Is your attention span really shrinking?

Where did the goldfish claim come from?

A 2015 Microsoft marketing report, which took the figure from a company called Statistic Brain. When the BBC investigated in 2017, Statistic Brain could not produce a credible source for it. The research the report was based on never mentions eight seconds and never mentions goldfish.

There is also no scientific basis for the claim that a goldfish attends for nine seconds. Both halves of the comparison were invented.

Microsoft has since taken the report down. The statistic outlived it and is still quoted in presentations, articles and product pitches.

So has nothing changed?

Something has, and it is more specific than the myth. Gloria Mark, an informatics researcher at the University of California, Irvine, has tracked attention on screens across workplaces and student populations for around two decades.

In 2004 people spent an average of about two and a half minutes on a screen before switching. That figure is now roughly 47 seconds. Other researchers measuring the same thing found 44 and 50 seconds, and the average across those studies is about 47.

Note carefully what that measures. It is how long people actually stay on one thing before switching, not how long they are capable of staying. Those are different claims, and the second one is what the myth asserts.

Why does the distinction matter?

Because it changes what you would do about it. A shrinking capacity would be something wrong with you. A shrinking habit is something about your environment, and environments can be changed.

People still read long books, watch three-hour films, and work through difficult problems for hours. Those would be impossible if the capacity had genuinely collapsed to seconds.

What has changed is the availability of an alternative. Every moment of difficulty now sits next to something easier, and switching costs nothing.

What does switching cost?

More than it feels like. A long line of laboratory work shows performance worsens and errors rise when people alternate between tasks compared with doing them one after another. Mark's work also finds that after attention is diverted from a work project, it takes roughly 25 minutes to return to it.

That figure is the useful one. If you check a message every ten minutes, you may never fully return to the thing you were doing at all.

This connects to the noise research on this site, where sound helped sustained attention and interfered with holding several things in mind. Attention is not one capacity, and different demands behave differently.

What actually helps?

Remove the alternative rather than resisting it. The switch is cheap because the other thing is one tap away, so distance and friction do more than resolve. Phone in another room, notifications off, one window.

Work in stretches long enough to get past the uncomfortable start. If refocusing costs 25 minutes, a 20 minute block spends its whole life in recovery.

And notice that difficulty is the trigger. Most switches happen at the moment work gets hard, which is precisely the moment worth staying.