Your Circadian Rhythm, Fixed: When Morning Light Helps — and When It Backfires
Sleep researcher Greg Potter takes apart the blanket advice to chase sunlight the moment you wake up: whether that helps depends entirely on which direction you want your body clock to move. The conversation covers chronotypes, practical light protocols for shifting earlier or later, creatine's effect on sleep drive, why REM percentages are a shaky target, and where to start if you keep waking at 3 a.m.
Overview
Siim Land hosts sleep and circadian researcher Greg Potter for a listener Q&A that starts with the most repeated piece of sleep advice online — get bright light as early as possible — and shows why it is directional, not universal. Potter explains chronotype as a bell-shaped distribution shaped by both genetics and light exposure, pointing to camping studies where a few days without artificial light pull everyone's rhythms closer together. He describes rare gene variants that produce entire families of extreme early risers, and small studies where matching shift schedules to chronotype improved sleep, self-rated health and work satisfaction.
From there he gives concrete light protocols: bright light soon after waking if you want to move earlier, light concentrated in the late afternoon and evening if you want to move later, and dim warm low-level lighting in the final hour before bed either way. The middle of the episode turns to creatine — its modest effects on hormones and creatinine, the mouse work suggesting high doses lower sleep drive, and why taking it with the first meal of the day remains the sensible default.
Potter then pushes back on optimising REM sleep, citing recordings showing different brain regions scored in different sleep stages at the same moment, and argues for duration, quality, timing, regularity and daytime function as the metrics worth watching. He also notes evidence that losing sleep at the start of the night appears kinder than losing it at the end. The episode closes with whey protein and food processing, sleep position, herbal sleep aids, melatonin, and a practical starting point for people who wake at 3 a.m. and cannot get back to sleep.
Key quotes
5For some people, that's good advice. For some people, it's terrible advice.
Over 30% of the time those two brain regions were scored as being in different sleep-wake states at the same time.
More REM sleep is not necessarily better.
It's probably in many instances better to lose sleep at the start of the night than it is at the end of the night.
Save your bed for sex and sleep only.
Key ideas
9Chronotype is a curve, not two boxes
Plot mid-sleep time across a large group of the same age and sex and you get a bell-shaped distribution. Most people sit in the middle; the extreme larks and extreme owls are the tails, not the norm.
Camping narrows the spread
When people spend several days with minimal artificial light, the gap between their sleep timing and melatonin rhythms shrinks considerably. A lot of apparent owlishness is an indoor lighting story, not a genetic one.
Rare variants produce extreme larks
Gene variants found in well under 1% of people can drive very early or very late sleep timing. Entire families have been identified who naturally wake well before 5 a.m., even as young adults.
Schedules can be matched to chronotype
Small studies removed the harshest shift for each chronotype — night shifts for larks, early shifts for owls. People slept better, rated their health higher and were more satisfied at work.
Light advice is directional
Bright light early accelerates the clock; light late in the biological day slows it. Before choosing a protocol, decide whether you want to move earlier, move later, or stay where you are.
Intensity and placement matter
If sunlight isn't available, a therapy lamp emitting at least 5,000 lux — most now deliver around 10,000 — placed roughly half a metre away and slightly to the side for about half an hour is the described setup.
Creatine and the drive to sleep
Mouse work from around 2017 found that very high creatine doses raised brain phosphocreatine, reduced sleep, lowered sleep depth and blunted rebound sleep. Human brain stores respond far less to supplementation, so effects in people are likely much subtler.
Sleep isn't one state across the brain
Simultaneous surface and deep recordings found the two regions scored in different sleep-wake states over 30% of the time. That patchiness also helps explain concentration lapses during long tasks.
Melatonin sits in its own category
No clear evidence of tolerance or withdrawal has emerged, and it doesn't meaningfully shift the proportions of sleep stages. Potter's caveat is dose: most products on shelves are stronger than would be ideal, especially long term.
Practical takeaways
7- 1
Pick a direction before you pick a protocol 11:40
Ask whether you want your clock earlier, later, or left alone. The same morning light habit is helpful for a late-shifted person and unhelpful for someone already waking at 4 a.m.
- 2
Half an hour of daylight, most days 14:00
The general guidance is at least thirty minutes outdoors in daylight daily, often more, while avoiding sun exposure when the UV index is very high.
- 3
Drop the overhead lights in the evening 16:20
For the three hours or so before bed, use warm white bulbs in low lamps rather than bright overhead lighting, and dim further in the final hour.
- 4
Take creatine with your first meal 26:50
That has long been Potter's default. During a loading phase, splitting the daily amount across the first hours of the day suits people better than evening doses.
- 5
Protect the end of the night 37:20
When sleep has to be shortened, the evidence discussed favours a later bedtime with a fixed wake time over a normal bedtime with an early forced awakening.
- 6
The fifteen-minute rule 57:10
If you've been awake around fifteen minutes, get out of bed, go to another room, do something relaxing in dim light and return only when sleepy — repeated as often as the night requires.
- 7
Rescue the day after, without extra caffeine 58:20
After a broken night, keep the fixed wake time, get plenty of bright light, stay physically active, and resist stacking caffeine to compensate.
Topics & chapters
15The advice that isn't universal
Potter opens by questioning the blanket instruction to get bright light the moment you wake, and previews the finding that losing sleep early in the night beats losing it late.
Nature, nurture and the chronotype curve
Circadian rhythms are driven by an intrinsic pacemaker, but where you land on the chronotype distribution reflects both genes and your light-dark environment. Camping studies show how much the environment contributes.
Rare genes, extreme larks
Dominant gene variants present in under 1% of people can produce families of natural pre-5 a.m. risers. Their timing is far more resistant to change than most people's.
Shift work matched to chronotype
Removing the worst-fitting shift for each chronotype improved sleep, self-rated health and job satisfaction in small trials. Flexible modern schedules make this more feasible.
When to get sunlight: it depends
The same early-light advice is excellent for a night owl on an alarm clock and counterproductive for someone in their seventies already waking at 2 a.m.
Shifting earlier: the light protocol
Bright light soon after waking, from the sun or a lamp of at least 5,000 lux at around half a metre, then reduced overhead lighting in the final three hours of the evening.
Shifting later: the light protocol
Avoid light in the first couple of hours after waking, load bright light between four hours and one hour before target bedtime, then wind down and dim.
Creatine, DHEA and DHT
Neither host has seen convincing evidence of a DHEA effect, and a recent study found no change in DHT or androgens. Creatinine rises, which is expected and rarely a concern.
Creatine and the drive to sleep
Mouse studies at very high doses reduced sleep need and rebound sleep. Human evidence is thin, retrospective and modest, and Potter would like to see polysomnography used properly here.
Dosing creatine well
First meal of the day is the default. Loading doses are often split to avoid GI distress, and newer studies scale intake to body weight rather than a flat 3-5 grams.
REM sleep: what the numbers hide
Claims that REM predicts mortality run into a measurement problem: brain regions can be in different sleep states simultaneously, and depression alters REM in the opposite direction to the simple story.
Lose sleep at the start, not the end
Most REM sits in the later hours, so early awakenings cost more. Metabolic and performance data also favour a late bedtime over a forced early rise at equal duration.
Whey protein, processing and bars
Potter dismisses the ultra-processed label as the wrong question for whey, while noting that too much can crowd out other foods. Most protein bars, he argues, remain poor products.
Sleep position, herbs and melatonin
Inclined beds may help some people who snore but suit few; glymphatic sleep-position claims rest on mouse work. Herbal aids lack long-term data, while melatonin shows no clear tolerance.
The 3 a.m. wake-up
The phenotype Potter finds most stubborn. He recommends starting with stimulus control therapy, adding relaxation work, and reserving longer-acting options for later.