If a fixed nighttime boundary fits your routine, you can download 9to9 from the App Store and choose the apps and websites you want quiet from 9 PM to 9 AM.
Ask someone why their phone is bad for their sleep and they will probably say “the blue light.” It is the explanation that won — the one that got into the magazines and the phone settings, the one with a whole industry of filters and glasses built around it. It is also, according to the actual research, only part of the story — and probably not the biggest part. The more likely culprit is the less photogenic one: the scroll itself, the engaging content, the hour the feed quietly steals. This article sorts the two explanations against each other, because they suggest very different fixes.
Why the wrong explanation won
It is worth pausing on how the blue-light story came to dominate, because it explains why the more important fix gets ignored. Blue light is a mechanism you can sell a solution to — a filter, a night mode, a pair of glasses — and the fixes are easy, purchasable, and require nothing of you. The scroll's mechanism is harder to sell: the fix is a boundary, a habit change, a thing you have to do rather than a thing you can buy. The explanation that won is not the one the evidence most supports; it is the one that was most marketable. That is worth knowing, because it means the fix that gets recommended most is probably not the one that matters most.
What the blue-light theory actually claims
The mechanism is real and worth stating fairly. Light in the blue part of the spectrum — which screens emit — acts on the circadian system, suppressing melatonin and delaying the signal that it is time to sleep. Controlled studies have shown that bright light at night shifts circadian timing, and screens contribute to that exposure. So the theory is not wrong; light at night does affect the body clock.
The question is the size. The dose of light a phone delivers is modest compared to the exposures that produce large circadian shifts — a dim screen in bed is a very different light environment than a bright room. The consensus among sleep researchers is careful on this point: light from screens is one factor among several, and its effects are real but depend on brightness, duration, distance, and timing. It matters; it is just rarely the main thing.
Sources: The impact of screen use on sleep health across the lifespan: consensus statement (PMC)
The scroll's case: the harm is the time and the arousal, not the photons
Now the competing explanation, which fits the evidence at least as well and probably better. The scroll harms sleep two ways that have nothing to do with light. The first is displacement: the feed is infinite and the hour is not, so the scroll extends the night — the bedtime that was 11 becomes 12:30 because the feed never told you to stop. This is not a physiological mechanism; it is arithmetic. The light could be perfectly warm and the hour is still gone.
The second is arousal: the content itself is activating. An argument in a comment section, a work email, a cliffhanger, the low-grade alertness of the feed — these keep the mind engaged in a way that makes sleep harder to reach even after the phone is down. A review of media and sleep notes that the effects depend on what the content is doing to the user, not just that a screen is on — which is the doomscroll's mechanism in a sentence.
It is also worth naming the mechanism that is neither light nor content: the replacement of a wind-down. Whatever the feed is doing to your arousal, it is also doing something simpler — it is occupying the hour that used to wind down. The bath, the book, the quiet — the scroll does not add stimulation on top of those; it replaces them. Even a perfectly calm feed, at a perfectly warm color temperature, is still the thing standing where the transition to sleep used to be.
Sources: How does modern media use impact sleep? (PMC)
The case where light really is the issue
To be fair to the light: it is not nothing, and for a specific person it is the main thing. If you are on the phone for a short, calm stretch before bed — a quiet read, a dim screen — and the sleep is still disrupted, the light exposure is the plausible culprit, and the fix is the brightness and the timing rather than the content. The blue-light problem is real; it is just smaller and rarer than the version where the feed took the hour. The mistake is not caring about the light — it is fixing the light and leaving the scroll.
How to tell which one is your problem
The distinction is not academic — it changes what you should fix. The diagnostic is simple: which fails first when you use the phone in bed, the sleepiness or the clock?
If you can be on the phone for a while and still fall asleep quickly when you put it down — the light and the content do not seem to keep you wired, but the scroll keeps you up an extra hour — the problem is displacement. The fix is a boundary on the time, not the light.
If the phone leaves you alert and stirred even in short sessions — the content wakes you up rather than eating the time — the problem is arousal, and the fix is a boundary on the content, not just the hour.
And if neither seems to matter — if the phone is beside the bed but the sleep is fine — then this whole question is not your problem, and the honest answer is to leave it alone.
What is actually worth changing
Putting the two explanations together produces a priority order that the blue-light story usually gets backwards. First, fix the time: the scroll's main harm is the hour it takes, so the highest-value change is a boundary — a scheduled block, a fixed 9-to-9 window — that closes the feed before the bed. That addresses displacement directly.
Second, fix the content: the stimulating, interactive, personally engaging stuff is the arousal problem — the comment section, the news, the work channel — and it belongs outside the last hour even when the device stays.
Third, and last, the light: dim the screen, use the night mode, keep the brightness low. It is the smallest of the three and the easiest to do, which is probably why it got the most attention — it is the fix that asks the least of you. It is worth doing and it is not the fix.
What a night that handles all three looks like
To make it concrete: the feeds close at 9 PM — the schedule, not a decision — so the displacement is handled before it starts. The stimulating content is out of the last hour because it is behind the block, so the arousal is handled by absence rather than restraint. And the screen that remains — for the ebook, the audio, the thing with an ending — is dim and in night mode, for whatever that smaller contribution is worth. None of the three is dramatic; together they are the version of the fix the evidence actually points to, which is mostly a boundary and a little bit of dimming, rather than a filter on a scroll that never ends.
The questions that follow
Does night mode help? Some — it reduces the blue-wavelength exposure, which addresses the smaller part of the problem. It does not address the scroll's time or arousal, so it is a complement to a boundary, not a substitute for one.
Is reading on a phone different from scrolling? Likely — a calm, finite activity on a dim screen is the better end of the spectrum, and probably closer to reading a book than to a doomscroll. The difference is the content and the ending more than the device.
So which is it — light or scroll? For most people, the scroll. The light is real and worth dimming; the scroll is the thing that took the hour. If you can only fix one, fix the one that was stealing the time.
A note on the glasses and the filters
One more honest point, because the products deserve a direct answer: blue-light-blocking glasses and screen filters are not useless — they reduce the exposure — but they are a fix for the smaller problem sold as a fix for the bigger one. If the scroll is still taking the hour, the glasses are dimming the light of a session that should not be happening at all. The order matters: end the feed, then worry about the light. The filters are a finishing touch, not the fix.
The honest summary: the blue light got the blame because it is a mechanism you can buy a fix for. The scroll got less attention because the fix for it is a boundary, which is harder to sell. The research points mostly at the second. Dim the screen if you like — but end the feed first.