SleepEvidence 2026

What this world covers

Sleep is not a function among others — it is the foundation on which all other worlds work. Anyone chronically too short or too poor on sleep cannot reach the performance, cognition, hormones or recovery of their well-sleeping counterpart via any substance or method. This world covers the lifestyle levers (sleep hygiene, light regulation, timing), the few evidence-based supplements (glycine, apigenin, magnesium, L-theanine, melatonin in specific indications), and the tracker technology (Oura, Whoop, 8sleep).

Important: sleep architecture is complex. REM, deep sleep (N3), light sleep (N1, N2) and wake phases alternate in cyclic 90-minute patterns. A substance that raises deep sleep can simultaneously suppress REM — and harm long-term. This world works with that complexity instead of reducing it to a score.

Why the order matters

In no other world do so many people try to compensate a lifestyle-driven problem with a substance. Anyone going to bed at midnight, scrolling on their phone until 1 AM, drinking three espressos in the afternoon and then searching for the right sleep supplement is searching in the wrong toolbox.

The order is always:

1. Consistent wake time (±30 min, even on weekends) — the most important circadian anchor
2. Light hygiene (daylight in the morning, dimming in the evening)
3. Caffeine strategy (nothing after 2 PM given 5–7 h half-life)
4. Room temperature (16–19 °C) and darkness
5. Exercise + strength training (improves sleep quality in nearly every RCT)
6. Alcohol and meals (both end 3+ h before bed)
7. Only then: glycine, apigenin, magnesium specifically

These seven points beat any supplement stack — and they cost nothing.

The most important levers

Sleep hygiene + architecture

Biggest effects come from routine stability, not optimization tricks. Most important factors with study data:

• Consistent wake time: stabilizes circadian rhythm (Wright et al.)
• Bright daylight in the morning: 5–10 min > 1000 lux within the first hour of waking
• Darkness in the evening: < 50 lux 2 h before bed (dimmed, warm white)
• Room temperature 16–19 °C: core body temperature must drop for sleep
• No caffeine after 2 PM: half-life 5–7 h, blocks adenosine buildup
• Strength training or cardio: improves deep sleep, reduces onset latency
• Alcohol avoidance: destroys REM sleep even at moderate amounts

Glycine

The best-documented sleep supplement with low side-effect profile. Multiple Japanese RCTs reproducibly show:

• Faster entry into deep sleep
• Reduced daytime fatigue after sleep deprivation
• Improved subjective sleep quality

Protocol: 3 g L-glycine in water or tea, 30 min before bed. Slightly sweet, well tolerated.

Apigenin

Flavonoid from chamomile, GABA-A modulator. Animal data strong, human studies thin but consistent anecdotal effects on deep sleep phases.

Protocol: 50 mg, 30–60 min before bed. Combines well with glycine and magnesium.

Magnesium (for sleep)

Magnesium bisglycinate is the form of choice — good bioavailability, double effect via glycine, minimal GI side effects.

Protocol: 300–400 mg elemental magnesium, 30–60 min before bed. Strongest effect with deficiency — more moderate with sufficient status.

Melatonin (in specific indications)

Not a sleep aid in the classical sense, but a circadian signal.

Sensible for:

• Jetlag: 0.5–1 mg 30–60 min before desired sleep at destination
• Shift work
• Delayed sleep phase syndrome (late chronotype)

Not sensible as a general sleep aid — supraphysiological doses (3–10 mg, US standard) shift the rhythm rather than induce sleep. Low doses (0.3 mg) preserve endogenous pulsation.

How we rate evidence

Sleep outcomes are methodologically delicate — subjective and objective measures correlate only partially. We weight:

1. Polysomnography RCTs (gold standard but rare in supplement studies)
2. Actigraphy + subjective scales (standard for medium studies)
3. Validated questionnaires (PSQI, Insomnia Severity Index)
4. Consumer trackers (Oura, Whoop, 8sleep): useful trend data, individual nights with caution
5. Subjective self-reports (most common form, most placebo-prone)

Important: a supplement that improves subjective sleep quality without changing sleep architecture can still have value — the psychological experience of sleep is a real variable.

Most common effects and interactions

Sleep substances interact with stress axis, thyroid and other sleep hormones:

• Glycine + magnesium bisglycinate: synergistic (double glycine), classical combo.
• Apigenin + benzodiazepines: theoretically additive via GABA-A — watch for prescription interactions.
• Melatonin + SSRIs: SSRIs can amplify melatonin's effect — dose lower.
• L-theanine + caffeine: blocks jitter, but theanine in the evening cannot offset the sleep-disturbing effect of caffeine residues — avoid caffeine.
• Alcohol + sleep supplements: all sleep supplements less effective under alcohol, some with cardiovascular risks (magnesium with hypotension).

What does NOT belong in this world

• Classical sedatives (Z-drugs, benzodiazepines) → medical indication, not self-optimization
• CBD and THC → separate legal and evidence discussion; CBD effects on sleep heterogeneous in RCTs
• Adaptogens for stress → World 05 (Cognition) or World 09 (Mental)
• Mental health substances like SAMe, saffron → World 09 (Mental)

Magnesium is a micronutrient (World 01) but is specifically addressed here for sleep — due to form, dose and timing significance.

How Biohacking AI operationalizes this

This world uses the tracker data many users already have:

1. The Sleep Tracker Integration imports Oura, Whoop or Apple Health data and correlates sleep architecture with lifestyle variables (training, caffeine, alcohol, meal timing). You see what actually works for you.
2. The Studies database filters per sleep supplement by endpoint (sleep onset, sleep efficiency, REM, deep sleep, subjective quality). Effects are context-dependent.
3. The Forum collects tracker data and stack protocols — moderated, with obligation to objective tracking instead of just "slept better."
4. The Coach does sleep diagnostics: your data show whether sleep onset, maintenance or wake-time problems dominate — then suggest the matching lever.

The goal is not "better sleep scores." The goal is: wake up consistent, rested, without substance dependency — and know the few sleep levers that actually work in your personal dataset.