Infinite Scroll Psychology: Why Apps Hold Attention
Photo: Sanket Mishra / Pexels
Every book has a last page. Every newspaper has a back cover. Every TV show has credits. Infinite scroll, the design pattern that loads new content automatically as a user nears the bottom of a feed, was built specifically to remove that boundary. Its inventor, designer Aza Raskin, has said the pattern wastes roughly half a million human lifetimes of scrolling every month, and that he never anticipated how it would be combined with algorithmic ranking to hold attention far past what any single user intended (NPR's Planet Money). This piece explains the specific psychological mechanisms, tested in laboratories decades before smartphones existed, that make a feed with no bottom so hard to put down.
The design problem infinite scroll was built to solve
Before infinite scroll became standard, web pages used pagination: click "next" to see the next ten results. That single click mattered more than it looked. It was what behavioral scientists call a decision point, a moment where a person has to consciously choose to continue rather than simply keep doing what they were already doing.
Raskin's own account of designing infinite scroll frames it as a usability fix. In his words, describing the original reasoning: "Every time I ask the user to make a choice they don't care about, I have failed as a designer... if you're scrolling down a set of blog posts or a set of search results and you haven't seen what you're looking for, you keep scrolling, then don't make me click the next button" (NPR). That is a reasonable engineering goal for search results. Applied to a ranked feed of content chosen to maximize engagement, it produces something very different: a mechanism with no natural exit.
Removing the click did not just make browsing more convenient. It removed the moment where a person reflects, even briefly, on whether to keep going. Raskin later summarized the lesson this way: "optimizing something for ease-of-use does not mean best for the user or humanity" (NPR).
Compare pagination to a staircase with landings. Every landing is a natural place to stop, catch your breath, and decide whether to keep climbing. Infinite scroll replaces the staircase with an escalator that never ends: you can step off any time, technically, but nothing ever tells you it is time to.
How unpredictable rewards keep the thumb moving
The most consequential piece of the mechanism was not discovered by a social media company. It comes from mid-20th-century behaviorism and, later, neuroscience.
B.F. Skinner's operant conditioning research found that behaviors reinforced on a variable-ratio schedule, meaning the reward arrives after an unpredictable number of repetitions rather than a fixed number, are more resistant to extinction than behaviors reinforced every single time. This is the same schedule that makes slot machines and lottery tickets so effective at sustaining play: the person cannot predict which pull, spin, or scratch will pay off, so they keep going. A feed works the same way. Most posts are unremarkable. Occasionally one is genuinely funny, surprising, or emotionally charged. The unpredictability of which scroll produces that post is the reinforcement schedule, not the content itself.
Neuroscientist Wolfram Schultz's work on dopamine, based on recordings from macaque monkeys performing reward-prediction tasks, gives the biological half of the story. Schultz found that dopamine neurons fire most strongly not when a reward arrives, but when a reward arrives unpredictably ("unpredicted reward"), and that neuron activity actually drops below baseline when an expected reward fails to appear (Hollerman & Schultz, Nature Neuroscience, 1998, PDF via Harvard Medical School; Schultz, Dialogues in Clinical Neuroscience, 2016). In plain terms: the brain's reward system is tuned to fire hardest on uncertainty, not on the payoff itself. This is why anticipation, the half-second before a page refreshes and new content loads, is doing more neurological work than the content that follows it.
Why does infinite scroll feel more compulsive than reading a magazine?
A magazine has a fixed, known amount of content and a physical end. Infinite scroll's feed is algorithmically re-ranked and effectively bottomless, and each new item arrives on a schedule the brain cannot predict. That combination of unlimited supply and unpredictable payoff maps directly onto the variable-ratio reinforcement schedule that produces the most persistent, extinction-resistant behavior in operant conditioning research.
The Zeigarnik effect: why an unfinished feed nags at you
There is a second, less-discussed mechanism working alongside the reward system: the brain's handling of incomplete tasks.
Soviet psychologist Bluma Zeigarnik documented in the 1920s that people remember interrupted or unfinished tasks better than completed ones. The effect, now named after her, is explained as a function of the brain's goal system: once a goal is activated, it stays cognitively "open" and continues to draw attention until it is either completed or consciously closed off (nesslabs.com summary of the Zeigarnik and Ovsiankina effects). A related finding, the Ovsiankina effect, describes the strong pull people feel to resume an interrupted task even without any reward attached to finishing it.
Follow-up research by Roy Baumeister and E.J. Masicampo found that unfinished goals generate intrusive thoughts that specifically degrade performance on unrelated tasks, and that the intrusion is reduced not by finishing the task but by simply making a concrete plan for it (summarized via CogniFit; nesslabs.com). A feed is, structurally, an unfinished task by design: there is always more below the fold, and the app deliberately gives no signal of "you have now seen everything." Closing the app does not resolve the open loop the way finishing a chapter does. That is one reason people report picking the phone back up minutes after intentionally putting it down, not because anything new happened, but because the "unfinished" feed is still cognitively active.
Social comparison and validation: the second reward system
Variable rewards explain why scrolling itself is sticky. They do not fully explain why the content people are scrolling through, specifically posts from other people, carries so much pull. That requires a second, older piece of psychology: social comparison.
Leon Festinger's 1954 social comparison theory proposed that people have an inherent drive to evaluate their own opinions and abilities by comparing themselves to others, particularly others they see as similar to themselves. Feeds built around curated self-presentation, filtered photos, highlight-reel life updates, engagement counts visible on every post, are close to a purpose-built engine for that comparison drive. Every scroll surfaces a new data point for "how am I doing relative to this person."
Internal Meta research made public through the 2021 disclosures by former Facebook product manager Frances Haugen gives a concrete, sourced measurement of this effect at scale. One internal slide, based on Meta's own user research and reported by The Wall Street Journal and later covered by CBS's 60 Minutes, stated the company's own finding: "We make body image issues worse for one in three teen girls" who already have body image concerns (CBS News, 60 Minutes). The same internal documents, submitted in testimony to the U.S. Senate, showed 13.5% of teen girl users said Instagram made suicidal thoughts worse, and 17% said it worsened their eating disorders (CBS News; congressional testimony documented via NPR). These are the platform's own researchers' findings, not outside critics' estimates, which is part of why the disclosure prompted congressional hearings rather than a routine news cycle.
Is infinite scroll designed to exploit these mechanisms deliberately?
The mechanisms (variable rewards, unfinished-task psychology, social comparison) predate social media by decades and were documented independently in behavioral labs. What is deliberate is the pairing: product teams A/B test features like autoplay, pull-to-refresh, and algorithmic ranking specifically to maximize time-on-app metrics, which means these well-known psychological levers get selected for and reinforced by engagement data even without any single designer intending to replicate a slot machine.
How much time this actually adds up to
The mechanisms above are not abstract. Common Sense Media's long-running census of teen and tween media use gives a direct measure of the outcome. Across the 2015–2021 period the organization tracked, average daily screen media use among U.S. teens aged 13 to 18 rose from 6 hours 40 minutes in 2015 to 8 hours 39 minutes in 2021, an increase of nearly two hours per day over six years (Common Sense Media, "Landmark Report"). Within that total, girls reported averaging about 5.3 hours of social media use per day compared to 4.4 hours for boys, according to the same census data.
That growth tracks closely with the period in which infinite scroll and algorithmic ranking became the default design across nearly every major platform, rather than an optional feature on a handful of apps. Correlation between a design pattern's adoption and a rise in usage time is not the same as controlled proof that the pattern alone caused the increase; screen time also rose due to smartphone ownership rates, app variety, and, during 2020 to 2021, pandemic-era isolation. But the mechanism-level evidence (variable-ratio reinforcement, dopamine response to unpredictability, the Zeigarnik effect, and measured social-comparison harms) gives a specific, testable account of why a bottomless feed would produce exactly this kind of session-length growth, rather than leaving the increase unexplained.
| Mechanism | Origin of the research | What it does to attention | Documented evidence |
|---|---|---|---|
| Variable-ratio reinforcement | B.F. Skinner, mid-20th-century operant conditioning | Makes the behavior resistant to stopping because reward timing is unpredictable | Same schedule used in slot machine design; foundational behaviorism |
| Dopamine prediction-error response | Wolfram Schultz, macaque dopamine neuron recordings | Anticipation of an uncertain reward produces a stronger neural response than the reward itself | Hollerman & Schultz, Nature Neuroscience, 1998 (PDF) |
| Zeigarnik / Ovsiankina effect | Bluma Zeigarnik, 1920s Berlin school of Gestalt psychology | An endless, un-completable feed stays cognitively "open," pulling attention back after the app closes | Baumeister & Masicampo replication on intrusive thoughts (summary) |
| Social comparison | Leon Festinger, 1954 social comparison theory | Curated content triggers automatic self-evaluation against peers, driving repeated checking | Meta's own internal research on teen body image, disclosed 2021 (CBS News) |
| Removed decision point | Aza Raskin's infinite scroll design, 2006 | Deletes the "click next" moment that once forced a conscious choice to continue | Raskin's own account and testimony (NPR) |
Table: The psychological mechanisms behind infinite scroll's attention hold, and where each one was independently documented.
What people get wrong about "addictive" apps
The most common misconception is that infinite scroll works through some novel, uniquely modern manipulation technique invented by app designers. It does not. Every mechanism above (variable-ratio schedules, dopamine prediction-error signaling, the Zeigarnik effect, social comparison theory) was documented in psychology and neuroscience labs before smartphones existed, in some cases before the internet existed. What is new is the delivery system: a feed can serve millions of personalized variable-ratio schedules simultaneously, re-ranked in real time based on what kept that specific person scrolling last time.
A second misconception is that willpower is the correct lever to pull. Telling someone to "just put the phone down" treats the problem as a character flaw rather than a designed system. The Baumeister and Masicampo research on unfinished tasks is useful here precisely because it points to a structural fix rather than a willpower fix: making a concrete plan (deciding in advance when a session ends, or what specific thing you are checking for) closes the open loop the same way finishing a task does, without requiring the task to actually be finished. That is closer to how effective interventions against variable-ratio design tend to work: adding friction back into the system (a scroll-limit prompt, a hidden like count, a "you're all caught up" marker) rather than asking users to out-willpower a system built by product teams running continuous A/B tests against exactly that willpower.
A third misconception is that this is only a teen problem. The Common Sense Media data cited above focuses on teens because that population is tracked most rigorously and because adolescent brains, still developing prefrontal-cortex regulation, are more vulnerable to reward-driven behavior. But the underlying mechanisms (variable-ratio reinforcement, dopamine prediction error, Zeigarnik-effect open loops) are not adolescent-specific. They describe how the human reward and attention systems work generally, which is exactly why the same feed design is effective across age groups. App categories vary widely, from feeds to relationship and couples apps.
What this means
Infinite scroll is not one trick but a stack of well-documented psychological mechanisms, each independently validated decades before social media existed, applied simultaneously and personalized in real time. The unpredictability of the reward keeps dopamine-driven anticipation high; the absence of a "you're done" signal keeps the Zeigarnik effect's unfinished-task pull active; the content itself, mostly curated self-presentation from peers, activates social comparison on nearly every scroll. None of that makes the pull imaginary or a personal failing, and none of it requires a conspiracy: it requires only that product teams optimize for time-on-app and let A/B testing find, and keep, whatever features move that number, which happen to be the same features psychology labs identified generations earlier as maximally attention-holding.
Read next:
- /psychology/cognitive-biases-explained
- /psychology/how-memory-can-be-false
- /experiments/the-asch-conformity-experiment
Frequently Asked Questions
Q: What is infinite scroll and who invented it? A: Infinite scroll is a web and app design pattern that automatically loads more content as a user approaches the bottom of a page, removing the need to click "next page." It was created by designer Aza Raskin in 2006 to reduce friction in browsing search results and blog archives, not originally for social feeds (NPR).
Q: Why does infinite scroll feel addictive? A: It pairs an unpredictable reward schedule (some posts are engaging, most are not, and you cannot predict which) with a feed that never signals completion. Variable-ratio reinforcement, documented by B.F. Skinner, is the most extinction-resistant reward schedule known in behavioral psychology, and Wolfram Schultz's dopamine research shows the brain's reward neurons fire hardest on unpredictable, not predictable, rewards.
Q: What is the Zeigarnik effect and how does it relate to scrolling? A: The Zeigarnik effect, documented by psychologist Bluma Zeigarnik in the 1920s, describes the brain's tendency to keep unfinished tasks cognitively "active," making them intrude on attention until completed or consciously closed off. An infinite feed has no completion point by design, so it stays open in a user's mind even after the app is closed (nesslabs.com).
Q: Did Facebook or Instagram know their apps affected teen mental health? A: Internal Meta research disclosed by whistleblower Frances Haugen in 2021 found that Instagram worsened body image issues for one in three teen girls with existing body image concerns, and that 13.5% of teen girls said the app worsened suicidal thoughts while 17% said it worsened eating disorders. These were the company's own researchers' findings, presented to the U.S. Senate (CBS News).
Q: How much has screen time actually increased? A: Common Sense Media's census found U.S. teen media use rose from 6 hours 40 minutes per day in 2015 to 8 hours 39 minutes per day in 2021, an increase of nearly two hours daily over six years, spanning the same period infinite scroll and algorithmic feeds became standard across major platforms (Common Sense Media).
Q: Is willpower enough to stop compulsive scrolling? A: Research on the Zeigarnik effect suggests structural fixes work better than willpower alone. Baumeister and Masicampo found that making a concrete plan for an unfinished task reduces the intrusive pull of that task nearly as effectively as finishing it, which is why time-limit prompts and deliberate stopping rules tend to outperform simply resolving to use an app less.
Q: Does infinite scroll affect adults the same way it affects teens? A: The core mechanisms, variable-ratio reinforcement and dopamine prediction-error response, are general features of the human reward system and are not adolescent-specific. Adolescents are considered more vulnerable because the prefrontal cortex regions that regulate reward-seeking behavior are still developing, but adults show the same underlying response patterns.
Q: Is infinite scroll the same thing as a slot machine, psychologically? A: The reinforcement schedule is functionally the same: both deliver rewards unpredictably after a variable number of actions, which is the schedule B.F. Skinner's research identified as most resistant to extinction. The content differs (social validation and novelty versus money) but the structure driving repeated action is the same variable-ratio pattern.
Q: Can app design features reduce infinite scroll's pull without removing the feature entirely? A: Yes. Features that reintroduce a stopping cue, such as end-of-feed markers, scroll-time reminders, or hidden engagement counts, add back the decision point that infinite scroll originally removed. Research on the Zeigarnik effect specifically supports this: closing the "open loop" with a clear signal reduces the pull even when nothing about the underlying content changes.
Q: Why do social media platforms keep infinite scroll if it has documented harms? A: Product teams optimize primarily for time-on-app and engagement metrics through continuous A/B testing, and features that leverage variable-ratio reinforcement and social comparison reliably outperform alternatives on those metrics. Aza Raskin has argued that optimizing purely for ease-of-use and engagement does not equate to optimizing for user wellbeing, which is the core tension behind ongoing litigation and regulatory scrutiny of feed design.
Sources
- NPR, "Why infinite scroll's inventor wants to kill his creation," 2026
- KOB 4, "Inventor of infinite scroll says he regrets it at New Mexico Meta trial"
- Hollerman, J.R. & Schultz, W., "Dopamine neurons report an error in the temporal prediction of reward," Nature Neuroscience, 1998
- Schultz, W., "Dopamine reward prediction error coding," Dialogues in Clinical Neuroscience, 2016
- nesslabs.com, "The psychology of unfinished tasks: the Zeigarnik and Ovsiankina effects"
- CogniFit Blog, "Stuck on Unfinished Tasks? How the Zeigarnik Effect Drives Memory, Attention, and Productivity"
- CBS News, 60 Minutes, "Facebook knew Instagram was pushing girls to dangerous content," 2022
- NPR, "Whistleblower tells Congress that Facebook products harm kids and democracy," 2021
- Common Sense Media, "Landmark Report: U.S. Teens Use an Average of Nine Hours of Media Per Day"
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles
Motivated Reasoning: Why Facts Do Not Change Minds
Motivated reasoning explains why smart people reject strong evidence: the brain protects identity first and accuracy second. Here is the research behind it.
False Memory Psychology: Why Confidence Is Not Proof
Human memory is reconstructive, not a recording. This guide explains false memories, suggestion, confidence, emotion, and eyewitness mistakes.
The Psychology of Social Media: Why It's So Addictive
Social media platforms lean on variable rewards, cue-triggered dopamine wanting, and social comparison. Here is what the research shows, and where it falls short.
How Social Media Rewires the Brain
Feeds and notifications work on the brain's reward system in ways that resemble variable-reward schedules. Here is what imaging studies, experiments, and large surveys actually show about how social media affects the brain, and where the evidence runs out.
The Science of Curiosity: Why We Want to Know
How curiosity works: the information-gap theory, dopamine reward circuits, three types of curiosity, the brain regions involved, and why the drive to learn persists across the lifespan. The neuroscience and psychology of the desire to know.
Why Do We Get Bored? The Science Behind Boredom and Motivation
Boredom is not simply lack of stimulation but a complex cognitive and emotional state signaling that current activities are not meeting psychological needs. Research suggests boredom arises when attention demands are too low (monotony, routine) or when the gap between attention available and attention demanded is mismatched. The attention restoration theory proposes that boredom serves as a signal that the mind needs a change. The default mode network (brain regions active during rest) becomes hyperactive during boredom, associated with mind-wandering and future planning. Boredom is not uniformly negative: susceptibility to boredom varies with personality traits (novelty-seeking, sensation-seeking) and predicts creativity, risk-taking, and innovation. Some of history's greatest scientific and artistic achievements came from bored minds seeking stimulation. Understanding boredom reveals it as a motivational signal that pushes people toward novel goals, exploration, and self-improvement.