↵ select ↓ ↑ navigate esc close

Paper Feed: Principles of Vasocomputation: A Unification of Buddhist Phenomenology, Active Inference, and Physical Reflex (Part I)

Paper Feed ·

Listen: https://serve.buildtall.systems/00eae0278527fd5af3a63b1fa06e8a193035a6c2a121a0b82146093291524b73.mp3

Source: https://serve.buildtall.systems/144fcc3cf03c6a4bf06f209b03212f786e73e25a3049100f540b8922c9c30d4b.pdf


This episode covers Principles of Vasocomputation: A Unification of Buddhist Phenomenology, Active Inference, and Physical Reflex (Part I), a long essay published on the Opentheory website. The author argues that a Buddhist idea about craving, tanha, can be explained by the brain's prediction machinery and by the muscle cells wrapped around blood vessels.

This is a long theoretical essay, not a study. Its text reports no experiment and no data set, and the author says outright that the health examples and the central mechanism are proposals rather than findings supported by data. The essay is the first part of a series, and the author promises later posts.

The question is why the brain produces tanha, the Buddhist term for the small, reflexive movements of grasping and aversion that the author links to most suffering. The author's answer is that tanha is a side effect of active inference, the strategy by which the brain creates predictions and then acts to make them true. On this account tanha is unskillful active inference: the brain generates predictions too quickly, in domains it cannot control, or out of step with its own world model, and then holds on to them. The essay then proposes a mechanism. Vascular smooth muscle cells, the thin sheath around every blood vessel, are said to do much more than move blood. Their contractions are proposed to compress neural patterns and freeze them, and when a contraction is held long enough, a latch-bridge mechanism common to smooth muscle locks it in place as a durable hyperprior. The author calls this vasocomputation and offers three linked hypotheses: a compressive vasomotion hypothesis, a vascular clamp hypothesis, and a latched hyperprior hypothesis.

The support offered is mostly argument. The essay relies on an evolutionary case that the vascular system, older than neurons and in contact with every part of the body, would be surprising to have served only as plumbing, and on the claim that vascular smooth muscle is the muscle best placed to tune the nervous system. Its strength is that it connects three areas, Buddhist phenomenology, active inference, and physical reflex, and makes predictions that could in principle be checked, such as that cessations in meditation would show almost no vasomotion and that EEG may pick up vasomotion more than neural activity. Its weakness is that nothing in the text is tested. The examples about migraines, diet, sauna and cold plunges, and psychedelics are offered as speculation. The text as presented here does not state a funding source or a data disclosure.

The essay is published on the Opentheory website, according to its own front matter. According to OpenAlex, the index has no record of this document: no work in the index matches its title and authors as of October 18, 2026. This episode therefore makes no claim about how often the essay has been cited.

The author begins with tanha, the term the Buddha used for what he called micro-motions of greed and aversion. By default, the author explains, the brain tries to grasp and hold onto pleasant sensations and push away unpleasant ones. The author notes that the Buddhist consensus seems to hold that this accounts for roughly ninety percent of suffering.

The author then surveys how teachers have rendered the word. Romeo Stevens suggests "fused to," "grasping," or "clenching," a sense that the mind is trying to make sensations feel stable, satisfactory, and controllable. Nick Cammarata calls it a "fast grabby thing" that happens within about 100 milliseconds after a sensation enters awareness, and Daniel Ingram suggests the grab can occur in as little as 25 to 50 milliseconds. Uchiyama Roshi describes its cure as "opening the hand of thought," and Shinzen Young suggests "fixation." In the vipassana view, tanha is something the mind instinctively does, and meditation lets a person see the process as it happens so that they can stop doing it.

The author's central complaint is that tanha is not yet a topic of study in affective neuroscience, and should be. The author argues that neuroscience is usually hard because complex dynamics tend to come from complex mechanisms, and the most valuable findings are the exceptions, where complex dynamics arise from elegantly simple core mechanisms. The author's bet is that Buddhist phenomenology is very careful about what it calls dependent origination, so the concepts it treats as basic building blocks of experience are especially likely to have simple implementations in the brain. The claim is stated plainly:

I don't think tanha has 1000 contributing factors; I think it has one crisp, isolatable factor.

The author suggests that finding this factor could reorganize systems neuroscience on the scale of earlier shifts brought by cybernetics, predictive coding, and active inference.

The second step is to ask what tanha is trying to do for us. The author's answer is that tanha is a side effect of a normal and generally effective strategy the brain uses a great deal, which is active inference. In this account, we move ourselves to action by first creating a predicted sensation, such as a sweet taste in the mouth or the sense of not standing near a dangerous-looking person, and then holding that prediction until our actions make it true, at which point the tension can release. Active inference therefore treats our to-do lists as stored predictions, which are untrue sensory observations that we act to make true.

The author calls the resulting hypothesis "tanha as unskillful active inference," or TUAI. The claim is that this process often goes wrong in three ways. First, the rate of generating normative predictions can outpace our ability to make them true, overloading a finite system. In the author's words,

Basically we try to control too much, and stress builds up.

Second, we generate predictions in domains we cannot control, such as expecting the taste of cake to linger forever, or expecting that we did not drop our glass of water on the floor. The author frames this as a predictive processing version of a story Romeo Stevens has told. Third, the system that represents the world model can fall out of sync with the system that maintains predictions as operators on that model. When the basis of the world model shifts relative to the basis of the predictions, the predictions become nonspecific or nonsensical noise, and that noise produces stress. The author adds a catch-all fourth category for cases where the prediction machinery is impaired outside any particular meaning, for example through metabolic insufficiency.

The third step asks what the brain's basic unit of organization is. The author favors Adam Safron's proposal that the nervous system is built from what Safron calls Self-Organized Harmonic Modes, or SOHMs. The author likes this idea because the mathematics of harmonic modes allows a lot of interesting computation to arise more or less for free. Safron suggests these modes work as autoencoders, which the author believes are functionally the same as symmetry detectors. The author also notes that evidence increasingly suggests SOHMs are organized around the brain's physical resonances at least as much as around its connectivity, which the author describes as a surprising result.

The author gives an intuitive picture of these modes. At high frequencies they act as feature detectors. At lower frequencies they are more like wind chimes: the presence and absence of particular modes, and their interactions, give us a subconscious feeling about what kind of environment we are in and where its rewards and dangers lie. The author expects that SOHMs are arranged to maximize the differentiability of likely world states, to minimize crosstalk, and in aggregate to form a world model, which in the Neural Annealing, REBUS, and ALBUS framework is called a belief landscape.

On this picture, to be in tanha-free open awareness, without greed, aversion, or expectation, is to feel "the undoctored hum of your SOHMs." But the author argues that we doctor our SOHMs all the time. When a pleasant sensation arises, we reflexively try to grab it and stabilize the resonance. When something unpleasant arises, we try to push it away and deaden the resonance. Social expectations to act normal and be useful also draw on the same finite resource pool. The author says this active SOHM management is effortful and unpleasant in rough proportion to how many modes must be managed and how long they must be managed. That leaves two questions the author wants to answer: how the brain manages SOHMs at all, and, if the Buddhists are right that this creates suffering, why the brain bothers to try.

The fourth step proposes that tanha is an artifact of what the author calls compression pressure. The idea is that a continual process pushes on the complexity of what is, meaning the neural patterns represented by undoctored SOHMs, and tries to collapse it into a simpler configuration. Sometimes it holds that simplification in place until we act to make it true. The result conflates what is, what could be, what should be, and what will be, and the author says this conflation is the source of a great deal of moral and epistemological confusion. Tanha is therefore both the pressure that collapses complexity into simplicity and the ongoing stress of keeping up the counterfactual parts of that collapse, which the author calls compression stress.

The author frames the question as a balance between two costs. Applying compression pressure has metabolic and epistemic costs, both immediate and ongoing. But the brain is a finite system, and if it does not continually compress patterns away, sensory chaos becomes unmanageable. The author's conclusion is blunt:

The right amount of compression pressure is not zero.

The author also describes tanha as an excessive forcefulness in metabolizing uncertainty. Drawing on Erik Hoel's work on energy, information, and uncertainty as equivalent and conserved quantities, the author compares the nervous system's task to digestion. It extracts value from sensations, then excretes the leftover information, leaving a low-information, low-uncertainty clean slate for the next sensation. We go wrong, the author says, when we assess context poorly, try to extract too much or too little certainty, or seek forms of certainty that do not suit the sensation. A person's personality, aesthetic, and much of their phenomenology can be described by their characteristic ways of metabolizing uncertainty: their habits for initial probing, for digestion and integration, and for excreting or externalizing waste, along with the particular reagents they cannot supply themselves and must find in the world.

This leads to the mechanism question. The author proposes that evolution coupled two very different systems to apply compression and prediction pressure selectively, in a way that keeps perception reliable. One system is focused on perception, with undoctored SOHMs serving as ground-truth perception. The other handles compression, judgment, planning, and action. This arrangement allows a near-unlimited capacity for adjustment and hypotheticals without corrupting perception. The author notes that neuroscience has usually located executive functions in particular brain areas, but suspects the core logic is hiding much closer to the action.

The fifth step turns to vascular smooth muscle cells, or VSMCs, the thin muscle sheath wrapped around every blood vessel. The author grants the standard picture: a branching network of ever-narrower channels, powered by the heart, that delivers nutrients to the brain, stomach, limbs, and other organs and carries away waste. The sheath undulates in a process called vasomotion, which the standard view takes to help keep blood moving and to help adjust blood pressure. The author accepts that this is true, but argues it reflects what has been easy to measure and misses most of what these cells do, which the author puts at around ninety percent.

The author's reasons are evolutionary. Evolution works in layers, and the oldest base layers often carry rudimentary versions of more specialized capacities, along with deep control hooks into newer systems built around them. The vascular system predates neurons and has co-evolved with the nervous system for hundreds of millions of years. Its muscle cells have physical access to every part of the body and can flex in arbitrary patterns and rhythms. The author concludes it would be surprising if evolution had used this system for nothing more than plumbing. The author also argues by following the money: the vascular system controls the nutrients and waste disposal for the nervous system, and so would likely win any contest over how the two systems are co-regulated. The author expects VSMC contractions to influence nearby neurons through mechanisms such as ephaptic coupling, by reducing blood flow, and by adjusting local physical resonance, and expects them to be triggered by local dissonance in the electromagnetic field.

The author then offers three related hypotheses about the computational role of these cells, which together make up what the author calls vasocomputation. The first is the Compressive Vasomotion Hypothesis. Here the vasomotion reflex works as a compression sweep on nearby neural resonances, collapsing and merging fragile, ambivalent patterns, which the author links to the "Bayesian blur" problem, into a more durable and definite state. The author holds that vasomotion, reflexive reactions to uncertainty, and patterns of tanha are equivalent.

The second is the Vascular Clamp Hypothesis. Vascular contractions freeze local neural patterns and plasticity for as long as the contraction lasts, in a way the author compares to collapsing a superposition or probability distribution, clamping a harmonic system, or pinching a critical network into a definite circuit. Particular vascular constrictions correspond to particular predictions within active inference, and function as a kind of medium-term memory.

The third is the Latched Hyperprior Hypothesis. If a contraction is held long enough, it engages the latch-bridge mechanism common to smooth muscle cells. This durably freezes the nearby circuit, isolating it from conscious experience and from global updating, and leaves it with a much smaller dynamical repertoire. In effect, the body makes a lasting commitment to a specific hyperprior. The local vasculature unlatches once the prediction the latch corresponds to is resolved, which restores the nearby networks' ability to hold a wider superposition of possibilities.

The author summarizes the sequence this way: the initial contractive sweep jostles the neural superposition of interpretations into specificity, the contracted state temporarily freezes the result, and if the contraction is sustained, the latch-bridge mechanism engages and cements the freeze as a hyperprior. In the author's phrase,

With one motion the door of possibility slams shut.

The author concludes that we collapse our world "into something less magical but more manageable, one clench at a time," and adds, "Tanha is cringe."

The author draws three implications. For the free energy principle and active inference, the claim is that the motifs of smooth muscle, particularly in the vascular system, may be where the brain's top-down predictive models are hiding, which the author describes as an open mystery in that field. Specific predictions are held as vascular tension, and that tension is released by action, consolidated by Neural Annealing, or made superfluous by neural remodeling, since a pattern held in place long enough becomes the default. In the language of the Deep CANALs framework, the neural weights that give rise to SOHMs form the learning landscape, and the SOHMs together with vascular tension form the inference landscape.

For Theravada Buddhism, the author argues that the motifs of the vascular system can be understood as the means by which we try to manipulate our sensations. Vasomotion is an attempt to pin a sensation down, which is tanha. Muscle contraction freezes patterns, and smooth-muscle latches block out feelings of possibility and awareness of that part of the body. Progress on the contemplative path, the author says, should show up as using these forms of tension less and needing them less. The author expects that cessations would correspond to an almost complete absence of vasomotion, and suggests that EEG may pick up vasomotion more than neural activity.

For practical health, the author argues that smooth muscle tension, especially in VSMCs and especially latched tension, is a system that science knows relatively little about, yet it is involved in an unusually wide range of problems. The latch-bridge mechanism is central. Smooth muscle cells can enter a discrete state in which they attach their myosin heads to actin in a way that locks the tension without requiring ongoing energy. Latches take seconds to minutes to form and dissolve, and can persist for minutes, hours, days, months, or years depending on the prediction they stabilize. The author suggests the sum of all latches likely accounts for the majority of bodily suffering, and that if you are holding tension in your body, you are subject to the mechanics of the latch-bridge.

The author offers a number of speculative examples. Migraines and cluster headaches are described as almost certainly inappropriate VSMC latches, and all hollow organs are surrounded by smooth muscle that can latch. A long-term diet of poor food, the author suggests, including seed oils, leads to random latch formation and what the author calls "lumpy" phenomenology. Sauna and cold plunges are offered as ways to force the clench-release cycle, and simply attending to the body and putting attention into latched tissue can release them. Psychedelics can force latches open. The author also describes "latch spirals," in which a latch reduces blood flow to an area, which reduces the energy available to those tissues, which in turn prevents the latch from releasing, because releasing it takes activation energy and a freely cycling cell spends energy at a higher rate. The author says many problems in neuropathy and psychiatry may follow this pattern. None of these claims is supported by data in the text. They are the author's proposals.

The footnotes add several points. The author suggests that the active inference term precision weighting could be split into the amount of sensory clarity available in stimuli and the amount of grabbiness in the compression system, and that the Kullback-Leibler divergence between these distributions might give tanha a numerical value. The author also suggests the arrow of compression points away from three attributes that Buddhism names: the brain pushes its SOHMs toward patterns that are stable, meaning dissonance minimization, satisfactory, meaning harmony maximization, and controllable, meaning compression maximization. Skeletal muscle, other smooth muscle, and fascia likely also distribute physical load and probably cognitive and emotional load as well. The author singles out vascular smooth muscle for three reasons: it has the best physical access to neurons, it regulates blood flow, and it has the latch-bridge mechanism. The author writes that VSMCs seem to be the only muscle with the means, motive, and opportunity to finely puppet the neural system.

The author closes by saying the work describes a bridge between three levels of abstraction: a central element of Buddhist phenomenology, the core accounting system of active inference, and a specific muscular reflex. The author suggests this may offer a functional route to combine the active inference view with Michael Levin's work on distributed stress minimization. The author promises later posts on why this mechanism has been overlooked and how its dynamics connect to human problems and capacities, and describes the research program as integral to both human flourishing and AI alignment.

The essay's main takeaway is that tanha may rest on one isolatable mechanism, and that tension held in smooth muscle, especially latched tension in blood vessels, is where the brain's predictions may be stored and where much bodily suffering may sit. Progress on the contemplative path, the author suggests, should show up as needing these forms of tension less and less.

The limits the essay states are its own. The claims are proposals, the health examples are speculative, and the mechanisms have not been tested. Open questions include whether vasomotion and tanha are truly equivalent, whether latch-bridge tension accounts for the majority of bodily suffering, and whether EEG can separate vasomotion from neural activity. The author promises later posts on why this mechanism has been overlooked, how its dynamics connect to human problems and capacities, and how the research program relates to Michael Levin's work on distributed stress minimization and to AI alignment. Listeners should treat the essay as a set of hypotheses to be tested rather than a settled result.

  1. Principles of Vasocomputation: A Unification of Buddhist Phenomenology, Active Inference, and Physical Reflex (Part I)