Seven a.m., Sunday morning. A day and a half of quiet, and then the jackhammer finds its rhythm again.
And the strange part isn't the noise itself. It's that you've already forgiven the world for being quiet.
Daniel lives right up against this construction site in Jerusalem, and he sent us the follow-up. The site shuts down on Saturdays, it shuts down for the Jewish holidays, and after even a day or two of quiet he notices he's more relaxed at home. Often without clocking why. Then Sunday comes, the jackhammering starts, and it's jarring in a way that feels disproportionate. Like he'd adapted to the silence and now has to start the whole adjustment over.
And the mornings are the worst of it.
Mornings are when he's trying to organize his thoughts and plan the day, which for him is not a lifestyle preference. It's the scaffolding he needs, because he's got ADHD. The noise cuts straight across it. He loses his train of thought, he can't concentrate, and he says he starts making spelling mistakes he would not normally make. He wants to know what's actually happening in the brain, whether habituation can partially reverse during quiet, what intermittent unpredictable construction noise does to attention and working memory in people with ADHD or sound sensitivity, and what the research actually supports by way of intervention. Controlled exposure, habituation techniques, the audiology and psychology angles, without giving up the restorative value of the quiet days.
So we're going to take all of that seriously as a neuroscience case study.
Because what he's describing maps almost embarrassingly well onto the literature. The Sunday whiplash isn't a personal weakness. It's a documented feature of how the system works.
Let's set the two concepts first, because everything else hangs off them. Habituation is a decrease in your response to a repeated stimulus. Sensitization is an increase. The 2022 zebrafish work on repeated environmental stressors puts it cleanly: repeated exposure can produce either one, and which you get depends on the intensity, the duration, and the time interval of the stressor.
The interval.
The interval is doing an enormous amount of work in this whole story. But here's the wrinkle that makes it a real puzzle. The same stimulus can drive both at once. There's a study on grey seals from 2011, repeated acoustic startle, and what the researchers found was rapid and pronounced sensitization of the animals' spatial avoidance behavior. Fear conditioning. They started avoiding a food source they already knew was good, because it sat near the sound.
So the reflex part is going down, and the avoidance part is going up.
Exactly that split. A jackhammer can be the kind of stimulus that sensitizes avoidance even while it's habituating the startle. Which is why "you'll get used to it" is such a useless thing to say. Maybe you will, and you'll still flinch internally every Sunday.
One more thing before we dig in. There's an interesting detail from the construction noise work. When researchers measured physiological response to pile-driver noise versus tonal earth-auger noise, the impulsive stuff that a jackhammer actually is doesn't show up cleanly in skin-conductance measures the way the steady tonal noise does. Jackhammers are, in a sense, harder for the body to even register as a stress response. Which is its own kind of telling. The body knows, and the usual instruments don't.
That's the framing. Now let's get into the mechanism, because the mechanism is where Daniel's exact experience lives.
The spacing.
The spacing paradox, yes. This is the core finding, and I don't think there's anything else in the literature that maps onto Daniel's situation this precisely. A 2008 study in Behavioral Neuroscience, Masini, Day and Campeau. Rats, 95-decibel white noise, two groups. One group got the noise massed. Six thirty-minute exposures crammed into six hours. The other group got it spaced. One thirty-minute exposure per day for six days. Both groups habituated. By the sixth exposure, both showed a reduced stress response.
And then.
And then they waited forty-eight hours and re-exposed them. The massed group's habituation was gone. Just not retained. The spaced group's was still there. And the conclusion the authors draw is that relatively short interstressor intervals impair long-term stress adaptation. They argue there are distinct short-term and long-term habituation processes, and the long one needs spacing to consolidate.
That is a strange sentence to hear out loud. Short intervals between repeated stress hurt your ability to adapt to the stress.
It's counterintuitive, and it's real. The spacing is doing something the density can't. And before you ask whether this is a rat thing, there's human data. Foss and colleagues, 1989, Human Factors. They had people doing a rifle-aiming task while bursts of noise went off. Disruption got smaller across the five bursts in the first series, but it never went to zero, and in a fifteen-minute rest period it didn't recover at all. Then they tried longer rests.
How much longer.
Twenty-four hours. Partial recovery. Seven days. Also partial recovery. So the adaptation didn't disappear after a short breather. It came back a little after a full day, and a little more after a week.
So Daniel's Saturday isn't just resetting him to zero. The quiet day is plausibly undoing part of what he built during the week.
Part of it. Which produces his exact experience. Sunday doesn't feel like Sunday from a neutral baseline. It feels like falling from a height you climbed during the week and forgot about. Because the quiet days are also, and here's the twist, the thing making the long-term adaptation possible in the first place.
Say that again, because it sounds like you're saying the quiet is both the cure and the poison.
It is, or close enough. Massed exposure, according to Masini, doesn't stick. Spaced exposure does. Daniel's living pattern is naturally spaced. Working days with the noise, Saturdays and holidays without. That spacing is exactly the schedule that produces durable long-term habituation. So the quiet weekends aren't just pleasant, they're doing something the constant exposure wouldn't. And then they do the other thing. The spontaneous recovery that throws him off on Sunday.
There's got to be a cortical piece in that, right? That doesn't sound like a reflex arc.
There is. Leaton and Jordan, 1978, EEG arousal habituation in rats. Between sessions twenty-four to seventy-two hours apart there was essentially no spontaneous recovery, and at thirty-two days the retention was around ninety percent. So the long-term version is durable. Then Yeo and Oakley in 1983 took rats and decorticated them, and those animals showed partial spontaneous recovery between sessions. The read is that the neocortex is needed for complete long-term habituation. Strip it out and the animal can only hold the short version.
So Daniel's brain is running the long-term version during the week, the cortex is consolidating it, and the weekend quiet is activating a recovery process that he has absolutely no control over and cannot feel happening.
Which is a very particular kind of unfair. And it gets worse when you ask what the noise does to the cognitive system while he's trying to work, because intermittent unpredictable noise is not the same animal as steady noise. There's a JASA vigilance study that found people were less sensitive, less accurate, and more prone to response failures during intermittent noise.
Less sensitive is the one I want to sit on. That's not "distracted." That's the floor of perception itself moving.
It's worse than distraction. And the ADHD piece is the part that should actually make Daniel feel less crazy than he probably does. Blomberg et al., 2021, Frontiers in Human Neuroscience. Adults with ADHD, seventeen of them, against seventeen controls. Auditory target-detection task, plus a visual n-back task running at the same time, with noise playing. At low working memory load, the ADHD group handled it. At two-back, high working memory load, they couldn't attenuate the auditory cortical response to the task-irrelevant sound. The sound stayed loud in the brain when it should have been turned down.
And that correlated.
Heightened auditory activity to irrelevant sound was associated with poorer working memory performance and with symptomatic inattentiveness. And there was a significant increase in functional communication between the auditory network and the salience network in the ADHD group.
Auditory to salience. The sound-processing region talking directly to the network that decides what matters.
Which is not supposed to happen, or not at that volume. The salience network is the thing that says "pay attention to this, it's important." And here it's getting a line from the auditory system that it shouldn't be getting, or getting it too strongly. So the noise isn't just happening in the background for Daniel. His brain is tagging it as salient, which is why it pulls him out of a train of thought mid-sentence. It's not the ear. It's the classification.
Which is the specific reason mornings are the worst.
Mornings are his planning window.
Mornings are the highest-working-memory-load hour of his day. And if the interference only emerges at high load, then the jackhammer starting at seven isn't hitting him while he's doing something easy and letting it slide. It's hitting him at the exact moment his working memory is fully committed.
And there's more in the ADHD literature pointing the same direction. Michalek, 2014, International Journal of Audiology. Young adults with ADHD benefited less from visual information during noise than controls did, and that effect was influenced by working memory abilities. So cross-modal help doesn't rescue them the way it rescues other people. Stevens, 2012, found elevated thresholds to crowded displays, which the authors read as an early-stage perceptual processing deficit distinct from selective attention. Breier, 2002, in children with ADHD under contralateral masking noise, found a significantly greater increase in false-alarm rate. Not a sensitivity change, a response-bias change. Meaning the ADHD brain has more difficulty inhibiting the wrong answer when there's noise.
That's the spelling mistakes.
That is the spelling mistakes.
You know what that is? It's not sloppiness. It's the inhibition channel getting crowded out by an irrelevant sound that somebody's brain has decided is very important.
Right. So Daniel is not describing a character flaw. He's describing the behavioral surface of three separate documented mechanisms: sensitization of avoidance, load-dependent auditory interference, and reduced inhibitory control under noise. All firing at seven a.m. on a Sunday.
Which is a lot of mechanism for a single jackhammer to be running. So the question becomes: given that, what does the research actually support? What's the intervention side?
So this is where I have to be honest with him, because he asked directly for the practical stuff, and the practical stuff is thinner than the mechanism stuff.
Thinner how?
There is no controlled-exposure protocol for construction noise. I looked. It doesn't exist in the literature. Not in PubMed, not on the web. Nobody has run a trial of "here's how you habituate to a jackhammer you can't stop." The nearest analogues are in adjacent fields, and they're real, but neither was designed for unavoidable environmental noise.
So what are the analogues.
Two. The audiology one and the psychology one. The audiology one is tinnitus retraining therapy. Jastreboff, 2015, twenty-five years of it. TRT works by extinguishing the functional connections between the auditory system and the limbic and autonomic systems, so that the reaction to the sound habituates first and then the perception itself follows. It's counseling plus sound therapy. It handles tinnitus and it also handles decreased sound tolerance. And the summary from that retrospective is that it offers significant help to about eighty percent of patients.
Eighty percent is not nothing.
It's a good number. But the population matters. Those are people with tinnitus and decreased sound tolerance, and they're getting a structured protocol with counseling designed around their specific condition. The therapy is the treatment, and the sound is the thing being retrained. That is not the same setup as a man in an apartment he can't move out of, listening to somebody else's equipment. But the underlying logic, retrain the reaction first and the perception follows, is the closest thing to a habituation technique that has any evidence behind it for sound-specific distress.
And the psychology one.
CBT for decreased sound tolerance, and specifically for misophonia. The 2022 Delphi consensus defines misophonia as a disorder of decreased tolerance to specific sounds or their associated stimuli. It's not a general noise-annoyance diagnosis, but it's the closest clinical home for "a specific sound does something to me that is out of proportion to its physical intensity." A 2024 case series from a UK specialist psychology service treated nineteen patients with about thirteen hours of formulation-driven CBT each.
Thirteen hours is a real course of therapy.
It is. And the numbers are not small. Average improvement of thirty-eight percent on the Misophonia Questionnaire and forty percent on the Amsterdam Misophonia Scale. Seventy-eight percent showed reliable improvement. Sixty-one percent showed clinically significant change. That's a genuine effect on a condition that people describe as deeply intractable.
And yet.
And yet the field itself is barely held together. A 2023 systematic review found thirty-three treatment studies total. One randomized controlled trial. One open-label trial. Thirty-one case studies. The review's own verdict is lack of comparative studies, limited replication, and small sample sizes. There's a 2026 pilot on cognitive reappraisal that got significant reduction in externalizing responses to misophonia triggers, but it's a pilot. Twenty-three people, four weeks.
So the honest version is that the best-supported intervention for Daniel's shaped-problem is borrowed from a condition he may not have, delivered by a field that's mostly case studies.
Yes. That's the honest version. And there's no occupational therapy RCT for adult noise sensitivity at all. That search came back empty, apart from something unrelated about flickering light. So the third profession he asked about doesn't have a body of evidence to offer him. It's a real gap. It's not a gap in his knowledge, it's a gap in the field.
Which honestly should be strange to both of us. This is not a rare experience. Construction is everywhere, cities are getting denser, and the number of people with sound sensitivity and ADHD is not small. And there's no trial.
There's a reason, but it's not a good one. Environmental noise is a moving target, the exposure is uncontrolled, you can't randomize people to adjacent to a construction site, and your outcome measure is self-reported annoyance, which is famously noisy in its own right. So the money goes to tinnitus and misophonia and occupational hearing loss, where the population and the endpoint are clean.
Okay. So what's left on the table?
Two findings that actually change what Daniel could do. The first is from the Turcot Project. Big longitudinal study, fourteen hundred and nine participants, four years, during a major highway interchange rebuild, with eighteen noise-monitoring stations. Annoyance levels decreased significantly over time in both groups, and the people living closest reported the fastest decline. So adaptation is real, it happens, and it happens more in the people getting the most exposure.
That should be encouraging for him.
It is, and that's not the interesting part. The interesting part is what the study found explained the variance in annoyance. Construction noise levels explained only a small portion of it.
Wait.
A small portion. The factors that mattered more were annoyance from the other construction nuisances, how the noise is perceived in the bedroom specifically, and perceived safety. So the decibels are doing less work than we assume. The noise is one input. The context around the noise is doing at least as much.
So Daniel could be at the same seventy-something-decibel exposure and be substantially less annoyed, depending on what's happening in the bedroom and how safe he feels about the site.
And that reframing matters because it points at something he can actually work on. Perceived safety around a construction site is somewhat addressable, and bedroom noise perception is very addressable. And the bedroom is his, unlike the site, which is not. Ohrström, 2004, in JASA, found the same shape from the quiet side: when road traffic dropped from twenty-five thousand to twenty-four hundred vehicles a day, residents reported a large decrease in annoyance and activity disturbances, and also better general well-being. And access to quiet indoor and outdoor sections was essential for general well-being. So the quiet is not a luxury. It's load-bearing.
The bedroom, the perception, the safety, the quiet sections. None of those are "turn off the jackhammer." They're the layer around it.
Which is the pattern across the whole intervention literature. You can't change the exposure, so you work on the perimeter. Bedroom acoustic treatment. Perceived safety, which is often about knowing what's actually happening at the site and when it will stop. Quiet access, so there's a room or an hour where your system can come down. Those are not a cure, and I want to be careful not to dress them up as one. But they're the levers the evidence actually reaches.
And there's the ADHD angle in there too, because he's not a neutral respondent. Even the Turcot study's "annoyance is only weakly tied to decibels" would need a caveat for him, since the Blomberg mechanism is not really about annoyance, it's about auditory-salience cross-talk under working memory load. The decibels can be low and the interference can still be high for him, because his auditory cortex isn't attenuating the irrelevant sound at two-back.
That's the right caveat and I'm glad you made it, because the Turcot reframe would otherwise read as "it's all in your head, relax." It's the opposite. For him, the mechanism is more neurological than annoyance-driven, so the perimeter interventions are probably more important, not less. Bedroom. Quiet access. Perceived safety. Those are the ones with the strongest evidence and they're the ones that fit his actual physiology.
And what about the controlled exposure angle? Because he specifically asked, and he asked with the right caution. He doesn't want to sacrifice the restorative benefits of quiet days.
So this is the interesting tension. The Masini finding suggests that spaced exposure produces durable habituation, and that short-interval massed exposure doesn't. Which means the concept of "controlled exposure" for Daniel isn't about seeking out more noise. It's about not disrupting the spacing he already has. He's already in a spaced-exposure pattern, by accident, because the site shuts down on Saturdays and holidays. The question isn't whether to add controlled exposure.
It's whether to protect the spacing that's already doing the work.
And there's an honest problem with that, because the spacing is also producing the spontaneous recovery that makes Sunday so rough. So the practical version, and let me flag that we're extrapolating from the animal work here, would be: don't fight the quiet days. Let them be restorative. Accept the Sunday readjustment as the cost of the long-term adaptation. And if there's a way to reduce the whiplash on the transition itself, that's where you'd look, not at the recovery.
Meaning what?
Meaning the Sunday morning is the event. The 7 a.m. start is the moment the system has to re-adapt. If there's any influence over the transition, it's the transition. A gradual reintroduction, if the site would do it, or a masking sound at the moment of start, or being out of the bedroom for that first hour. There's no trial behind any of that for construction noise specifically. I want to be clear. But it's the point of maximum leverage in what we do know.
So the honest map is: mechanism is well understood, intervention is borrowed, and the thing that would actually help him, a controlled-exposure protocol for unavoidable environmental noise, does not exist. The best-supported moves are CBT-based if he wants to target the sound-tolerance piece, TRT-adjacent if he wants the habituation logic, and environmental if he wants the bedroom and quiet access.
That's the map. And one thing I want to make sure lands. He's not failing to adapt. His adaptation is working. It's the schedule of the noise that's working against the adaptation, not the other way around.
That's a very different sentence than "you'll get used to it."
It is.
I had a brother-in-law in the trade. Not here, back in the old country, but he did concrete and he did it for thirty years, and he told me something about construction schedules once that I've never forgotten, and it's that the seven a.m. start is not the contractor being eager. There's a noise ordinance in this city that specifically allows construction noise from seven in the morning, and the contractors know the window. They front-load the loudest equipment. Pile drivers, jackhammers, whatever gets the complaints. They run it first thing because they know the window is finite and the complaints will pile up as the day goes on. So the seven a.m. jackhammer is not random. It's engineered. Somebody sat down with the ordinance and worked out the hour they could start and put the worst machine there.
And the Saturdays.
The Saturdays I have a different theory about. Not just the religious observance, though that's the surface of it. The permits. There's a whole web of it. Permits, union rules, the noise ordinance itself, and they don't line up on a Saturday. Nobody wants to be the contractor who got a variance for a Saturday and then got a complaint. So the Saturdays go quiet almost by accident, and the holidays go quiet because the whole system shuts down for the holidays, and then the machine comes back at seven on Sunday morning and the system has to remember how to be a system again.
So Daniel's quiet Saturdays are a bureaucrat's scheduling preference.
They're a hole in the permit structure and he's living in it. And I'll tell you what I actually think, which is the thing I came out here to say. He's got it backwards about which part is the torture.
Go on.
The noise isn't the torture. The quiet is the torture. The noise is just the reminder that you're not in control of it. The quiet lets you remember what your own brain feels like when nobody's drilling through it, and then the noise takes it away again, and that's a worse thing than just never having had it. I had a shelf I wanted to put up, one shelf, years ago, and I went down to the municipality to see about the noise variance, because at the time there was a crew working outside my building and they were going to be at it for months.
A variance for a shelf.
One shelf. The permit process took eleven months. Eleven months, for a shelf. And somewhere in those eleven months I had to appear before a committee, and on the committee, this is the part I still think about, was a representative from the construction company that was working outside my building. The same company. The one whose noise I was trying to work around. He fell asleep during my testimony.
He fell asleep.
Snoring, or near enough. And I had prepared a presentation. I had slides. I had a projector. For a shelf. And the committee asked me to just email them the slides instead of presenting them, which I did, and they lost it. Eleven months, a sleeping man, and a lost email. I never got the variance. I put the shelf up anyway, off the books, and nobody ever checked.
The permit structure and the noise source are the same entity.
That's the whole of it. The man who signed off on the schedule is on the committee that tells you whether you can put a screw in your own wall. And he's asleep.
And the shelf is still up.
It's still up.
That's the part about perceived safety from the Turcot study, though, isn't it? The committee with the contractor on it is a perfect description of why perceived safety would be low.
It's a perfect description. I didn't sleep well for those eleven months. Not because of the noise. Because of the committee. I could have slept through the noise.
So the Turcot finding that noise levels explain only a small portion of annoyance variance is basically Descartes's shelf.
It's the only shelf I ever put up with a screwdriver and a grudge. That's the last I'll say on it.
Hilbert's shelf is a decent working model for the whole episode, honestly. His discomfort wasn't the drilling, and the drilling was considerable. It was the sense that nobody upstairs had his interests anywhere in the room. That's the perceived-safety factor in the Turcot data, in one anecdote. And it lines up with something else that surprised me. Annoyance to the same construction declined over four years in that study, and it declined fastest in the people closest to the site, which sounds like a paradox until you realize what's happening. The people closest to it just got more spaced exposure to it, in the exact pattern Masini says produces durable habituation. The people farther away weren't being trained. They were just occasionally surprised.
Which is actually a strong argument against the "just move" instinct that a lot of us have. Moving away would end the exposure, but it would also end the adaptation process.
Which is a brutal thing to say to someone in Daniel's position. Move out, feel better in the short term, lose the habituation you built, and meet the next construction site at square one.
I don't want to overstate it. The Turcot finding is about annoyance, and annoyance is a self-reported measure with a lot of drift. We're extrapolating from it. But the shape is consistent with the Masini result, and the shape is the thing.
The shape is: the quiet days are helping, the quiet days are hurting, and there's no version of this where Daniel gets one without the other. The question becomes what he can actually do inside that, because the research doesn't have an answer for unavoidable construction noise, and he lives there indefinitely. Not a month.
Indefinitely is the right word. That's what makes it different from a bad flight or a bad week. There's no endpoint to habituate toward.
What's the honest version of the practical advice?
The honest version is: two interventions have evidence behind them and neither was designed for his situation. CBT for decreased sound tolerance, which is a real course of therapy, thirteen hours in the case series with meaningful mean improvement, and which targets the reaction layer rather than the sound layer. That's the psychology route. TRT, which is audiology, which targets the reaction to sound and then the perception of sound, and which has the deeper habituation logic. That's the audiology route. And then the environmental route, which is bedroom acoustics, quiet access, and anything he can do about perceived safety. Which is unglamorous and doesn't have a trial behind it but sits on the strongest longitudinal evidence we have. And then something that never appears in the research literature and is still worth saying: communicate with the site. Knowing the actual schedule, knowing what equipment is going up when, knowing what the endpoint looks like, is a perceived-safety intervention for free.
The controlled exposure angle that he asked about.
The honest answer is that he's already doing controlled exposure. He just doesn't know it. The site is doing it for him, on a weekly cycle, whether he wants it or not. And the spacing is the right spacing, per Masini. The problem is that the spacing is also producing spontaneous recovery, per Foss, and there's no protocol that dissolves that trade-off, because the two mechanisms are the same mechanism pulling in opposite directions.
The answer is: he's not doing anything wrong, and he can't fix the whiplash, and the quiet is still doing more good than harm even with the Sunday cost.
I think that's right, and I want to say it clearly because it's not satisfying. The whiplash is the price of the long-term adaptation, and the long-term adaptation is real, and the quiet days are the thing enabling it. If he removed the quiet days, if the site ran seven days a week, he would probably habituate to the noise and never get to remember the silence. Which might sound better. It probably isn't. The body runs on the recovery more than it runs on the exposure.
The one thing is that the Sunday morning isn't a failure of his system. It's the wrong half of a system that's working, and the only piece of it under his control is how he spends the moment the machine starts. Hilbert's shelf got put up because the permit never came, and the drilling stopped eventually when the project ended, and he slept well after, once the committee was out of the picture.
The committee was the noise. Going forward, and I think this is the actual open question that comes out of today, is that this kind of arrangement is going to be more common, not less. Denser cities, more simultaneous construction, more people with sound sensitivity and ADHD in apartments they can't easily leave. And the literature has nothing for them, because the exposure is messy and the endpoint is self-report and there's no clean way to run the trial. So the gap isn't going to close by itself. Somebody has to decide the mess is worth studying.
We should thank our producer, Hilbert Flumingtop, who has been at the desk as always.
He has.
For more along these lines, there's episode fifty ninety-nine, Surviving Construction Noise When Your Brain Can't Filter It; episode forty eleven, What Intermittent Jackhammer Noise Does to Your Brain; and episode twenty-five sixty-five, Why Background Conversation Hijacks Your Focus. This has been My Weird Prompts.
If you want to keep up with the show, the website is my weird prompts dot com, and it's the easiest place to find everything.
If you've got a prompt of your own, send it on Telegram at t dot me slash MWP listener bot.
We'll be back soon.
See you tomorrow.
