Short answer: sleep architecture is the shape of your night — the sequence and proportion of stages you cycle through. It explains why two people can both log eight hours and wake up in completely different states. Deep sleep is front-loaded and physically restorative; REM is back-loaded and cognitively restorative. Losing either one has a different cost.
The sleep stages
Each cycle runs roughly 90 minutes and moves through four states.
N1 (light sleep). The transition out of wakefulness. Heart rate slows, and you are easily woken. A small share of the night.
N2 (established sleep). Heart rate and breathing stabilise, sleep spindles appear on EEG, and memory consolidation begins. This is the largest single share of total sleep.
N3 (deep sleep, slow-wave). The most physically restorative stage: growth hormone release, tissue repair, immune consolidation. Concentrated in the first half of the night.
REM. Dreaming, emotional processing, associative and procedural memory. Concentrated in the second half, with the longest REM periods just before waking.
For normative proportions, the reference work is the meta-analysis by Ohayon and colleagues in Sleep (2004), which pooled quantitative sleep parameters across the lifespan in healthy individuals. It puts slow-wave sleep at roughly 13–23% of total sleep in healthy adults and shows that share declining steadily with age.
An earlier version of this page attributed that figure to a review about pre-bed bathing, which was wrong. The correct source is above.
Why deep sleep carries the physical load
Most growth hormone release is tied to slow-wave sleep rather than simply to time in bed — the physiology is set out in the review by Van Cauter and Plat in the Journal of Pediatrics (1996), which describes GH pulses as tightly coupled to slow-wave onset.
Immune consequences track sleep too. Research from UCSF in Archives of Internal Medicine (2009) found adults sleeping under seven hours were substantially more likely to develop a cold after controlled viral exposure than those sleeping eight or more.
What disrupts architecture without shortening sleep
This is the important category, because it is invisible on a duration tracker.
Alcohol. Suppresses REM early and causes rebound fragmentation later. The review by Ebrahim and colleagues in Alcoholism: Clinical and Experimental Research (2013) summarises the effects on normal sleep. A previous version of this page described that source as a meta-analysis of 27 studies; it is a review, and we have corrected the description.
Elevated evening cortisol. Chronic insomnia is associated with round-the-clock HPA axis activation, with the largest differences in the evening and first half of the night — shown by Vgontzas and colleagues (2001). We previously stated that evening cortisol administration dose-dependently reduced slow-wave sleep, citing a paper that turned out to be about something else; that claim has been removed. See High Cortisol at Night.
A warm bedroom. Core body temperature must fall for sleep onset and deep sleep. A systematic review in Sleep Medicine Reviews (2019) found that passive body heating — a warm shower or bath one to two hours before bed — improved sleep onset latency and quality, by accelerating the subsequent heat loss and core temperature drop.
Magnesium insufficiency. Magnesium supports GABA receptor function, the inhibitory system that allows entry into and maintenance of slow-wave sleep. Abbasi and colleagues (2012) found magnesium supplementation improved sleep efficiency and onset latency in older adults with insomnia.
Supporting architecture rather than sedating it
Sedation and restoration are not the same thing — a sedative can lengthen time asleep while flattening the stages that produce recovery. The AE·ORA REST approach targets the conditions that let normal architecture happen: magnesium glycinate at 275 mg elemental for GABA function, standardised ashwagandha for elevated evening cortisol, and a melatonin-free calm blend for the wired-at-bedtime pattern.
Practical levers, in order of effect size, are in How to Increase Deep Sleep.
Frequently Asked Questions
Quick answers to the questions readers most often ask.
What is sleep architecture?
It is the internal structure of your sleep — the sequence and proportion of N1, N2, N3 and REM stages you cycle through each night. It determines the quality of recovery, which is why total hours alone is a poor measure.
How much deep sleep do you need per night?
Normative meta-analysis puts slow-wave sleep at roughly 13 to 23 percent of total sleep in healthy adults, around one to two hours across a full night. The proportion declines steadily with age, so comparing yourself to a younger norm is misleading.
What happens if you do not get enough deep sleep?
Deep sleep is when most growth hormone is released and most physical repair occurs. Consistently reduced slow-wave sleep works against tissue repair and immune consolidation, and the effect compounds across consecutive disrupted nights.
Does magnesium help with deep sleep?
Magnesium supports GABA receptor function and NMDA modulation, the systems involved in entering and holding deep sleep, so adequate status is a precondition for normal regulation. Randomised evidence for improved sleep measures comes from older adults with insomnia.
What is the difference between REM and deep sleep?
Deep sleep is primarily physical restoration — growth hormone, tissue repair, immune function — and is concentrated in the first half of the night. REM is primarily cognitive and emotional processing and memory consolidation, concentrated in the second half. Going to bed late costs deep sleep; waking early costs REM.
References
Every reference below has been verified against PubMed.
- Van Cauter E, Plat L. Physiology of growth hormone secretion during sleep. J Pediatr. 1996.
- Vgontzas AN et al. Chronic insomnia is associated with nyctohemeral activation of the hypothalamic-pituitary-adrenal axis. J Clin Endocrinol Metab. 2001.
- Ohayon MM et al. Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals. Sleep. 2004.
- Cohen S et al. Sleep habits and susceptibility to the common cold. Arch Intern Med. 2009.
- Ebrahim IO et al. Alcohol and sleep I: effects on normal sleep. Alcohol Clin Exp Res. 2013.
- Abbasi B et al. The effect of magnesium supplementation on primary insomnia in elderly. J Res Med Sci. 2012.
- Haghayegh S et al. Before-bedtime passive body heating by warm shower or bath to improve sleep: a systematic review and meta-analysis. Sleep Med Rev. 2019.
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure or prevent any disease.
