It is 9pm, you have not eaten, and the thing you did for four hours is not the thing you sat down to do.
You opened the laptop at five to send one email. A question about the spreadsheet caught you on the way, and the next four hours went into a formula nobody asked for. The email is still a draft. Dinner is cold on the counter. You feel two things at once: the good tiredness of deep work, and the drop in your stomach when you remember the email.
Hyperfocus is real, and it runs on the same rule as not starting
Hyperfocus has been in every ADHD forum for years and in very few labs.
In 2021 two researchers went looking for the science and called their review Hyperfocus: the forgotten frontier of attention. Their finding was plain. Hyperfocus is real, it is understudied, and it is more often reported in ADHD than in other people. It follows interest: the thing you lock onto is the thing you find interesting, not the thing you chose. And the word has no agreed definition, so every study measures something a little different (Ashinoff and Abu-Akel, 2021). So we say it the way they did: a long, deep lock on one interesting thing, with the rest of the room turned down, and no clock.
Now put it next to the unstarted email. They feel like opposites, too much attention and none. But they run on one rule. The psychiatrist William Dodson, who saw adults with ADHD in his practice for decades, describes a nervous system engaged by interest and challenge and not by importance. He calls the locked-in state "The Zone", and in his model the same brain gets into the zone on a thing it finds interesting and cannot get a grip on a thing it does not, however much the second thing matters. A clinical model from years of patients, not a trial, so we say it plainly and no bigger.
An ADHD brain does not start on importance. It starts on interest, and hyperfocus is what interest looks like when it lands. The spreadsheet was a puzzle, so it started itself and did not stop. The email is from Tuesday, and an email from last Tuesday is not new. The four things this brain does start on are here.
You were never lazy. You have one brain with one rule, and the rule chose the spreadsheet.
What hyperfocus costs
Hyperfocus is not a gift with no bill.
People like to say otherwise, because the hours feel good and the work is often good. But you have paid the bill and you know the items on it. The dinner gone cold. The meeting at three you heard about at four. The friend who texted twice. And the biggest item, the task next to the task: the formula instead of the email, the perfect folder structure instead of the tax return, the new codebase instead of the invoice. Hyperfocus picks the target, it picks by interest, and the invoice is never interesting.
Then there is the coming out. You surface at nine, hungry and a little sick, with the email still open and nothing left for it. The good hours were real. So is the evening they cost, and the plain way to say it is this: hyperfocus is attention without a steering wheel.
The first time it cost me a task, I was 11
I wrote my first program at 11, in QBASIC, to check my maths homework.
The program worked. I did not do the homework. The computer was new and the sums were not, and my brain went where the new thing was and stayed there until the evening was gone. I have done a version of this every week since. The interesting problem gets my whole head for hours, and the boring task beside it waits for a deadline. Every tool I built for myself after was an attempt to put the boring task in front of my head at the moment my head arrived.
I have ADHD. When something interests me, I hyperfocus. That is how I have been shipping code since I was 11. But when a task is boring, I know exactly what to do, and I still can't start. So I built Kaapi.
Point it, do not fight it
You cannot switch hyperfocus on, but you can choose what is in front of you when it arrives.
Three things help, and none of them is willpower. First, one task in view when you sit down, so the lock has one place to land. Ten open tabs are ten candidates, and interest will pick the wrong one. Second, a clock you can see, counting up. Hyperfocus has no clock of its own, so the hours have to be shown to you. Third, a small voice every few minutes saying the name of the task you started, so the moment you are on the task next to the task, you hear it. Not a scold. A name.
Then let it run. When the lock lands on the right thing, the session should have no end you did not choose. A timer ringing at 25 minutes breaks the one state this brain is good at, and a page of its own explains why.
How Kaapi does it
A focus session is one task, one clock counting up, and no default end.
Start focus takes the window. The clock counts up from zero, and nothing stops it but you, because the state you are protecting does not run to a timer. Every 5 minutes by default, Timi, the pebble at the bottom right, says one line naming the task you started: "Still on 'Send the invoice'?", "You're 15 minutes in.", "15 minutes on 'Send the invoice'. Still going?" When you are on the invoice, it is company. When you are three tabs deep in the spreadsheet, it is the name of the thing you left, said once, in the corner. Back or Esc leaves the session running, so a glance at the list does not end the run.
Each session opens under a different sky, drawn in code: Cloud gazing, Slow sunrise, Fog on hills and Still sea by day; Star gazing, Aurora, Fireflies and Night lake by night. The sky comes from a random bag and is never the same twice running. "Another sky" works up to three times, then it stays, so the newness goes into starting and not into choosing. A Scene switch turns it off. Music is opt-in: one lo-fi loop we made ourselves.
How to do this without Kaapi
Before you sit down, decide what the lock is allowed to land on.
- Write the one task on a card and put it under the screen. Close the other tabs before you open the first one.
- Start a stopwatch, not a timer. Let it count up, and read it each time you look up.
- Set one alarm for the thing hyperfocus will eat tonight: dinner, the school run, the meeting at three. Label the alarm with the thing.
- When you surface and find yourself on the task next to the task, do not argue with the four hours. Read the card, and start. Starting is the win.
Frequently asked questions
What is ADHD hyperfocus?
Hyperfocus is a long, deep lock of attention on one thing you find interesting, with the rest of the room turned down and no sense of the clock. A 2021 review found it is real, understudied, and more often reported in ADHD, and the word still has no agreed definition. It follows interest, not importance.
Is hyperfocus a good thing?
It is attention without a steering wheel. The hours can produce good work, and they can also eat dinner, a meeting and the task you sat down to do. Calling it a gift hides the bill. The useful stance is to choose what is in front of you before it arrives, and to keep a visible clock while it lasts.
Why can I hyperfocus on games but not on work?
Because the lock follows interest, and an ADHD brain does not start on importance. A game is new, it answers back, and it has a clock. A report has none of those unless you add them. The brain is not choosing the game over you. It is following the only rule it has.
How do I stop hyperfocus from eating the whole evening?
Give it a clock and an exit. A stopwatch you can see counting up, one alarm for the thing you cannot miss, and a line every few minutes naming the task you started. You will still lose some evenings. You will lose fewer, and you will know where they went.
Sources
- Ashinoff, B. K., and Abu-Akel, A. (2021). Hyperfocus: The forgotten frontier of attention. Psychological Research, 85(1), 1–19. doi.org/10.1007/s00426-019-01245-8. A review; the word has no agreed definition.
- Dodson, W. (updated 2025). How adults with ADHD think: Uncomfortable truths about the ADHD nervous system. ADDitude. additudemag.com/adhd-in-adults-nervous-system. A clinical model from practice, not a trial.