Focus / does-multitasking-work

Does Multitasking Work? What the Research Shows

12 min readchecked

For two tasks that both need your attention, multitasking does not work the way it feels like it does. Brain-imaging research suggests you can make only one decision of this kind at a time, so what looks like doing two things at once is really switching quickly between them. Lab studies measure the time cost of a single switch in tenths of a second, but the bigger costs show up as mistakes, lost place and less remembered. In a classroom experiment published in Computers & Education, students who multitasked on a laptop during a lecture scored 55% on a comprehension test, against 66% for students who did not.

Multitasking does work when one of the tasks is close to automatic, such as walking while you talk or folding laundry while you listen to a podcast. The trouble starts when both tasks need thought, and a phone is very good at turning a single task into two. Below is what happens in your brain when you try, where the most quoted numbers come from, what the research on heavy multitaskers actually found, whether anyone is genuinely good at it, and how to set up your phone so you switch less.

What happens in your brain when you multitask

The core problem is a bottleneck. In a 2006 brain-imaging study in Neuron, Paul Dux and colleagues at Vanderbilt University gave people two simple tasks at nearly the same time. The second response was usually delayed, and the scans traced the hold-up to the lateral prefrontal cortex, a region that appears to handle only one "what do I do now?" decision at a time. The authors described it as a central bottleneck of information processing. How rigid that bottleneck is remains debated, because practice on very simple task pairs can loosen it, but the cost of combining tasks that need real thought is well documented.

Because of that bottleneck, the brain handles two demanding tasks by switching between them, and switching has a measurable cost. In a 2001 paper in the Journal of Experimental Psychology, Joshua Rubinstein, David Meyer and Jeffrey Evans ran four experiments in which young adults either repeated one task or alternated between two, such as sorting shapes by different rules or solving different kinds of arithmetic problem. Switching always took longer than repeating. The cost grew when the rules were complex or unfamiliar, and shrank when a cue warned people which task was coming.

The authors explained the cost as two steps: shifting your goal to the new task, then loading the rules for doing it. The American Psychological Association's summary of the research puts the cost at sometimes a few tenths of a second per switch. That is small enough to go unnoticed, and the time alone is rarely the main problem: the studies below find larger costs in errors and in what people take in.

Where the "40% of productive time" figure comes from

You will often see the claim that multitasking costs up to 40% of your productive time, attributed to this research. It is not a result from the 2001 paper. The figure appears on the APA's page, which says Meyer "has said" that brief mental blocks from switching can cost as much as 40% of someone's productive time. It is an estimate from one of the researchers, not something the experiments measured.

The experiments themselves measured fractions of a second on simple lab tasks. The 40% figure is an estimate of what those costs might add up to in a heavily interrupted day, but nobody has tested it directly. A similar figure, the claim that it takes 23 minutes to refocus after an interruption, turns out to have the same history: a number repeated far more often than it was measured.

Do heavy multitaskers have worse attention?

Possibly, but the effect is much smaller and less certain than the first study suggested. The most cited paper on this is a 2009 study in PNAS by Eyal Ophir, Clifford Nass and Anthony Wagner at Stanford. They surveyed 262 students about how often they used several media at once, such as texting while watching television, then brought the heaviest and lightest multitaskers into the lab.

The heavy multitaskers were more easily distracted by irrelevant information, and, surprisingly, worse at switching tasks: in the task-switching test, their cost per switch was 167 milliseconds larger than the light multitaskers'. The finding was widely reported as proof that multitasking damages your ability to focus. But the lab groups were small, 15 people each in the task-switching test, and the study could not show which came first.

Later work has made the picture more modest:

  • The effect shrank on re-analysis. A 2017 meta-analysis by Wisnu Wiradhany and Mark Nieuwenstein pooled 39 effect sizes and found a small difference between heavy and light multitaskers (Cohen's d of 0.17). After correcting for the tendency of small studies with striking results to get published, it fell to between 0.01 and 0.07, which was not statistically significant. Their own two replication attempts mostly failed to reproduce the original findings.
  • A ten-year review found little progress. A 2021 meta-analysis by Douglas Parry and Daniel le Roux, covering 118 assessments from 46 studies, found a small overall link. It was strongest when people rated their own attention problems and weakest on objective lab tasks, where it disappeared after the same bias correction. The authors concluded that, ten years on, the field was "no closer to understanding" the relationship.
  • A larger study found no difference. In a 2020 study of 261 people, heavier media multitaskers were no more distractible and had no more lapses of attention than lighter ones.

A 2018 review in PNAS by Melina Uncapher and Anthony Wagner, one of the original authors, takes a middle line. It concludes that the balance of evidence suggests heavier media multitaskers perform worse in some areas, including memory and sustained attention, while noting that many studies find no difference. Of at least 24 working-memory tests in adults, about half found heavy multitaskers did worse, and none found them doing better. The studies are correlational, and the 2021 meta-analysis found every one it included was cross-sectional, so it remains possible that people who already find it hard to focus are drawn to multitasking, rather than multitasking causing the problem.

The fair summary is that habitual multitasking is linked to slightly worse attention in some studies, but the claim that it permanently rewires your brain goes well beyond the evidence. The research on short-form video and attention has a similar shape.

Can some people multitask well?

A very small number can, and most people who think they are one of them are not. In a 2010 study in Psychonomic Bulletin and Review, Jason Watson and David Strayer at the University of Utah had 200 people drive in a simulator while doing a demanding spoken memory and arithmetic task. Most got worse at both. Five people, 2.5% of the sample, showed no decline on either task, and the authors called them "supertaskers". Five people is a tiny group, so the true share in the population is uncertain.

Your own sense of how good you are is not a guide. In a 2013 study in PLOS ONE, David Sanbonmatsu and colleagues tested 277 students on a lab measure of multitasking ability and asked them to rate their own. The self-ratings had no meaningful relationship with actual performance. The average rating was 63 on a scale where 50 meant average, and 193 of the 277, about 70%, put themselves above average.

The students who scored lowest on the multitasking test reported multitasking the most, including using their phones while driving. Impulsivity and sensation-seeking also predicted more multitasking, so the people who multitask most tend to be the ones least equipped for it.

What multitasking does to learning

The clearest real-world evidence comes from classrooms, and it shows two costs: one to the person multitasking and one to the people around them. In a 2013 study by Faria Sana, Tina Weston and Nicholas Cepeda at York University and McMaster University in Canada, students watched a 45-minute lecture and took notes on a laptop.

In the first experiment, 40 students were split into two groups. Half were also given online tasks to do during the lecture, the kind of browsing a bored student might do. On a comprehension test afterwards, the multitaskers scored 55% against 66% for those who only took notes.

In the second experiment, 38 students took notes with pen and paper. Half sat where they could see the screens of other students who were multitasking on laptops. Those with a view of the screens scored 56%, against 73% for students with no multitaskers in view.

Comprehension scores after a lecture, with and without laptop multitasking

02040608055%Multitasked66%Did not multitask56%Saw a screen73%Saw no screen
Experiment 1 compared students who multitasked with those who did not. Experiment 2 compared students who could see a multitasker's screen with those who could not. Source: Laptop multitasking hinders classroom learning for both users and nearby peers, Computers & Education (2013): n=40 and n=38.
Show the numbers
Comprehension scores after a lecture, with and without laptop multitasking: underlying values
Multitasked55%
Did not multitask66%
Saw a screen56%
Saw no screen73%

Both experiments were small and used a single simulated lecture, so the exact gaps should not be taken as typical. A 2019 study in Educational Psychology by Arnold Glass and Mengxue Kang at Rutgers University tested the idea in a real course. Over a term, 118 students in two sections of a psychology course were allowed devices in some lectures and not others. Phones and laptops did not lower scores on quizzes taken in class, but material from device-allowed lectures was remembered less well in unit and final exams weeks later. That was true for the whole class, including students who did not use a device, which fits the second Sana experiment.

If you study with your phone nearby, the guide to studying without phone distraction covers ways to keep it out of reach.

Even a few seconds of interruption costs something

You do not need to hold a full conversation for a switch to hurt. In a 2014 study in the Journal of Experimental Psychology: General, Erik Altmann at Michigan State University and colleagues interrupted people working through a multi-step task. Interruptions averaging 2.8 seconds roughly doubled the rate of errors where people lost their place in the sequence, and interruptions averaging 4.4 seconds roughly tripled it. Nobody has measured phone glances this way, but a few seconds is shorter than most notification checks.

For many people, those interruptions arrive constantly. Microsoft's 2025 Work Trend Index, based on anonymised Microsoft 365 data, found that the most interrupted fifth of its users were interrupted by a meeting, email or chat every two minutes during core working hours, or 275 times a day. That is not the average worker, and the report comes from a company that sells workplace software, but it shows how fragmented a working day can become.

The fragmentation is not only imposed by others. Gloria Mark, a researcher at the University of California, Irvine, who has tracked how long people stay on one screen before switching, reports in an interview that the average fell from about two and a half minutes in 2004 to 75 seconds in 2012 and 47 seconds in recent years. Those figures are summarised from several of her studies rather than one, and come from tracking office and information workers, so treat them as an indication of the trend rather than a precise measurement.

When multitasking does work

Multitasking works when only one of the tasks needs your active attention. The bottleneck in the Dux study is about deciding what to do, and well-practised tasks such as walking, washing up or stirring a pan need very little of that decision-making. That is why you can usually hold a conversation on a walk, while combining two tasks that both need thought, such as reading email in a meeting, tends to go badly. This is an inference from the bottleneck research rather than something one study has tested across everyday tasks, and even an easy task gets harder when the other one becomes demanding.

Two cases need more care:

  • Talking while driving is not one of the easy combinations. Driving feels automatic, but in a 2001 simulator study in Psychological Science, David Strayer and William Johnston found that phone conversations doubled the rate at which drivers missed simulated traffic signals, and the result was the same whether the phone was handheld or hands-free. The guide to stopping texting while driving covers the phone settings that help.
  • Background music depends on the task. Music without lyrics is often fine for routine work, while lyrics compete with reading and writing. The evidence is covered in does music help you focus.

One finding sounds as if it supports multitasking but does not. In a 2018 paper in Psychological Science, Shalena Srna, Rom Schrift and Gal Zauberman ran 32 studies with 8,242 participants and found that people performed better on the same activity when it was described to them as multitasking. The authors put this down to people engaging more. The activity itself did not change, so it is evidence that feeling busy can raise effort, not that doing two demanding things at once works.

How to multitask less with your phone

The aim is fewer switches, not more willpower. These steps follow from the research above, though none has been tested as a package:

  1. Batch your notifications. Fewer interruptions means fewer forced switches. On an iPhone, go to Settings > Notifications > Scheduled Summary and add non-urgent apps, so their alerts arrive two or three times a day. On Android, turn off notifications for individual apps in Settings > Notifications > App notifications. The notification detox guide goes further.
  2. Use a focus mode for single-task blocks. On an iPhone, set up a Work Focus in Settings > Focus and allow only the people and apps you need. On Android, use Settings > Modes on recent Pixel phones, or Settings > Modes and Routines on Samsung. Calls from favourites can still come through.
  3. Put the phone out of sight. In the Sana study, seeing other people's screens lowered scores. Nobody has tested your own phone in the same way, but keeping its screen out of view during work that needs thought is a reasonable precaution.
  4. Work in timed blocks. Committing to one task for 25 or 50 minutes, then checking messages in the break, turns constant switching into a few planned ones. The evidence for this approach is covered in does the Pomodoro technique work, and the Deep Work Sprint gives you a five-day structure. A single-task block is also where absorption has room to build, and the research on how to get into a flow state covers what else helps.
  5. Close what you are not using. One window and one tab for the task at hand removes the cue to switch. If you find you reopen sites without thinking, a blocker such as Freedom can lock them for a set period on both phone and computer.

Common questions

Is multitasking bad for your brain?

There is no good evidence that it causes lasting damage. Heavy media multitaskers do slightly worse on some attention and memory tests in some studies, but the effect is small, has not replicated well, and may reflect people with weaker focus multitasking more rather than the other way round. The clearer cost is in the moment: switching between demanding tasks makes each one slower and more error-prone.

Why does multitasking feel productive?

Because you are busy the whole time and the costs are hidden. Each switch costs only a fraction of a second, so you do not notice it, and a 2018 series of studies found that simply believing you are multitasking can make you try harder. Most people also overrate their own ability: in one study, about 70% of students rated themselves above average.

Can you train yourself to multitask better?

Only narrowly. The 2001 experiments found that switch costs shrink with familiar rules and clear cues, and in a 2001 study in Psychological Science, some participants learned to perform two basic reaction tasks together almost without interference after relatively modest practice. Those gains apply to the practised pair, not to multitasking in general, and the 2013 PLOS ONE study found that the people who multitask most in daily life tend to be the worst at it.

Is it multitasking to listen to a podcast while you work?

Yes, and how well it works depends on the work. Because a podcast, reading and writing all involve processing language, a podcast is likely to get in the way of work with words. That is an inference rather than something a study of podcasts has tested. Paired with something routine, such as tidying or a walk, it is the kind of combination that usually works.

When to see a GP

Finding it hard to stick to one task in a world of constant notifications is normal, and usually improves when you reduce interruptions. See your GP if trouble concentrating has been with you since childhood, shows up across work, home and relationships, and does not improve when distractions are removed. Those can be signs of ADHD, which affects adults as well as children and can be treated. The NHS page on ADHD in adults explains the symptoms and how to ask for an assessment, and if you already know you have ADHD, the guide to focusing with ADHD when your phone is right there covers what tends to help. Also see your GP if poor concentration comes with low mood, poor sleep or exhaustion, which can have other causes worth checking.

Advertisement

sources for this page

  1. Isolation of a central bottleneck of information processing with time-resolved fMRINeuron · 2006 · Lab study using simple, speeded responses; shows the bottleneck exists rather than how it plays out in everyday tasks.
  2. Executive control of cognitive processes in task switchingJournal of Experimental Psychology: Human Perception and Performance · 2001 · Four experiments with young adults · Small samples and simple lab tasks; switch costs measured in fractions of a second, not as productivity.
  3. Multitasking: Switching costsAmerican Psychological Association · Research summary. The 40% figure is a statement attributed to David Meyer, not a measured result.
  4. Cognitive control in media multitaskersPNAS · 2009 · 262 students surveyed; lab groups of 15 to 22 heavy and light multitaskers · Very small lab groups; correlational; later replications found much weaker effects.
  5. Cognitive control in media multitaskers: Two replication studies and a meta-analysisAttention, Perception, & Psychophysics · 2017 · 39 effect sizes pooled, plus two small replications · Pooled effect (d = 0.17) fell to 0.01 to 0.07 and lost significance after correcting for publication bias.
  6. "Cognitive control in media multitaskers" ten years on: A meta-analysisCyberpsychology: Journal of Psychosocial Research on Cyberspace · 2021 · 118 assessments from 46 studies · All included studies cross-sectional; the link with objective lab measures disappeared after bias correction.
  7. Media multitasking, mind-wandering, and distractibility: A large-scale studyAttention, Perception, & Psychophysics · 2020 · 261 participants · Media multitasking measured with a short questionnaire.
  8. Minds and brains of media multitaskers: Current findings and future directionsPNAS · 2018 · Narrative review co-written by an author of the 2009 study; notes many studies find no difference and causality is unknown.
  9. Supertaskers: Profiles in extraordinary multitasking abilityPsychonomic Bulletin & Review · 2010 · 200 participants in a driving simulator · Only five supertaskers identified, so the population share is uncertain.
  10. Who multi-tasks and why? Multi-tasking ability, perceived multi-tasking ability, impulsivity, and sensation seekingPLOS ONE · 2013 · 277 University of Utah undergraduates · Student sample; correlational.
  11. Laptop multitasking hinders classroom learning for both users and nearby peersComputers & Education · 2013 · 40 students (experiment 1) and 38 students (experiment 2), Canadian undergraduates · Single simulated 45-minute lecture; small samples. Gaps are percentage points.
  12. Dividing attention in the classroom reduces exam performanceEducational Psychology · 2019 · 118 students in two sections of one Rutgers course · One course at one university; devices did not affect in-class quiz scores, only later exams.
  13. Momentary interruptions can derail the train of thoughtJournal of Experimental Psychology: General · 2014 · Lab task measuring sequence errors; not a measure of everyday productivity.
  14. Breaking down the infinite workdayMicrosoft Work Trend Index · 2025 · Anonymised Microsoft 365 signals; interruption figure based on the top 20% of users by volume · Industry report from a software vendor; describes the most interrupted users, not the average.
  15. Can't pay attention? You're not aloneUniversity of California · 2023 · Interview with Gloria Mark; the 47-second figure is not traced to a single peer-reviewed paper.
  16. Driven to distraction: Dual-task studies of simulated driving and conversing on a cellular telephonePsychological Science · 2001 · Driving simulator study with students, not on-road driving.
  17. Virtually perfect time sharing in dual-task performance: Uncorking the central cognitive bottleneckPsychological Science · 2001 · Basic choice-reaction tasks in the lab; only some participants reached near-perfect time sharing.
  18. The illusion of multitasking and its positive effect on performancePsychological Science · 2018 · 32 studies, 8,242 participants · About perceiving a single activity as multitasking, not about doing two demanding tasks at once.
  19. ADHD in adultsNHS · Patient information page.