weakest output of these devices. [VERIFY: epoch-by-epoch REM agreement vs PSG varies by device and firmware and is commonly well below wake/sleep agreement.] Read trends over weeks, not a single night.
Questions logged on this protocol
What is REM sleep?
REM (rapid eye movement) sleep is a distinct sleep stage marked by bursts of fast eye movement, a fast desynchronized 'awake-like' EEG, vivid narrative dreaming, and near-total skeletal-muscle atonia that stops you from acting out dreams. It was first described by Aserinsky and Kleitman in 1953 (Science). It alternates with the NREM stages (N1, N2, N3) across the night in a roughly 90-minute cycle, and it is sometimes called 'paradoxical sleep' because the brain is active while the body is effectively paralyzed.
How much REM sleep do I need?
About 20-25% of total sleep time in healthy adults, per the Ohayon et al. 2004 normative meta-analysis, which is roughly 90-120 minutes on a 7-8 hour night. Because REM is back-loaded toward morning and the last REM period can run 30-60 minutes, the final 1-2 hours of sleep contribute disproportionately to your REM total. There is no single target number to hit on a tracker: total sleep time and an unbroken late-night window are the variables that matter.
What is the difference between REM and deep sleep?
They are different stages with opposite EEG signatures and different jobs. Deep sleep (N3, slow-wave) shows high-amplitude slow delta waves (0.5-2 Hz), the highest arousal threshold, and drives declarative-memory consolidation and glymphatic clearance (Rasch & Born 2013; Xie et al. 2013). REM shows a fast 'awake-like' EEG, vivid dreaming, and muscle atonia, and supports emotional and procedural processing. Timing differs too: N3 dominates the first third of the night; REM dominates the last third. A night can be deficient in one while normal in the other.
Why do I get more REM toward morning?
Two systems stack. Slow-wave (N3) pressure is highest at sleep onset and dissipates with each cycle, freeing later cycles for REM. And REM propensity is gated by the circadian clock, peaking near the core-body-temperature minimum in the last hours before habitual wake (Carskadon & Dement). So the final cycles are both released from N3 competition and riding the circadian REM peak, which is why your longest, most vivid dreams and easiest awakenings cluster toward morning.
Can you increase REM sleep?
Mostly by not suppressing it and by not waking early. Because REM is concentrated in the final cycles, sleeping an extra hour in the morning yields a large REM gain, whereas going to bed earlier adds mostly deep sleep. Avoid alcohol before bed (it cuts first-half REM), and be aware that SSRIs/SNRIs pharmacologically suppress REM. After selective REM loss the brain produces a REM rebound, evidence the quota is defended. [VERIFY: consumer claims that specific supplements or devices reliably raise REM in healthy adults are mostly small-sample or unsupported.]
Is too little REM sleep dangerous?
Chronically short or fragmented sleep reduces REM and is associated with impaired emotional regulation and memory, but a few low-REM nights are not harmful on their own and the brain compensates with a rebound. If a tracker reports persistently low REM, the usual cause is short total sleep, alcohol, or a REM-suppressing medication rather than a primary REM disorder. [VERIFY: wearable REM percentages are inferred, not EEG-measured, and should not be used to self-diagnose.] Persistent unrefreshing sleep, loud snoring with witnessed pauses, or acting out dreams warrants a clinical sleep evaluation, not tracker tuning.
- [01]
- [02]
- [03]
- [04]
- [05]
Other stacks in this hub
Why You Wake at 3am
The cortisol awakening response, the two-process model, and the actual causes of mid-sleep awakening, with the protocols.
Sleep Onset Latency: What It Is and How to Measure It
Sleep onset latency (SOL) is the time from lights-out to the first epoch of N1. Why under-15-minute SOL is normal, what longer SOL signals, and how the actigraphy estimate misses by 10–20 minutes.
Caffeine and Sleep: Half
Caffeine's ~5-6h half-life, the ~10-12h quarter-life, and why the last dose belongs 8-10 hours before bed.