Attention Span Statistics 2026: Where Every Number Comes From

You need one number to open a talk, a lesson plan or a blog post about focus, and the first one you find is the famous line: humans now have an 8-second attention span, one second less than a goldfish. It's tidy, it's dramatic, and it sits on page one of Google. It also has no study behind it.

This page traces that claim to its source, then gives the numbers that replace it, each one linked to the original paper, survey or report, with what was measured and how many people took part. We also checked Google ourselves: in October 2026, the claim still appears as fact on 11 of the 39 top-ranking pages we could read.

Key Attention Span Statistics

• 11 of 39 readable top Google results for 12 attention-span searches state the 8-second goldfish claim as fact, and 7 of those 11 are addiction treatment or therapy providers (BestWebsiteTemplate.com study, October 2026).
• For the search "attention span statistics", 4 of the 8 readable top-10 pages repeat the claim as fact (our study, October 2026).
• Microsoft Canada's 2015 report credited the 8-second figure to the website Statistic Brain and dated it 2013; Microsoft's own research was a survey of 2,000 Canadians and EEG recordings of 112 (Microsoft Canada, 2015).
• 50% of UK adults believe the 8-second claim and only 25% know it's false (King's College London, survey of 2,093 adults, 2021).
• Office workers stayed on one screen for a median of 40 seconds (mean 47 seconds) before switching (Mark et al., CHI 2016, 40 workers).
• Interrupted work was picked back up after an average of 25 minutes 26 seconds, with about two other tasks done in between (Mark, Gonzalez and Harris, CHI 2005).
• People's minds wander during 46.9% of sampled waking moments (Killingsworth and Gilbert, Science, 2010, 2,250 adults).
• Sustained attention peaks at about age 43 (Fortenbaugh et al., Psychological Science, 2015, 10,430 people).
• Adults' scores on a standard concentration test rose between 1990 and 2021 across 287 samples and 21,291 people in 32 countries (Andrzejewski, Zeilinger and Pietschnig, 2024).

What Is the Average Human Attention Span?

There is no single average attention span, and no study has measured one in seconds. Attention depends on the task: people hold focus through a two-hour film and lose it in a dull meeting. What researchers can measure is narrower: how long people stay on one screen, how often their minds wander, or how well they do on a concentration test. Those numbers are below, each with its method.

The closest thing to the "seconds" figure people want comes from Gloria Mark at the University of California, Irvine, who has logged how long office workers stay on one screen before switching. In her own summary, that was about two and a half minutes in 2003 to 2004, 75 seconds in 2012 and 47 seconds in recent years (University of California).

The recent figure needs care. In the 2016 study behind it, 40 workers at a large US tech company were tracked for about 12 workdays. The mean time on a screen was 47 seconds, but the median, which the paper's abstract reports, was 40 seconds. That's screen switching at work, not the limit of human focus: switching to email to check one fact doesn't mean attention ran out.

Is the 8-Second Goldfish Attention Span Real?

No. The figure first appeared on Statistic Brain, a statistics website that named no study, only "National Center for Biotechnology Information, U.S. National Library of Medicine, The Associated Press". Microsoft Canada put it on one slide of a 2015 advertising report, labelled with that source. The press then reported it as Microsoft's finding. Goldfish research doesn't support the fish half either.

Here is the chain, from the documents themselves:

1. Statistic Brain (2014): an archived page lists "average attention span" as 12 seconds in 2000, 8 seconds in 2013 and 9 seconds for a goldfish, with a research date of January 2014 and no named study.
2. Microsoft Canada (Spring 2015): page 6 of its 52-page "Attention spans" report repeats those three numbers and credits Statistic Brain. Microsoft's own work in the report was an online survey of 2,000 Canadians and EEG brain recordings of 112 people. Neither measured attention in seconds.
3. The press (May 2015 on): news outlets reported that Microsoft had found humans now trail goldfish.
4. The BBC check (2017): the BBC's More or Less team asked the US National Library of Medicine and the Associated Press for the research behind the figure. Neither could find any. Statistic Brain did not respond.

Retellings keep drifting. Several pages we read say Microsoft measured the drop in 2015, when the slide dates the 8 seconds to 2013. Others turn it into a gap between millennials and Gen Z, but the slide compares years, not generations. One page credits the figure to a Microsoft study of task switching, which is not what the report contains.

As for the fish: goldfish are a standard species in learning and memory research. In one study, 8 goldfish learned a one-hour daily feeding window over several weeks and kept anticipating it for days after the food stopped. There is no study measuring a "9-second" goldfish attention span.

How Many Top Google Results Still Repeat the 8-Second Claim?

In October 2026, we read the top Google results for 12 attention-span searches. Of 39 pages we could read, 24 mention the 8-second or goldfish figure: 11 state it as fact, 12 call it a myth or unsupported, and 1 stays neutral. For "attention span statistics", the page writers are most likely to cite from, 4 of 8 readable results repeat it as fact.

Who repeats it matters. Seven of the 11 pages presenting it as fact belong to addiction treatment or therapy providers, the kind of health site readers tend to trust. The other four are two marketing agencies, a university admissions marketing page and a student newspaper. Pages that debunk the claim include a hospital system, two universities and The Conversation, which updated its 2025 article in September 2026 to say Microsoft's claim was unsubstantiated.

How you search changes what you find:

The pattern is plain: search for the myth by name and Google shows you the debunks. Search for a number ("average human attention span") and the top result can hand you the myth as fact. That's how the figure keeps reaching new writers.

Are Attention Spans Getting Shorter?

The evidence doesn't show a decline in the ability to concentrate. A 2024 meta-analysis of the d2 concentration test, covering 287 samples, 21,291 people and 32 countries from 1990 to 2021, found adults' scores rose moderately over time, while children showed no gains. What has clearly changed is how often people switch: between screens, apps and topics.

The switching data is real and worth citing correctly:

So the honest sentence is: people switch tasks and topics more often than they did 20 years ago, and heavy short-video use is linked with weaker attention scores, but there is no evidence of a population-wide collapse in the ability to focus.

How Long Does It Take to Refocus After an Interruption?

The widely quoted "23 minutes and 15 seconds" comes from a 2006 Gallup interview with Gloria Mark, not from a paper. Her 2005 study of 24 workers found interrupted work was resumed after an average of 25 minutes 26 seconds, and people did about two other tasks in the meantime. It measures time until return, not time lost refocusing.

The 2005 paper adds two figures writers rarely use: 57.1% of stretches of work on a project were interrupted, and 77.2% of interrupted work was picked up again the same day. If you cite the "about 23 minutes" figure, credit the interview; if you want the peer-reviewed number, use 25 minutes 26 seconds, and describe it as time before returning to a task.

How Long Do People Pay Attention Online?

On the web, the first seconds decide whether a visitor stays. A Microsoft Research study of 205,873 pages found the longer someone stays on a page, the less likely they are to leave, a pattern on 98.5% of pages. The best-known summary, from Nielsen Norman Group, is that users often leave web pages in 10 to 20 seconds.

For writers, the practical reading is that readers decide fast whether to stay, then read selectively. At an average reading speed of about 238 words a minute, 10 to 20 seconds is only 40 to 80 words, which is why the answer should sit at the top of a page. Our own study of how long a blog post should be found the median top-ranking page runs about 1,000 words, so length isn't the issue; burying the answer is.

How Much Time Do People Spend on Screens and Phones?

UK adults spent 4 hours 30 minutes online each day in May 2025, according to Ofcom, rising to 6 hours 20 minutes for 18 to 24 year olds. Smartphones made up 77% of that time. In a smaller logging study, young adults used their phones about 85 times a day, and more than half of those uses lasted under 30 seconds.

The phone study also found people's own estimates of their use didn't match the logs, which is a good reason to prefer measured figures over self-reported ones.

What Is the Average Attention Span by Age?

The best large study puts the peak of sustained attention at about age 43. Researchers tested 10,430 people aged 10 to 70 on a continuous attention task and found performance peaked in the early 40s, with people becoming more cautious in how they responded as they aged. The popular "2 to 3 minutes per year of age" rule for children has no research source we could find.

That rule appears on many parenting sites, with age tables that disagree with each other: one gives a 2-year-old 4 to 10 minutes, another 4 to 6. Treat it as a rough heuristic for planning activities, not a statistic.

For teenagers and screens, the strongest data is a JAMA study of 2,587 Los Angeles students aged 15 and 16, followed for two years. ADHD symptoms appeared in 4.6% of those with no frequent digital media use, against 9.5% of those with 7 frequent activities and 10.5% with 14. The authors report an association, not proof that media use causes the symptoms.

Do Students Lose Focus After 10 to 15 Minutes in a Lecture?

The research doesn't support a fixed limit. A 2016 review in Advances in Physiology Education found the primary data do not back a 10 to 15 minute attention limit, and that differences between teachers matter more than lecture format. A 2007 review in Teaching of Psychology reached the same conclusion: little support for a decline after 10 to 15 minutes.

Both reviews, by Bradbury (2016) and Wilson and Korn (2007), found few primary studies behind the 10-minute figure. Bradbury notes that the most-cited source barely discusses student attention, and Wilson and Korn found the studies relied on indirect measures such as note-taking and self-reports while ignoring differences between students. Breaking a lecture into parts can still help; it's just not a law of the brain. If you are planning a talk by the clock, our average speaking rate study shows how many words fit into each minute.

Which Attention Span Statistics Are Safe to Cite?

Use this table before you quote a number. "Avoid" means the figure has no study behind it or is commonly described wrongly.

If you write about AI tools and attention, our AI writing statistics page uses the same rule: every figure linked to the study that produced it.

Methodology

On 25 October 2026 we collected the top 10 US desktop Google results for 12 attention-span searches (103 results, 64 unique pages). We skipped 17 forum, social and video pages and could not load 8 more, leaving 39 readable pages. We searched each for the 8-second and goldfish claims, then read every mention and labelled the page as stating the claim as fact, debunking it, or neutral.

Limits: one day, one country, one device type, and pages that block automated visits were left out. One person labelled the pages; the raw results and labels are kept on file. All other figures on this page come from the primary source linked beside them, opened and checked in October 2026. Where a figure comes from a company that sells related services, we say so.

How to Cite This Page

You're welcome to use any figure or table here. Please credit BestWebsiteTemplate.com and link to this page. For the study figures, we suggest also linking the original source shown next to each number, so your readers can check it themselves.