V2R: Resonant Convergence in a Dynamically Coupled Phase Field Revisited — Claim-Bounded Numerical Audit of Model Closure, Associative Completion, Distributed Carrier Reorganization, and Directional Rejection

Abstract

V2R is a structured numerical revision of the historical V2 study, Resonant Convergence in a Dynamically Coupled Field (2025). The original work contained several under-specified elements, including an undefined feedback function, a quartic spatial expression without a unique dynamical generator, and a Gaussian-like sampling kernel without a complete decoder. V2R therefore does not attempt a literal reconstruction. Instead, it separates historical definitions from explicitly declared classical surrogate closures and evaluates all resulting claims through a gated numerical evidence programme. The revision establishes an internally auditable phase-field surrogate and characterizes bounded associative completion, multi-template recognition and open-set rejection, cue-assisted stress survival, finite tested sequential capacity, time-dependent distributed carrier reorganization, historical phase-key function, phase-key/late-γ functional dissociation, sparse competitor veto, opponent-indexed negative-space rejection, and signed directional susceptibility. It also records several negative and superseded results: the historical quartic localization claim is not supported in the tested regime; simple state transplantation does not establish portable memory identity; opponent-exclusive territory is not a unique memory locus; and historical state packets or source directions are not freely transplantable. The final exact-geometry 14-seed replication supports role-conditioned local re-instantiation of the recipient-to-competitor direction at another territory with the same relational role. This result is explicitly interpreted as a geometry-constrained relational susceptibility, not as portable memory. V2R also corrects the historical interpretation of a reported 50.1% accuracy after exact 15% corruption: because such corruption leaves the input at 85% accuracy by construction, the historical value cannot demonstrate restoration relative to the corrupted-input baseline. The reported results apply only to the declared classical numerical surrogate. They do not demonstrate a biological memory mechanism, quantum or plasma dynamics, or a physical hardware implementation. Exact configurations, seed schedules, source code, raw records, and implementation-sensitive reconstruction details are not publicly distributed.