You are writing a difficult email. A message comes in; you reply in ten seconds. You go back to the email and reread the last paragraph to remember where you were. Another one comes in. An hour later the email is done, you have answered twelve messages, and it feels like a productive hour.
It was not. It was an hour in which you paid, twelve times, a tax that shows up nowhere: the cost of switching tasks. Multitasking does not exist. What exists is rapid switching — and every switch charges.
The brain does not do two things: it switches
Research on task switching is old and consistent. In a set of experiments published in the Journal of Experimental Psychology, researchers at the University of Michigan measured what happens when people alternate between tasks: each switch costs time — the switch cost — and that cost grows when the tasks are more complex or less familiar.
What happens in the switch is mechanical. The brain has to deactivate the rule set of the previous task and load the new one. That takes fractions of a second on simple tasks and far longer on tasks that require context — a line of reasoning in progress, a text half written, a problem you were two steps from solving. When you come back, part of that context is gone and has to be rebuilt.
The cost you do not see
Lost time is the small part. A study presented at CHI, the main conference on human-computer interaction, had people do an office task with and without interruptions. The result was surprising: those who were interrupted finished the task faster — they compensated by speeding up. But they paid for it with more stress, more frustration, more time pressure and more perceived effort.
That explains the feeling of the "productive" hour in which you answered twelve messages: you did deliver. Only in a state that is not sustainable, and the cost shows up later, as tiredness that does not match what got done.
There is a third cost, the most worrying. Stanford researchers compared people who do a lot of media multitasking — several screens, several conversations at once — with those who do little. The heavy multitaskers were worse at filtering irrelevant information, worse at holding things in working memory and, ironically, worse at switching tasks. Practice does not train the skill. It seems to wear it down.
Why it is so hard to stop
If the cost is so clear, why does everyone keep doing it? Because switching has three immediate advantages that focused work does not.
It gives a quick reward. Replying to a message takes ten seconds and produces a feeling of completion. Writing the difficult email takes an hour and produces anxiety until the end.
It is a legitimate escape. When the main task gets uncomfortable, handling the interruption looks like work — and it is. You are not procrastinating; you are "getting things done". It is the most respectable procrastination there is.
It comes with a built-in cue. The message buzzes. The email appears. The meeting lands on the calendar. Deep work has none of that: it depends on you deciding to go back, and deciding costs.
How to get out of multitasking without dropping the work
The problem that remains
The parking rule works — if the place where you park does two things: accepts the note in seconds, with no form, and brings the item back at the right time without you having to remember to check. Without the first, you do not write it down. Without the second, the car park becomes a graveyard, you stop trusting it — and go back to handling everything on the spot.
Where I wanted to end up
I spent years believing I was good at multitasking and ending my days drained without knowing why. What changed was not willpower; it was having a place to send everything that arrived at the wrong time — one sentence, in seconds, with the guarantee that it would come back when the time was right.
That is what I wanted ASB to be: the place where the interruption waits, so you do not have to pay for it now.