Understanding Sleep Stages: NREM, REM, and Why They Matter for Sleep Apnea

By Dr. Boris Zusin · Published September 8, 2026

A normal night of sleep isn't one uniform state that just gets lighter toward morning — it's a repeating cycle of distinct stages, each with its own brainwave pattern, muscle activity, and physiological purpose. Understanding this cycling, sometimes called sleep architecture, is what makes concepts like "REM-predominant" sleep apnea or "deep sleep" actually make sense, rather than just being terms that get thrown around.

The Two Basic Types of Sleep

Sleep divides into two broad categories: non-REM (NREM) sleep and REM sleep, named for the presence or absence of the rapid eye movements that mark the latter. These alternate in cycles across the night, with each full cycle lasting roughly 90 to 120 minutes — so a typical night includes somewhere between 4 and 6 complete cycles. Polysomnography, the formal sleep study used to diagnose sleep apnea and other sleep disorders, identifies these stages by combining brainwave activity (EEG), eye movement (EOG), and chin muscle tension (EMG) under standardized scoring criteria set by the American Academy of Sleep Medicine.

The Stages of NREM Sleep

NREM sleep is further divided into three stages of progressively deeper sleep. Stage N1 is the lightest stage — the brief transition into sleep, with a low arousal threshold (something as simple as hearing your name called can wake you). Stage N2 is deeper, marked on EEG by distinctive brief bursts of activity called sleep spindles and K-complexes; it's also linked to memory consolidation, particularly for procedural (skill-based) memory. Stage N3, often called slow-wave or deep sleep, shows high-amplitude, slow delta brainwaves and is associated with minimal cognitive activity during the stage itself — it's also the most physiologically restorative stage, tied to tissue repair, growth hormone release, and immune function, and it's the stage from which NREM parasomnias like sleepwalking arise. Together, NREM sleep makes up roughly 75–80% of total sleep time in a healthy adult, broken down further into approximately 2–5% N1, 45–55% N2, and 10–20% N3.

REM Sleep: Not Just Dreaming

REM sleep looks almost paradoxical on an EEG — brain activity closely resembles wakefulness, even though the body is, in a sense, more disconnected from it than at any other point in the night. This is because REM sleep produces near-total skeletal muscle atonia (muscle relaxation), alongside the rapid eye movements that give the stage its name. REM accounts for an estimated 80% of dreaming and makes up roughly 20–25% of total sleep time in a healthy young adult. Its neurochemical profile is distinct from NREM sleep as well — dominated by acetylcholine, with the brain's serotonergic and noradrenergic (aminergic) systems largely suppressed — which is part of why REM-suppressing medications like many antidepressants change dream recall and sleep quality. REM sleep also plays a documented role in cognitive and emotional processing and in a further, separate form of memory consolidation from what happens in N2.

How the Night Actually Unfolds

Sleep onset always occurs in NREM, not REM, and the first REM period doesn't arrive until roughly 80 minutes in. That first REM episode is typically short — 10 minutes or less — but each successive REM period across the night grows progressively longer. The practical result is a lopsided distribution: slow-wave (deep) sleep is concentrated in the first half of the night, while REM sleep becomes progressively more dominant in the second half, particularly in the hours before natural waking. This is exactly why a therapy that only covers the first several hours of sleep can end up systematically undertreating a REM-concentrated condition — a point covered in more detail in our piece on REM-predominant sleep apnea.

Why Sleep Changes With Age

Sleep architecture isn't fixed across a lifetime. With increasing age, the proportion of lighter N1 and N2 sleep rises while N3 (deep, slow-wave sleep) declines — a shift toward lighter, more fragmented sleep overall. REM, by contrast, tends to stay relatively constant as a proportion of the night. This is part of why older adults commonly report lighter, more easily disrupted sleep even without a diagnosed sleep disorder, and it's a normal (if sometimes frustrating) part of the aging process rather than necessarily a sign of pathology on its own.

Two Systems Keep This Cycle on Schedule

Two separate, synergistic biological processes regulate when and how you sleep. The homeostatic drive builds progressively with each hour spent awake — it's the mechanism behind feeling more tired the longer you've been up, and behind "sleep debt" catching up with you. The circadian process is a roughly 24-hour internal rhythm, governed by the brain's suprachiasmatic nucleus and reinforced by pineal melatonin secretion, that promotes wakefulness during the day and counterbalances the building homeostatic drive until the two align to produce a consolidated period of nighttime sleep. Disruption to either system — shift work, jet lag, irregular schedules — can desynchronize them and fragment sleep even when total sleep time looks adequate on paper.

Why This Matters for Sleep Apnea and Other Conditions

Sleep architecture isn't just a physiology curiosity — deviations from these normal proportions and timings can be diagnostically meaningful. Obstructive sleep apnea disrupts this architecture directly, and because muscle atonia during REM sleep makes the upper airway more prone to collapse, breathing events often cluster more heavily in that stage, which is the physiological basis for the REM-predominant OSA phenotype discussed elsewhere on this site. Sleep architecture is also altered by a range of other conditions and exposures worth knowing about: depression and many antidepressant medications characteristically suppress REM sleep, chronic pain disrupts sleep continuity and reduces slow-wave sleep, and various neurodegenerative and movement disorders are associated with specific architecture abnormalities, including NREM and REM-stage parasomnias. None of this is something we diagnose or manage directly in a dental setting, but understanding the architecture underneath a sleep study report is part of understanding why your particular case of OSA is being evaluated and treated the way it is.

If you've had a sleep study and are curious what your own report's breakdown by stage actually means for your case, that's a good question to bring to your evaluation — the stage-by-stage detail often explains more about your symptoms than the overall AHI number alone.