Infinite Scroll and Autoplay: How Endless Feeds Change Stopping Cues and Time Awareness

2 days ago

16 min read

Author: Ukrainian Psychological Hub · Published: September 27, 2026 · Editorial Policy

Infinite scroll and autoplay are interface mechanisms that make digital content continue with fewer explicit stopping points. Infinite scroll keeps loading new items as a person moves through a feed. Autoplay starts the next video, episode, or media item without requiring a fresh play decision. Their psychological importance lies less in a mysterious effect on the brain than in a simpler change to the structure of choice: a natural endpoint is weakened or removed, while continuation becomes easy, immediate, or automatic.

Current research supports a measured conclusion. Interface structure can change how long people stay, what they remember, how attentive they feel, and how much agency they report. A 2025 experiment with U.S. Netflix users found that disabling autoplay reduced average daily watching and average session length. A 2026 laboratory study of TikTok-style feed design found that replacing autoplay with explicit playback increased perceived agency and reduced time distortion, while pagination produced more subtle quantitative effects. These findings show that defaults and stopping opportunities can matter. They do not show that one interface feature inevitably creates addiction or causes a mental disorder.

The strongest way to understand endless feeds is therefore as an interaction among design, goals, habits, content, context, and individual differences. This article owns the mechanism question—what infinite scroll and autoplay do to stopping cues and time awareness. Broader questions about why platforms compete for attention belong to the attention economy, while broader problems of phones and apps pulling focus away from a current goal are covered in digital distraction.

What Are Infinite Scroll and Autoplay?

Infinite scroll and autoplay are often grouped together because both can create continuity, but they are different interface mechanisms. Keeping them separate matters because the evidence for one should not automatically be transferred to the other.

Infinite scroll

Infinite scroll is a feed architecture in which additional content loads as the user approaches the apparent end of the current set. There may be no final page, no visible last item, and no required “next page” action. The person still performs actions—scrolling, swiping, tapping, pausing, opening comments—but the feed itself does not regularly present a structural endpoint that asks whether another unit should begin.

A one-week field study of 46 social-media users described infinite scrolling as part of a recurring “loop” in which sessions could become longer than intended and later feel regrettable. Importantly, the Rixen and colleagues study also found that people stopped for many reasons involving the app, their situation, and their own goals. Infinite scroll influenced the conditions of use; it did not erase human agency.

Autoplay

Autoplay changes a different boundary. When one media item ends, the next begins with little or no additional action. In a manual-play design, the end of an item naturally creates a decision point: nothing further happens until the person chooses to continue. Under autoplay, doing nothing can mean continuing. Stopping becomes the action that must interrupt the sequence.

This shift is behaviorally meaningful. In the Netflix autoplay experiment, 76 U.S. users were studied with autoplay enabled or disabled. Disabling autoplay significantly reduced average daily watching and average session length. The same study also found divided user preferences: some valued autoplay’s convenience, while others viewed its influence on continued watching more negatively. That split is useful because it shows why autoplay cannot be summarized as simply “good” or “bad.”

What they share

Infinite scroll and autoplay both reduce the frequency of moments in which continuation requires a new, explicit commitment. That shared property is the core of the stopping-cue question. It is also why the two features can feel similar in use even though one organizes scrolling and the other organizes media playback.

What Is a Stopping Cue?

A stopping cue is a boundary in an activity that creates a plausible moment to stop, switch tasks, or reconsider what to do next. It does not need to be a pop-up warning. The bottom of a page, the end of an episode, the completion of a finite list, a “load more” button, or a required play action can all function as boundaries because the current unit ends before another begins.

The psychological value of a boundary is that it creates a decision point. In a finite sequence, the environment periodically asks, in effect, “continue?” In a continuous sequence, the environment can keep supplying material without that question becoming salient. The person can still stop at any moment, but stopping depends more heavily on an internally generated cue—remembering another goal, noticing the time, feeling fatigue, receiving an external interruption, or deliberately deciding that enough is enough.

This distinction also helps explain why a stopping cue is not the same thing as an interruption. A 2020 laboratory and field study comparing pagination and infinite scrolling in social-commerce interfaces found that interruptions changed different forms of user contribution in different directions: they reduced content appreciation but increased content creation, with user characteristics moderating effects. The Zhang, Liu, and Ho study is a reminder that adding friction can support one goal while hindering another. A boundary is a design tradeoff, not an automatic psychological benefit.

How Endless Feeds Change the Default From Stopping to Continuing

Interfaces influence behavior partly through defaults. With manual playback, the default after a video ends is no playback. With autoplay, the default can be the next video. With pagination, reaching the bottom produces a boundary and requires an explicit next-page action. With infinite scroll, additional content can appear before that boundary is reached. These are small differences in interaction cost, yet small differences repeated dozens or hundreds of times can meaningfully change the shape of a session.

The mechanism does not require the assumption that users are passive victims of technology. People arrive with motives: entertainment, social connection, boredom relief, news seeking, work avoidance, curiosity, habit, or a specific informational goal. Content quality and personal relevance also matter. A person looking for one practical answer may leave quickly; someone relaxing after work may actively want a long, uninterrupted stream. The same interface can therefore support intentional immersion in one context and unwanted overextension in another.

For the broader economic and design context in which platforms may benefit from retaining attention, see Attention Economy: How Digital Platforms Compete for Your Time and Focus. DLA-39 focuses on the narrower behavioral architecture: what happens when the interface supplies fewer natural opportunities to stop.

What Research Shows About Infinite Scroll

Research directly isolating infinite scroll is still smaller than the public discussion around it. The existing evidence includes field studies, controlled interface experiments, and HCI research on absorption and design friction. That mix is useful, but it does not justify claims that every endless feed produces the same effect in every person.

Longer-than-intended use can emerge as a loop

In the 2023 field study by Rixen and colleagues, participants used social media in everyday life for a week while researchers examined reasons for continuing and stopping. Infinite scrolling was associated with the experience of being caught in a loop and with sessions that could be regretfully prolonged. The study is valuable because it studied behavior in natural settings, yet its sample was modest and it does not prove that infinite scroll alone caused every extended session.

Adding friction can change attention and memory

A 2024 experiment by Ruiz, Molina León, and Heuer compared a conventional infinite-scroll condition with a design that required participants to react to a post before further content became available. Thirty participants showed better recognition memory in the friction condition. At the same time, many found the added friction frustrating. The result fits a broader design principle: forcing a pause can increase deliberate engagement, but it can also reduce smoothness and satisfaction.

Feed design can influence absorption and perceived control

Baughan and colleagues deployed a custom Twitter client to 43 U.S. participants and studied what they called normative dissociation—a state of deep absorption accompanied by reduced self-awareness and a weaker sense of agency. Participants sometimes experienced such absorption as a welcome break and sometimes as passive, unsatisfying scrolling. Interventions including reading-history labels, time-limit dialogs, usage statistics, and custom lists reduced reported normative dissociation.

That study is especially useful for avoiding a common category error. Feeling absorbed, losing track of what one has read, or realizing that more time passed than intended does not by itself establish addiction, pathology, or neurological damage. Absorption can occur in many ordinary activities. The clinically relevant questions are different: distress, loss of control, persistence despite harm, and functional impairment require their own evidence and assessment.

What Research Shows About Autoplay

Autoplay has somewhat clearer experimental evidence because it can be switched on or off while other parts of the service remain relatively stable. Even here, effects depend on the platform, the content, and what outcome is measured.

Autoplay can extend viewing sessions

The 2025 Netflix study provides direct behavioral evidence that autoplay can increase consumption. When autoplay was disabled, average daily watching and average session length fell. This does not mean every user watched more because of autoplay, nor does it show that longer viewing was necessarily harmful. It shows that the interface default itself changed aggregate behavior.

Manual play can preserve attention to the transition

A 2024 experiment with 394 participants compared manual playback, fully passive autoplay, and an “interpassive” form of autoplay in which users retained a toggle. The Chen, Kang, Sajjadi, and Sundar study found that interpassive autoplay was preferred to the other modes, yet it also produced more inattentiveness than manual play. Inattentiveness and a heuristic sense of control were associated with stronger perceptions of having gone down a content “rabbit hole.” The result complicates a simple control narrative: an interface can make people feel that control is available while still changing how attentively transitions are processed.

Explicit playback may increase perceived agency

A 2026 preliminary laboratory study manipulated TikTok-style feed features and found that replacing autoplay with explicit playback significantly increased perceived agency and reduced time distortion. Replacing infinite scroll with pagination produced subtler quantitative changes, although participants tended to favor pagination. Because this was a preliminary controlled study, its results are best treated as evidence that small feed-level changes can matter rather than as a universal estimate of how much they matter.

Infinite Scroll, Autoplay, and Time Awareness

“I lost track of time” can describe several different experiences. A person may underestimate elapsed minutes, notice time less frequently, remember little of what happened during the interval, or simply discover that a session lasted longer than planned. These are related experiences, but they are not the same measurement.

The 2026 TikTok-feed experiment linked explicit playback to less time distortion and greater perceived agency. By contrast, a 2025 experimental study of short-video viewing and time estimation with 56 college students found that task type—watching short videos versus reading public articles—did not significantly affect time estimation; task duration mattered more. Problematic short-video watching and estimated weekly use were also not significantly related to the study’s time-distortion measure.

Taken together, these studies support a cautious conclusion. Some interface conditions can affect subjective agency and temporal experience, but “endless feeds make the brain unable to tell time” is far stronger than the evidence. Time awareness depends on how it is measured, how long the task lasts, what content is shown, how engaged the person is, and the exact interface conditions.

This distinction matters for everyday self-observation. Realizing that an hour passed faster than expected is meaningful information about how an activity fits one’s goals. It is not a neurological diagnosis. It may reflect absorption, low-friction continuation, an engaging task, fatigue, boredom relief, or a combination of factors.

Why Endless Feeds Can Feel Hard to Leave

Fewer moments of re-evaluation

Every explicit boundary offers a chance to compare the current activity with competing goals. Should I keep watching, go to sleep, return to work, answer a message, or make dinner? When boundaries are rare, that comparison may occur less often unless the person creates it deliberately. Infinite scroll and autoplay are therefore best understood as reducing environmental prompts for re-evaluation, not as eliminating the capacity to choose.

Attentional absorption

Absorption can narrow awareness of the surrounding environment and of competing intentions. The normative-dissociation research suggests that social-media use can sometimes take this form. Crucially, absorption is not automatically negative: participants described both restorative and unsatisfying experiences. The practical question is whether the experience matches the person’s current goal.

Repeated low-effort continuation

Repeated actions become easier to perform with less deliberation. In an endless feed, the next unit often requires a familiar, minimal action such as a swipe. In autoplay, it may require no action at all. Over time, this can combine with learned routines: opening an app in bed, during a commute, between tasks, or whenever boredom appears. The interface supplies continuity while context supplies the cue.

Repeated checking belongs to a related but distinct mechanism. See Compulsive Phone Checking: Why We Keep Reaching for the Phone for the role of cues, uncertainty, habit, boredom, and checking behavior. Repeated phone checking by itself is not evidence of an OCD compulsion or a formal addiction diagnosis.

Content and goals still matter

An endless feed full of irrelevant material is easy to abandon. A finite page full of highly relevant material can hold attention for a long time. Personal relevance, novelty, social meaning, emotional salience, and the user’s reason for being there all shape persistence. Interface design changes the cost and timing of continuation; it does not replace the psychology of the person using it.

Does Infinite Scroll or Autoplay Cause Addiction?

Infinite scroll and autoplay are not diagnoses, and exposure to these features does not by itself establish a behavioral addiction. Research may describe problematic, compulsive, habitual, excessive, or dysregulated patterns of technology use, but those constructs are not interchangeable. A long session can be intentional. Frequent use can be functional. A habit can be strong without producing clinically significant impairment.

Likewise, “infinite-scroll addiction” is not an official DSM or ICD diagnosis. The evidence reviewed here shows that interface features can influence session structure, attention, memory, perceived agency, and continued consumption under specific conditions. It does not establish that infinite scroll or autoplay inevitably produces an addictive disorder.

The same precision applies to popular neurobiological language. Descriptions such as “dopamine addiction,” “fried brain,” or “rewired attention span” are not adequate summaries of the studies above. The experiments manipulate interface conditions and measure behavior or experience; they do not demonstrate a single dopamine mechanism that explains everyday scrolling.

Infinite Scroll Is Not the Same Thing as an Algorithm

Infinite scroll describes how content is delivered. A recommendation algorithm describes how content is selected or ranked. The two are often combined, but either can exist without the other. A chronologically ordered feed can use infinite scroll. A personalized recommendation system can present results in pages.

Keeping these mechanisms separate prevents causal overreach. If a study finds that autoplay changes viewing duration, that does not tell us whether recommendation quality caused the same effect. If a study finds that personalized content is engaging, that does not establish that infinite scroll was responsible. Real platforms bundle many features, so research that isolates one feature is especially valuable.

Infinite Scroll, Dark Patterns, and Persuasive Design

Infinite scroll and autoplay can appear in debates about dark patterns and persuasive design, but the categories should not be collapsed. A feature’s behavioral effect can be studied without inferring the motives of a specific company or designer. Whether a particular implementation qualifies as deceptive or manipulative depends on its context, disclosure, user expectations, available controls, and the definition being used.

DLA-39 therefore stays at the mechanism level. The broader structural question—why digital products compete for time and attention—is covered by the attention economy article. Broader interface-choice questions belong to the cluster’s separate dark-pattern and persuasive-design owners.

How to Reintroduce Stopping Cues Without Quitting Technology

A practical response to endless feeds is to restore decision points. This is an interface and habit strategy, not a universal treatment. Research does not show that one digital rule automatically improves anxiety, depression, sleep, attention, or well-being for everyone.

Turn off autoplay when the option exists

Disabling autoplay changes the default from “continue unless I stop it” to “stop unless I start the next item.” The Netflix experiment shows that this change can reduce viewing time and session length in at least one real-world service context. It is a direct way to restore a boundary without banning the service.

Choose finite views when available

Some platforms offer search results, saved lists, subscriptions, playlists, or chronological views that are more finite than a recommendation feed. A finite unit gives the session a visible shape. The goal is not to make every interaction laborious; it is to create points where continuing becomes a fresh choice.

Use a deliberate session goal

Before opening a feed, a concrete goal can supply an internal stopping cue: check three messages, watch one saved video, find one recipe, read updates from selected accounts. This technique will not prevent every extended session, but it makes the end condition easier to recognize than an open-ended instruction such as “check social media.”

Create an external boundary

A timer, calendar block, bedtime routine, app limit, or planned transition can provide an external stopping cue when the interface does not. The important distinction is that such tools create a reminder to reconsider, not a guarantee of behavior change. People can ignore reminders, and rigid restrictions can be annoying or counterproductive in some contexts.

Reduce unrelated interruptions when focus is the goal

Endless feeds are only one route by which digital environments pull attention away from a task. Alerts create a different mechanism: interruption. For that evidence, see Notifications and Attention: How Alerts Interrupt Focus and Task Performance. For the broader interaction of devices, apps, task switching, and environmental cues, see Digital Distraction.

When Is Endless Scrolling Actually a Problem?

The useful threshold is functional rather than purely numerical. A person can spend substantial time with digital media for work, education, relationships, creativity, or chosen entertainment without that duration alone proving harm. Concern becomes more meaningful when use repeatedly conflicts with the person’s own goals, displaces important activities, causes significant distress, or continues despite clear negative consequences.

Examples include repeatedly missing sleep because an intended ten-minute session becomes two hours, being unable to complete necessary work because feed use repeatedly interrupts it, or feeling persistent loss of control that produces meaningful impairment. Even then, the interface feature is only one part of the picture. Stress, mood, sleep, attention difficulties, social circumstances, habits, and the specific content being consumed may all matter.

For a broader framework that evaluates technology by how it fits a person’s goals, functioning, and context rather than by screen minutes alone, see Digital Well-Being: What It Is, What Shapes It, and What Research Shows.

What the Evidence Can—and Cannot—Support

The current evidence supports five practical conclusions. First, infinite scroll and autoplay alter the architecture of continuation. Second, removing autoplay can reduce viewing duration in an experimental real-world service context. Third, feed-level design changes can influence memory, attentional experience, perceived agency, and sometimes time distortion. Fourth, adding friction can have costs, including frustration and reduced smoothness. Fifth, effects vary with measurement, task, content, context, and user characteristics.

The evidence does not support five stronger claims. It does not show that every endless feed is intentionally designed to harm users. It does not show that infinite scroll or autoplay inevitably causes addiction. It does not show that these features cause ADHD, depression, anxiety, or another clinical disorder. It does not establish a universal amount of scrolling after which harm begins. And it does not justify treating subjective “lost time” as proof of neurological damage.

The most accurate model is interactional: design changes the environment in which choices occur; people bring motives, vulnerabilities, habits, and goals; content changes what is worth continuing; and context changes what continuing costs. Studying those parts separately produces better explanations than reducing the whole experience to willpower or to an all-powerful algorithm.

Frequently Asked Questions

Why is infinite scroll hard to stop?

Infinite scroll can be hard to leave because it supplies new content without regularly presenting a clear endpoint. That reduces the number of environmental moments that invite re-evaluation. The effect interacts with content relevance, habits, goals, fatigue, boredom, and social context, so difficulty stopping varies across people and situations.

Does infinite scroll make people lose track of time?

It can contribute to subjective time loss in some settings, but the evidence is not universal. A 2026 controlled TikTok-feed study found less time distortion when autoplay was replaced with explicit playback. A 2025 short-video experiment did not find a significant effect of short-video versus reading task type on time estimation. Measurement and context matter.

Is infinite scroll addictive?

“Addictive” is often used casually to mean compelling or hard to stop. Scientifically, that language should be more precise. Infinite scroll is an interface feature, not a diagnosis. Studies can show effects on continued use, attention, memory, agency, or session length without establishing an addictive disorder.

Does autoplay make you watch more?

In at least one strong field experiment, yes. The 2025 Netflix study found that disabling autoplay reduced average daily watching and average session length. That is evidence of a causal effect of autoplay on those viewing outcomes in that study context, not proof that every form of autoplay has the same effect.

Is pagination always better than infinite scroll?

No. Pagination introduces boundaries and can create opportunities to stop, but it also interrupts flow. Research shows tradeoffs: design friction can improve recognition memory while frustrating users, and pagination-versus-infinite-scroll effects can differ depending on whether the goal is consuming, appreciating, or creating content. “Better” depends on the outcome being optimized.

What is the difference between infinite scroll and doomscrolling?

Infinite scroll is an interface mechanism. Doomscrolling is a behavioral pattern centered on prolonged consumption of distressing or threatening news or information. Infinite scroll can make continued consumption easier, but a person can doomscroll on finite pages and can use an infinite feed without consuming negative news.

What is the difference between autoplay and recommendation algorithms?

Autoplay determines what happens at the end of one media item: the next item begins automatically. Recommendation algorithms determine which item is selected or ranked next. A service can use either mechanism independently, though contemporary platforms often combine them.

How can I stop scrolling without deleting every app?

Restore boundaries rather than relying only on willpower: disable autoplay where possible, use finite lists or saved content, open the app with a specific goal, set an external time boundary, and choose views that require an explicit decision to continue. Treat these as experiments in self-regulation and environment design, not as guaranteed treatments for mental-health symptoms.

References

Baughan, A., Zhang, M. R., Rao, R., Lukoff, K., Schaadhardt, A., Butler, L. D., & Hiniker, A. (2022). “I Don’t Even Remember What I Read”: How Design Influences Dissociation on Social Media. CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3491102.3501899

Chen, C., Kang, J., Sajjadi, P., & Sundar, S. S. (2024). Preventing users from going down rabbit holes of extreme video content: A study of the role played by different modes of autoplay. International Journal of Human-Computer Studies, 190, 103303. https://doi.org/10.1016/j.ijhcs.2024.103303

Jiang, Y., Yan, Z., & Yang, Z. (2025). Losing Track of Time on TikTok? An Experimental Study of Short Video Users’ Time Distortion. Behavioral Sciences, 15(7), 930. https://doi.org/10.3390/bs15070930

Monge Roffarello, A., & De Luca, A. (2026). Am I in Control? How the Design of the TikTok Feed Shapes Users’ Sense of Agency. CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3772363.3798790

Rixen, J., Meinhardt, L. M., Glöckler, M., et al. (2023). The Loop and Reasons to Break It: Investigating Infinite Scrolling Behaviour in Social Media Applications and Reasons to Stop. Proceedings of the ACM on Human-Computer Interaction, 7(MHCI), 1–22. https://doi.org/10.1145/3604275

Ruiz, N., Molina León, G., & Heuer, H. (2024). Design Frictions on Social Media: Balancing Reduced Mindless Scrolling and User Satisfaction. Mensch und Computer 2024. https://doi.org/10.1145/3670653.3677495

Schaffner, B., Ulloa, Y., Sahni, R., Li, J., et al. (2025). An Experimental Study Of Netflix Use and the Effects of Autoplay on Watching Behaviors. Proceedings of the ACM on Human-Computer Interaction, 9(2). https://doi.org/10.1145/3710928

Zhang, Y., Liu, L., & Ho, S. Y. (2020). How do interruptions affect user contributions on social commerce? Information Systems Journal, 30(3), 535–565. https://doi.org/10.1111/isj.12266

Psychological Encyclopedia