A digital distraction is any prompt from a device that redirects your attention away from what you chose to do. That covers the obvious interruptions — notifications, calls, messages — and the quieter ones: badge counts, autoplay, an open tab you keep glancing at, and the silent phone sitting face-down beside you.
On this page
Why willpower is the wrong tool
The standard advice for digital distraction is a variation on trying harder: be more disciplined, be more present, want it more. It fails so consistently that the failure deserves an explanation.
Resisting a distraction draws on executive control — the effortful system that holds you on one thing while something else competes for you. It's limited, it depletes across a day, and it's the same system you need for the work itself. Spending it on resistance means spending it twice.
Meanwhile the thing you're resisting doesn't get tired. It runs continuously, it was refined against millions of people, and it costs nothing to deploy again a minute later. That's not a fair contest, and losing it says nothing about you.
You're spending a depletable resource against one that never runs out.
The productive move is to stop competing on resistance and start changing the conditions. Every mechanism below has a structural countermeasure that costs you nothing to maintain once it's in place.
Notifications and the checking habit
The mechanism is a variable-ratio reinforcement schedule, documented by Ferster and Skinner in 1957.[1] When a reward arrives unpredictably, the behavior that produces it becomes more frequent and far more resistant to extinction than when the reward is reliable.
Your phone is close to a pure implementation. Most alerts are noise. A few genuinely matter. You can't tell which without looking. So you look, and each time the uncertainty resolves, the loop tightens slightly.
This explains something people find confusing about their own behavior: you check when you're fairly sure nothing has happened. Of course you do. Expectation was never the driver. Uncertainty was.
The countermeasure that works: turn notifications off rather than resisting them. Not silenced, not batched into a summary — off, for everything that isn't a person who might genuinely need you. Resisting costs executive control every single time. Not receiving costs nothing, so the benefit compounds while the discipline required stays at zero.
Notification fatigue is the second-order effect. Once volume passes a threshold, you stop reading alerts and start dismissing them, which means the genuinely important one gets swiped away with the rest. High notification volume doesn't only cost attention. It costs you the signal you kept notifications on for.
Infinite feeds and the missing endpoint
Infinite scroll works by deleting the stopping cue.
A list that ends gives you a decision point. You reach the bottom, you see "next page," and for a moment your intention gets a vote. Usually you notice you're finished.
A feed that loads forever never gives you that moment. Continuing becomes the default state and stopping becomes the effortful one. You don't decide to keep going; you just don't decide to stop, repeatedly, and the session ends when something external interrupts it rather than when you chose.
Aza Raskin, who designed the pattern in 2006, has said publicly for years that he regrets it. Autoplay applies the same logic to video. Pull-to-refresh applies the slot-machine lever with better animation.
The countermeasure: reintroduce an endpoint. An external timer works. So does switching a feed to a chronological or "following only" view where you eventually reach content you've seen. So does deciding the stopping condition before you start — a number of items, a single specific thing you came for. Any endpoint restores the decision the interface removed.
Doomscrolling
Doomscrolling stacks two mechanisms, which is why it's harder to break than ordinary scrolling.
The first is negativity bias: threatening or distressing information holds attention more strongly than neutral information. That's an old and generally useful feature of human attention, and it long predates any technology. Something that might hurt you should be harder to look away from.
The second is the infinite feed. Combine them and you get a system that surfaces the most alarming available material into a container with no bottom. Your attention is doing exactly what it evolved to do, in an environment that supplies an unlimited quantity of the trigger.
The feeling afterward — drained, anxious, and no better informed — is the tell. Genuine information-seeking terminates when you've learned the thing. Doomscrolling doesn't terminate, because resolution was never what it was offering.
The countermeasure is a container rather than a ban. Decide when you'll read the news, from which sources, for how long. A fixed window supplies the endpoint the feed removed, and it separates being informed — which is worth something — from being churned, which isn't.
Ambient distraction: the phone that does nothing
Ward, Duke, Gneezy, and Bos tested whether a phone affects you while it's doing nothing at all.[2] Participants completed working-memory and fluid-intelligence tasks with their phone on the desk, in a bag, or in another room. Every phone was silenced. Performance was best when the phone was in another room.
Not ringing. Not lighting up. Present. Some portion of capacity goes to not attending to it, and that portion isn't available for the task.
This is the reason the widely-repeated advice to put your phone in another room outperforms putting it face-down. Face-down still leaves it in the field of things you're actively not looking at. Distance is doing real work that willpower can't substitute for.
Self-interruption, the one nobody counts
Research on workplace interruption has repeatedly found that a large share of interruptions are self-generated.[3] Nobody messaged you. Nothing buzzed. You switched anyway.
This matters because it sets a ceiling on what environmental fixes can achieve. You can silence every device you own and still switch tasks every four minutes, and blaming the phone at that point is aiming at the wrong target.
Self-interruption usually signals one of three things. The current task is unclear, so there's nothing concrete to work on. The current task carries emotional weight you haven't named. Or you've hit a genuine capacity limit and need a break rather than a different task.
The three are worth distinguishing, because the fixes differ. How to start a task when your brain won't works through each one.
What actually works
Ordered by leverage — how much they return for how little ongoing effort they cost.
- Turn off notifications. One-time cost, permanent benefit, breaks the variable-ratio loop at its source. Nothing else on this list comes close.
- Put the phone in another room during focused work. Not face-down. Another room. The research is specific on this.
- Give open-ended things an endpoint. A timer, an item count, or a decided stopping condition restores what infinite scroll removed.
- Decide what you're doing before you sit down. An unclear task is the most reliable cause of self-interruption, and it's not fixed by any environmental change.
- Externalize the priority. After an interruption you want to look something up, not reconstruct it. Working memory is what interruption takes first.
- Batch the checking. If something genuinely needs monitoring, check it at fixed times. Scheduled checking has an endpoint. Ambient checking doesn't.
What doesn't work reliably: trying to want it less, relying on a blocker without deciding what fills the gap, or any approach whose ongoing cost is willpower. The test for a good countermeasure is whether it gets cheaper over time. Structural changes do. Resistance doesn't.
The short version
- Notifications work through uncertainty, not content. Turn them off rather than resisting them.
- Infinite scroll removed the stopping decision. Put one back with a timer or a decided endpoint.
- Doomscrolling stacks negativity bias onto an endless feed. Use a fixed window, not a ban.
- A silenced phone in the room still costs measurable capacity. Distance beats discipline.
- Much of the interruption is self-generated, and no device setting fixes an unclear task.
Clear the trigger. Then know what to do next.
Cutting distractions leaves a gap. Mass Distractions Task Planner fills it with the thing you already decided mattered — sorted by temperature, hot to cold, so you never have to re-derive it.
Start your free trial30 days free. No credit card needed.
Frequently asked questions
What counts as a digital distraction?
Any prompt from a device that redirects your attention away from what you chose to do. That includes notifications and calls, and also badge counts, autoplaying video, an open tab you keep glancing at, and a silent phone on the desk.
What is doomscrolling?
The compulsive consumption of an endless feed of distressing content. It combines a negativity bias that makes threatening information harder to look away from with an infinite feed that never presents a stopping point.
Why do I check my phone when nothing has happened?
Because checking has been reinforced on a variable-ratio schedule. Sometimes there's something, usually there isn't, and you can't tell without looking. The checking isn't driven by expecting something. It's driven by not knowing.
Does turning off notifications actually help?
Yes, and it's the highest-leverage single change available. Resisting a notification costs executive control, which depletes. Never receiving one costs nothing, so the benefit compounds rather than demanding more discipline over time.
Is a silent phone on the desk still a distraction?
Yes. Performance on working-memory and fluid-intelligence tasks was measurably better when the phone was in another room than when it was silenced on the desk. Some capacity goes to not attending to it.
References
- Ferster, C. B., & Skinner, B. F. (1957). Schedules of Reinforcement. Appleton-Century-Crofts.
- Ward, A. F., Duke, K., Gneezy, A., & Bos, M. W. (2017). Brain drain: The mere presence of one's own smartphone reduces available cognitive capacity. Journal of the Association for Consumer Research, 2(2), 140–154.
- Mark, G., Gonzalez, V. M., & Harris, J. (2005). No task left behind? Examining the nature of fragmented work. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '05), 321–330. See also Mark, G., Gudith, D., & Klocke, U. (2008). The cost of interrupted work: More speed and stress. CHI '08, 107–110.