You have probably noticed something embarrassing about yourself and never said it out loud: you are a more disciplined person when someone can see you.

Alone in your kitchen, you open the document, read the same sentence four times, and then you are somehow reading about a shipwreck from 1892. But put a stranger at the next table in a café — someone who does not know your name, will never read your work, and is not paying you the slightest attention — and you write for ninety minutes without once reaching for your phone. Nothing about the task changed. Nothing about your willpower changed. The only thing that changed was that a human being existed nearby.

This makes a lot of people feel like frauds. If your focus depends on an audience, is it even yours? But the discomfort is built on a bad assumption — that attention is supposed to be a private, self-generated resource, and that needing help is a character flaw. It isn't. Your attention system evolved in the constant presence of other people. It has always been calibrated to them. The lonely, sealed-off room is the strange condition, not the café.

The oldest finding in social psychology

In 1898, Norman Triplett noticed that cyclists rode faster when racing against others than when riding alone against the clock. He followed it up in the lab with children winding fishing reels, and found the same thing: they wound faster in pairs than solo. It is often cited as the first experiment in social psychology, and the effect it named — social facilitation — turned out to be one of the most durable and most confusing results in the field.

Confusing, because for decades the studies contradicted each other. Sometimes an audience made people better. Sometimes it made them dramatically worse. Cockroaches ran a simple maze faster with other cockroaches watching, and a complicated maze slower. Students recited well-learned material more fluently in front of a crowd and fumbled unfamiliar material more badly.

Robert Zajonc resolved this in 1965 with an explanation so clean it still holds up. The presence of others, he proposed, raises general arousal — physiological alertness, a readiness to act. And arousal doesn't make you smarter. It makes you more likely to do whatever your dominant response is in that situation.

If the task is simple or well-practiced, your dominant response is the correct one, and arousal pushes you toward it faster. You wind the reel quicker. You type faster. If the task is novel or complex, your dominant response is often wrong — a fumble, a guess, the obvious-but-mistaken move — and arousal pushes you toward that instead. Hence: an audience helps you perform what you already know, and hurts you while you are still learning it.

Later work refined the mechanism. Nickolas Cottrell argued the arousal isn't automatic from mere presence but comes specifically from evaluation apprehension — the anticipation of being judged. Blindfolded observers who couldn't see the performer produced far weaker effects. And Robert Baron's distraction–conflict theory offered a third account: another person creates an attentional conflict, because part of you wants to monitor them and part of you wants to do the task. That conflict narrows attentional focus onto the central task and squeezes out peripheral information — which is exactly why it helps on simple tasks and wrecks complex ones that need a wide field of attention.

Notice what all three explanations share. The other person is not helping you. The other person is changing what your attention is willing to ignore.

What this means for the way you actually work

The internet has independently rediscovered this and named it body doubling — a term that came out of the ADHD community, where people found that simply having someone else present, working on something entirely unrelated, made previously impossible tasks start. It has since spread to virtual co-working rooms, silent video calls, and libraries full of strangers wordlessly typing.

Body doubling gets described as "accountability," but that's not quite right, and getting it wrong will make you use it badly. There is no accountability with a stranger in a café. Nobody will ask what you produced. What actually changes is subtler and more mechanical:

Your dominant response gets amplified. This is the whole ballgame. Social presence is a multiplier, not an additive boost. If you sit down in a co-working space with a clear, well-defined, already-practiced task — write the section you already outlined, respond to twelve emails, clean the dataset — presence pushes you into it hard. If you sit down with a foggy, novel, cognitively demanding problem — figure out what this essay is even about — presence often makes you worse, because your dominant response under pressure is to perform the appearance of working. You will open the document, arrange it beautifully, and think nothing.

Your peripheral attention narrows. The distraction–conflict account predicts that presence buys you tunnel vision. That is a gift when the distraction you need to suppress is your own phone, and a curse when the solution requires a loose, wandering, associative mind. This is why writers often draft in cafés and think in the shower.

Self-monitoring goes up. Being observable makes you self-aware, and self-awareness makes you compare your current behavior to a standard. Under someone's eye, the gap between what I am doing and what I said I would do becomes visible in a way it simply is not at 2pm alone in your apartment. That is uncomfortable. That discomfort is the mechanism working.

So the practical rule is not "work near people." It is: match the presence to the phase of the work. Use it for execution, protect solitude for exploration.

Your next moves

  • Split your task list into two columns before you decide where to work. One column: tasks you already know how to do, that just require doing. The other: tasks where you don't yet know the shape of the answer. Take column one to the café, the library, or the co-working call. Do column two alone, ideally on a walk.

  • Book one two-hour silent co-working session this week for a task you have been avoiding for more than ten days. Not a meeting — a video call or shared room where nobody talks. Say the task out loud at the start, then mute. The saying-out-loud converts vague presence into evaluation apprehension, which is the active ingredient.

  • If you can't get a human, get a proxy — but be honest that it's weaker. A timer running visibly, a session you have to explicitly abandon, a shared doc where someone might glance at the edit history. Anything that makes your behavior observable to a future observer recruits some of the same self-monitoring.

  • Never body-double the hard thinking. The next time you sit in a busy place trying to solve a genuinely novel problem and produce nothing, don't conclude you're undisciplined. Leave. That's the effect working exactly as Zajonc described it.

  • Before any co-working block, write the single sentence that defines "done." Presence amplifies your dominant response, so the entire question is whether you've made the right response dominant before you sit down. A vague goal under social arousal produces beautifully formatted nothing.

The deeper thing here, the part that is worth sitting with: your attention has never been purely yours. It is a social organ. It has always been partly held by other people, shaped by who is watching, tuned to a room. Building a life where you sit alone in silence and expect the machinery to run on internal fuel alone is not a test of character. It's a design error.

What you can control is the structure around the work — the visible commitment, the defined edge, the thing that makes drifting away a decision rather than an accident. That's the part Reclaim is built for: it turns a focus session into something with a stated intention, a visible boundary, and a cost to abandoning it, so the accountability lives in the structure even when nobody's in the room. If your best hours keep dissolving alone at your desk, give your focus something to answer to.