Attention Span Statistics 2026: Goldfish Myth & 47 Seconds
The claim that humans now have an 8-second attention span, shorter than a goldfish’s 9, was repeated by Time, the Guardian and the New York Times after a 2015 Microsoft Canada marketing report. Microsoft didn’t measure it. The report credited a website called Statistic Brain, whose listed sources told the BBC they had no record of the research, and no goldfish study behind the 9 seconds has ever turned up. What researchers have measured is narrower: office workers keep one window on screen for 47 seconds on average before switching (Mark et al., CHI, 2016). Here are the attention span statistics worth knowing in 2026, each sourced inline with the year noted.
The 8-second attention span traces to a website, not a study
8 seconds in 2013, down from 12 seconds in 2000, against 9 seconds for a goldfish: that’s the chart on page 6 of Microsoft Canada’s “Attention spans” report, and the line under it reads “Source: Statistic brain” (Microsoft Canada, archived copy, 2015). Microsoft’s own research was an online survey with attention games taken by 2,000 Canadians in late 2014, plus EEG recordings of 112 people. Neither produced a figure in seconds.
The same 8 seconds was labeled 2012 before it was labeled 2013. Statistic Brain’s page in February 2012 listed “The average attention span in 2012: 8 seconds,” citing only The Associated Press (Statistic Brain, archived, 2012). By March 2015 the row read 2013, and the National Center for Biotechnology Information and the U.S. National Library of Medicine had been added as sources (Statistic Brain, archived, 2015).
Neither listed source could find the research. When the BBC’s More or Less contacted the National Center for Biotechnology Information and the Associated Press, “neither can find any record of research that backs up the stats,” and the attention researchers it interviewed had no idea where the numbers came from (BBC, 2017). Open University psychologist Gemma Briggs told the program attention is “very much task-dependent,” which makes a single average close to meaningless.
The only numbers on the Statistic Brain page with a real source are its web-browsing figures, such as 17% of page views lasting under 4 seconds. Those come from Jakob Nielsen’s analysis of 59,573 page views logged by 25 users in a study by Weinreich and colleagues (Nielsen Norman Group, 2008).
Microsoft’s report argued against its own headline. Its foreword opens “Think digital is killing attention spans? Think again,” and its survey found that age “isn’t significantly tied to sustained attention”: 31% of 18-to-34-year-olds scored in the top third for sustained attention, against 35% of those 55 and older (Microsoft Canada, 2015). What the survey did capture was habit: 77% of Canadians aged 18 to 24 said the first thing they do when nothing is occupying their attention is reach for their phone, against 10% of those 65 and older.
No goldfish study backs the 9-second figure
There is no evidence that goldfish, or fish in general, have short attention spans or memories, the BBC concluded after checking (BBC, 2017). Fish behaviour researcher Felicity Huntingford told the program, “Goldfish can perform all the kinds of learning that have been described for mammals and birds,” and pointed out that goldfish are a model species for studying how memory forms.
Goldfish remembered a one-hour daily feeding window for weeks. In a 1994 experiment, eight goldfish learned to press a lever that dispensed food only during 1 hour in every 24, a schedule kept for 4 weeks. The fish started responding in anticipation of feeding time, and kept that pattern for several days after the dispensers were switched off (Gee, Stephenson & Wright, Journal of the Experimental Analysis of Behavior, 1994).
The older myth gave goldfish a three-second memory. The 9-second attention span is newer, and Statistic Brain named no study for it.
The measured number: 47 seconds on a screen before switching
47 seconds is the average time 40 information workers spent on one computer window before switching, logged by software over two work weeks, and the median was 40 seconds (Mark, Iqbal, Czerwinski, Johns & Sano, Neurotics Can’t Focus, CHI, 2016). Email windows averaged 61.8 seconds, and the average worker switched between apps 273 times a day.
About two and a half minutes in 2004, 75 seconds around 2012, 47 seconds since: that’s the trend Gloria Mark of UC Irvine describes from two decades of studies (American Psychological Association, 2023). She says other researchers have replicated the result at 44 and 50 seconds (University of California, 2023), and that the figure has held at “a steady state” for five or six years (APA, 2023).
The 2004 paper itself reports 2 minutes 53 seconds per stretch of PC use, and 2 minutes 11 seconds on any device or paper, from 14 people at an investment firm shadowed by researchers with stopwatches (González & Mark, Constant, Constant, Multi-tasking Craziness, CHI, 2004). Mark’s two-and-a-half-minute figure sits between the two. The 2012 figure is 75.5 seconds per window for 13 information workers whose computers were logged (Mark, Voida & Cardello, A Pace Not Dictated by Electrons, CHI, 2012).
About 10 and a half minutes is how long people spend on one project before switching, when Mark groups their screen switches by project (University of California, 2023). A switch to a related document for the same task still counts against the 47 seconds.
What 47 seconds measures: the 2016 paper calculates focus duration by dividing each person’s daily computer time by their number of window switches. It tells you how long a screen holds someone, not how long a person can concentrate. The early figure came from observers with stopwatches and the later ones from logging software, so the decline also compares two methods. What switching costs, like the time it takes to resume an interrupted task, is covered in context switching statistics.
Lab tests put peak focus at about a minute, and vigilance slips after half an hour
76 seconds was the longest stretch young adults stayed “in the zone” during a 6-minute-15-second attention test, against 67 seconds for older adults and 30 seconds for children (Simon et al., Frontiers in Cognition, 2023). The study covered 262 people aged 7 to 85. Children’s attention span fell 27% over the course of the task, while adults’ didn’t change significantly.
Sustained attention peaks in the early 40s and declines gradually after that, in 10,430 people aged 10 to 70 who took a 4-minute attention test online (Fortenbaugh et al., Psychological Science, 2015).
After the first half hour on a monitoring task, people reliably miss more targets. Norman Mackworth documented this “vigilance decrement” in 1948, and his follow-up work found that a phone call, a half-hour break or performance feedback reduced or eliminated it, while urging people to pay closer attention did nothing (Klein & Feltmate, The vigilance decrement: its first 75 years, 2025).
46.9% of waking hours go to thinking about something other than the task at hand, from 250,000 smartphone check-ins with 2,250 volunteers. Minds wandered at least 30% of the time during every activity except one (Killingsworth & Gilbert, reported by the Harvard Gazette, 2010).
The “10 to 15 minutes” lecture attention limit isn’t supported by primary data. A review found that the most cited source for it barely discusses student attention, and that the biggest differences in student attention came from the teacher, not the lecture format (Bradbury, Advances in Physiology Education, 2016).
Little evidence says attention capacity is shrinking
The BBC’s 2017 investigation found no evidence that human attention spans are shrinking (BBC, 2017). The closest tests of the question point the same way.
Children wait longer than they used to. Among 840 U.S. children given the marshmallow test, those tested in the 2000s waited 2 minutes longer on average than children in the 1960s, and 1 minute longer than children in the 1980s, the opposite of what surveyed adults predicted (Carlson et al., Developmental Psychology, 2018). Delay of gratification is self-control, not attention, but it’s a closely related skill measured the same way across four decades.
The media-multitasking effect may not exist. In two replication studies, only 5 of 14 tests found heavier media multitaskers more distractible, and only 2 held up under a stricter Bayesian analysis. A meta-analysis of 39 effect sizes found a weak association that turned nonsignificant after correcting for small-study effects (Wiradhany & Nieuwenstein, Attention, Perception & Psychophysics, 2017).
Heavier media multitaskers do score worse on some tasks, especially ones that depend on sustained attention, though many studies find no difference and the direction of cause is unknown (Uncapher & Wagner, PNAS, 2018).
What has measurably sped up is collective attention. A hashtag in Twitter’s hourly top 50 stayed there for 17.5 hours on average in 2013 and 11.9 hours in 2016 (Lorenz-Spreen et al., Nature Communications, 2019). The authors’ model attributes this to more content competing for limited attention. It measures how fast topics turn over, not individual attention spans.
Most web visits get under 15 seconds of active attention
55% of visitors spent fewer than 15 seconds actively on a page, across 2 billion visits Chartbeat tracked over a month, and on article pages the share was 1 in 3 (Tony Haile, Chartbeat CEO, in Time, 2014). Visitors held for three minutes were twice as likely to return as those held for one.
On 98.5% of 205,873 web pages, the chance a visitor leaves is highest at the start and falls the longer they stay (Liu, White & Dumais, Microsoft Research, 2010). The researchers call it “screen-and-glean”: people decide fast whether a page is worth it, then settle in. The data came from opted-in browser logs, counting only pages with at least 10,000 visits.
Users read at most 28% of the words on an average web page, and 20% is more likely, and they spend only 4.4 seconds more for each additional 100 words (Nielsen Norman Group, 2008). The underlying data was collected in 2005, so treat it as a baseline.
Without email, workers stayed on each screen 75% longer
77 times a day is how often 40 information workers checked email, spending almost an hour and a half a day in it (Mark et al., Email Duration, Batching and Self-interruption, CHI, 2016). On days with more email time they rated their productivity lower, and in hours with more email time their measured stress was higher. 41% said they mostly checked email on their own initiative, not because a notification prompted them.
Time per window rose from 75.5 to 131.9 seconds when 13 workers went five workdays without email, a 75% increase by our arithmetic, and window switches fell from 37.1 to 18.2 an hour (Mark, Voida & Cardello, CHI, 2012).
Checking email less without switching it off
Few jobs can go a week without email, and a self-started check is usually a look to see whether anything needs you. Carly answers that question for you. It triages and labels mail across all of your inboxes, drafts replies, and texts you only when something urgent arrives, so you aren’t checking email every few minutes. It also books meetings without the back-and-forth, which removes another reason to keep the inbox open. For where the rest of the day goes, see productivity statistics, and for why the hard task keeps getting pushed back, procrastination statistics.
FAQ
What is the average human attention span? There isn’t one measured average. On a computer, office workers spend 47 seconds on one window before switching (Mark et al., CHI, 2016). On a focused lab task, young adults’ longest “in the zone” stretch averaged 76 seconds (Simon et al., 2023). The widely quoted 8 seconds has no study behind it (BBC, 2017).
Is the human attention span shorter than a goldfish’s? No. The 8-second human and 9-second goldfish figures appeared in a 2015 Microsoft Canada report that credited Statistic Brain, and neither traces to research (BBC, 2017). Goldfish in a 1994 study learned a one-hour daily feeding window over four weeks and kept anticipating it for days after the food stopped (Gee, Stephenson & Wright, 1994).
How long is the average attention span in minutes? Under a minute on a screen: 47 seconds per window (Mark et al., 2016). About 10 and a half minutes on one project before switching (University of California, 2023). On monitoring tasks, people start missing more targets after the first half hour (Klein & Feltmate, 2025).
Are attention spans getting shorter? Screen switching has sped up, from about two and a half minutes in 2004 to 47 seconds (APA, 2023), but that measures behavior, not capacity, and the methods changed along the way. Children now wait 2 minutes longer on the marshmallow test than children in the 1960s (Carlson et al., 2018), and the evidence linking media multitasking to distractibility is weak (Wiradhany & Nieuwenstein, 2017).
Related: Time Management Statistics 2026 · Email Statistics 2026 · Deep Work · Best AI tools for deep work
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
