Body doubling is the practice of doing your work while another person is present. They are usually doing their own thing, sometimes something completely different. They are just there. Somehow that is enough to get you started and keep you going.
If you have never needed it, it sounds like superstition. If you have, you know it is one of the few focus techniques that reliably works on the days when nothing else does.
The trouble is the logistics. It requires a second human being, awake, available, and willing to sit through your Tuesday afternoon. That is a lot of coordination for something you need at unpredictable times. It also fails exactly when you need it most, which is late at night on a task you have been avoiding for a week.
So it is worth asking what is actually doing the work, and how much of it survives without the second person.
What is actually happening
Mostly three things, and you can pull them apart.
The session has edges. Someone else being there gives the work a defined start and a defined end. "I should get to that today" has no beginning, so it never begins. A session is a block with a shape. A surprising amount of the effect comes from the work getting scheduled into a container.
Leaving becomes visible. When you are alone, drifting to your phone goes unnoticed, even by you. With a witness, backing out is an event. They may never say a word, but the exit becomes visible. An exit you cannot see is one you make over and over without noticing.
The deciding is already done. You told them what you were going to work on. That decision was made out loud, before the session started. You sit down to execute something already chosen.
That third one does more of the work than it gets credit for, and it needs no second person at all.
What survives alone
Take those three apart and most of it is reproducible.
The edges are just a timer, which is why Pomodoro works for some people. But a timer only supplies edges. It answers how long, and it assumes you have already settled what. If you have not, the timer starts and you spend the first several minutes choosing, which is the same stall you had before, now with a countdown running.
The decision can be made in advance, by you, for you. Settle the order at a time when you are under no pressure to start, and the working session begins with an instruction already in hand. It is also free.
The visible exit is the one that needs a mechanism. Drifting away happens quietly, so you cannot catch yourself in the moment. You can count it, though. A record of how many times you left the session, and how many times something pulled you out, turns an invisible habit into a number you can look at afterwards. That gets closer to a person in the room than it sounds.
Everyone knows they get distracted. Seeing that it happened fourteen times in fifty minutes is the surprise.
Why we built it this way
Tunl Focus goes after the same three mechanics directly, without a stranger on a video call.
The Reckoning is where the deciding happens, deliberately separated from the doing. Every task gets a verdict, so by the time you sit down, what is next is already settled. The session starts with an instruction.
The Tunnel supplies the edges and the visibility. It runs one task, full screen, with nothing underneath it. It also counts the two things that end sessions: how often you back out, and how often something interrupts you. Those numbers act as the witness. They are a plain record, which is most of what the witness was providing anyway.
The honest limit
Some of body doubling is genuinely social, and that part belongs to people. If what actually helps you is another person's company, or the gentle pull of wanting to look busy in front of someone you like, a person is still the answer. A counter on a screen gets you part of the way.
What you can reproduce is the structural half: decide beforehand, give the work edges, make leaving visible. That half is available on a Tuesday at eleven at night with nobody around, which is exactly when most people need it.
If the reason you cannot start is further upstream than focus, the fuller version of that argument is in why task paralysis happens. If you want the case for stripping the screen to one thing, that is one task at a time.