Infinite scroll is hard to stop because the feed has no ending, so the decision to stop has to come from you or your surroundings. In field research, people mostly left when another task, a person, sleep, boredom or their own decision broke in, not because the feed ran out. A clear, explicit stop signal worked fastest. Decide why you are opening the app, set a stop signal outside it, and name where you go next.
The feed removes the moment you would decide to stop
A paged website ends at the bottom of a page. To see more, you choose to tap “next.” An infinite feed loads the next posts as you move, so that choice point disappears. The infinite scroll glossary entry explains the design itself; this article covers what researchers saw when they watched people use it in daily life.
The consequence is simple: with no built-in ending, a short check can quietly turn into a long session, and the session often lasts until something else interrupts it. Doomscrolling, the habit of scrolling through bad or upsetting news, usually happens on exactly this kind of feed. The studies below measured infinite scrolling on social apps in general, not negative news in particular.
What a week of tracked social media sessions showed
Jan Ole Rixen and colleagues at Ulm University and TU Darmstadt published “The Loop and Reasons to Break It” in Proceedings of the ACM on Human-Computer Interaction in September 2023. For one week, an Android app logged when 46 U.S. adults opened and closed Facebook, Instagram, TikTok, Reddit and Twitter. After some sessions, it asked what they had done in the app, why they left and how they felt.
- Scrolling filled most sessions. Across 1,064 sessions with questionnaires, only 22% included no feed scrolling. More than 85% of sessions lasting longer than a minute included some, including many that began with something else, such as a message or a story.
- Scrolling-only sessions felt worse. People rated their feelings less positively after sessions that were only scrolling than after sessions that mixed in chatting, posting or stories. Sessions were not negative on average, and differences between people mattered more than what a session contained.
- The feed rarely ended the session. Most reasons for leaving had nothing to do with the app. In sessions of 10 minutes or more, the most common reasons were leaving to do another task (20%), boring content (13%), deciding to stop (13%), switching to another app (11%) and going to sleep (9%). Old or boring content together explained only about one in five exits.
The authors describe two loops: an outer loop of opening the app again and again, and an inner loop inside a session where the feed keeps it going. Sessions that started from a notification and ended once that task was done had the least scrolling.
Study limits. This was a small observational study of Android users recruited online: 143 people installed the study app and 46 finished with usable data. Questionnaires were spaced at least three hours apart for most sessions, reasons were self-reported, and the study did not test any intervention. It was supported by a state PhD scholarship (Landesgraduiertenförderung).
What happened when researchers interrupted the scroll
Part of the same research group then tested interruptions. “Can’t Stop,” by Luca-Maxim Meinhardt and colleagues, appeared in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies in September 2026, after an arXiv preprint posted July 17, 2026. For seven days, 104 adult Android users in the U.S. and U.K. ran an app that watched infinite scrolling on TikTok, Instagram, Facebook and YouTube Shorts. After 15 minutes of continuous scrolling, it randomly showed one of three interventions:
- a pop-up suggesting a break, which could be dismissed;
- a visual overlay that made the content steadily harder to see over 3 minutes 31 seconds;
- a vibration that grew stronger over the same period.
Across 1,294 interruptions, the plain pop-up worked fastest. People stopped scrolling after a median of 7 seconds, compared with 28 seconds for the vibration and 56 seconds for the visual overlay. People with high impulsivity or low self-control responded best to the explicit pop-up; for people with high self-control, the type of interruption mattered less. Location, stress and the other situations the researchers measured made little difference.
The pop-up had a weakness, though: its ratings fell quickly when people ignored it, while the slower visual overlay kept steadier ratings. After an interruption, the median break before the next scrolling session was 3.4 hours, but about one in seven sessions resumed within five minutes.
Study limits. Of 792 people who registered, 214 completed the week and 104 experienced all three interruptions and were analyzed. The main measure was time to stop, which favours a sudden cue over a gradual one. Ratings were collected only after people stopped, the study lasted one week, and the pop-up resembled break reminders people had seen before. It did not follow screen time, sleep or well-being beyond the week. The German Research Foundation (DFG) funded it, and the authors published their app, data and analysis code. Neither study tested Unscroller.
What this means for your own scrolling
Put the two studies together and a practical picture appears. The feed will not end the session for you. Sessions ended when another task, a person, sleep or boredom broke in, or when someone made a clear decision to stop. A direct, explicit signal to stop got a fast response from many people, especially those who find it hardest to stop on their own.
So instead of hoping you will notice the time, decide the ending before you start. The checklist below turns the most common reasons people left into cues you set in advance. It is an original planning aid based on these findings, not a tested programme.
Build an exit before you open the feed
Name the reason for opening
One message, one post, one account, one search. In the 2023 study, sessions that people left after checking the notification they came for had the least scrolling. When your task is done, that is your exit.
Set a stop signal outside the app
Start a timer in your phone’s Clock app, or ask someone you are with to tell you when time is up. The 2026 study found an explicit, clearly worded prompt worked faster than a subtle one. A signal you set yourself is not the same intervention, but it gives the session the ending the feed leaves out.
Name where you go next
The most common reason long sessions ended was another task. Write the next task down before you open the app, for example “reply to Ana, then start the laundry,” so the interruption is already waiting.
Treat “this is boring” as the exit
Boring content was one of the top reasons people left. When the posts stop being interesting, close the app instead of scrolling further to find a better one.
Give bedtime its own boundary
Going to sleep ended nearly one in ten long sessions, which means the feed often runs until bedtime. If that is when it happens to you, the guide to stop doomscrolling at night sets a phone-down time and a charging spot.
Copy this exit card
I am opening: ______ (app)
To do: ______ (one task)
Stop signal: timer for ______ minutes, or ______ (person)
Next I will: ______
If the feed turns boring: I close it.
A worked example
Maya opens Instagram at 9:40 p.m. to answer one message from her sister. Before opening it, she sets a five-minute timer and writes “pack lunch for tomorrow” on her card. She answers the message in two minutes, then catches herself scrolling Reels. When the timer goes off, she sees “pack lunch” on the card, closes the app and goes to the kitchen. If the timer had gone off mid-video, the card would still have told her where to go. This is a made-up example, not a customer result or a recommended time limit.
When the exit keeps losing to the feed
If you keep overriding your own stop signal, change what you open rather than relying on a better reminder. Compare a whole-app schedule with a narrower boundary in our app limits and feed blockers guide, and use the practical plan to stop doomscrolling for the wider routine. If short videos are the main pull, see how to block Reels and Shorts on iPhone. Keep essential contacts, emergency alerts, accessibility tools and work or school access available.
For adults managing their own use, Unscroller offers supported focused social browsing on iPhone and iPad. It reduces supported feed and short-video surfaces while keeping useful actions where supported, so you can open a specific account, message or link without an endless feed beneath it. These are web flows; they do not change every native social app. Coverage depends on the platform, device, permissions and third-party changes.
Check setup and plan options during onboarding; completing it does not unlock every paid feature. Standard accounts start at age 16, subject to a higher regional minimum. Parent-managed setups have separate age and Family requirements. Review the privacy policy before choosing features.
- Rixen et al., Proceedings of the ACM on Human-Computer Interaction 7(MHCI), Article 228 (September 2023): The Loop and Reasons to Break It
- Meinhardt et al., Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 10(3), Article 140 (September 2026): Can’t Stop
- Meinhardt et al., arXiv preprint 2607.15818 (posted July 17, 2026)
- Meinhardt et al.: open study app, data and analysis code
- Unscroller: current age and Family requirements
This guide is educational and describes current product behavior where noted. Unscroller is a productivity and digital-wellbeing tool, not medical treatment.
Keep going
Want the message without the endless feed?
Review Unscroller’s supported browsing tools, setup and plan options for your device. Open the account or link you need without a feed underneath it.