Performing a version of yourself, and what it costs
The short version
- Some people run a live check on themselves in company — tone, face, volume, timing — and correct it as they go.
- It works. That is the whole reason anyone does it, and any account that skips the benefit is not describing the real decision.
- It is also expensive in a specific way: you end up unable to tell who likes you, only who likes the version.
Most advice about social difficulty assumes the difficulty is a shortfall — you did not know what to say, so learn what to say. There is a different situation, which is that you know exactly what to say, you say it, and the saying costs something the other people in the room are not spending.
The work is monitoring. Checking your expression against what the moment seems to want. Hearing your own volume from outside and bringing it down. Noticing the second before you interrupt and holding the sentence. Choosing which of the things you actually think is the one that will land. Doing all of that while also following the conversation, which is the part everyone else is spending their attention on.
Why it works
It works because it is skilled. People who do this are often read as easy company, and that reading is not a mistake — the adjustments are real adjustments, made in real time, and they genuinely smooth the room.
It also buys concrete things. Interviews go better. Meetings go better. A group that would otherwise have found you hard to place finds you legible instead, and being legible determines who gets invited back, who gets asked to lead something, who gets treated as reasonable during a disagreement. Anyone telling you to simply stop is asking you to give up all of that, usually without offering to cover the difference.
Where it costs
It costs during. Running a second process alongside the conversation uses capacity, and the capacity is finite, so long social stretches drain in a way that looks disproportionate from the outside. Three hours can be as expensive as a working day, and the recovery afterwards is not sulking.
It costs after, in a way that is harder to name. If a version of you was in the room, the warmth in the room is attached to that version. Being liked stops functioning as information. You can be surrounded by people who enjoy your company and still hold the private position that none of them have met you, which is not insecurity — it is an accurate reading of the situation you set up.
And the version tends to harden. Do it long enough with the same people and dropping it becomes a disclosure rather than a relaxation, so the cheapest option is to keep going, which is how the tiredness becomes permanent.
The mismatch runs both ways
There is a useful idea from Milton's 2012 paper on the double empathy problem: when two people with different processing styles find each other hard to read, the difficulty is a two-way mismatch rather than one person's shortfall. It is described in the research literature as mutual.
In practice the adjusting is not mutual. One person does nearly all of it, usually the one who has learnt that the room costs less when they do. Naming that does not solve it, but it does change what the effort is. It is not compensation for a gap in you. It is one side of a shared mismatch, done entirely by one side.
What the setting can do
Reduce how much real-time performance the setting requires, which is nearly always possible and rarely considered.
Concretely: send the agenda and the questions in advance, so contribution does not depend on composing under observation. Accept written input as equal to spoken input, and say so out loud, rather than treating the person who typed it as having opted out. Run smaller groups — six people is a conversation, twenty is an audience. Make cameras genuinely optional and stop reading the choice. Build in real breaks at events, in a room where nobody is expected to talk, and let people leave early without producing a reason.
Any group can adopt one norm that helps more than all of this: stop treating quiet as a problem to be solved. A great deal of performance is generated by other people's discomfort with silence.
Small things to try this week
- Pick one low-stakes setting and drop one specific adjustment. Not all of them. One — the fixed expression, or the pre-editing of sentences.
- After your next long social stretch, note the cost in hours rather than judging it. Cost data is more useful than a verdict.
- Tell one person one true and slightly inconvenient thing. Small. The point is to get a single piece of evidence about how the unedited version lands.
If nothing changes after a few weeks, that is information too, and it usually says more about how much adjusting a particular room demands than about how well you are doing it.
Written for Thought Club as general guidance, not medical advice. Free to read, print, and share.