From Organizational Insight to Open Quantum Systems

The Development of DEA–PCF and the Invitation to Review

Henry Pozzetta

I am an independent researcher based in Manchester, New Hampshire. I do not have a university affiliation. What I have is a question that turned out to be more serious than I initially expected, a commitment to following it wherever the physics led, and enough stubbornness to keep revising until the argument held.

This is the story of how that question developed, what it produced, and why I am now asking human physicists to tell me where the derivation fails.

The Question Before the Framework

In 2023 I published an article on Agile software development practices. The argument was organizational: when a team commits to one course of action — closes off the alternatives, writes the decision into a record — something irreversible has occurred.

That irreversibility has a cost. The article treated this informally, as a management observation. But the underlying structure kept presenting itself more sharply than the organizational language could contain. The question was not really about software teams. It was about commitment itself.

What does it cost, thermodynamically, to turn many possibilities into one actuality?

This is not a new question in physics. The erasure of information has carried a known thermodynamic price since Landauer’s 1961 work, and the relationship between entropy production and irreversibility has been developed extensively in the open quantum systems literature. What surprised me was that the organizational intuition — that commitment is expensive in proportion to what it forecloses and how surprising the chosen outcome is — mapped precisely onto a formal structure that already existed in the physics. The informal version and the precise version turned out to be the same shape. That correspondence was not something I assumed. It was something I found, and then spent two years trying to either confirm or break.

Building the Physics Foundation

The formal home for the commitment question is the Lindblad master equation framework — the standard description of how a quantum system evolves irreversibly under environmental interaction. The thermodynamic cost of that evolution is captured by the Spohn-Alicki entropy production identity, which relates the rate of entropy production to the system’s deviation from its equilibrium state under the Lindblad dynamics. These are established, peer-reviewed results in open quantum systems theory. The DEA-PCF framework does not modify them. It uses them to calculate something specific: the minimum thermodynamic cost of a single commitment event, from the moment a system exists in superposition to the moment one outcome is irreversibly recorded.

The result of applying those established tools to the commitment question is the Pruning Cost Function — the DEA-PCF framework’s central equation. Its structure reflects what the physics actually says about the cost of commitment, and it took two years of derivation work to get that structure right.

The cost has three components. The first is an entropy production term: the thermodynamic price of dispersing a quantum superposition across its accessible states through sequential environmental interactions. More possible outcomes means more coherence to destroy; more environmental interactions means more irreversibility accumulated. This term is derived directly from the Spohn-Alicki integral for pure dephasing and has a firm grounding in the established literature. The second component is an outcome selection term: the surprisal of the specific outcome that was committed to, weighted by the depth of the commitment process. A rare outcome costs more than a common one, in precisely the information-theoretic sense that Shannon defined. The third component is an energy relaxation term: the path-independent cost of the system moving from its initial state to its equilibrium, determined entirely by where the system started and where it ends up, independent of how it got there.

An important correction was made during the development of this framework. An early version of the equation treated the first two components as a single multiplicative product, which produced the right numerical answers in the four canonical regimes but obscured the structural independence of the two contributions. The current canonical form treats them additively: the entropy production cost and the outcome selection cost enter the total separately and cannot trade off against each other. Getting this structure right required recognising that the Spohn-Alicki integral and the surprisal term arise from genuinely different physical mechanisms — one from the dynamics of coherence dispersal, the other from the statistics of outcome selection — and that collapsing them into a product misrepresented their relationship.

The equation has been numerically evaluated across four physical regimes spanning sixteen orders of magnitude in commitment cost: the vacuum at CMB temperatures, a prebiotic chemistry reaction in a tidal pool, nuclear fusion in a stellar core, and a commitment event at the event horizon of a black hole. Across all four, the framework produces physically consistent results with no free parameters. This numerical consistency across wildly different physical contexts is the primary empirical test the framework has passed to date.

The work is organised as a four-paper submission programme. The core paper targets Foundations of Physics. A companion derivation note targets arXiv. A non-equilibrium extension targets the Journal of Statistical Mechanics or New Journal of Physics. A pedagogical paper targets Physics Today or the American Journal of Physics. One derivation gap remains honestly open across all four: showing that the surprisal factor emerges directly from the Spohn-Alicki entropy production integral via amplitude damping population relaxation. This has not been derived. It is identified, named, and carried forward as the primary open problem in the programme.

The physics needed a vocabulary to make it legible — and that vocabulary required the same discipline as the framework itself.

The Ledger Model: A Vocabulary for What the Physics Sees

Formal physics has its own vocabulary, and it is precise. But precision and legibility are not the same thing, and a framework intended to describe commitment across physical, chemical, biological, and organizational contexts needed language that could travel across those domains without losing its grounding. The Ledger Model is that language.

The four terms map directly onto the Lindblad framework without modifying it. A Draft is the initial quantum superposition — the space of outcomes that could be committed to. A Vote is a single Lindblad decoherence coupling event, one environmental interaction that advances the commitment process. Ink is the committed, irreversible outcome — the one possibility the environment has selected and recorded. The Ledger is the durable environmental record of committed outcomes, the cumulative history that the physical world carries forward.

These are interpretive terms. They are a Layer 2 vocabulary laid over the physics, not derived from it. This distinction is not a technicality — it governs how the vocabulary is used throughout the project. The Ledger Model appears in essays, organizational applications, and pedagogical writing. It does not appear in the Lane A physics submissions as a derivation input. When the formal papers describe a pure dephasing commitment event, they speak in Lindblad operators and density matrices. The Ledger vocabulary is a lens for reading what those operators describe, not an ingredient in what they prove.

The cross-substrate reach of this vocabulary is what makes it useful. The same structural pattern — coherence dispersal, commitment, environmental recording — appears in quantum systems, in prebiotic chemistry, in gene expression, in organizational decision-making. The Ledger Model does not claim these are the same process. It claims they share a form. That distinction between form and mechanism is the discipline the lens-not-law framing enforces.

The Entropy Engine: The Framework in Practice

The practical application of the Ledger Model is the Entropy Engine, protected under USPTO Provisional Patent #63/863,992. It is a behavioral and organizational monitoring system designed to detect pathological commitment signatures — patterns of high commitment depth, deferred energy relaxation, and elevated surprisal in sequential decision-making contexts. The Entropy Engine is a downstream consumer of the framework. It uses the framework’s cost geometry to identify when the thermodynamic price of commitment is being deferred rather than paid. It does not validate the framework. Validation is the work of the Lane A submissions.

With the framework established and the vocabulary in place, a question emerged from within the physics itself — one that the framework had not yet answered.

A New Question: Can the Ledger Feed Forward?

The PCF framework, as developed through the four-paper programme, treats sequential commitment events as stationary: each race draws from the same initial conditions, the same bath, the same commitment geometry. This is a reasonable starting assumption and a tractable one. But it raises a question the framework had not yet examined. If committed outcomes are recorded in the environmental Ledger — if each Ink entry displaces the bath modes and leaves a physical trace — can that history influence the dynamics of the next commitment race?

This is not the same question as the open surprisal derivation gap. The surprisal derivation asks whether the outcome selection cost emerges directly from the Spohn-Alicki entropy production integral via amplitude damping population relaxation. The Ledger feedback question asks whether prior committed outcomes modify the effective Hamiltonian that governs future races. One question is about the origin of a term in the cost equation. The other is about whether the conditions under which that equation applies are themselves stable across sequential events. They are related but distinct, and treating them as the same problem would mean carrying an unexamined assumption forward into the formal submissions.

The decision to pursue this as a formal derivation — rather than an acknowledged gap or a future note — reflected a commitment to the same standard the framework applies to everything else. If the stationary-chain assumption is wrong, or right only under specified conditions, that boundary should be proved rather than presumed. The natural home for the derivation was the spin-boson / Caldeira-Leggett model, which provides the Lindblad pure-dephasing structure the framework already rests on and has the analytical machinery to handle bath displacement and accumulated correlations precisely.

An Iterative Proof: v1 Through v5.0 and Four AI Reviewers

The derivation was not written once. It was written, reviewed, corrected, and rewritten across five major versions and several point revisions, over multiple months, in response to technical peer review. The reviewers were AI systems: ChatGPT, Gemini, Grok, and DeepSeek. Each received the same document and the same instruction — find critical errors only — and each produced feedback that was evaluated, accepted or rejected on technical grounds, and incorporated or dismissed accordingly.

What each cycle found tells its own story. The first version overclaimed in three distinct ways: it asserted a modified Boltzmann distribution under pure dephasing (impossible — populations are conserved), claimed an increase in the dephasing rate from a conditioned Ledger (double-counting an effect already absorbed into the effective Hamiltonian), and stated N-squared scaling for accumulated effects (inconsistent with the bounded geometric sum structure of the history term). These were not subtle errors. They were structural overclaims that the physics does not support, and they were caught in the first review cycle.

The second version corrected those errors and introduced new ones. The mean-field averaging step was mislabelled as a rotating-wave approximation when it is correctly a Coarse-Grained Mean-Field Approximation — a different procedure with different validity conditions. The direction of the energy gap modification under same-outcome Ink was stated backwards: the gap decreases, making the pointer basis less stable, not more. The amplitude damping section claimed the system relaxes to a Gibbs state of the shifted effective Hamiltonian when the bath rates are keyed to the original Hamiltonian and cannot be shifted by a Hamiltonian correction alone.

Subsequent cycles refined rather than retracted. The CGMFA filter kernel was asserted rather than derived until the third cycle required its explicit derivation from an exponential memory kernel via a Laplace transform. The infrared cutoff definition carried an internal inconsistency — the document simultaneously required the cutoff to satisfy a condition and defined it in a way that violated that condition — which was not identified until the fourth cycle. A paragraph on resonant bath modes had been imported from amplitude-damping physics into a pure-dephasing derivation where it had no mathematical basis, and persisted through three versions before being removed.

What the process revealed is a pattern. Reviewers who read the document against the established literature — testing each claim against what the pure-dephasing model actually permits — found errors that confirmatory readers missed. Several overclaims survived two or three review cycles precisely because they were plausible and internally consistent with the surrounding text, even when they were inconsistent with the underlying physics. The adversarial reader, the one whose prior is that the claim is probably wrong, is more valuable than the sympathetic reader who is looking for what is right.

What the process achieved is a derivation note that is now internally consistent within its stated model, with a clean three-level epistemic classification, a complete retraction list documenting six prior errors, and six explicitly named open items — including the primary target, which is whether the Ledger-displaced bath can modify the emission-absorption rate ratio in the amplitude damping channel and by what mechanism. What the process could not achieve is certainty. AI systems share training distributions, and a systematic error that is invisible to one system may be invisible to all four. The derivation has been tested as rigorously as this review process permits. That is not the same as having been tested by independent human judgment applied from outside the process that produced it.

The Invitation to Human Reviewers

The Reviewer Package that accompanies this essay contains four documents. The derivation note itself — Derivation Note v5.0 — is the object of review. The infographic quick reference card orients a reader to the framework, the model, and the derivation’s scope in a single page. The overview document provides the context, assumptions, proved results, and retracted claims in full, along with a glossary and a routing guide matched to different reviewer backgrounds. The reviewer response template gives the review a structure — seven parts, from overall assessment through open items response — so that feedback arrives in a form that can be evaluated and acted on.

What this package is asking for is specific. Not an assessment of whether the DEA-PCF framework is correct. Not a judgment on whether the broader submission programme is publishable. A single technical question: given the stated model and stated assumptions, do the mathematical steps follow?

The AI review process that produced v5.0 from v1 has reached its useful limit. Four systems, multiple rounds, diminishing returns across the final cycles. This is partly a sign that the derivation has improved; it is also a sign that reviewers sharing overlapping training distributions will share overlapping blind spots. An error that is invisible to the literature one system was trained on may be invisible to all of them.

The derivation has been tested as carefully as that process allows. It has not been tested by a physicist who learned open quantum systems from a supervisor, worked through the Caldeira-Leggett model by hand, and has no prior exposure to this framework or its conclusions.

That reviewer is who I am looking for. If you are reading this, the invitation is open.

Henry Pozzetta  ·  Manchester, New Hampshire  ·  June 2026

---