In 2015 headlines around the world announced that the human attention span had fallen to eight seconds, shorter than a goldfish's nine. The figure came from a marketing report by Microsoft Canada, which cited a statistics website rather than original research. Psychologists who tried to trace it found nothing behind it, and goldfish, for their part, can be trained to remember things for months. The goldfish statistic is a myth with no study behind it. Attention span is not one fixed number; how long people can focus depends on the task, the person and the setting. The real research on switching and interruption is interesting enough without the myth.

The trouble with multitasking

For tasks that need thought, people do not really do two things at once. They switch rapidly between the tasks, and every switch carries a cost. In laboratory studies, such as those by Robert Rogers and Stephen Monsell in 1995 and by Joshua Rubinstein, David Meyer and Jeffrey Evans in 2001, people were slower and made more errors when they alternated between tasks than when they repeated one task. These switch costs may be fractions of a second each, but they add up, and they grow when the tasks are complex or unfamiliar. Simple, well-practiced actions, like walking while talking, combine more easily, which is part of why multitasking feels possible.

Heavy media multitaskers

In 2009 Eyal Ophir, Clifford Nass and Anthony Wagner at Stanford compared people who reported heavy media multitasking, such as messaging while watching video and browsing, with light multitaskers. The heavy multitaskers were worse at filtering out distractions, which surprised the researchers, who had expected practice to make them better. The study was widely reported as proof that multitasking damages the mind. Later work was more cautious. Several replications found smaller or inconsistent effects, and a 2018 review by Melina Uncapher and Wagner concluded that heavy multitaskers often do somewhat worse on some attention tests but that cause and effect remain unclear. People who are easily distracted may simply be drawn to multitasking.

Interruptions and attention residue

The informatics researcher Gloria Mark has spent two decades observing how people work at computers. Her team found that average time on one screen before switching fell from about two and a half minutes in 2004 to about 47 seconds in studies from the late 2010s onward. A 2008 study she led found that people who were interrupted often finished tasks just as fast, by working faster, but kept pace at the price of more stress, frustration and effort. Interruption has a hidden cost even when output looks the same. Mark's observations also suggest that people interrupt themselves nearly as often as others interrupt them.

The management researcher Sophie Leroy described another reason switching is costly. In 2009 she named it attention residue: when people move from an unfinished task to a new one, part of the mind stays on the unfinished task, and they perform worse on the next one. Finishing a task, or jotting a quick note about where to resume, appears to reduce the residue.

The phone on the desk

A 2017 study led by Adrian Ward reported that simply having a smartphone nearby, even switched off, lowered scores on tests of working memory and reasoning, an effect the authors called "brain drain." It became one of the most cited findings in the field. Yet later attempts to replicate it, including several studies published in the early 2020s, mostly failed to find the effect. This is a useful lesson for anyone reading about attention research: a striking single finding deserves interest, and also patience until it has been confirmed by others.

None of this means the worries are baseless. Switch costs and the stress of constant interruption are well established. What the record shows is that the strongest claims, the ones that make the best headlines, are often the weakest scientifically, while the solid findings are less dramatic and more useful.

Worked example: an afternoon of report writing

Priya has two hours to write a report. In the first hour her phone sits face up and her email is open. She checks messages eleven times. Each check is short, but each return to the report costs a few seconds of finding her place, a switch cost, and her mind keeps drifting to an unanswered email, a case of attention residue. She finishes one page and feels frazzled, which fits Mark's finding that interrupted workers keep pace at the price of stress. In the second hour she closes her email, puts the phone in a drawer and writes a short note listing what she must answer later. She writes two pages and feels calmer. One afternoon proves nothing scientifically, and the failed brain drain replications suggest that the drawer may matter less than the closed inbox. But the pattern fits the better established findings: fewer switches and fewer open loops leave more attention for the work.