--- title: "Why people keep scrolling: habit, infinite feeds, and the evidence against the dopamine explanation" kind: "Article" topic: "Attention & feeds" date: "2026-10-09" url: "https://canopticon.com/articles/why-people-keep-scrolling" author: "Chris Telford" license: "CC BY 4.0" --- You open an app to check one thing. A long while later you are still there, and you could not say what you were looking for. The usual explanation is that the feed is a slot machine: every swipe might pay out, and the brain keeps pulling the lever. The research behind this article says that story is mostly wrong about scrolling, and that the real causes are duller and easier to design against. ## Habit does most of the work The best causal study is a randomised experiment by Allcott, Gentzkow and Song, which found that temporary incentives to reduce social media use have persistent effects, suggesting social media are habit forming, and that allowing people to set limits on their future screen time substantially reduces use. Their model attributes 31% of use (about 48 minutes a day) to self-control problems.

caption: How much social media use comes from self-control problems - Self-control problems | 31% | 31 | #2f55c4 - Everything else | 69% | 69 | #cfc9bb note: About 48 minutes a day, on the study’s model Habit also hides itself. People do not notice how much today's scrolling builds tomorrow's: they are "inattentive to habit formation," meaning they underestimate how much today's scrolling shapes tomorrow's. > **Objection: Surely it is the dopamine, though?** > > Partly, for one part of social media. Posting does behave like reward learning: it "conforms qualitatively and quantitatively to the principles of reward learning." People time their posts to collect likes. But that is posting, not scrolling. Neither paper shows that passively scrolling a feed works like a variable-ratio slot schedule. the passive-scroll analogy is intuitive but largely untested.

“Their model attributes 31% of use (about 48 minutes a day) to self-control problems.” The effect is far from even: for 22% of participants the excess was under 10 minutes a day. Posting, meanwhile, turns out to be more mixed than pure reward-chasing; a newer study found posting is a hybrid of reinforcement learning and habit. Who chases which differs: Younger users and women were more reward-driven; older users and men were more habitual. ## The feed has nowhere to stop A page, a chapter or the end of an episode is a natural moment to decide whether to keep going. Infinite scroll removes that moment. The mechanism, a removed decision point, is well documented. Its effect size is not. Remarkably, nobody has measured it directly: the research found no randomized trial that compares infinite scroll with pagination on time spent or wellbeing. Without an ending, something else has to end the session — usually life. In a field study of thousands of app sessions, "most reasons to break SM sessions were not related to the app but the user's general context." People leave because the real world interrupts, not because the feed lets them go. Even the history is contested. Microsoft filed a patent on 14 February 2006 (Hugh Williams, Julie Farago, Nick Craswell), while Aza Raskin also claims it. The most repeated statistic about it is not data; Raskin's figure of "200,000 human lifetimes wasted daily" figure is a back-of-envelope estimate, and the bottomless-soup study often used to explain stopping cues comes from a lab whose head has had 18 papers retracted. One psychological pull is real, though not the one usually cited. The famous claim that unfinished tasks stick in memory has largely failed: Ghibellini and Meier's 2025 meta-analysis found "no memory advantage for unfinished tasks but found a general tendency to resume tasks". People do not remember open loops better; they feel pulled to close them. ## Doomscrolling is checking that never finishes Bad-news scrolling has its own driver. People who find uncertainty hard to bear do more of it, and the research reads it as a search for reassurance that the feed never delivers: doomscrolling is information-seeking that never reaches closure. > **Objection: So doomscrolling causes anxiety** > > It is linked to it, strongly and consistently — but the direction is not known. A 2026 meta-analysis of 44 studies found it associated with anxiety and distress, yet its authors rate certainty "very low" for every outcome, and "the evidence does not establish causality." Anxious people may scroll more as much as scrolling makes people anxious. The link to uncertainty has been measured, if only in surveys. In one study, intolerance of uncertainty was "positively associated with doomscrolling (β = 0.29, p < .001)," and the research's interpretation is that under threat, negative news feels like useful surveillance, and the scroll becomes a checking ritual that briefly relieves anxiety and then brings it back. ## The ranking shows people what they say they do not want The feed's order matters as much as its length. In a preregistered experiment on Twitter, engagement-based ranking amplified "emotionally charged, out-group hostile content that users say makes them feel worse about their political out-group." And the users did not want it: "users do not prefer the political tweets selected by the algorithm." > **Objection: Then just switch to a chronological feed** > > That changes how long people stay, not what they end up believing. In the large 2020 election study, a chronological feed "substantially decreased the time they spent on the platforms." But it also raised content from untrustworthy sources by 68.8% on Facebook. Ranking changes what you see and how long you stay far more than what you believe.

caption: Time spent, compared with the average US user - Algorithmic feed | +73% | 91 | #2f55c4 - Chronological feed | +37% | 46 | #9fb1e4 note: Facebook and Instagram, 2020 US election study The amplification is not politically neutral either. Using Twitter's own long-running chronological holdout group, researchers found "in six out of seven countries studied, the mainstream political right enjoys higher algorithmic amplification than the mainstream political left," and some parties were amplified by more than 200%. A ranking built on what people say they want behaves differently, though not perfectly: it cut angry, partisan content but "may reinforce echo chambers." ## What actually reduces compulsive scrolling The interventions that have been tested show a clear pattern. The research's table: | Intervention | Study | Effect | Caveat | |---|---|---|---| | Friction before opening an app (breathing pause) | Grüning, Riedel & Lorenz-Spreen, PNAS 2023 ("one sec") | 36% of opening attempts abandoned; attempts down 37% by week 6; actual openings down 57% after six weeks | ~280 opt-in users, no field control group; co-author is the app's founder (disclosed) | | Newsfeed "diets" | Purohit et al., CHI 2023 (N = 138) | Time on Facebook −64% (strict), −39% (self-regulated) | Only the self-regulated diet preserved sense of control | | Removing the newsfeed / goal reminders | Lyngs et al., CHI 2020 (N = 58) | Both kept users on task; reminders cut daily time; newsfeed removal shortened visits | Reminders "often annoying"; newsfeed removal triggered FOMO | | Commitment devices (self-set limits) | Allcott, Gentzkow & Song, AER 2022 | "Substantially reduces use" | Effects concentrated among a subset of users | | In-feed stopping prompts after 15 min | CHI 2025 "Scrolling in the Deep" (316 participants); 2026 preprint "Can't Stop" (N = 104) | Users took an average of 3 min 31 s to stop after an intervention; context mattered little; explicit pop-ups beat gradual friction for low-self-control users | Preprint not yet peer-reviewed | | "You're All Caught Up" marker | Instagram, 2 July 2018 | Appears "when you've seen every post from the last two days" | Instagram has never published usage data | *Copied from the research’s table of interventions.[Details](doc:diagnosing-the-feed-why-scrolling-works-why-x-feels-like-a-town-square)* The conclusion is about where to intervene: friction and finite containers work; polite reminders mostly annoy. The large effects come from changing the container itself: The biggest effects come from changing the container (no feed, a bounded feed, a delay before entry), not from asking people to behave better inside an infinite one. And one obvious candidate has never been tested: No peer-reviewed study I found directly tests a daily digest. ## Regulators have started treating the endless feed as a design problem The question has moved from psychology to law. In February 2026 the European Commission preliminarily found TikTok in breach of the Digital Services Act over "addictive design." It suggested "disabling key addictive features such as 'infinite scroll' over time." TikTok called the findings "categorically false and entirely meritless." A European Parliament brief adds that in March 2026, a US jury found Meta and YouTube liable for addictive design. The research cautions that the European findings are preliminary and disputed. ## What this explains It explains why willpower advice fails: you are not fighting a craving, you are fighting a habit in a place with no exits. It explains why limits you set in advance work better than resolutions in the moment — the decision is made before the habit starts running. It explains why the end of a scroll is usually an interruption rather than a choice. And it explains why a feed can be compelling without being wanted: what keeps people longest is not what they would pick. A feed that ends, ranked by what people say they value, would lose some time spent. The research suggests it would lose very little that anyone wanted to keep.