Most of what we call multitasking is really task switching: moving attention back and forth between jobs, fast enough that it feels simultaneous. Each switch has a small cost in time and accuracy, and those costs add up. True parallel performance only happens when one of the tasks is so practised it runs almost automatically, like walking while talking. Here is what the research says about switch costs, what multitasking games really measure, whether you can train the skill, and how to organise your day so switching costs you less. Try the games on our multitasking games page as you read.
In task-switching experiments, people respond more slowly and make more errors on the trial right after a switch than on a trial that repeats the same task. This switch cost is typically a fraction of a second per switch, and it persists even when people know a switch is coming. Stephen Monsell's 2003 review describes it as one of the most reliable findings in the field.
Joshua Rubinstein, David Meyer and Jeffrey Evans (2001) showed that switch costs grow when the tasks are more complex or less familiar. The American Psychological Association summarises this research with the estimate that heavy switching can cost as much as 40% of productive time. That figure is an illustration rather than a measurement of your day, but the direction is consistent across studies.
| Stage | What happens |
|---|---|
| Goal shifting | Deciding to change what you are doing now |
| Rule activation | Loading the rules of the new task and suppressing the old ones |
| Residual interference | Bits of the old task linger and slow the new one, sometimes for several trials |
One task is automatic. Walking, stirring a pot or typing a familiar word needs little attention, so it can run alongside another task.
The tasks use different channels. Listening to music while sorting socks is easier than listening to a podcast while reading, because the second pair both use language.
Neither task is urgent. As soon as both need decisions at the same moment, one has to wait.
Driving shows the limit clearly. In a 2001 study by David Strayer and William Johnston, people in a driving simulator missed more traffic signals and reacted more slowly while on a phone call, and hands-free phones were no better than handheld ones. The bottleneck was attention, not the hands.
Each game is a good measure of how much you can juggle in that game. Research on multiple-object tracking suggests most people can follow about three to five moving objects at once, and fewer when they are split across two areas. That is why these games feel easy up to a point and then suddenly fall apart.
| Game | What you juggle | Main skill |
|---|---|---|
| Sort & Count | Sorting items while keeping a running count | Protecting working memory during interruptions |
| Dual Tracker | Moving targets in two separate zones | Divided visual attention |
| Traffic Control | Several vehicles needing timely decisions | Scanning and prioritising |
| Priority Target | Targets with different values and timers | Deciding what to do first |
| Dual N-Back | A position stream and a letter stream together | Updating two working memory streams |
Partly. With practice, switch costs shrink for the specific tasks you practise, because the rules become more automatic. That is real and useful: pilots, air traffic controllers and chefs get better at juggling the tasks of their own job.
The evidence is weaker for a general "multitasking ability" that transfers to everything. Large reviews of brain training, such as Simons and colleagues (2016), find that practice mostly improves the practised tasks. There is also a surprising twist: a 2009 Stanford study by Eyal Ophir, Clifford Nass and Anthony Wagner found that people who reported heavy media multitasking did worse at ignoring distractions and switching tasks than light multitaskers. Doing a lot of multitasking does not seem to make people good at it.
So play multitasking games because they are a fun challenge and a clear measure of divided attention, not because they will turn you into a better multitasker at work.
Batch similar tasks. Answer all your messages in one block rather than between other jobs.
Finish a step before switching. Stopping at a natural break reduces the interference that lingers.
Write down where you were. A one-line note ("next: check row 40") cuts the restart cost when you come back.
Silence notifications during focused work. Each alert is an invitation to switch.
Use a timer. Work in fixed sessions with planned breaks. Our free focus timer does this with 25-minute sessions.
Make routine tasks automatic. The more practised a task is, the less it costs to combine it with something else.
For more habits with good evidence behind them, read how to improve focus and concentration and our guide to screen time and attention.
Monsell, S. (2003). Task switching00028-7). Trends in Cognitive Sciences.
Rubinstein, J. S., Meyer, D. E., & Evans, J. E. (2001). Executive control of cognitive processes in task switching. Journal of Experimental Psychology: Human Perception and Performance.
American Psychological Association. Multitasking: switching costs.
Strayer, D. L., & Johnston, W. A. (2001). Driven to distraction. Psychological Science.
Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive control in media multitaskers. PNAS.
Simons, D. J., et al. (2016). Do "brain-training" programs work? Psychological Science in the Public Interest.
Only partly. For tasks that both need attention, the brain switches between them rather than doing them at once. True parallel work happens when one task is automatic.
The extra time and errors on the first moments of a task after switching from another one. It is usually a fraction of a second per switch but adds up over a day.
They improve your performance in those games. Evidence that the gains carry over to work or driving is weak, so treat them as a challenge and a measure.
Research on multiple-object tracking puts the typical limit at about three to five moving objects, and fewer when they are spread across separate areas.
A small number of people show unusually low costs in dual-task studies, but most people, including those who multitask a lot, perform better when they do one thing at a time.
Multitasking is mostly fast switching, and every switch costs a little. Batch your tasks, protect focused blocks, and enjoy multitasking games as a test of divided attention rather than a cure. Try Sort & Count or browse all our multitasking games.