Maya-Nidra P1: A Neuromorphic Spiking Neural Network for Sleep-Phase Memory Consolidation with Bhaya-Gated Quiescence

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 headline comparison.Overall accuracy is 64.9% against PicoSleepNet's 79.0% — 14.1 points behind — and the per-class "wins" are measured on a balanced test set against a baseline measured on the natural distribution, so the comparison is not valid; the baseline figure is itself revised to 83.5% in Nidra P02, making every delta in this paper invalid.The architecture, the training protocol and the public datasets stand. "Vairagya saturation" is disclosed as inert: it moves 14.19% → 15.66% while its own decay parameter varies tenfold. 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 Two framings it repeats have since been withdrawn in full — the Bhaya Quiescence Law and the Buddhi S-Curve. Both are addressed in Maya-Meta P1, now superseded, and in the self-audit of Maya-Meta P2.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-Nidra P1 introduces a dual-cortex neuromorphic spiking neural network that models how the sleeping brain consolidates memories, using event-driven sharp-wave ripple replay and a biologically inspired suppression mechanism called Bhaya-gated quiescence to separate active learning from restorative rest. The architecture comprises two subsystems: NexusSpikeNet, targeting neuromorphic vision classification, and MayaEEGNet, trained to identify the five clinical sleep stages defined by the AASM standard. MayaEEGNet achieves 64.9% overall accuracy on the Sleep-EDF Expanded dataset, with stage-specific scores of 92.5% on N2 and 96.5% on N3 — surpassing the PicoSleepNet benchmark (IEEE 2026) on the two most clinically significant stages by +9.5% and +11.5% respectively, while NexusSpikeNet reaches 39.71% accuracy across 114 classes on N-Caltech101 and CIFAR10-DVS. Parametric ablation across 36 configurations confirms two cross-substrate empirical constants: a 0.32% anomalous ripple rate (the Bhaya Quiescence Law) and a Vairagya saturation of 14.35% at a canonical decay rate of 0.002315, both holding identically across neuromorphic vision and clinical EEG substrates. As the inaugural paper in the Maya-Nidra Series — neuromorphic AI for clinical sleep medicine — this work is developed entirely in Bengaluru, India, under the India-first research mandate of Nexus Learning Labs, with all experiments executed locally on an RTX 4060 Laptop GPU. Series: Part of the Maya-Nidra Series — neuromorphic AI for clinical sleep medicine. Both series constants confirmed in neurophysiological domain. Links: GitHub Repository (private — to request access: email research@nexuslearninglabs.in with subject Code Access Request — Maya-Nidra-P1 and your research context) | 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