Sensory overload in an adult body

The man across the table was shouting the word “intimate.” He was not angry. He was describing the restaurant, and he had to shout because shouting was the only working volume in the room. A few minutes earlier someone at the next table had announced that it was really loud in here. Everyone agreed out loud. Then the volume went up, because the pause had left a gap and the gap wanted filling.

The hostess had called the room lively.

I was not having a bad evening. I was running out of capacity. That is a different event with a different mechanism, and the mechanism has a shape you can learn to see.

Key takeaways

• Sensory overload in adults is a capacity problem. Input arrives faster than the system clears it, and the system takes power from everything else it was running.
• Comprehension holds the whole time. Speech and social response are what get cut, which is why it reads as withdrawal from outside.
• It accumulates across a day rather than arriving as one event, so the visible failure lands hours after the cause.
• The usable variable is the room, and the room is a decision somebody already made.

What is actually happening during sensory overload in adults

Sensory overload in adults is a capacity problem. The nervous system is clearing less input than is arriving, so it starts taking power away from everything else it was running. Comprehension holds. Speech, social response and conversational tracking are what get cut first, because they are the most expensive processes on the list.

By the time I stopped contributing to that dinner, my system had assigned everything it had to managing sound. The social processing that would have let me answer the question I had just been asked was not available. Silence was an output, not a choice.

A circuit breaker works the same way and gets less credit for it. It does not weigh which appliance matters. It trips when the load passes what the wire can carry, and the room goes dark regardless of what you were doing in it. Nothing about the breaker is an opinion about your evening.

Why sensory overload is not a preference

A preference is something you can trade. Sensory overload is not available for trade, because the input arrives before any decision does. The sound is already in the room. The fluorescent flicker is already overhead. The processing cost is incurred whether or not you have decided the evening is worth having.

Two numbers make the asymmetry concrete. They do not come from the same kind of source, so I will say which is which. Restaurants in major cities have been reported measuring eighty to ninety decibels at peak hours, a figure from a magazine investigation into restaurant acoustics rather than from an official statistics series (The Atlantic, How Restaurants Got So Loud). The eighty-five decibel figure is regulation. It is the action level at which an employer has to run a hearing conservation program, measured as an eight-hour time-weighted average in an occupational setting (Occupational Safety and Health Administration, Occupational Noise Exposure, 29 C.F.R. § 1910.95).

Two hours at dinner is not an eight-hour shift, and the comparison does not transfer cleanly. It is not meant to. The point is narrower than that. A sound level that triggers a regulatory duty when a person is paid to stand in it gets sold as atmosphere when a person is paying to sit in it.

The design decision was made before anyone arrived, by people for whom the volume is the product. The noise is not a defect in their evening. It is the reason they left the house. And the person who goes quiet in that room is then assessed by the people who picked the room.

What are some common sensory issues experienced by people with autism?

The ones most commonly described are sound, light, texture, smell and interoception, in combinations that differ from person to person. Autistic adults have been describing these for decades, in their own forums and in their own words, well ahead of the research literature. The description belongs to them. What I am adding is one vantage point, which is thirty-seven years on the other side of a desk.

One piece of my own wiring is relevant here and I only found it after the autism diagnosis. I am also what researchers call a highly sensitive person, which accounted for a second set of things I had filed under being difficult. Loud rooms felt like assault when I was a child. They still do.

How the cost accumulates across a day

Sensory overload in an adult working day does not arrive as one event. It accumulates. Every meeting, every open-plan hour, every corridor conversation draws on the same supply, and the supply does not reset in between. So the failure shows up late in the day, in the least structured part of it, which is also the part where people form their impressions of you.

Before a work dinner I start running calculations three days out. Which restaurant. How loud. The lighting. Whether it is a long table where I am fixed next to one person for two hours, or a round table where six people talk at once. Whether the background music is somehow louder than the foreground. Whether I can sit against a wall so the sound arrives from one direction. Whether the menu is online, so I am not choosing food while also holding the room.

The person I am meeting prepares for the same dinner by deciding whether to drive. In my next book, out in April 2027, I put it this way: I arrive at the restaurant having already completed a mission briefing. We are both expected to have had the same evening. When I have not, the shortfall gets recorded as mine.

The arithmetic of that is worth sitting with. By the third hour of a long dinner, one person at the table is running full sensory processing, social performance monitoring and conversation tracking at the same time, on a system that passed its threshold an hour ago. Everyone else is having a nice night out. The one working hardest is the one who looks disengaged, and the table reaches its verdict at exactly the wrong moment.

There is a further cost and it has the least measurement behind it of anything on this page. Suppressing a visible response is a form of camouflaging, and camouflaging has been tested against exhaustion directly in at least one study. In a cross-sectional survey of ninety-two autistic adults recruited through autistic-led groups, burnout-exhaustion partially mediated the association between camouflaging and depression (Autism in Adulthood, Camouflage, Burnout-Exhaustion, and Depression in Autistic Adults). Two features of that study matter as much as its result. Everything was measured at a single timepoint by self-report, so the mediation is a decomposition of correlations rather than a sequence anyone watched unfold. And the exhaustion measure had to be adapted from a parental burnout scale, because no validated autistic-burnout instrument existed to use instead. The second fact is the more useful one. This field is still borrowing its instruments, which is worth knowing before anyone tells you the question is settled. What that bill looks like when it finally comes due is a separate piece.

The same processing architecture runs in both directions, which is why a subject can hold for thirty years and a restaurant can be unsurvivable for ninety minutes. One ledger, two entries.

The environment edits that changed mine, and what they cost

The edits that worked were environmental, and none of them was a coping strategy. Lighting I control. Sound I control. A door that closes. A schedule with recovery built into it rather than hoped for. Each one cost me something real, usually convenience or a particular kind of job, and each one bought back capacity that had been going to the room.

The first one I ever found was the school nurse’s office. My classroom had fluorescent lights that buzzed, twenty-five children generating chaos, and noise that arrived with no warning. The nurse’s office had a chair and a calm adult in it. I traded the classroom for that room on a regular basis and I would make the trade again today. At six I had already worked out that the room was the variable. It took me another fifty years to get the vocabulary for it.

The reason I ended up running companies is less inspiring than it sounds. A founder sets the lighting. I was not climbing. I was finding places where I could breathe.

Which brings me to the argument I find hardest to get around. No company has ever posted a job requiring the ability to work eight hours without a bathroom break. No performance review has ever dinged someone for excessive biological needs. We accepted a long time ago that biology requires infrastructure, and we built the infrastructure. Then we stopped. The same building that provides stalls for one biological system asks people to override a different one all day, and files the results under character.

What I would do instead of a protocol

What I would suggest is an accounting you can run on yourself in two weeks.

Write down the hour of the day you stop being able to do the easy things. Not how you felt. The hour. Then write down what room you had been in for the two hours before it. Do that for ten working days and read the two columns together.

Most people find the hour is fairly stable and the room is what moves. That is the finding, and it is the only part of this anyone can act on. The hour is a symptom. The room is a decision somebody made.

The room was never neutral. It was just never on the invoice.

Sources

• Wagner, K. (2018). How Restaurants Got So Loud. The Atlantic, 27 November 2018. Reported decibel measurements from a magazine investigation into restaurant acoustics, not an official statistics series. The Atlantic, How Restaurants Got So Loud
• Occupational Safety and Health Administration. Occupational Noise Exposure, 29 C.F.R. § 1910.95. The 85 dBA action level is an eight-hour time-weighted average in an occupational setting. Occupational Safety and Health Administration, Occupational Noise Exposure
• Benatov, J., Sarel-Mahlev, E., & Bar Yehuda, S. (2025; print issue 2026). Camouflage, Burnout-Exhaustion, and Depression in Autistic Adults. Autism in Adulthood. doi:10.1089/aut.2024.0147. Online survey, n=92, recruited through autistic-led groups; CAT-Q, PHQ-9, and an exhaustion subscale adapted from the Parental Burnout Assessment. Cross-sectional, so the mediation is a decomposition of correlations measured at one timepoint. Autism in Adulthood, Camouflage, Burnout-Exhaustion, and Depression in Autistic Adults
• Mann, P. A. (2026). The Perception Revolution. Fast Company Press. The nurse’s office, the highly sensitive person material, and the environment-design argument.
• Mann, P. A. (April 2027). The Projection Fallacy. Fast Company Press. The restaurant passage and the pre-dinner calculation; the accumulation across a long evening.

Peter Allen Mann is the author of The Perception Revolution (Fast Company Press, 2026) and The Projection Fallacy (April 2027). 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.