Hannah spends her days designing buildings people walk through and notice. Daniel spends his days designing systems nobody sees at all, and the two of them have been arguing about what a room should look like for about as long as they've been married.
Which is a great place to start, because that argument is not just a marriage thing. It's the whole field in miniature.
Daniel sent us a prompt — a long one, in text, and I'll get to it — but here's the shape of it. His wife Hannah is an architect, he's a technologist, and he'd noticed something he thinks the rest of us overlook: that architecture is one of the few professions that straddles two completely different worlds. There's the version everybody knows, the visual one, buildings designed for how they look. And then there's the version we use in tech when we say we're architecting a system — designing for how a thing actually functions.
And he thinks the ambiguity there is a feature, not a bug.
Right, and he's explicit about it. There's no single correct way to build a building, and there's no single correct way to build a complex agentic pipeline, and he finds that liberating rather than frustrating. But where he and Hannah differ is aesthetics. Daniel is squarely in the minimalist utilitarian camp — he's come at a laptop with a spudger to get a manufacturer sticker off it, because the sticker served no function.
That's the most Daniel sentence in the whole prompt.
And he admits it. He says he might be at the pretty extreme end. But even he concedes a bare white wall is a little stark and depressing, and he thinks there are commercial and industrial designers who manage to blend functionality with aesthetics rather than choosing. Lighting is on his list too — how light affects mood and circadian rhythm, which he figures architects are going to be forced to engage with. And then two actual questions. One: give us an overview of the thinkers, architects, and currents in the movement that specifically design space for ergonomics and organization. Two: is an ergonomically-informed architect actually needed, over and above an ergonomist? And is any of this still in its inception, or is it already gaining traction?
Two questions, and the second one turns out to be harder than it sounds.
So let's start with the word itself. What does it actually mean to architect something?
The building sense is the original one — designing a physical structure for people to occupy. The technical sense is borrowed, and it's borrowed because it's apt. When you architect a system, you're deciding where the load-bearing walls go. What's foundational. What can be moved later and what can't. What talks to what.
And in both cases there's no canonical answer, which is the thing that drives engineers crazy.
Crazy. I was looking at a thread this week where people were arguing about agentic coding, and the phrases that kept coming up were — and I'm paraphrasing — that agents need architectural guidance or they'll turn your codebase to mush, that the architecture is the thing that matters, not the objective function. And the same thread had someone saying there's not really a workable architecture for a real app in what they were discussing. So you've got the word being used as a tool, as a boundary, and as a dismissal, all in one conversation.
That's the word doing three jobs and none of them precisely.
And that looseness is exactly what Daniel's pointing at. There's no one right way to design an agentic pipeline, which is why the arguments never resolve — the arguments are about taste and constraints, not about a correct answer that somebody's hiding.
But here's the tension I want to sit in for the rest of this. When most people hear architecture, they think aesthetics. The facade, the silhouette, whether the building is beautiful. And there's a whole other dimension that's about whether the building actually works — whether the light is right, whether you can find your way around, whether your body is comfortable in it, whether your brain is.
And the question Daniel is really asking is whether that second dimension has a movement behind it. A named strand. People he could go read. Practitioners he could hire.
Which needs to be answered, because it does exist.
It exists and it has a name. Neuroarchitecture. It's an established interdisciplinary field — it's been growing since roughly two thousand, and it integrates neuroscience and architecture to study how built environments shape behavior, emotion, and cognition. It sits at the intersection of what the brain does and what the room does to it.
So Daniel's instinct wasn't wrong. There is a current.
There's a whole current. And a few figures worth knowing. Juhani Pallasmaa is the Finnish architect — his line is that buildings internalize the world and externalize the mind. Fred Gage at the Salk Institute is the one who puts it bluntly: changes in the environment change the brain, and therefore they change our behavior. Sarah Robinson argues there's a neuroscience revolution in architecture that's comparable to Galileo or Darwin in scale, which is a big claim, but it tells you the ambition. Nancy Kanwisher identified the parahippocampal place area — the part of the brain that lights up for rooms with furniture, for scenery, for city streets, more than for objects or faces. John O'Keefe won a Nobel for place cells. Isabelle Sjövall is the Swedish neurodesigner who's optimized property portfolios worth about one point two billion pounds using these principles.
One point two billion. That's not a research lab, that's a market.
And there's institutional scaffolding underneath it. There's the Academy of Neuroscience for Architecture. There's a dedicated neuroarchitecture lab at the Universitat Politècnica de València. There's an organization called NeuroArchi whose framing is that we spend ninety percent of our time in buildings that were designed without a deep understanding of the people inside them.
It's the mission statement of the whole field. And it's roughly accurate. Ninety percent of your life is inside a structure somebody drew.
So take the mechanism seriously for a second, because I don't think this is decoration. What's actually happening when a well-organized room is well-organized?
The parahippocampal place area is the cleanest example. It fires more for organized spatial scenes than for individual objects. Your brain has a dedicated system for processing a room as a room — the layout, the boundaries, the arrangement — rather than as a pile of things.
So an unorganized room isn't just annoying, it's harder to represent internally.
That's the implication. Your brain has to do more work to build a stable picture of a space that keeps shifting. And the flip side is wayfinding. Landmark cues, distinct colors, lighting that shifts appropriately, minimizing decision points — those aren't aesthetic choices, they're load-reduction. They're the spatial equivalent of a good interface.
And there's a specific one there that maps directly onto Daniel. Minimizing decision points.
Every doorway where you have to stop and think which way to go costs you something. The field treats that as a budget, and design either spends it well or wastes it.
Now, the anterior cingulate cortex — Herman, you mentioned people walking through spaces with curvature.
The ACC activates significantly when people move through built environments, and especially with curved forms. That's a stress and conflict-monitoring region. So how a space is shaped is being tracked by a system that also handles error detection and emotional regulation.
Which is a slightly uncomfortable thought if you've ever been in a building that felt wrong before you could explain why.
And then there's the mirror neuron work, which is where the third-skin metaphor comes from. Skin is the first. Clothing is the second. The building is the third. And Reggio Emilia has the version I like best — the environment is the third teacher.
Say that one again.
The environment is the third teacher. Their claim is that after the parent and the teacher, the room itself is doing pedagogy. The layout is teaching.
That's the bridge to Daniel's actual thesis, and I want to flag it now, because it's the thing that makes the rest of this episode interesting. If the environment is the third teacher, then designing for organization is designing for cognition. You're not just storing objects, you're storing cognitive capacity outside the skull.
That's intention offloading, and it's a real research construct. Using calendars, leaving objects strategically placed, phone alerts. It's a documented strategy, not a personal quirk.
Now here's where I want to bring in the aesthetics argument, because Daniel wants ammunition and I'm happy to hand it to him, but the field complicates it.
Give me the complication.
Bruce Mau's critique is that we've let two sensory domains, sight and sound, dominate our design imagination, and that we create almost exclusively for one sense. The visual. So the man who wants the sticker off his laptop has a serious designer making his argument for him.
But.
But the neuroarchitecture people are not arguing for barren. Pleasant places activate the brain more than unpleasant ones. So the bare white wall Daniel admits is depressing isn't a neutral default that minimizes distraction. It might be actively working against him.
That's the tension, and I want to defend the white wall for one second before we let it die.
Go ahead.
The argument for the bare wall isn't that it's pleasant. It's that it's low-information. If your problem is that you can't hold a task in your head, a room with fewer competing signals is a room with fewer things competing for the same resources. The mistake isn't the white wall itself. The mistake is assuming the white wall is free.
Say more.
If a room is barren, you've traded stimulus for clarity. That's a real trade, and sometimes it wins. But it's not a free choice, and if you never think about the cost, you end up with a room that's calm and slightly hostile at the same time.
Which is Daniel's own admission. He says a bare white wall is a little bit stark and depressing. He's already feeling the cost.
And the research backs the feeling. The field's whole position is that environments are not passive — they're actively modifying what the brain does. Fred Gage's line again. The building changes the brain.
Now, Daniel asked specifically about ADHD and working memory. What are the numbers?
Sixty-two percent of adults with ADHD report significant working memory impairment. That's Barkley. And the comparison point is that neurotypical working memory holds somewhere in the range of five to seven items at a time, and ADHD working memory caps out lower.
Five to seven items. That's the entire buffer.
That's the desk you're working on. Everything else is off the desk or it's gone.
And this is why the organizing systems matter in Daniel's life. Not for tidiness. For the desk.
Right. Externalizing is the strategy. Every object that has a fixed home is one less thing riding in working memory. Every system that catches a task for you is one less thing you have to hold.
And I want to note the emotional piece, because it doesn't get enough attention. There was a study of ADHD adults and task management, twenty-two interviews, and the finding was that the struggle isn't willpower. It's an emotional and relational misalignment between how the person's cognition works and the structures they're expected to fit into.
That's the part where architecture comes in. If the built environment is one of those structures, then it's doing some of the misalignment.
So let's talk about whether anyone is designing for this. Here's what I want from the rest of this conversation. Daniel asks whether an ergonomically-informed architect is needed — over and above an ergonomist. And whether it's inception or traction. Those are two different questions and I think they have two different answers.
The mechanism part is that the knowledge exists but it's fragmented. The neuroarchitecture field is real, the institutions are real, the commercial practitioners are real. But the specific strand Daniel is pointing at, ergonomics plus organization, doesn't have a named standard the way it should.
Let's do the traction question first, because I think there's a cleaner answer there.
There is. Ieva Honkanen Architecture sells what it calls neuro-spatial architecture as a productized service. Remote spatial audits start around thirteen hundred euros. On-site biometric audits, using heart-rate variability and galvanic skin response wearables plus lux mapping, run three and a half to six thousand. Full design work runs three hundred to four hundred fifty euros per square meter.
So there's a price list. That's the traction answer.
There's a price list and a positioning statement. Honkanen's line is that traditional architecture is aesthetic and neuro-spatial architecture is biological. And there's another one that's better: architecture is a neurological language, every wall whispers to your biology.
That's either profound or a marketing line, and I haven't decided which.
It's both, and that's fine. But it's evidence the field is past inception. Somebody is charging real money for this. And the framing, human capital return, is the part I'd expect architects to be skeptical of, because it turns a design philosophy into an HR pitch.
So traction, yes. Now the harder question. Is an ergonomically-informed architect actually needed over and above an ergonomist?
This is where the honest answer is that the role Daniel is describing doesn't exist yet as a formal thing.
No credential, no named movement.
Not as a distinct professional category. The closest real thing is neuroarchitecture and neurodesign, which is well-established but framed around neuroscience and wellbeing broadly. Not specifically ergonomics plus organization, and definitely not specifically ADHD.
And the ADHD design literature, for what it's worth, is about digital task-management tools. It's about apps, not buildings. There's a paper from this year about user-authored executive scaffolding, which is a nice way of saying ADHD people building their own interaction architectures, and it's all about software.
That's the gap. The ergonomics knowledge is genuine but it's siloed, and the specific application to physical design for ADHD is underserved. Daniel's question is open.
Let's make the argument for the architect over the ergonomist, because I think Daniel's right and it's worth saying why.
It's integration. An ergonomist can tell you the optimal desk height and the optimal reach envelope, and that's real expertise. But an ergonomist doesn't sit in the drawing meetings. They don't decide where the window goes, they don't decide the floor plan, they don't decide whether the deep floor plate is even a viable idea. The architect does, and the ergonomic advice arrives after a hundred decisions are already locked in.
So the ergonomist is consulted, the architect is the one who decides what the ceiling can hold.
And they're the one who carries the liability, which is why they get to draw.
There's a lighting piece here too, and Daniel flagged it, and it's the strongest case for the architect-ergonomist hybrid.
The WELL Building Standard is the codified one. Circadian lighting design: at least two hundred fifty equivalent melanopic lux at seventy-five percent or more of workstations, sustained for four or more hours a day. Evening lighting capped at fifty or under. And classrooms with under twenty-fives need one twenty-five at seventy-five percent of desks.
That's a number a lighting designer can hit, but the number is downstream of the floor plan.
Which is the point. The fluorescent lamp plus air conditioning let architects design deep floor plates with workstations nowhere near a window. That was a legitimate design decision at the time. Now it's understood as circadian-harmful, and the reason is that the deep floor plate was baked in before anyone knew to care.
And EML isn't lumens. It's weighted to a specific cell type.
ipRGCs. Intrinsically photosensitive retinal ganglion cells. That's how light gets into your circadian clock — it's a different channel than the one you see with.
And how dominant is light, compared to other environmental factors?
Light-induced melatonin suppression is over ninety percent. Electromagnetic fields are in the twenty to fifty percent range. So this is the dominant environmental lever on the circadian clock, and it's the one architects control most directly.
Ninety against fifty. Not a close call.
There's a proposal now to treat it like hygiene. Light hygiene. The argument is that irregular twenty-four-hour light patterns weaken circadian robustness and increase health hazards, so it belongs in the same category as sleep hygiene or dental care. And a doctoral thesis out of Aalborg is pushing a biological lighting protocol because current practice is still focused on visual metrics and energy codes.
Visual metrics and energy codes. So the profession is optimizing for appearance and electricity cost, and the biology is unoptimized by default.
That's exactly the thesis.
Now let's do the offloading paradox, because this is the part of Daniel's prompt I think he doesn't realize is aimed back at him.
The scarppy part.
Explain it.
The premise of offloading is that you move a memory task out of your head into the world. Calendar, list, notification, fixed location for an object. That works. The question is what happens to the capacity you moved.
The answer from recent work is uncomfortable. Scarampi and Gilbert and colleagues published work showing that offloading weakens the neural representation of the thing you were supposed to remember. The internal trace gets thinner.
There's more. A study from earlier this year showed offloading reduces relational and item-specific processing — meaning the depth of encoding drops. And there's a pair of papers from this year, one by Maier and colleagues with seven hundred four participants, showing metacognitive feedback reduces AI cognitive offloading with an odds ratio of point four seven and improves test performance with an odds ratio of one point five one. The title is designing against deskilling.
The framing is that offloading is a tool that can scaffold or atrophy, and which one depends on how you use it.
Thomas Koberda's paper on training mathematicians makes the distinction I think is most useful. Productive offloading versus premature offloading. It's not about whether you offload. It's about whether you offload the thing you were supposed to build capacity for.
Which means for Daniel, the question isn't whether to use the modular storage system. It's which capacities he wants to keep in-house.
Working memory isn't a single thing. Offloading the location of your keys is probably free. Offloading the structure of a problem you're supposed to be getting better at solving is not free.
That's the distinction I'd want to hand to Daniel. Not stop offloading, but offload the trivium and keep the thing you're training.
That's the exact bridge back to buildings. If the environment is the third teacher, then a well-organized environment is offloading to a teacher. And if the teacher is doing the work for you forever, you don't learn it, and if the teacher builds the scaffold and then steps back, you do.
The productive question isn't whether we should externalize cognition. It's when externalizing serves us and when it erodes us. And that question is identical whether you're designing a room or an agentic pipeline.
Identical, and neither field has a good answer yet.
Hilbert, you've been sitting there. What have you got?
Hilbert: I agree with almost all of that. I want to push back on one small piece, which is the idea that ergonomic design is new. There was a place I worked, a commercial kitchen outfit, and my job for about six months was drawing nothing but the clearance radius for walk-in refrigerators. Thirty-six inches for a walk-in door. Thirty inches for a reach-in. I could still draw those to scale on a napkin.
This is not new.
Hilbert: It isn't new. It's just not called neuroarchitecture. It's called kitchen workflow, or factory floor planning, and the people who did it had been doing it for a very long time. And the funny thing was — I was a junior draftsman, and the ergonomics knowledge wasn't coming from an architect. It came from equipment specialists and foodservice consultants. They knew that the whole game was throughput. How many steps does a line cook take per order. How many times does a server cross the pass. How many degrees does the walk-in door swing before it hits the prep table.
Thirty-six inches is a number somebody worked out empirically, not from a paper.
Hilbert: Somebody worked it out on their feet, yes. And you can write it on a drawing and the drawing becomes the knowledge. Anybody who reads it is making the same decision every time, whether or not they understand why.
That's the thing I hadn't thought about. The knowledge is encoded in standards and specs, not in the training.
Hilbert: Which is the gap. There were architects involved, but the ergonomics knowledge lived with the equipment people and the consultants. The architect drew the envelope and the consultants drew the function into it. Nobody had both.
The silo Daniel is pointing at is real and it's older than the neuroarchitecture movement.
Hilbert: Much older. And the piece that bothers me is that the specification outlives the reasoning. Somebody writes thirty-six inches and two decades later nobody at the firm remembers why, and they either keep the number because it's the number or they change it because it's inconvenient. I watched a project change a clearance to save a few inches, and the same people were surprised when the kitchen got slow.
That's a good way of putting it.
It's the exact thing we've been circling. The knowledge exists, it's just not in the hands that draw the room.
Hilbert: I should say — the tools they gave me weren't made for anybody with claws. Every drafting pen I ever used, the grip was wrong. You learn to adapt.
If commercial kitchens have had ergonomic specialists for a century, why doesn't the rest of architecture?
Hilbert: Because the kitchen has a number attached to it. You can count orders per hour, so the ergonomics gets paid for. A living room has no such number, so it doesn't.
That's the whole episode in one sentence.
Hilbert: That's the whole job. If you can't count it, it doesn't get designed, and if you can count it, somebody else owns it.
Misconception time. Herman, what's the wrong belief people walk away with?
That architecture is mostly about how a building looks, and that function is a professional afterthought somebody handles later.
Correct it.
The function is being decided at the same time as the look, and the functional decisions are the expensive ones. The deep floor plate, the window placement, the circulation. Once those are drawn, the ergonomics can only negotiate with what's already there.
The second wrong belief, the one aimed at Daniel?
That offloading is free. Moving a memory out of your head into the world works, but the internal trace gets thinner, and the question is whether you're offloading the trivium or the thing you're supposed to be building.
One forward thought from all of this, and it's Hilbert's point, actually. If the commercial kitchen has a countable number and the living room doesn't, then the future of ergonomically-informed architecture might depend less on whether the field gets a credential and more on whether somebody finds a number worth optimizing. If the field has a price list now — and it does, three hundred to four hundred fifty euros a square meter — that number is doing exactly the work the kitchen's orders-per-hour did.
Somebody's decided it's worth counting.
Thanks to Hilbert Flumingtop, our producer. This has been My Weird Prompts.
If you enjoyed this one, leave a review on your podcast platform of choice. It helps other listeners find the show.
We'll be back soon.
