Peter Allen Mann

Masking Autism: Why the Bill Arrives a Day Late

Two days of meetings. A conference talk, a customer visit, a supplier audit.

The work went well. By any measure anyone in those rooms would use, I was good at it.

Then I got home and lost the next day. Not a slow morning. A full day where the things that are normally automatic were not available.

Masking is expensive, and the bill arrives late. This essay is about the mechanism that makes that true, and about why the people best at it get the largest bill.

Key takeaways

  • Masking is continuous self-monitoring and performance. It is neither shyness nor politeness, and it runs on the same limited resource the work runs on.
  • The cost is subtracted from capacity rather than added alongside it, which is why output can hold right up until it does not.
  • The better the performance, the longer the invoice is deferred and the larger it gets. Competence is what buys the delay.
  • Unmasking is not a decision you make once, and it is not safe everywhere. Treat it as a cost calculation per setting.
  • The concept, the word and the critique came from autistic adults. The research followed them.

On this page

What masking consists of

Masking is the continuous production of an acceptable social surface. Monitoring your own face and correcting it. Calibrating how long to hold eye contact. Scripting small talk in advance and running the script. Suppressing a response that would come naturally and selecting a different one. Checking, while you are talking, how the talking is landing.

Hull and colleagues interviewed ninety-two autistic adults about this and the descriptions converge on the same thing: intensive concentration, self-control, and the management of discomfort, held continuously for as long as other people are present.

Read that list again and notice what it is not. None of it is the work. All of it is the apparatus around the work, running at the same time as the work, on the same equipment.

Why it works, which is the reason it persists

Masking persists because it is effective. The meeting goes well. The customer signs. The interview converts. No one in the room detects anything to accommodate, which was the objective, and the performance is rewarded in exactly the currency organizations use: promotion, scope, the benefit of the doubt.

Any account of masking that treats it as a mistake is describing something people do not do. People mask because it works, and because in most rooms the alternative has a price that is visible immediately while the price of masking is not visible for a day.

That asymmetry is the whole design. One cost arrives in the room, in front of the people whose opinion determines your career. The other arrives at home, alone, on a Saturday.

The cost is subtracted, not added

Here is the part that took me longest to understand, and it is the part that changes what you do about it.

Self-monitoring draws on working memory. Working memory is the same limited workspace you use to reason, hold a problem in mind, and do complex work. So, the energy spent navigating a room is not spent alongside the work. It comes out of the exact resource the work depends on.

Which means there is no version of this where you mask and also perform at full capacity. You perform at capacity minus the tax. If the output still looks full, you are covering the gap with something else: longer hours, no recovery, the weekend.

Meanwhile the output holds, and holds, and holds. As long as it holds, no one asks, including you.

The delayed invoice

The bill for a masked performance does not arrive during the performance. It arrives hours or days later, and it arrives as a loss of ordinary function rather than as tiredness. Speech is harder. Decisions that are usually automatic require deliberation. Plans made in good faith get cancelled. From outside it reads as flakiness, and from inside it reads as a character defect, which is how I read it for about four decades.

I used a credit line without knowing it was a credit line. The statements were arriving the whole time. I filed them under personal failings because that was the only category I had.

The twenty-four hours after a trip are not a failure of resilience. It is the invoice. It was always the invoice.

Why being good at it makes it worse

The intuition most people have is that a skilled masker pays less. The opposite holds, for two reasons that compound.

The first is demand. A person who performs well in rooms gets sent to more rooms. Competence is routed toward the exact settings that cost the most, and the routing is done by people acting reasonably on the evidence in front of them.

The second is detection. The better the surface, the less anyone sees, and the less anyone sees, the longer the situation continues without adjustment. Support arrives in response to visible difficulty. Producing no visible difficulty is what masking is for.

So the skill buys you a larger bill on a longer deferral. I spent thirty-seven years in leadership and the thing that kept the arrangement running was that it kept working.

How is masking different from the things everyone does at work?

Everyone adjusts at work. The difference is in what the adjustment costs and whether it ever stops. A professional register is a layer you put on and take off, and it runs on the parts of social processing that are automatic. Masking is a layer that has to be computed continuously, from the outside, because the automatic version is not available.

That distinction matters because the usual response to someone describing masking is that everybody does it. Everybody does do a version of it. What hardly anyone else does is run it as a deliberate calculation for eight hours and then find the ordinary parts of their evening are gone.

Three things separate the two, and all three are observable rather than felt:

  • It is deliberate. The adjustments are selected, one at a time, in real time. Where to look, when to speak, what expression to hold.
  • It does not switch off with familiarity. A professional register relaxes with people you know well. The computation often does not, because the thing being computed is other people’s reactions rather than your own comfort.
  • It has a measurable aftermath. There is a recovery period, it is longer than the event, and it takes out functions unrelated to the event.

That third one is the test I would offer anyone trying to work out whether this describes them. Not whether the day was tiring. Whether the day after was unusable, and whether the things that stopped working were things that have no obvious connection to talking to people.

What the organization sees while this is happening

From the employer’s side the sequence is invisible and then abrupt. The output is good. The reviews are good. The person is given more of the settings they appear to handle well, which is a reasonable decision made on the evidence available. Then something changes and it looks like it changed without warning.

I spent most of my career on the side of that table, and the thing I would tell any executive reading this is that the failure is in the instrument. A performance system measures the structured hours. The cost is incurred in the unstructured ones and paid at home. There is no column for it, so it does not exist, right up until the person stops being able to produce the surface and the whole thing reads as a sudden decline.

Where this ends if nothing changes

Sustained masking is associated with anxiety, stress and a disturbed sense of self in Hull’s interviews, and it sits in the causal account autistic adults give of burnout. Raymaker and colleagues, working with autistic adults rather than about them, define autistic burnout as chronic exhaustion, loss of skills and reduced tolerance to stimulus, and locate the cause in a sustained mismatch between expectations and capacity without adequate support.

Note what that definition does not say. It does not say burnout is caused by working too hard. The load in question is the load of operating inside an environment built for a different processing style, and the masking is the mechanism by which that load gets paid.

If you are already there, the honest answer is that there is no protocol, and I have written about why at length in recovery is not a protocol.

What unmasking does and does not mean

Unmasking has acquired a romance it does not deserve. It is not a single decision, it is not a personality reveal, and it is not safe in every room. Some rooms punish it. The punishment is real, and it falls on the person rather than on the room.

What it is, practically, is a set of per-setting decisions about which parts of the performance to stop running and what that choice costs on each side. Dropping eye contact in a one-to-one with someone who already knows is nearly free. Dropping the scripted small talk at a customer dinner may not be. The calculation is different in every setting and it is yours to make.

When I disclosed my diagnosis at the company I was running, the effect I did not predict was on everyone else. Energy that had been going into performance came back and went into the work. I will not tell you that generalizes. I had the authority to make the room safe first, and most people reading this do not.

That asymmetry matters enough to say plainly. Advice to unmask, given to someone with no power in the room, is advice to absorb a cost on behalf of an environment that will not change in return.

What to measure instead of trying harder

The useful move is not a coping strategy. It is a record, because the whole problem with this cost is that it is invisible, delayed, and therefore unattributable.

Three columns, for a month:

  • The setting. What it was, how long, how many people, and how much of it was unstructured.
  • The recovery. Hours or days before the automatic things were automatic again.
  • What specifically was gone. Speech, cooking, decisions, the ability to answer a message. The list is more specific than “tired” and the specificity is what makes it usable.

A month of that tells you something no amount of effort will. It tells you which settings are expensive, and expensive settings are a property of the environment rather than of you. That is the only part of this that anyone can actually change, and it is often changeable at close to zero cost: an agenda in advance, a decision recorded in writing, a room with a door.

The word for the trait list is yours. The word for the environment is design.

Why this is a property of the room

The cost of masking varies enormously by setting, and the thing that varies is the setting rather than the person. The same adult can hold a two-hour technical review at close to zero cost and be wrecked by forty minutes of unstructured networking. Same brain, same day. What changed is how much of the interaction had to be computed in real time.

That is a design fact, and it points at the part of this that is actually fixable. An agenda circulated in advance removes the computation of what is coming. A decision recorded in writing removes the need to have been reading the room correctly while it was being made. A door removes the sensory load. None of these is an accommodation in the sense of a concession, and all of them are cheap. What they have in common is that they reduce the amount of a job that has to be done through live social inference.

I built a lot of this at the companies I ran, for selfish reasons, long before I had a diagnosis to explain why. The implicit version of an organization was unreadable to me and I had no alternative. What it produced was a set of environments where other people also did better, which I did not anticipate and did not design for.

So the honest end of this essay is not a technique. The amount of masking a given day requires is set mostly by how much of the work runs on things that were never written down, and that number is decided by whoever designs the day. If that is not you, the record described above is how you make the case to the person it is.

What this essay is and is not

I will be direct about what this is. I am not a clinician, and nothing here is diagnostic or clinical advice. I am an operator who was diagnosed autistic at fifty-five after thirty-seven years of running organizations, and I am describing a mechanism I ran on for most of that time without knowing its name.

The name is not mine either. Masking, camouflaging, the whole account of what it costs and why people keep doing it, came out of autistic adults describing their own lives, and it was circulating in that community for years before the research caught up with it. The insight is theirs. The vantage point is mine, and the only thing I add is having watched it from the side of the table that decides who gets promoted.

If the label is what brought you here, the companion piece is high-functioning is a description of other people’s convenience, which is the same mechanism seen from outside.

Sources

  • Hull, L., Petrides, K. V., Allison, C., Smith, P., Baron-Cohen, S., Lai, M.-C., & Mandy, W. (2017). “Putting on My Best Normal”: Social Camouflaging in Adults with Autism Spectrum Conditions. Journal of Autism and Developmental Disorders, 47(8), 2519–2534. Qualitative study, 92 autistic adults. Journal of Autism and Developmental Disorders, “Putting on My Best Normal”
  • Raymaker, D. M., Teo, A. R., Steckler, N. A., Lentz, B., Scharer, M., Delos Santos, A., Kapp, S. K., Hunter, M., Joyce, A., & Nicolaidis, C. (2020). “Having All of Your Internal Resources Exhausted Beyond Measure and Going Back to Baseline”: Defining Autistic Burnout. Autism in Adulthood, 2(2), 132–143. Community-based participatory research with autistic adults. Autism in Adulthood, Defining Autistic Burnout
  • Alvares, G. A., et al. (2020). The misnomer of “high functioning autism”: Intelligence is an imprecise predictor of functional abilities at diagnosis. Autism, 24(1), 221–232. Autism, The misnomer of “high functioning autism”
  • Schmader, T., & Johns, M. (2003). Converging Evidence That Stereotype Threat Reduces Working Memory Capacity. Journal of Personality and Social Psychology, 85(3), 440–452. The measured finding that managing a threat to how one is perceived reduces working-memory capacity. Journal of Personality and Social Psychology, Converging Evidence That Stereotype Threat Reduces Working Memory Capacity
  • Mann, P. A. (2026). The Perception Revolution. Fast Company Press. Chapter 11, on the mask providing no advantage once contribution is measured by outcomes.

Peter Allen Mann is the author of The Perception Revolution (Fast Company Press, 2026). He spent thirty-seven years in leadership, including four years as a Navy Surface Warfare Officer and two companies founded and run, and was diagnosed autistic at fifty-five. He writes for executives and system designers at SHIELD Perception.