It’s 1:42pm on a Thursday. You are eating a sandwich, debugging a function, half-listening to a podcast at 1.5x speed, and watching a Slack thread about a meeting tomorrow. You have been doing this for forty minutes. You feel productive in the way a hummingbird looks productive — high movement, unclear progress. The function is still broken. You’re not sure what the podcast was about. The Slack thread has scrolled past you twice. The sandwich is gone, which is probably the only thing you actually finished.
You will tell yourself, when you stand up, that you “got a lot done.” You will not have gotten a lot done. You will have run four overlapping streams of partial attention, and the cost of running them in parallel is invisible to you because you’ve done it so many times it feels like a normal mental state.
It isn’t. It’s a brain repeatedly losing its place.
The thing called multitasking is actually rapid switching
Strictly, your brain is not multitasking when you do this. It’s switching. The conscious-attention system is serial. It can do one cognitive thing at a time. When you appear to be doing two cognitive tasks at once — coding and reading Slack, writing and listening to a podcast with words — what’s happening is that attention is jumping between them, and each jump has a cost.
David Meyer’s lab at the University of Michigan made this concrete two decades ago. Rubinstein, Meyer, and Evans (2001), published in the Journal of Experimental Psychology: Human Perception and Performance, ran a series of experiments showing that switching between tasks consistently produced “switch costs” — measurable slowdowns and increased error rates compared to doing the tasks in sequence. The American Psychological Association’s summary on multitasking puts the practical estimate this way: shifting between tasks can cost up to 40% of someone’s productive time. Forty percent. Not in extreme cases. In normal office task-switching.
The deeper mechanism is something called attention residue. Sophie Leroy’s 2009 paper “Why Is It So Hard to Do My Work? The Challenge of Attention Residue” in Organizational Behavior and Human Decision Processes demonstrated that when you switch from Task A to Task B, a fragment of Task A persists in working memory for minutes after — your cognitive resources are still partially allocated to the thing you just left. You are not fully on Task B. You are on Task B with 70% of your processor and 30% still chewing on the Slack message you read three minutes ago.
This means the “tax” on switching isn’t just the second of jumping. It’s the diminished cognitive capacity for the next several minutes after every jump. If you switch tasks every two minutes, you are essentially never operating at full capacity. You’re operating at 70%, all day, while feeling like you’re working hard.
The honest version of what humans can actually do: one cognitive task plus one automatic task. Walk and listen to a podcast — fine, walking is automatic. Drive a familiar route and sing along to a song — fine, both are partially automatic. Fold laundry and watch a show — fine. The combination that doesn’t work, the one that feels like multitasking but actually isn’t, is two cognitive tasks at once. Reading a thread while writing an email. Listening to a meeting while debugging a function. Coding while keeping half an eye on a chat window. These are serial switching disguised as parallel work, and the disguise is what makes them so costly — you don’t notice the tax because you don’t notice the switching.
What to do during your next focus block
The intervention is unglamorous and works immediately: close everything except the task.
Tomorrow, when you sit down for a 90-minute work block on something that requires real thinking, do this before you start. Close every browser tab that isn’t for the task. All of them. The “I’ll just leave Gmail open in case something urgent comes in” tab is a switching invitation that will get accepted within fifteen minutes. The Slack tab is the same. Quit Slack, not just minimize it — minimized apps still ping, and the ping is the switch. If you’re on a Mac, quit it from the menu bar. If you’re on Windows, close it. Phone face-down or in another room. Notifications off at the OS level, not just within apps.
This sounds extreme. It is extreme compared to the normal way you’ve been working. It is also unremarkable compared to how anyone who does serious cognitive work for a living actually works. Writers don’t write with seven tabs open. Surgeons don’t operate with a chat window. The reason your office worker version of focus has fifteen tabs open is not that office work is different — it’s that nobody enforced the focus condition because it didn’t visibly cost the company anything in any single moment. The cost shows up in your output over the week, which nobody is measuring closely.
The other practical move is sequencing. If you have a 90-minute block, do not put two different cognitive tasks in it. One task. Finish or get to a clean stopping point on Task A before opening Task B, even if you think “they’re related.” Attention residue says they aren’t related to your brain — they’re two different working-memory loads, and switching between them mid-block costs you more than batching them.
A useful self-test: at the end of a working day where you “felt busy,” ask whether you finished any one thing cleanly. Most people who feel busy have started six things and finished zero. The completion rate is the diagnostic. If you can’t point to one thing that actually got to a stopping point, the day was almost certainly fragmented switching, not multitasking, and the output reflects that.
The reason this is hard isn’t that the technique is complicated. It’s that the muscle of staying on one task for 90 minutes has atrophied for most people who have spent a decade in chat-based work. The first few blocks will feel uncomfortable in a specific way — your hand will keep reaching for the phone, your cursor will keep drifting toward the closed Slack tab, you’ll notice an itch to “just check” three or four times in the first twenty minutes. That itch is the muscle complaining. Don’t switch. Stay. The itch fades around minute 25 if you let it.
If your environment keeps recreating the switching cues — the phone arriving at your desk at 9am face-up, the Slack tab reopening because it’s pinned — the deeper fix is removing those cues from the environment so the urge doesn’t form in the first place. And once you’ve held a clean 90-minute block, the natural follow-up is taking a real break before starting the next one, because the focused work depletes resources you need to refill before the next round.
Multitasking isn’t a skill you’re bad at. It isn’t a thing. The real skill is closing the other tabs and staying with one thing long enough that the work actually gets done.