VSA-TOE

How does a located point of view hold together? The Self-Return Framework models the Proprioceptive-Vestibular-Cortical (P-V-C) loop as the minimal embodied control system maintaining a stable perspectival address. A thread on the architecture. 🧵👇

Stuart, I posted the Self-Return Framework to Zenodo: a falsifiable proprioceptive-vestibular-cortical (PVC) architecture where the α/θ ratio is actively regulated toward φ via a dissipative circle map (not KAM), with degradation ordering derived from the return operator’s eigenvalue structure rather than assumed. Deafferentation cases (IW, GL) anchor the empirical claims, and §15 lays out a pre-registered perturbation-and-return covariance test. Scope is deliberately bracketed to Address/Occurrence. Would value your read on the vestibular-as-DC-ground framing in particular. zenodo.org/records/21909260

Mine is the Self-Return framework (PVC): a mechanistic account of how a brain keeps a stable Address and returns to it. Proprioceptive, vestibular, and cortical signals form a loop, with the α/θ ratio regulated toward φ by a dissipative circle-map attractor on a 2-torus. It makes testable predictions, including a perturbation-and-return study.

The Self-Return framework (PVC) treats the self as a process rather than a thing: a nested recursion in which a body-anchored reference repeatedly returns to itself. Its core object is the return operator ℛ, which acts on a proprioceptive-vestibular-cortical (P-V-C) triad. The vestibular node has two arms, a sustained otolithic DC reference grounded in tonic afferent baseline discharge, and a high-pass canal arm that contributes a relaxing eigenvalue rather than an oscillatory mode. Proprioceptive afference, with fascia as structural substrate, and cortical dynamics close the loop. The cortical α/θ ratio ρ is regulated toward φ by a dissipative circle-map attractor on a 2-torus, and the loop's stability depends on two incommensurate frequencies. The framework claims only two things, Address (what the loop is anchored to) and Occurrence (the first-person occurring itself), and explicitly brackets Awareness and the binding problem. The loop is specified in substrate-neutral dynamical terms, so the claim concerns the organization, not the material. The manuscript sorts every claim into evidential tiers (Established, Inference, Novel, Descriptive, Owed) and states openly that φ cannot be separated from nearby Fibonacci convergents at achievable resolution. Its most distinctive contribution is a degradation ordering derived from ℛ's eigenvalue structure, which predicts how self-return should fail as each node is compromised. The priority experiment is a perturbation-and-return covariance study: ratio-preserving covariance should appear at slow regulatory timescales and be absent within the fast return window. I would value your view on where this scope-limited, classical dynamical account meets or diverges from Orch OR, and on whether the testable predictions seem well posed. zenodo.org/records/21909260

Self-Return: The Architecture of Embodied Coherence

A recursive, self-return characterisation of the updating process is compatible with process-oriented accounts in which no fixed substrate is required: reafferent loops can continuously revise which factors currently exert the strongest organising pressure without those factors becoming permanent primitives or being localised to any single anatomical site.

A lot of the information from the Self-Return framework is in this video. teapaw.org/silentaltitudes/status…

Why don't you fall apart when you close your eyes? Not metaphorically. Literally. Close your eyes right now and stand on one foot. You don't dissolve into confusion about where your limbs are or who's in charge of them. Something is holding you together as a single, located point of view, moment to moment, without you doing anything to make it happen. That "something" is what I've spent the last few years trying to model. I call it Self-Return. Here's the short version, then I'll back it up. The claim Your sense of being a located "you" — a single point of view sitting somewhere, looking out — isn't a static fact about your brain. It's an active, continuous process. Your nervous system is constantly measuring how far it has drifted from a reference state and pulling itself back, faster than the world can knock it off course. When that return process is fast enough, you get a stable self. When it slows down, things get strange. When it's cut, things fall apart in very specific, predictable ways. The system doing this work is a loop between three things: your proprioceptive sense (the fine-grained readout of where your body parts are, fed by receptors in muscle and connective tissue), your vestibular sense (the inner-ear system that tracks orientation relative to gravity), and your cortex (which does slower, predictive correction on top of the other two). I call it the P-V-C loop. The evidence isn't hypothetical — it already happened to someone There's a man named Ian Waterman. At 19 he caught a rare autoimmune illness that destroyed the nerve fibers carrying proprioceptive information from his neck down. He didn't lose strength — his muscles still work fine. He lost the sense of where his body is in space. With his eyes closed, he cannot stand up. He simply collapses, like a puppet with the strings cut, even though nothing is wrong with his muscles or his balance organs. With his eyes open, he can move — but only by consciously, effortfully watching every limb and calculating its position, like piloting a mech suit instead of just having a body. Decades of practice never made it automatic. It's still a full-time cognitive job, every second, for the rest of his life. That case is the anchor for everything else. It tells you, directly, that a fast automatic body-sense is doing something no amount of visual feedback or willpower can substitute for. Vision can replace the controller. It cannot replace the reference. The part that surprised even me: your inner ear can't lie, but it can be ignored Your vestibular organs (the otoliths, specifically) do one very narrow job: they transduce sustained linear acceleration — for someone standing still, that's just gravity's pull, constant in magnitude and direction. Your brain cannot alter what that signal says. There's no wiring that lets you edit that signal from the inside. That makes it a rare thing in a self-regulating body: a reference point the system did not author. What your brain can do is turn the volume up or down on how much it listens to that signal versus the others (proprioception, vision). That single adjustable dial — trustworthy content, adjustable weight — turns out to be exactly what you need to have a stable frame of reference that isn't just circular self-consistency. If the brain could edit the content, "orientation" would just mean "whatever the brain currently says," which isn't orientation at all. If it couldn't adjust the weighting, you couldn't compensate when one of your senses becomes unreliable (in the dark, on a boat, in freefall). You need both properties at once, and the vestibular system is the only channel in the body that has them. This also predicts something specific and testable: turning up the trust in your vestibular signal isn't always good. There's a known failure mode — pilots experience it — where sustained acceleration fools the inner ear into reporting a false "down," and if you're leaning too hard on that signal, you follow it into a wrong orientation with total confidence. More trust in the reference channel is only better up to a point. That's a real prediction, not a hand-wave: it should be possible to find the point where more weighting starts hurting rather than helping. Even your sense of balance has a blind spot, and it's exactly where the theory says it should be Blindfold someone and ask them to walk in a straight line. They can't. They walk in loops, completely convinced they're going straight (this is a real, published result). But give them the sun to look at, and they hold a straight line easily. Why loops and not, say, random drift? Because your vestibular system gives you a hard reference for two of your three rotational axes (tilting forward/back, tilting side to side) but nothing for the third — turning left or right around your own vertical axis. That axis has no gravity-based anchor. It has to be tracked by integrating rotation over time, which drifts, the way a ship's dead-reckoning navigation drifts without a fixed star to check against. If a theory says "stability requires an external, unauthored reference," this is the sharpest possible test of it: the one axis that lacks such a reference should be the one axis where people actually get lost. It is. That's not a metaphor I'm reaching for — it's a specific, falsifiable structural prediction that already has supporting data. The brainwave part — stated carefully, because this is where people usually go wrong There's a long-documented pattern in EEG: the ratio between your alpha rhythm (~10 Hz, associated with a relaxed, idling brain) and your theta rhythm sits close to the golden ratio, φ ≈ 1.618, more often than it sits at a "nicer" number like a clean 2:1. I want to be direct about what I am and am not claiming here, because this is exactly the spot where good ideas get ruined by overreach. I am NOT claiming φ is some mystical universal constant, or that your brain is "wired to sacred geometry." The actual mathematical reason φ shows up is almost the opposite of mystical: φ is the number that is hardest to approximate with a simple fraction. In coupled-oscillator systems, ratios that are easy to approximate with simple fractions get "captured" — the two rhythms lock together. φ is the ratio two oscillators can hold the longest without locking. So a system whose whole job is staying maximally ready to couple with anything, while being captured by nothing, has a mathematical reason to sit near φ specifically — not because φ is special, but because being hard to lock is functionally valuable, and φ is the extreme case of hard-to-lock. That's the honest version. It's more interesting than the mystical version, and it's actually testable, whereas the mystical version isn't. And here's the part I insist on saying even though it costs the theory some swagger: with current EEG resolution, we cannot yet tell the difference between the true golden ratio and its close cousins (5/3, 8/5). The gap is smaller than our measurement noise. So the specific number φ is not something I can currently prove. What I can test — and what actually matters — is whether that ratio is being actively defended: if you experimentally nudge one brainwave frequency, does the other one track it to preserve the ratio? That's the real signature of a regulated system versus two rhythms that just happen to sit near each other. That test is designed, pre-registered, and ready to run. Why this matters, and what it doesn't try to explain I'm not claiming to solve the hard problem of consciousness — why there's subjective experience at all. I'm bracketing that entirely. What this framework tries to explain is narrower and, I think, more tractable: why a located, first-person point of view holds together as a stable, coherent thing instead of falling apart moment to moment, and what specifically breaks when it doesn't. That's a real, empirical, falsifiable question. Sever the right channel and the theory predicts exactly what should happen and what shouldn't. Perturb the right rhythm and it predicts whether another rhythm should follow it or ignore it. It stakes out clear ways to be proven wrong, and it has already survived several rounds of trying to break it — including one where a factor-of-two math error in the core equations got caught and fixed rather than buried. The deafferentation case is real. The blind-walking result is real. The EEG ratio regularity is real. What's new here is the claim that they're all readouts of the same underlying loop — and that claim is the part still waiting on data. Full technical writeup, with the equations, the failure modes I've already ruled out, and the experiments designed to break it, is on Zenodo. This is the plain-language version of the same argument. If you close your eyes and don't fall apart, something is working very hard, very fast, to keep it that way. I'd like to know exactly what. zenodo.org/records/21909260

Self-Return Framework

Architecture of the Self

In the Self-Return framework, the body is not a peripheral input channel feeding a centrally-located self, it is the substrate through which Address and Occurrence are continuously constituted — the proprioceptive-vestibular-cortical triad doesn’t report a boundary condition to an inner locus, it generates the return trajectory (ℛ) that Address consists in. Because the α/θ ratio ρ is regulated toward φ via the body’s own dissipative circle-map dynamics — not toward some free-floating locus outside the triad — there is no coherent Address prior to or independent of that regulation. On this view, embodiment isn’t a prior locus having a body as an accessory or a sensory periphery it happens to inhabit — the vestibular DC reference, proprioceptive afference, and cortical integration are structural terms within the architecture the model uses to generate Address and Occurrence, such that degrading them doesn’t diminish access to something that persists elsewhere, it degrades the return process itself, in the ordered sequence your eigenvalue structure predicts.

The Engine of Identity

Self-Return predicts something most “consciousness” frameworks don't commit to: which axis stays bounded when return capacity degrades, and which one doesn't. Not degradation in general — degradation that's anisotropic. Heading drifts unbounded; roll and pitch stay bounded. The axis is named in advance, derived from the return operator's eigenvalue structure, not fit after the fact. A substrate-neutral rival has no reason to predict this. If the widening eigenvalue gap shows up on a different axis, or isotropically, this part of the framework is wrong. zenodo.org/records/21909260