Theoretical analysis demonstrates non-equilibrium criteria for self-maintaining patterns in active media, revealing distinct spectral boundaries between passive persistence and agency.
This article follows from one sentence about the world: that the substrate is a diffusion whose fluctuations are nowhere degenerate. That single assumption turns out to carry a surprising amount. The geometry of the state space is not imposed on it but is the covariance of its own noise; the second law appears as the positivity of a quadratic form rather than as an extra law; and the demand that a pattern hold its own order becomes a demand for a circulating current, which no equilibrium can supply.What follows is an attempt to say, without leaving that ground, what distinguishes a pattern that keeps itself from one that merely lasts. Observation enters as a choice — a chart, an interval, a transversal — and the content of the theory is what survives changing it. Three groups act on the resulting transfer matrix, and their invariants differ; a quantity reported without its group is not a measurement of anything. On those invariants rest the two spectral criteria and the bridge between them, the margins and the gate between spiral and node, and the two ways a pattern can stop: slowly, by settling into a fixed point nothing can leave, or suddenly, by a fluctuation it could not absorb. Both are the same barrier read at different rates.Later parts follow the same object upward and outward. A lineage is a return map on a different section, so the whole algebra applies again with generations for returns. A collective has two spectra rather than one, and its two failures are distinct: it can die as an object, or it can cease to be one object while every member survives. The last part asks what of all this a recorded series can actually deliver, and answers, in several places, less than one would like.Three things are introduced as definitions with explicit existence hypotheses and never as postulates: the separation of agent from medium, the discrete layer of gates and resets, and the criterion of agency itself. That last is deliberate. A criterion written as an axiom cannot be contradicted by measurement, and a theory whose central term is beyond refutation has bought its certainty at the only price that matters.
No takes yet. Share an insight, caveat, or question.
Alexander V. BALABA (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: