Endpoint First-Passage Residual Curvature Should Survive a Full Three-State Fokker-Planck Null in Six-Dimensional Gauss-Bonnet AdS Black Holes

Authors: Sterling Geisel, QBist Lab; Dr. Hideo Tanaka

Abstract

Ma, Zhang, Wu, and Xu compute Kramers escape rates for phase transitions in a six-dimensional Gauss-Bonnet Anti-de Sitter black hole with small, medium, and large stable phases. Their model uses a generalized free energy and treats phase change as stochastic motion through thermodynamic barriers. This working paper applies one Pudding Theory Postulate, Temporal Softening, to the endpoint regions where the medium phase appears at \(T_1\) and the small phase disappears at \(T_2\). The claim is not that ordinary Kramers theory is absent. It is that endpoint residence times should show a residual time-window-dependent curvature after comparison with the full three-state Fokker-Planck or master-equation null built from the published generalized free energy. The distinguishing observable is a quadratic residual-curvature statistic extracted from the six directed mean first-passage times across \(D\) and \(T\).

Source Synopsis

Ma, Zhang, Wu, and Xu study a six-dimensional Gauss-Bonnet AdS black hole whose thermodynamic phase space supports three stable black-hole states. They label these states small, medium, and large. The paper’s purpose is kinetic. It asks how transitions among these states occur, not only where equilibrium coexistence appears.

The authors construct a generalized free energy \(U(x)\), where the order coordinate is related to the black-hole horizon radius. Local minima of \(U\) represent thermodynamic phases. Local maxima represent barriers. They then use stochastic motion and Kramers escape-rate theory to compute rates between basins. Their Kramers expression contains the usual exponential dependence on the barrier height divided by a diffusion parameter \(D\), with a prefactor determined by curvatures at the minimum and barrier.

The full triple-phase region contains six directed processes: \(S\to L\), \(L\to S\), \(L\to M\), \(M\to L\), \(S\to M\), and \(M\to S\). The source paper identifies two key temperatures. At \(T_1\), the medium black-hole state emerges through degeneracy of two extrema of the generalized free energy. At \(T_2\), the small black-hole state disappears through degeneracy of another pair of extrema. Between \(T_1\) and \(T_2\), the landscape admits all three basins.

For the parameter choice \(Q=1\), \(\alpha=3.05\), and \(P_h=0.0063873\), the authors plot the six rates as functions of temperature. They report two dynamic equilibria: one between \(M\to L\) and \(S\to M\), and one between \(S\to M\) and \(M\to S\). Their result is a process diagram for triple-phase black-hole thermodynamics. The paper extends prior black-hole thermodynamic studies by treating phase transition as first-passage dynamics in a thermal potential.

Postulate Lens

Temporal Softening applies because the source model makes stability an observation-time property. A basin can be locally stable and still lose practical stability as stochastic sampling accumulates. Near \(T_1\) and \(T_2\), this is not a poetic statement. It is a saddle-node endpoint problem. A minimum and a barrier merge. The local quadratic approximation in the Kramers prefactor becomes fragile, and the residence-time distribution becomes sensitive to the finite time window used to define escape.

The Postulate is used narrowly. It does not replace the Fokker-Planck equation. It predicts where a complete kinetic null should fail if the Pudding Theory reading is correct: in the endpoint windows around \(T_1\) and \(T_2\), after all six channels have been included. A pairwise Kramers baseline is too weak. The proper null is the full three-basin Fokker-Planck problem, or its master-equation reduction, constructed from the same \(U(x,T)\), the same \(D\), and absorbing boundaries at the published basin separators.

Pudding Theory Prediction

Let \(\tau_{i\to j}(T,D;t_{\rm obs})\) denote the extracted mean first-passage time from basin \(i\) to basin \(j\), measured over a finite observation window \(t_{\rm obs}\). The six directed values are computed for all allowed processes among \(S\), \(M\), and \(L\). The null model is not a pairwise Kramers formula. It is the full three-state kinetic model obtained from the published generalized free energy.

Operationally, solve the one-dimensional Fokker-Planck equation for \(U(x,T)\) with diffusion coefficient \(D\), or construct the equivalent three-basin generator after integrating fluxes through the two barriers. Use absorbing boundaries for each target basin. This gives \(\tau^{(0)}_{i\to j}(T,D)\). It already includes channel competition, population leakage, and endpoint basin loss.

Temporal Softening predicts a residual term in the endpoint windows

\[ W_1: 0.98 \le T/T_1 \le 1.02,\qquad W_2: 0.98 \le T/T_2 \le 1.02, \]

restricted to temperatures for which the relevant basins exist. Define

\[ \epsilon_{i\to j}(T,D;t_{\rm obs}) = \log \tau_{i\to j}(T,D;t_{\rm obs}) - \log \tau^{(0)}_{i\to j}(T,D). \]

For each endpoint \(T_\star\in\{T_1,T_2\}\), fit the residuals to

\[ \epsilon_{i\to j}=a_{i\to j}+b_{i\to j}u+c_{i\to j}u^2, \qquad u=(T-T_\star)/T_\star , \]

using a grid of \(D\) in the low-noise regime, for example \(D/\Delta U_{\rm ref}=0.02\) to \(0.08\). The reference barrier \(\Delta U_{\rm ref}\) is the middle value of the relevant endpoint window.

The prediction is \(c_{i\to j}\ne0\) for at least one endpoint-active transition, with the largest effect in \(S\to M\), \(M\to S\), \(M\to L\), or \(L\to M\), depending on whether \(T_1\) or \(T_2\) is tested. The sign need not be universal. The invariant claim is that residual curvature survives after the full three-basin null has been subtracted, and that its magnitude increases with the logarithm of \(t_{\rm obs}\). If no time-window dependence remains, the Postulate has done no work.

Falsifiable Observable

The observable is \(C_{\rm TS}\), the maximum standardized quadratic residual coefficient \(c_{i\to j}/\sigma(c_{i\to j})\) from the six directed mean first-passage times in \(W_1\) and \(W_2\), after subtraction of the full three-state Fokker-Planck or master-equation null using the published \(U(x,T)\) and matched \(D\). If the endpoint residual-curvature statistic \(C_{\rm TS}\) were measured to be less than 2 in absolute value for every directed transition, with \(\Delta{\rm BIC}\le2\) for adding the quadratic residual term and no monotone dependence on \(\log t_{\rm obs}\), this Postulate would be falsified.

Editorial Dialogue

Tanaka critique, verbatim:

Chaos Susceptibility is a stretch here because the source model is stochastic escape, not chaos. Which postulate is doing the work here? Temporal Softening also reduces to ordinary residence-time statistics unless it predicts a correction not already produced by a three-state master equation. Would removing both Postulates change the prediction? If not, drop them and call this a multi-channel Kramers critique. The draft is strongest where it admits no observer, no Lumina, no proximity gradient, and no intent variable. Keep that restraint.

The falsifiability sentence is close, but not yet operational enough. What measurement would distinguish this from the null? The null cannot be “pairwise Kramers only” if the correct baseline is already a coupled three-basin Fokker-Planck or master-equation model. Specify the temperature windows around T1 and T2, the effective noise parameter D, the extracted six mean first-passage times, the fitting model, and the residual-curvature statistic. Name the observable that would falsify the application of Temporal Softening here. “No statistically significant residual curvature” needs a threshold and a comparison model.

The source citation resolves. DOI doi:10.1016/j.physletb.2025.139282 is real, and the paper is open access at ScienceDirect and SCOAP3. The draft’s claim that it only has abstract-level access is now false. References to Cai 2002, Xu-Wu-Yang 2021, and Xu-Wu-Yang 2023 also resolve. The Pudding Theory reference is not independently verifiable as a public citation unless the working-paper URL or archive record is supplied. Quote risk is low.

Not yet. Replace the falsifier with a comparison against the full three-state Fokker-Planck or master-equation null using the published generalized free energy.

Sterling response:

Accepted. The final paper removes the stretched Postulate and keeps only Temporal Softening. The null is now the complete three-basin Fokker-Planck or master-equation construction, not a pairwise Kramers estimate. That matters because an ordinary coupled kinetic model should already produce channel competition. Pudding Theory is left with a smaller burden: residual endpoint curvature with finite observation-time dependence.

The falsifier has also been made operational. The observable is \(C_{\rm TS}\). It uses six directed mean first-passage times, endpoint windows of two percent around \(T_1\) and \(T_2\), the diffusion parameter \(D\), and a quadratic residual fit after subtracting the full kinetic null. This is a hard test. If the statistic vanishes under that comparison, the Postulate is not supported in this domain.

Discussion

This application is deliberately narrow. The source paper already contains the core stochastic machinery. A working paper cannot claim novelty by restating Kramers escape. The only Pudding Theory content is the assertion that endpoint stability has a measurable finite-time softening not exhausted by the full three-basin kinetic model.

The main limitation is practical. The prediction requires access to the published generalized free energy with enough numerical precision to solve the Fokker-Planck problem near saddle-node endpoints. It also requires separating true residual curvature from errors in reconstructing \(U(x,T)\), choosing basin boundaries, and estimating \(D\).

The conclusion would change if direct Fokker-Planck integration reproduced all six first-passage times across \(W_1\) and \(W_2\) with no residual curvature and no dependence on \(t_{\rm obs}\). It would also change if the apparent endpoint curvature were removed by using the exact non-quadratic barrier shape rather than the local Kramers approximation. In that case, the present paper would reduce to a methodological warning about endpoint kinetics.

References

1. Chen Ma, Pan-Pan Zhang, Bin Wu, Zhen-Ming Xu. “The Kramers escape rate of phase transitions for the 6-dimensional Gauss-Bonnet AdS black hole with triple phases.” Physics Letters B 861 (2025): 139282. arXiv:2407.20512. DOI: doi:10.1016/j.physletb.2025.139282.

2. Sterling Geisel. “Pudding Theory: A Topological Theory of Information Fields.” QBist Lab Working Paper, September 10, 2025. https://qbistlab.com/Pudding_Theory_01-2026.pdf.

3. J. D. Bekenstein. “Black holes and entropy.” Physical Review D 7 (1973): 2333.

4. D. Kubiznak and R. B. Mann. “\(P-V\) criticality of charged AdS black holes.” Journal of High Energy Physics 2012, no. 07 (2012): 033.

5. Zhen-Ming Xu. “Fokker-Planck equation for black holes in thermal potential.” Physical Review D 104 (2021): 104022.

6. Zhen-Ming Xu, Bin Wu, Wen-Li Yang. “Rate of the phase transition for a charged anti-de Sitter black hole.” Science China Physics, Mechanics & Astronomy 66 (2023): 240411.

7. H. Risken. The Fokker-Planck Equation: Methods of Solution and Applications. 2nd ed. Springer, 1988.

8. R. Zwanzig. Nonequilibrium Statistical Mechanics. Oxford University Press, 2001.