The eye has a second job nobody mentioned
For a century, the textbook account of the eye had two light detectors. Rods for dim light and movement, cones for color and detail. Between them, vision.
Then researchers noticed something that did not fit. Mice engineered to have no working rods and no working cones, completely blind by any visual measure, still reset their body clocks to a light and dark cycle. Something in the eye was detecting light and reporting it somewhere other than the visual system.
In 2002 that something was identified: a small population of cells called intrinsically photosensitive retinal ganglion cells, or ipRGCs. They carry a pigment called melanopsin. They make up a tiny fraction of the cells in the retina, they contribute almost nothing to what you see, and they wire directly to the hypothalamus, to the cluster of neurons that runs the body clock.
Their function is not vision. It is timekeeping.
What the clock actually controls
Calling it a sleep clock undersells it. The circadian system sets the daily rhythm of core body temperature, cortisol, melatonin, growth hormone, insulin sensitivity, blood pressure, digestive enzyme output, immune cell trafficking and cellular repair. Nearly every tissue in the body carries its own clock genes, and the light signal arriving through these cells is the master timing input that keeps them synchronized to each other and to the outside world.
This is why light at the wrong time is not merely a sleep issue. Sleep is the symptom people notice. The timing signal reaches considerably further than that.
Why blue specifically
Melanopsin is most sensitive to light at a wavelength of about 480 nanometers, which sits in the blue part of the visible spectrum. That is not arbitrary. It is close to the color of open sky, which is what this system evolved reading.
So the receptor that tells your brain it is daytime is tuned to the color of daytime sky. For all of human history that was a reliable signal, because the only thing producing large amounts of 480 nanometer light was the sky, and the sky went dark on schedule.
LED screens and LED bulbs produce white light by starting with a blue emitter, typically peaking around 450 nanometers, and using a coating to fill in the rest. That is an efficient way to make white light. It also means the light coming off a phone held at arm's length is disproportionately weighted toward the exact band this system reads as noon.
One correction to how this is usually explained. It is not only about color. Researchers now quantify this as melanopic irradiance, which accounts for intensity, wavelength and duration together. A dim blue screen and a bright warm lamp can deliver similar signals. This matters because it means the fix is not simply "make the screen orange." Brightness and how long you look are doing at least as much work as color.
What was measured when people read in bed
The clearest experiment on this is straightforward enough to describe in a sentence. Chang and colleagues at Harvard published it in PNAS in 2015. Participants lived in a controlled laboratory and spent four hours before bed reading either a light-emitting e-reader or a printed book, then swapped.
Compared with the printed book, the e-reader nights produced:
- ●Suppressed evening melatonin, the hormone that signals biological night
- ●A delayed circadian clock, meaning the whole internal schedule shifted later
- ●Longer to fall asleep, and less REM sleep
- ●Reduced alertness the following morning, even after a full night in bed
That last point is the one people miss. The cost did not land only at night. It landed on the next day.
The mechanism, in order
Melatonin is produced by the pineal gland and its release is governed by the light signal arriving through those retinal cells. Light in the evening does not simply reduce melatonin, it postpones the entire release curve. The body is told that dusk has not happened yet.
The consequence is a phase delay. You become sleepy later, wake at the same alarm, and lose sleep off the front end. Repeat that nightly and the clock sits permanently shifted against the schedule the rest of life is run on, which researchers describe as social jetlag: living in one time zone biologically and another one practically.
A 2023 study in Communications Biology refined this usefully. Comparing displays at matched brightness but different melanopic content, lower melanopic light shortened the time to fall asleep, blunted evening melatonin suppression, advanced melatonin onset and reduced evening alertness. The melanopic dose, rather than the device, was what predicted the effect.
Which is good news, because melanopic dose is something you can change without giving up the device.
Melatonin is not only about sleep
Melatonin is a hormone, and it is also a potent antioxidant that crosses cell membranes and concentrates in mitochondria. It is involved in immune regulation and in the overnight repair processes that run while you are unconscious.
This is where the honest boundary sits. That evening light suppresses melatonin is settled and reproducible. That chronic suppression through screen use causes specific diseases in humans is not settled, and the observational literature linking night-shift work to disease is suggestive rather than conclusive because shift workers differ in many other ways. What can be said without stretching: a hormone with functions beyond sleep is being reduced nightly by an exposure that did not exist thirty years ago, and no one has run the long trial.
Why children are not small adults here
Three things make the same screen a larger exposure for a child, and only one of them is about behavior.
A clearer lens
The lens of the eye yellows gradually across life, and a yellow lens filters short-wavelength light. A child's lens is far clearer than an adult's, so substantially more blue light reaches the retina from the identical screen. The filter adults have partly acquired has not been built yet.
Wider pupils
Pupil size falls with age. Younger eyes admit more light overall, which raises the melanopic dose from the same source at the same distance.
A clock still being set
Circadian rhythm is not fully mature at birth. It consolidates across infancy and childhood, and shifts naturally later during adolescence. Evening light lands on a system in the middle of establishing its own timing rather than on one already established.
The adolescent squeeze
Teenage clocks shift later on their own. That is normal biology, documented across cultures, and it is not laziness. Evening screen exposure pushes an already-delayed clock further in the same direction, and school start times do not move.
The result is a teenager who cannot fall asleep at a reasonable hour and cannot function at the hour they are required to be alert. That is often treated as a discipline problem or medicated as one. Part of it is a light problem, and light is the part that is adjustable.
For the wider question of infant and pregnancy exposure, including the EMF component that travels with every screen, see the EMF in pregnancy and infancy article, and screen time and child development for what changes beyond the light itself.
What the evidence supports, and what is being sold
This topic has a large product industry attached to it, and the strength of the marketing does not track the strength of the evidence. Sorting them is the useful part.
Blue-blocking glasses: weaker than advertised
These are sold with considerable confidence. The trial evidence does not match it. A Cochrane review concluded the evidence remains inconclusive, with roughly half of trials showing benefit and the rest showing none. A 2025 systematic review and meta-analysis of randomized crossover trials using objective sleep measurement reached a similarly hedged conclusion, noting small samples and inconsistent protocols across the literature.
Part of the problem is that "blue-blocking" is not a specification. Lenses vary enormously in how much they actually filter and in which band, and a faintly tinted pair sold at a checkout may block very little. Amber and red lenses that visibly change what you see block considerably more than clear or lightly tinted ones.
They may help, particularly for people already prone to disrupted sleep. They are not established, and buying a pair while changing nothing else is the least effective option on this page.
Night mode and screen filters
Built-in warm modes shift the display's color balance toward amber. They do reduce the blue component. What most do not do is reduce brightness, and since melanopic dose depends on intensity as well as wavelength, a warm-toned screen at full brightness in a dark room is still delivering a substantial signal.
Dimming does more than tinting, and doing both does more than either. This is the least discussed and most effective adjustment available, and it costs nothing.
What has the strongest support
- ●Bright light in the morning. The most robust finding in the entire field, and it is the one nobody sells because it is free. Morning outdoor light anchors the clock and makes it more resistant to evening disruption. Outdoor light on an overcast day is still far brighter than any indoor lighting.
- ●Dimming everything in the evening, not only screens. Overhead lighting is often the larger source. Lamps at eye level or below, at low output, deliver a much smaller dose than a ceiling fixture.
- ●Distance. Light intensity falls with the square of distance. A phone at 30 centimeters delivers a far larger dose than a television across a room, which is why the small device usually matters more than the big one.
- ●Consistency. The clock entrains to regularity. Similar sleep and wake times, including at weekends, do more than any single evening intervention.
What this page is not claiming
That screens cause disease. That melatonin suppression from a phone has been shown to shorten life. That any product on this page is a treatment for a sleep disorder.
What it is saying is narrower and better supported: a timing system that reads one specific band of light is now receiving that band at hours when it never previously arrived, the effect on melatonin and sleep timing is measurable and reproducible, and nobody disclosed any of it when the devices were sold. If you have a diagnosed sleep disorder, that belongs with a clinician rather than with a pair of tinted glasses.
Research, Sources & Further Reading
The Receptor and the Mechanism
Melanopsin and intrinsically photosensitive retinal ganglion cells
Identified in 2002. Melanopsin absorption peaks at approximately 480 nanometers. These cells signal environmental illumination directly to the hypothalamic master clock and contribute little to image-forming vision.
Berson, Hattar and Provencio, 2002 onward
"Melanopic irradiance defines the impact of evening display light on sleep latency, melatonin and alertness"
Low melanopic light shortened sleep latency, attenuated evening melatonin suppression, advanced melatonin onset and reduced evening alertness compared with high melanopic light. Establishes that the melanopic dose, not the device, predicts the effect.
Communications Biology, 2023. PMC9974389
Screens Before Bed
Chang AM, Aeschbach D, Duffy JF, Czeisler CA. "Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness"
Controlled laboratory crossover. Four hours of e-reader use before bed suppressed melatonin, delayed circadian timing, lengthened sleep onset, reduced REM sleep and impaired next-morning alertness compared with a printed book.
PNAS, 2015
Do the Glasses Work
"Efficacy of blue-light blocking glasses on actigraphic sleep outcomes: a systematic review and meta-analysis of randomized controlled crossover trials"
Objective sleep measurement across randomized crossover trials. Evidence remains inconsistent, limited by small samples and heterogeneous protocols.
Frontiers in Neurology, 2025. PMID 41341515
"Interventions to reduce short-wavelength light exposure at night and their effects on sleep: A systematic review and meta-analysis"
Broader review of blue-light reduction interventions, reaching similarly qualified conclusions.
SLEEP Advances, 2020