Home › Blog › Doomscrolling

Doomscrolling · 5 min read ·

Why Can't I Stop Doomscrolling? The Honest Answer

By The doomscroll.ai team

The short answerYou can't stop doomscrolling because the feed is built to never give you a natural stopping point, and willpower is weakest exactly when you scroll: tired, bored or anxious. Research suggests about a third of social media use comes from self-control problems, not choice. What works is a limit set in advance that something else enforces.

It's 1:14 AM. You picked up your phone to check one thing. You can't remember what the thing was. Your thumb is still moving.

If you've ever asked yourself "why can't I stop doomscrolling?" while doing exactly that, you're not weak and you're not alone. You're up against something that was designed, tested and improved by thousands of very smart people to keep you exactly where you are.

Here's what's actually going on in your head, why the usual fixes fail, and what works instead.

What is doomscrolling, exactly?

Doomscrolling is scrolling long past the point where it's useful or enjoyable, often through bad news or endless short videos, while part of you wishes you'd stop. The "doom" originally meant grim news, but most people now use it for any feed that swallows their evening.

The defining feature isn't the content. It's the gap between what you meant to do and what you did. You opened the app for two minutes. It's been fifty.

Why can't I stop doomscrolling even when I want to?

Because wanting to stop and being able to stop are run by different parts of you, and the feed is designed to talk to the second one.

  • No stopping cues. A book has chapters, a TV episode ends. A feed never does. Your brain waits for a natural pause that never arrives.
  • Variable rewards. Most posts are dull, but every so often one is great. That unpredictability is the same pattern that makes slot machines hard to walk away from.
  • Bad timing. You scroll most when you're tired, bored or anxious, which is exactly when self-control is weakest.
  • It's always one more. Each swipe costs nothing, so stopping never feels urgent. Forty small decisions add up to an hour nobody chose.

Economists estimated that self-control problems cause about 31% of social media use: time people spend that they would choose not to, if they could decide in advance.

Source: Allcott, Gentzkow & Song (2022), "Digital Addiction", American Economic Review. The 31% is a model-based estimate from the working paper, not a direct measurement.

Is it a willpower problem?

Mostly no. Willpower is real, but it's a poor tool against a habit that fires automatically, dozens of times a day.

The same research found people are partly unaware of how habit-forming their phone use is. We expect tomorrow-us to scroll less than today-us did, and tomorrow-us rarely agrees.

People consistently underestimate how habit-forming their phone use is, so they keep planning to cut back "tomorrow".

Source: Allcott, Gentzkow & Song (2021), "Digital Addiction", NBER Working Paper 28936

That's why shame doesn't help. Feeling bad at 1 AM doesn't change what you do at 1 AM tomorrow. A different setup does.

Why don't screen time limits work?

Because ignoring them costs nothing. The built-in limits on iPhone and Android show a polite screen with an "Ignore limit" button, and after the first week most people tap it without reading.

ApproachWhy it failsWhat it gets right
Built-in app limitsOne tap to dismiss, no consequenceLimits set in advance
Deleting the appReinstalled within daysAdds real friction
Willpower aloneWeakest when you scroll mostFree
Grayscale or notifications offHelps a little, easy to undoMakes the feed less shiny

Notice the pattern: the parts that work are deciding in advance and adding friction. The part that fails is that nothing enforces the decision when it matters.

What actually helps you stop?

Things that work with your brain instead of against it. Start with these, today:

  1. Decide your limit in the morning, not at midnight. Pick a daily number for the one or two apps that get you. Decisions made in advance are far easier to keep.
  2. Move the phone out of reach at night. Charge it across the room or outside the bedroom. Distance is friction.
  3. Create a stopping cue. A podcast ending, the kettle boiling, a set alarm. Your brain needs the pause the feed won't give.
  4. Track it honestly for a week. Just measuring helps, and it shows you your real danger hours.
  5. Get an enforcer. Someone, or something, that notices when you go over and makes it awkward to carry on.

In one study, simply tracking their own social media use made people less anxious and less afraid of missing out.

Source: Hunt, Marx, Lipson & Young (2018), Journal of Social and Clinical Psychology. Both groups improved, so this is an observation, not a controlled effect.

When students capped social media at about 30 minutes a day for three weeks, their loneliness and depression dropped compared with students who carried on as usual.

Source: Hunt, Marx, Lipson & Young (2018), Journal of Social and Clinical Psychology. Small sample of undergraduates; the 30-minute figure is the total of three 10-minute caps, not a single limit. Self-esteem and general well-being did not improve.

How do you make a limit stick?

Give it a consequence that's harder to ignore than a button. That's the idea behind doomscroll.ai, a screen-time app for Android and iPhone with a pixel-art Grim Reaper.

You set your limits once. Go over on Instagram and stay on it, and he texts you 5 minutes in. Still there at 10 minutes? He phones you. Out loud. In a demon voice, by default. He only ever calls about what you're doing right now, so the moment you stop, it all stops.

You can argue with him for extra time, or tell him you'll stop and earn a halo. At midnight he gives you a verdict: angels for good days, ghouls for bad ones.

Key takeaways

  • Doomscrolling isn't a character flaw: feeds have no stopping cues and reward you unpredictably.
  • Research estimates about 31% of social media use comes from self-control problems, not choice.
  • Built-in limits fail because ignoring them costs one tap.
  • Decide limits in advance, add friction, create stopping cues and track your real use.
  • Limits stick when something enforces them at the moment you go over.

Frequently asked questions

Is doomscrolling an addiction?

Not in a medical sense for most people, and it's best not to self-diagnose. But it does share features with habits that are hard to break: it's automatic, it's rewarded unpredictably, and people keep doing it after deciding to stop. If your phone use is seriously affecting your sleep, work or mood, talk to a doctor.

Why do I doomscroll more at night?

You're tired, your self-control is lower, and there's nothing scheduled to interrupt you. Bed is also where the phone is closest and the room is darkest, so the screen is the most interesting thing around. Charging the phone outside the bedroom is one of the simplest fixes there is.

How long does it take to stop doomscrolling?

Most people notice a difference within a couple of weeks of changing their setup, such as moving the phone at night and setting limits in advance. In one study, students who cut social media to about 30 minutes a day felt less lonely within three weeks.

Does deleting social media apps work?

Briefly. It adds friction, which helps, but most people reinstall within days because nothing has changed about the moments that send them there. Limits you set in advance, with something that enforces them, tend to last longer than deleting and reinstalling.

Let Death keep count.

doomscroll.ai is free. Set your limits once, and if you go over and stay over, the Grim Reaper calls. Hang up any time. He'll remember.

Sources

  1. Allcott, Gentzkow & Song (2022), "Digital Addiction", American Economic Review. The 31% is a model-based estimate from the working paper, not a direct measurement.
  2. Allcott, Gentzkow & Song (2021), "Digital Addiction", NBER Working Paper 28936
  3. Hunt, Marx, Lipson & Young (2018), Journal of Social and Clinical Psychology. Small sample of undergraduates; the 30-minute figure is the total of three 10-minute caps, not a single limit. Self-esteem and general well-being did not improve.
  4. Hunt, Marx, Lipson & Young (2018), Journal of Social and Clinical Psychology. Both groups improved, so this is an observation, not a controlled effect.