Maya-Vaidya P1: A Neuromorphic Dual-Head Spiking Neural Network for Multimodal Pain Classification and Severity Estimation with BCM Chronification Tracking

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 neuromorphic-constant claim and restates the result.GEV is the model's own mean hidden firing rate, and the spike function carries no surrogate gradient, so the encoder never trains — an untrained network on Gaussian noise reproduces GEV ≈ 0.368; the constant therefore needs neither the data nor the training.Random Forest on the same features and protocol scores 0.5465 / 0.4707 against the SNN's 0.4302 / 0.3714, and that must be the stated result. I ran and published that baseline myself, which is what makes this correctable. 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 Two framings it repeats have since been withdrawn in full — the Bhaya Quiescence Law and the Buddhi S-Curve. Both are addressed in Maya-Meta P1, now superseded, and in the self-audit of Maya-Meta P2.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-Vaidya P1 introduces a dual-head spiking neural network (SNN) designed to simultaneously classify pain type and estimate pain severity from physiological signals, addressing the need for neurobiologically grounded computational models of clinical pain that can generalise across individuals. The architecture combines leaky integrate-and-fire (LIF) neurons with a BCM ChronificationTracker — a novel module that models the acute-to-chronic pain transition through a sliding modification threshold — trained and evaluated using rigorous Leave-One-Subject-Out cross-validation across 86 participants drawn from the PhysioPain dataset. Across all 86 independent LOSO folds, the model achieved a cross-subject GEV (Global Excitatory Voltage) convergence of 0.3693 ± 0.0197 — invariant across subjects, pain types, and all five ablation conditions — alongside 4-class pain classification accuracy of 43.0% (macro F1 = 0.371), headache detection F1 = 0.576, and pain severity MAE = 0.759 on a 0–5 Likert scale, outperforming an MLP baseline. The GEV invariance replicates the Bhaya Quiescence Law confirmed across 14 or more prior Maya substrates, and the embedded decay constant PAIN_DECAY_RATE = 0.002315 serves as a canonical IP anchor across both loss weighting and membrane dynamics. As the first entry in the Maya-Vaidya series, this work advances India-first neuromorphic AI for clinical pain analysis, offering an open research prototype toward accessible, physiologically interpretable pain assessment tools. This is a research prototype only and is not a licensed medical device; for medical emergencies in India, please contact 112. Series: Part of the Maya-Vaidya Series — neuromorphic AI for clinical pain analysis. India-first. Research prototype only — NOT a licensed medical device. For medical emergencies in India: 112. Links: GitHub Repository (private — to request access: email research@nexuslearninglabs.in with subject Code Access Request — Maya-Vaidya-P1 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