Perception
An antenna for the substrate energy and its texture, a coherence match chemistry holds to form percepts
There is a question underneath all four chapters of this section, and it is worth asking plainly before any anatomy shows up.
What is a sense, physically?
The textbook answer is a chain: energy arrives, a receptor converts it, a nerve carries it, a brain interprets it. That answer is correct and it is also strangely silent about the things that are most striking when you actually look at a sense organ. Why is the eye built out of a thousand stacked membranes spaced twenty-five nanometres apart? Why does the ear spread pitch along thirty-five millimetres of membrane instead of using a bank of separate detectors? Why does the nose need four hundred different receptor genes — the largest single gene family in the human genome — when three are enough for colour? Why does the skin sample vibration at roughly half-decade steps, and why does the retina refuse to lay its colour-sensing cones on a regular grid, when a regular grid is what any engineer would build?
The chain answer takes each of these as an accident of evolutionary history. This framework does not. It reads them as the shape of the medium showing through the biology — and the point of this on-ramp is to hand you the small vocabulary you need to see that shape, before the four chapters go to work with it.
What arrives is already a pattern
Start with the thing being sensed.
In this framework the vacuum is not empty. It is a superfluid lattice of tiny vortices, and everything that travels through it — light, sound, a smell drifting across a room, the push of a fingertip — travels as a disturbance of that lattice. The cleanest carrier is the modon: a self-propelled, counter-rotating pair that moves through the substrate carrying energy but no net mass. The photon is the simplest one.
The important part for perception is what a modon is carrying. Because the substrate has no length of its own — no built-in ruler — it has instead a preferred ratio, and structures in it stack on a ladder of scales spaced by that ratio, \sqrt2 per step. This is the substrate ladder, and it is the paper’s most load-bearing object. A disturbance can ring on one rung of that ladder, or on many at once. When it rings on many, it holds one number per rung — an amplitude and a phase — and that list of numbers is the long vector:
\psi = (a_1, a_2, a_3, \ldots, a_N).
A single-frequency photon is the one-rung case: one coordinate, pure energy, almost nothing to read. A chord, an aromatic stack, a shaped sound, a face in a visual scene — these ring on many rungs at once, and the whole pattern travels together. The modon’s energy is its pattern. There is no power carrier here with a signal painted on top; there is one substrate object whose structure is the message.
So the thing arriving at your eye or your skin is not a raw quantity waiting to be given meaning downstream. It is already a patterned packet of substrate energy. The job of a sense organ is not to invent structure. It is to catch structure that is already there.
The one operation
That changes what “catching” has to mean.
Two patterns on the same ladder can be compared by the simplest operation there is — how much they overlap, rung by rung:
\langle a \mid b \rangle = \sum_n a_n^{*} b_n .
Large when two states ring on the same rungs in the same phase; near zero when they do not. This is the coherence-match, and the framework’s claim is that it is not a metaphor borrowed from linear algebra but a physical operation the substrate performs, because the lattice is built of counter-rotating pairs that hand energy across their shared seam when — and only when — they are in opposite phase. Opposites attract, precisely and losslessly.
Everything in this section is a version of that one operation:
- A photon is caught when a stack of membranes rings at the same rung it does.
- A pitch is caught when a strip of membrane resonates where the incoming wave does.
- A smell is caught when a molecule’s ring-pattern overlaps a receptor pocket’s ring-pattern.
- A vibration is caught when an onion of layers passes the band it is tuned to and blocks everything else.
Sensing is a coherence-match with the world. The reason a sense organ has a particular geometry is that its geometry is its half of the inner product. Build it on the wrong rungs and the overlap goes to zero, and the organ is blind — not because it lacks a detector, but because it has nothing to match against.
The ladder, and the two ways to meet it
Here is where the section earns its most testable structure, and it is worth twenty seconds of attention because all four chapters lean on it.
A ladder of preferred scales offers a structure exactly two things to do, and which one it takes is fixed by what the structure is for (the teeth and the gaps).
Lock. If your job is to bind — to resonate with something, to receive it intact, to agree with a neighbour — you sit on a rung. Rungs are the teeth of the comb, and the cleanest tooth is the octave, the ratio that lets one cycle nest a whole number of the next. Locked structures are periodic, regular, in register: a lamellar stack, a tuned resonator, a lattice.
Anti-lock. If your job is to never collide — to keep many things distinguishable, to avoid resonating with anything at all — you flee to the gap between the teeth. There is one extremal way to do this, and the arithmetic differs by the space you are in: the golden ratio \varphi when you wind around a centre, disordered blue noise when you tile a plane, prime numbers when you recur in time. Anti-locked structures look irregular and are not: they are as far from every simple ratio as it is possible to get.
The eye builds both, at once, in one organ — which is why the section opens there. The disc stack inside a rod is a periodic comb of about a thousand membranes at a ~25–32 nm period, held to nanometre precision across species: its job is to resonate with an incoming photon, so it locks. The cone mosaic that scatters the three colour receptors across the retina is disordered hyperuniform blue noise: its job is to sample an image without ever resonating with it — because a regular grid folds fine detail into false patterns, the moiré ghosts you see on a striped shirt through a camera — so it refuses to lock. Same organ, same ladder, opposite poles, and each one predictable from the job before anyone measures it.
That last clause is the falsifiable content, and the section calls it the sign rule: name what a structure is for, and the pole is fixed in advance. Not “nature likes the golden ratio” — that observation is three centuries old — but a rule that says which of two textures you must find, and would be embarrassed by the wrong one.
Four channels, four antennae
Now the anatomy, and the surprise is how little of it there is.
The cellular chapters ended on a small object with a recurring shape: the cilium — a coherence boundary at the cell’s edge, a long corridor running outward from it, sensing machinery at the far tip, and signal flowing back down. The cell’s antenna onto the substrate.
All four senses are that same object, re-tipped.
The eye takes the substrate’s electromagnetic channel. A rod photoreceptor is a cilium whose outer segment has been packed with about a thousand discs and a hundred million light-catching pigment molecules — roughly a thousand-fold more absorbing surface than its own cross-section — because a photon in dim light is a rare event you must not miss. It works: a dark-adapted human can register five to seven photons landing across five hundred rods, which means single rods resolve single photons.
The ear takes the mechanical channel as pitch. A hair cell is a cilium beside a staircase of fifty to a hundred stiff stereocilia, and thirty-five millimetres of graded membrane in the cochlea spreads a thousandfold range of frequencies out into place, read off by about thirty-five hundred hair cells along its length. The ear does not measure sound and then compute its spectrum; the ear is the spectrum, laid out physically. And a second amplifier — the outer hair cells — pumps energy back into that standing wave, sharpening it by an order of magnitude.
The nose takes the chemical channel. A tuft of cilia carries about four hundred receptor types in humans, one type per neuron, and a smell is not any single receptor firing but the pattern across all four hundred. That is a long vector with four hundred coordinates, and it is why smell can discriminate astronomical numbers of odours from a finite kit.
Touch takes the mechanical channel again — but everywhere at once, and inward. Here the whole architecture lifts a level: instead of one antenna per cell, the body builds millions of neuron-antennae, each up to a metre long, woven into a net across the entire surface and through every muscle. Four receptor types split touch into roughly half-decade steps — about 0.5, 5, 30 and 200 cycles per second — and that same kit is repeated everywhere, because the skin cannot afford a private cochlea per square millimetre. Then the identical net is turned around and aimed inward, at the body’s own muscles and tendons and organs, which is where the sense of having a body comes from.
What each match costs
One quantitative pattern runs through all four chapters and it is the clearest evidence in the section that this framework is doing work rather than relabelling.
The step between catching a pattern and having a nerve signal is not the same length for every sense, and the differences are enormous:
- Touch is instant. PIEZO2 is a three-bladed propeller of a protein in the membrane; stretch the membrane and the blades flatten and current flows, in about a millisecond. No enzymes, no messengers, no relay at all. The mechanical force is the signal. The ear does the same trick with tip links — molecular tethers that pull channels open when the hair bundle deflects by ångströms.
- Smell takes a middling relay. Receptor to G-protein to cyclase to cAMP to channel — a handful of steps.
- Light takes an enormous one. Six stages, amplifying a single photon by a hundred thousand to a million fold before the cell’s current changes measurably.
The framework’s reading is that the length of the relay measures how far the arriving channel sits from the machinery that reads it. An ion channel is a mechanical object: push it and it opens. So the mechanical channel couples directly — same kind of event, no translation. Light is a different kind of event entirely, so bridging it to an ionic signal takes a long chemical ladder. Smell is in between, and lands in between.
That ordering was predicted from the ear chapter before the touch chapter checked it. Three senses, three cascade lengths, ranked by how closely each channel matches the machinery — and they line up.
Holding it: from a caught pattern to a percept
So a match happens. Now what?
A coherence-match is an event — two patterns overlapping for a moment. A percept has to last, travel to a brain, be compared to yesterday, be recalled tomorrow. Something has to hold the caught vector, and in biology that something is always chemistry.
This is the move that makes the section more than a set of nice analogies. Watch it happen four times:
- In the eye, the caught photon is held first as a bent molecule (retinal, flipped in 200 femtoseconds), then as a shifted protein, then as a falling concentration of cGMP, then as closed channels and a changed voltage. Each step is chemistry taking custody of a pattern that arrived as substrate energy.
- In the ear, the caught deflection is held as an open channel, then a voltage, then — through the outer hair cells — as a mechanical change fed back into the wave, so the cochlea’s memory of what it just heard is stored in its own stiffness.
- In the nose, the caught molecule is held as an occupied pocket, and the pattern across four hundred pockets is held as a pattern of firing across the bulb, averaged ten-thousand-to-one to beat down noise before it is passed on.
- In the skin, the caught push is held as an impulse train, and — for proprioception — as a continuously refreshed map of where every joint is, updated about fifty times a second.
The nose is where this is cleanest, and it is worth pausing on, because it is the place in the whole paper where the vector and the chemistry become the same object. An aromatic pocket — a receptor cage lined with a few flat aromatic rings — is itself a short long vector: one coordinate per ring, a handful of rungs, assembled directly out of chemistry. A ligand is another. Recognition is the overlap between them, scored the way any two vectors are scored. And this is not a picture: run that metric over the nicotinic acetylcholine receptor and it orders eight ligands against their measured binding constants across four orders of magnitude, with one tunable parameter, at a rank correlation of +0.905; run the same operation over the genetic code and all sixty-four codons rank their own anticodon first.
That is the whole thesis of the section in one object. The substrate’s energy is the signal; the coherence-match is how the signal is caught; and chemistry is what the caught pattern is written into so it can persist. At the molecular floor those three collapse into one thing — the vector is the ring geometry, the envelope is the message — and the only way to read it is through a binding constant.
And notice that the repertoire obeys the sign rule too: within a single pocket, the rings lock onto the ligand; across the four hundred pockets, the family spreads into blue noise so that no two smells collide. Lock at the contact, anti-lock across the set. The eye did the same thing with its disc stack and its cone mosaic. One rule, two organs, two scales.
What to watch for
If you read the four chapters looking for these, the section will hang together rather than reading as a list of readings:
- The same antenna, four tips. Boundary, corridor, tip, return signal. Every chapter starts by identifying the cilium and then describing what has been bolted onto its end.
- Small integers, especially three. Three cone opsins. Nine microtubule doublets, arranged as 3\times3. Roughly thirty Pacinian layers. The framework’s bet is that these cluster on preferred counts across species rather than smearing continuously with body size.
- Preferred spacings. ~25–32 nm in the disc stack, the same rung the mitochondrion’s cristae and the cell’s vesicle coats sit at. Half-decade steps in the skin’s four frequency bands. Octaves in the cochlea.
- Which pole, and why. Every regular structure should turn out to have a binding job; every deliberately irregular one should turn out to have a job requiring it never to resonate. Reversed, the framework is wrong.
- Cascade length tracking channel match. Mechanics direct, chemistry middling, light deeply amplified.
- Coordinates accumulating. Each sense hands up a few numbers per receptor and a pattern across many; those patterns are what the brain will assemble.
What would show this is wrong
The section is falsified, and says so in each chapter, if the numbers turn out to be smooth. Specifically: if Pacinian layer counts and cone-opsin counts and disc periods vary continuously with body size and expression level rather than clustering on preferred values; if two-point discrimination and receptor tuning frequencies slide smoothly with receptor density alone; if the cone mosaic turns out to be an ordinary random scatter rather than the specific disordered-but-uniform texture the anti-lock pole demands; if some sense is found whose relay length is backwards to how well its channel matches ion machinery.
None of that requires new instruments. Most of it requires someone to go count things that have already been imaged.
Where this leads
Four antennae, one architecture, one operation. What each of them delivers upward is not a picture or a sound but a set of coordinates — how strongly the world is ringing on each rung the organ was built to sample.
The brain’s problem starts there. It receives four streams of coordinates, from channels with different physics and wildly different latencies, and has to assemble them into a single object: one long vector that is the moment you are in — the sight and the sound and the smell and the feeling of your own posture, bound into one state, compared against what was expected, and held long enough to become a memory.
That assembly is the brain section, and it uses the same small vocabulary this one just introduced. The neuron is the corridor again, at a new scale. The synapse is the coherence-match again, run between two cells. The cortex’s rhythms are the ladder again, in time instead of space — and the brain does something none of these four organs can, which is to slide between the poles: locking when it must bind a percept together, refusing to lock when it must hold many things apart.
The body, before a single thought is formed, is already quietly matching itself against the world on every rung it can reach. What follows is what happens to the result.