# Is the Eight-Second Attention Span Real? Origins vs. Measurements > There is no research support for 'humans have an eight-second attention span, shorter than a goldfish.' It spread from a 2015 Microsoft Canada report citing a statistics site, and when the BBC traced the source in 2017 no underlying data could be confirmed. The measurement that does hold up comes from Gloria Mark's longitudinal work: average time attention stays on one thing on screen fell from about 2.5 minutes in 2004 to 47 seconds in recent years. That is not a ceiling on ability — it is the interval between switches. - URL: https://brainrank.jp/en/blog/attention-span-myth - Published: 2026-08-26 - Publisher: BrainRank Editorial Team (https://brainrank.jp) - Tags: Cognitive Abilities, Psychology The claim that modern humans can only focus for eight seconds, less than a goldfish's nine, has no research behind it. Trace the number and you arrive at a statistics site whose sources cannot be verified. A real measurement does exist, but it says something different. This article separates the myth's origin from the data. ## Tracing the eight-second claim The number went wide after a 2015 consumer research report from Microsoft Canada. The figure it cited came from a statistics site called Statistic Brain. According to [the BBC's 2017 investigation of the claim](https://www.bbc.com/news/health-38896790), the reporter could not get the statistics site's listed sources to produce any supporting research. Attention researchers interviewed for the piece could not point to a study supporting the figure either. The comparison — 'a goldfish manages nine seconds' — has no source at all. Goldfish are known to retain learning over months, so the nine-second figure is itself wrong. ## Attention duration is not one number The trouble with the eight-second claim starts before accuracy. Attention duration varies with the task, your interest, the reward, fatigue, and time of day. The same person drifts off after three minutes of a dull meeting and stays with a film for two hours. No constant of the form 'this person's attention span is N seconds' can describe that gap. The claim spread because it was simple and felt true. ## The number that was actually measured: 47 seconds There is a trustworthy measurement. Gloria Mark and colleagues at UC Irvine have tracked how long knowledge workers keep attention on a single object on screen for nearly two decades. In [an APA interview](https://www.apa.org/news/podcasts/speaking-of-psychology/attention-span), Mark describes the trend as follows. - 2004 studies: about 2 minutes 30 seconds - 2012 studies: about 75 seconds - Recent studies: about 47 seconds The key point is that this is not the limit of how long someone can concentrate. It measures the average interval before attention moves to the next object. The eight-second claim sounds like a statement about capacity; 47 seconds is a statement about behavior. They are not the same kind of fact. ## What shrank is the environment, not the ability Reading 47 seconds as evidence that humans have degraded is a mistake. There is no reason to think attention mechanisms changed in twenty years. What changed is the density of devices that invite a switch: notifications, infinite scroll, several chat threads, the assumption of being permanently reachable. And every switch carries a cost. We covered that in [is multitasking really inefficient](https://brainrank.jp/en/blog/multitasking). If a switch happens every 47 seconds, you are paying that cost dozens of times an hour. ## Four steps to audit your own attention Your impressions are unreliable. The way to see how your attention actually moves is to externalize it. 1. Log it: for one day, note the time and reason each time you break off 2. Split the interruptions: count those that came from outside (notification, someone speaking) separately from those you initiated (drifting to another tab) 3. Find when self-initiated breaks cluster: usually a sign of fatigue or boredom 4. Change one thing and measure again: notifications off, tabs closed — one change at a time If self-initiated breaks dominate, the cause is usually task design rather than temptation. When a task is too large for the next move to be visible, attention leaves. The logic of this audit is the same as [metacognition](https://brainrank.jp/en/blog/metacognition). ## As an exercise in doubting numbers The eight-second claim is a textbook case of an assertion that collapses when traced, yet circulates as fact through a chain of citations. Claims of that shape are everywhere. The procedure for separating claim, evidence, and assumption is in [how to build critical thinking](https://brainrank.jp/en/blog/critical-thinking), and the mechanisms that warp judgment are in [cognitive bias](https://brainrank.jp/en/blog/cognitive-bias). ## Measure what sits underneath attention Sustained attention cannot be separated from working memory and processing speed. [BrainRank's free IQ test](https://brainrank.jp/en/start) scores 20 questions across spatial reasoning, pattern recognition, logical reasoning, and classification (about 10 minutes) using item response theory (IRT). Taking it once in a high-focus window and once in a low one lets you see the environment's effect in your own data. Results are estimates for entertainment and self-understanding, not a medical or diagnostic assessment. ## Frequently asked questions ### Where does the eight-second claim come from? It spread from a 2015 Microsoft Canada consumer report that cited a statistics site called Statistic Brain. A BBC reporter traced the sources in 2017 and could not confirm any underlying research. The comparison figure of nine seconds for a goldfish has no source either. ### So attention spans are not shrinking? There is no evidence the underlying ability has declined. What changed is the environment. In Gloria Mark's longitudinal research, the average time attention stays on one thing on screen went from about 2.5 minutes in 2004 to 75 seconds in 2012 to 47 seconds recently. That is the interval between switches, not a limit on focus. ### Then how can people watch a two-hour film? Because attention duration is not a single constant — it varies enormously with the task, interest, and reward structure. The premise that one number can describe human attention does not hold in the first place. ## Sources - [the BBC's 2017 investigation of the claim](https://www.bbc.com/news/health-38896790) - [an APA interview](https://www.apa.org/news/podcasts/speaking-of-psychology/attention-span) --- Statistics and percentages are calculated from a normal distribution model with a mean of 100 and a standard deviation of 15. The tests on this site provide estimates for entertainment and self-understanding; they are not medical or clinical assessments, nor official intelligence tests. When citing, credit "BrainRank (brainrank.jp)" and link the article URL.