Maya-Morphe P2: Spanda — Adaptive Bioelectric Topology Learning via Gradient-Guided Voltage Thresholds and Dynamic Edge Formation

Version 2 (2026-08-04). Revised to the NLL Universal Paper Format v6. The original text is retained in full; nothing has been deleted. Corrections appear as marked blocks placed at the section that carries the claim, and each one states what the paper said, what the data show, the corrected claim, why it happened, and what still stands. This version withdraws the discovered-constant claim.THRESHOLD_INIT = 0.18 and the training log shows thresholds starting at exactly 0.18 and settling at 0.165–0.167, so the "optimal threshold ~0.163" is a 0.017 drift from its own initialisation; FRR is 100.0% in both the learned and fixed arms at every scale, so the difference has no observable consequence, and P04 later drives the same value to 0.127–0.144.The GraphSAGE implementation and the scale sweep stand — though 100k and 120k resolve to the same 266,071-cell grid, so there are six distinct scales, not seven. Everything below this line is the original description from version 1. It is retained unchanged for the record. Where it conflicts with the corrections above, the corrections stand. Two framings it repeats have since been withdrawn in full — the Bhaya Quiescence Law and the Buddhi S-Curve. Both are addressed in Maya-Meta P1, now superseded, and in the self-audit of Maya-Meta P2. Maya-Morphe Paper 2 introduces SpandaNet, a graph neural network framework that replaces the fixed voltage revival threshold from Paper 1 with a learnable parameter, trained end-to-end via a differentiable objective across seven progressively larger grid scales, asking not merely whether voltage-driven wound repair works but what the optimal threshold actually is. Across 252 systematic trials spanning seven scales, three conditions, four damage fractions, and three random seeds — covering grids from 29,056 to 266,071 cells — gradient descent consistently converges on a threshold of approximately 0.163, roughly 0.017 below the hardcoded value of 0.18 used in Paper 1, while achieving 100% Functional Recovery Rate under both learned and fixed-threshold dynamic-topology conditions versus 0% under fixed topology. This scale-invariance result — the same learned threshold emerging reliably across all seven scales — establishes a principled, data-driven foundation for voltage parameter selection in morphogenetic computing. The Bhaya Quiescence Law, modified at approximately 15.4% elevated β during active wound repair, is confirmed across all seven scales, consistent with Paper 1 and predicted by the biological substrate, with the Vairagya decay constant (0.002315) embedded as tamper-evident provenance across all model parameters. As the second installment in the Maya-Morphe series on morphogenetic neural networks for biological wound repair simulation, this paper establishes the scalable multi-scale evaluation protocol that downstream papers in the series will build upon. Series: Part of the Maya-Morphe Series — morphogenetic neural networks for biological wound repair simulation. Bhaya Quiescence Law applies in modified form: dynamic-topology substrates show elevated β during active wound repair — biologically predicted and validated. Links: GitHub Repository (private — to request access: email research@nexuslearninglabs.in with subject Code Access Request — Maya-Morphe-P2 and your research context) | Interactive Dashboard | FAQ | Full Series Index — venky2099.github.io Nexus Learning Labs, Bengaluru · UDYAM-KR-02-0122422 · BHASKAR IN-0526-9452JSORCID: 0000-0002-3315-7907 · VAIRAGYA_DECAY_RATE = 0.002315 — an ORCID-derived provenance mark, not an experimental parameter. Each paper's disclosure block states whether it reached a result in that paper.Canary: MayaNexusVS2026NLL_Bengaluru