Doomscrolling began as a word for compulsively reading bad news. It now describes something broader: moving through a feed for long stretches, taking in posts, clips, and headlines one after another, without stopping to think about any of them and without doing anything with them afterwards.
The feed feels like information. It has facts in it, opinions, tips, and news. The question this article asks is what a person actually keeps from those hours, what the research says the habit does, and what the same hours produce when they are spent looking something up on purpose.
The Short Version
| Scrolling a feed | Researching a question | |
|---|---|---|
| Who chooses the topic | The algorithm | You |
| What you do with each item | Glance and move on | Read, compare, take notes, test it |
| What it feels like | Being informed | Effort, sometimes confusion |
| What remains a week later | A mood and a few fragments | An answer you can explain and use |
| What 967 hours a year adds up to | Nothing you can point to | A language, or about 150 books |
How Much Time Goes Into the Feed
DataReportal's Digital 2026 report, drawing on survey data from GWI, found that online adults spend more than an hour a day on social networks such as Facebook and X. When video platforms such as YouTube and TikTok are added, the total rises to 18 hours and 36 minutes a week. Among women aged 16 to 24 it is 25 hours and 45 minutes a week.[1]
Put into a year, the average figure is about 967 hours. That is 40 full days and nights, or roughly 121 eight-hour working days. For the youngest heavy users it is more than 1,300 hours.
Not all of that time is doomscrolling. Some of it is messaging friends, watching something chosen on purpose, or following a tutorial. The share that is passive is what the rest of this article is about.
Why Scrolling Feels Like Learning and Is Not
Memory depends on what a person does with information, not on how much of it passes in front of them. Three findings explain why a feed leaves so little behind.
- Seeing is not retrieving. In a well-known experiment, students who practiced recalling a passage remembered substantially more of it a week later than students who simply read it again, even though the rereaders felt more confident about what they knew.[9] A feed is rereading at its most extreme: thousands of items, each seen once, none recalled.
- Access is mistaken for knowledge. Across nine experiments, Yale researchers found that people who had just looked things up online rated their own understanding higher, mistaking access to information for knowledge in their own heads.[7] An earlier study in Science found that when people expect information to stay available, they remember less of the information itself and more of where to find it.[8]
- Most of it is never even opened. An analysis of more than 35 million public Facebook posts found that around 75% of the links people forwarded were shared without the sharer clicking on them first.[6] The headline was the whole encounter.
The result is a specific kind of illusion. After an hour of scrolling, a person has the feeling of having caught up with the world, and almost nothing they could explain to someone else.
What the Research Links It To
The evidence on harm is real but should be read carefully, because most of it shows association, not cause.
- Mood. A 2026 systematic review and meta-analysis in Frontiers in Psychology covering 44 studies found doomscrolling associated with anxiety and depression (both r = 0.35), with psychological distress (r = 0.51), and with lower well-being and life satisfaction. The authors rated the certainty of the evidence as very low and stated that it does not establish causality.[2]
- Passive use specifically. One of the few experiments on the question found that people asked to use Facebook passively, scrolling without posting or commenting, felt worse afterwards than those who used it actively. A follow-up tracking people through their day found the same pattern and pointed to envy as the reason.[4]
- Attention. A 2025 meta-analysis in Psychological Bulletin, pooling 71 studies and 98,299 participants, found that heavier use of short-form video went with poorer cognition overall (r = -.34), most strongly for attention (r = -.38) and inhibitory control (r = -.41).[3]
- Remembering what you meant to do. In a controlled experiment with 60 participants, a break spent on TikTok significantly reduced people's ability to carry out a task they had planned beforehand. Breaks spent on Twitter, on YouTube, or resting did not.[5]
The honest reading is this: people who scroll more report worse mood and weaker attention, and the two experiments suggest part of that runs from the scrolling to the effect. It is also likely that people who are already anxious or distracted scroll more. Both can be true.
What the Same Hours Buy When You Research Instead
Research here does not mean academic work. It means starting with a question, looking for the best sources on it, reading them properly, and coming out with an answer you can use. The difference in output is large because every hour builds on the one before it.
| What 967 hours a year is equal to | The arithmetic |
|---|---|
| Professional proficiency in a new language | The US Foreign Service Institute budgets 24 weeks of 40 hours (23 in class, 17 of self-study), or 960 hours, to bring a diplomat to general professional proficiency in languages such as Italian, Portuguese, or Dutch[10] |
| About 150 books | At the average adult reading speed of 238 words a minute for non-fiction,[11] 967 hours covers about 13.8 million words, or roughly 150 books of 90,000 words each |
| Half a working year on one subject | 967 hours is about 121 eight-hour days, close to six months of full-time work |
Nobody will convert every minute of feed time into study, and nobody should try. The comparison matters because of how lopsided it is. Even a quarter of the average, about four and a half hours a week, comes to more than 240 hours a year: enough to read around 38 books or to work through a subject in real depth.
Why Research Sticks and the Feed Does Not
The gap is not about willpower or intelligence. It follows from four differences in what the mind is doing.
- A question comes first. A person researching knows what they want to find out, so each new fact has somewhere to go. A feed supplies answers to questions nobody asked.
- The same topic comes back. Research returns to one subject from several sources, which is repetition spaced over time. A feed changes subject every few seconds.
- The reader has to produce something. A note, a comparison, a decision, an explanation to someone else. Producing is retrieval, and retrieval is what builds memory.
- The result gets used. Research usually ends in an action: a purchase made, a problem fixed, a skill tried. Use is the strongest test of whether something was understood.
How to Turn Scrolling Time Into Research Time
- Keep a question list. Write down things you actually want to know. When you reach for the phone, open the list and pick one.
- Search instead of browsing. Go to a search engine or an AI assistant with a question, not to a feed with none. How to choose between them is covered in How People Search for Information Now.
- Write three lines afterwards. What you learned, where it came from, and what you will do with it. This single habit converts reading into retrieval.
- Save, then return. If a post in the feed is worth knowing, save it and come back with time to read the source. If you never come back, it was not worth knowing.
- Move the apps. Take feed apps off the home screen and turn off their notifications. The reasoning is set out in Digital Minimalism.
- Use the feed on purpose. Messaging people, following a tutorial, and watching something you chose are not the problem. Scrolling with no aim is.
For study methods that make the hours count, see Nine Methods to Learn Anything Faster.
The Bottom Line
The feed and the search box deliver the same raw material, and they produce opposite results. One hands over a stream chosen by someone else and asks for nothing back, so nothing stays. The other starts from your own question and makes you work, so the answer becomes yours. The hours are already being spent. The only choice is which of the two they go to.
Sources & Further Reading
- DataReportal. Digital 2026: Global Overview Report. View source
- "Doomscrolling, mental health, and psychological well-being in adolescents and adults: a systematic review and meta-analysis." (2026). Frontiers in Psychology, 17. View source
- Nguyen, L., Walters, J., Paul, S., et al. (2025). "Feeds, feelings, and focus: A systematic review and meta-analysis examining the cognitive and mental health correlates of short-form video use." Psychological Bulletin, 151(9), 1125-1146. View source
- Verduyn, P., Lee, D. S., Park, J., et al. (2015). "Passive Facebook usage undermines affective well-being: Experimental and longitudinal evidence." Journal of Experimental Psychology: General, 144(2), 480-488. View source
- Chiossi, F., Haliburton, L., Ou, C., Butz, A., and Schmidt, A. (2023). "Short-Form Videos Degrade Our Capacity to Retain Intentions: Effect of Context Switching On Prospective Memory." CHI 2023. View source
- Sundar, S. S., Snyder, E. C., et al. (2024). "Sharing without clicking on news in social media." Nature Human Behaviour. View source
- Fisher, M., Goddu, M. K., and Keil, F. C. (2015). "Searching for explanations: How the Internet inflates estimates of internal knowledge." Journal of Experimental Psychology: General, 144(3), 674-687. View source
- Sparrow, B., Liu, J., and Wegner, D. M. (2011). "Google effects on memory: Cognitive consequences of having information at our fingertips." Science, 333(6043), 776-778. View source
- Roediger, H. L., and Karpicke, J. D. (2006). "Test-enhanced learning: Taking memory tests improves long-term retention." Psychological Science, 17(3), 249-255. View source
- US Department of State. "Foreign Language Training." Foreign Service Institute. View source
- Brysbaert, M. (2019). "How many words do we read per minute? A review and meta-analysis of reading rate." Journal of Memory and Language, 109, 104047. View source
