We report a frozen diagnostic study of galaxy rotation curves in the SPARC sample within the Viscous Emergent Spacetime (VES) framework. The analysis uses the SPARC rotation-curve sample of 175 disk galaxies with Spitzer 3.6 µm photometry and accurate H I/Hα rotation curves, introduced by Lelli, McGaugh, and Schombert. The purpose of this paper is not to claim a discovery or a completed alternative to dark matter, but to document a reproducible Stage 5B diagnostic in which the response of an informational medium is represented by a conditional leakage-gated bidirectional susceptibility. The model is frozen after a metadata audit and is evaluated against a fixed baryonic baseline with no per-galaxy optimization in the diagnostic stage. In the full sample, the frozen Stage 5B proxy improves over the baryonic baseline in 172 of 175 galaxies, reducing the median RMSE from 40.95 to 33.87. Using an official SPARC metadata-derived purity audit, the clean subset contains 94 galaxies, all of which improve relative to the frozen baryonic baseline. In this clean subset, the median RMSE decreases from 46.98 to 39.32. The three remaining failures are concentrated in low-quality, sparse, or baryonic inner-overshoot cases rather than in a loss of outer support. These results should be read as a structured diagnostic of a model architecture and data-quality filter, not as physical validation. Required next steps include comparison against standard baselines, bootstrap and jackknife robustness tests, and replacement of the algebraic radial proxy by a full elliptic or radial PDE solver.
Mikheil Rusishvili (Sun,) studied this question.