What question did this study set out to answer?

To introduce Maya-OS, an innovative arbitration layer utilizing an affective spiking neural network for real-time operations.

March 24, 2026Open Access

Maya-OS: An Affective Spiking Neural Network as a Conversational Operating System Arbitration Layer

Key Points

To introduce Maya-OS, an innovative arbitration layer utilizing an affective spiking neural network for real-time operations.
Developed a four-neuron affective spiking neural network.
Implemented with real-time metrics reading from the operating system.
Conducted five experimental sessions focusing on system responsiveness and safety.
Demonstrated effective process scheduling based on conversational intents.
Showed evidence of Shraddha-mediated escalation suppression.
Confirmed discharge of LIF action potentials during operation.

Abstract

We present Maya-OS, the first implementation of a four-neuron affective Spiking Neural Network deployed as a real-time conversational operating system arbitration layer. Four LIF neurons — Bhaya (fear, τ=3), Vairagya (wisdom, τ=20), Shraddha (trust, τ=10), and Spanda (aliveness, τ=5) — continuously read live system metrics and conversational intent signals, producing an affective voltage vector that governs process scheduling, resource protection, and syscall arbitration decisions without hardcoded rules. Across five experimental sessions (133 ticks, Windows 11), we demonstrate autonomous threat accumulation, Shraddha-mediated escalation suppression as an emergent safety primitive, confirmed LIF action potential discharge, intent-driven neural modulation, and autonomous homeostatic recovery. No prior work has deployed an affective SNN as a live OS arbitration layer. Extends Swaminathan (2026) https://doi.org/10.5281/zenodo.19151563. Codebase: https://github.com/venky2099/Maya-OS

Maya-OS: An Affective Spiking Neural Network as a Conversational Operating System Arbitration Layer

Key Points

Abstract

Cite This Study