Scale-Free Metabolic Networks Should Amplify Weak Coherent Bias Under Nutrient Scarcity

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

Abstract

Mitsumoto and Ishihara derive a dynamical mean-field theory for dense catalytic reaction networks with arbitrary degree distributions. Their central result is topological and dynamical. Homogeneous networks undergo a metabolic-starvation transition under poor nutrient supply, while networks with scale-free out-degree distributions avoid this transition. The mechanism is not the usual hub-dominated ordering seen in epidemic or Ising models. It arises from chemical species with extremely small out-degree, which retain catalytic abundance and prevent cellular shrinkage. This Working Paper applies the Chaos Susceptibility Postulate of Pudding Theory to that result. The prediction is narrow. Weak coherent informational bias should be most visible near the low-nutrient starvation boundary, where the dynamical system already has high susceptibility. It should be suppressed in nutrient-rich overnutrition regimes. The distinguishing observable is a topology-dependent shift in steady growth rate under controlled coherent perturbation.

Source Synopsis

Mitsumoto and Ishihara study a minimal intracellular catalytic reaction network. A cell contains chemical species divided into unpenetrable metabolic products, penetrable metabolic products, and nutrients. Catalytic reactions have the form \(j+k\to i+k\), where \(k\) catalyzes conversion of species \(j\) into species \(i\). Nutrients can enter from the environment, penetrable products can leave, and all abundances are diluted by cell growth.

The authors analyze dense random networks using dynamical mean-field theory. This gives exact effective equations conditioned on in-degree and out-degree for arbitrary degree distributions. The fixed points fall into three regimes: metabolic, overnutrition, and starvation. The overnutrition transition occurs at high nutrient supply and is independent of degree distribution. The starvation transition occurs at low nutrient supply and depends strongly on out-degree topology.

For homogeneous Poisson out-degree distributions, the metabolic state disappears below a finite nutrient threshold. For increasingly heterogeneous uniform distributions, that boundary shifts downward. For the singular uniform case with full support at zero out-degree, and for power-law out-degree distributions, the starvation transition disappears. The growth rate becomes extremely small at poor nutrient supply but remains positive for any \(\alpha_3 g>0\).

The paper identifies the mathematical origin of this suppression. The function \(X(\mu^\ast,m^\ast_{\mathrm{cat}})\) diverges logarithmically as \(\mu^\ast\to0^+\) when the out-degree distribution includes sufficiently many species with \(v/c\to0\). These species lose little material through outgoing reactions. If they are unpenetrable metabolic products, they preserve catalytic activity inside the cell.

The authors also show that abundance statistics reflect in-degree topology. Since fixed-point abundance is proportional to in-degree, a power-law in-degree distribution produces a power-law abundance distribution with the same exponent. Numerical simulations on Erdos-Renyi and directed Barabasi-Albert networks support the dense-limit theory down to moderate connectivity.

Postulate Lens

This paper calls for the Chaos Susceptibility Postulate: highly chaotic systems are most susceptible to small coherent inputs.

The source model is not a consciousness experiment. It is a dynamical systems paper. The relevant point is susceptibility. Near the starvation boundary, the metabolic network is not merely noisy. It is close to a qualitative transition where small changes in catalytic balance determine whether growth remains positive. Mitsumoto and Ishihara show that topology changes this susceptibility. Power-law out-degree structure removes the finite starvation threshold by preserving mass in low-out-degree catalytic products.

The Postulate applies because Pudding Theory predicts that weak coherent bias should not move stable systems by direct force. It should alter branching probabilities most strongly where the target dynamics already amplify microscopic differences. In the present domain, the amplification coordinate is not a mechanical Lyapunov exponent reported by the source paper. It is the divergence structure of the low-nutrient mean-field equations and the persistence of positive \(\mu^\ast\) close to \(\alpha_3 g=0\).

Thus the source paper gives a clean theoretical substrate for a restricted Pudding Theory claim. Topology determines whether weak perturbations vanish or are promoted into macroscopic metabolic persistence.

Pudding Theory Prediction

Pudding Theory predicts a topology-dependent response to weak coherent informational perturbation in catalytic metabolic networks under nutrient scarcity. Consider an experimental or simulated catalytic network with fixed \(\alpha_1,\alpha_2,\alpha_3,d\), and with nutrient supply set just above the homogeneous-network starvation boundary. The perturbation must be small enough that it does not change reaction stoichiometry, nutrient concentration, permeability, or mean degree. It acts only as a weak bias on transition rates among near-degenerate catalytic pathways.

For homogeneous out-degree networks, the predicted shift in steady growth rate \(\Delta\mu^\ast\) should be small and should vanish rapidly as nutrient supply decreases below the usual starvation threshold. For scale-free out-degree networks, the same perturbation should produce a larger positive shift in \(\mu^\ast\), concentrated in species with small out-degree and nonzero catalytic contribution. The effect should grow as the measured low-nutrient susceptibility grows, not as total network size alone grows.

The prediction is also asymmetric across regimes. In nutrient-rich conditions near the overnutrition transition, the source theory shows that increased heterogeneity suppresses growth and that the transition boundary is independent of degree distribution. Pudding Theory therefore predicts little or no coherent-bias advantage there. The perturbation should not rescue metabolism from overnutrition, because the catalytic sector is being washed out by nutrient dominance rather than being balanced at a starvation edge.

The strongest test is not a single endpoint. It is a response surface. One should compare \(\Delta\mu^\ast(\alpha_3 g,d)\) across Poisson, broad uniform, and power-law out-degree ensembles with matched mean connectivity. The Pudding Theory signal is a selective excess response in the low-nutrient, high-susceptibility sector. It is not a universal growth increase.

Falsifiable Observable

The distinguishing observable is the low-nutrient susceptibility ratio \(R=\Delta\mu^\ast_{\mathrm{scale-free}}/\Delta\mu^\ast_{\mathrm{Poisson}}\), measured under matched weak coherent perturbation at equal \(N,c,\alpha_n,d,\alpha_3 g\), with \(\alpha_3 g\) chosen near the Poisson starvation boundary. If the low-nutrient susceptibility ratio R were measured to be 1.0 ± 0.1, this Postulate would be falsified. The expected Pudding Theory result is \(R>1\), with the excess localized to networks containing low-out-degree catalytic products.

Editorial Dialogue

Tanaka: The source paper already explains the phenomenon. Low out-degree products retain catalytic mass. The logarithmic divergence of \(X(\mu^\ast,m^\ast_{\mathrm{cat}})\) is sufficient. Nothing in the derivation requires Pudding Theory.

Sterling: Correct. The source result stands without Pudding Theory. The question here is different. It asks where weak coherent bias would be detectable if such bias exists. The answer is constrained by the source model. It should appear near the low-nutrient metabolic boundary and should depend on out-degree topology.

Tanaka: That sounds like relabeling sensitivity as a Postulate.

Sterling: It would be relabeling if no new observable were proposed. The proposed observable is comparative: a perturbation-induced \(\Delta\mu^\ast\) ratio between matched scale-free and homogeneous networks. Standard DMFT predicts the unperturbed topology effect. Pudding Theory predicts an additional coherent-bias response that scales with the same susceptibility structure.

Tanaka: And if no excess response appears?

Sterling: Then this application fails. The source paper gives the correct null model. That is why the test is useful.

Discussion

This Working Paper does not claim that Mitsumoto and Ishihara observed Pudding Theory effects. They did not. Their work supplies a dynamical map of where such effects should and should not appear.

The main limitation is operational. A coherent informational perturbation must be defined without changing ordinary biochemical control parameters. In simulation, this can be implemented as a small phase-locked bias in selected reaction propensities. In wet metabolism, the analogue would be harder to isolate because temperature, nutrient availability, enzyme expression, and growth dilution all produce ordinary shifts in \(\mu^\ast\).

The second limitation is topology. Real metabolic networks are not typical random graphs. They contain evolved modules, correlations, and chemically constrained hubs. Mitsumoto and Ishihara note that typical power-law ensembles are baselines, not complete models of living cells.

The conclusion would change if matched perturbation experiments found equal responses across Poisson and scale-free out-degree networks, or if the response appeared primarily in nutrient-rich overnutrition regimes. Either result would break the proposed link between susceptibility and coherent bias.

References

1. Kota Mitsumoto and Shuji Ishihara. “Starvation suppression in scale-free metabolic networks: Dynamical mean-field analysis of dense catalytic reaction networks.” arXiv:2603.19850, 2026. DOI: doi:10.48550/arxiv.2603.19850.

2. Sterling Geisel. “Pudding Theory: A Topological Theory of Information Fields.” QBist Lab, September 10, 2025.

3. S. A. Kauffman. The Origins of Order: Self-Organization and Selection in Evolution. Oxford University Press, 1993.

4. A.-L. Barabasi. Network Science. Cambridge University Press, 2016.

5. H. Jeong, B. Tombor, R. Albert, Z. N. Oltvai, and A.-L. Barabasi. “The large-scale organization of metabolic networks.” Nature 407, 651-654, 2000.

6. C. Furusawa and K. Kaneko. “Evolutionary origin of power-laws in a biochemical reaction network: Embedding the distribution of abundance into topology.” Physical Review E 73, 011912, 2006.

7. C. Furusawa and K. Kaneko. “Adaptation to optimal cell growth through self-organized criticality.” Physical Review Letters 108, 208103, 2012.

8. R. Pastor-Satorras and A. Vespignani. “Epidemic spreading in scale-free networks.” Physical Review Letters 86, 3200, 2001.