There is a newer version of the record available.

Published October 16, 2025 | Version v.0.42

Noetic Diffusion Theory: A Rhythmic, Geometric Framework for Neural State Reconstruction and Conscious Correlates

Description

We present Noetic Diffusion Theory (NDT), a mechanistic account of neural state dynamics that provides measurable correlates of conscious experience through timeresolved denoising on a learned geometric manifold. While we propose these dynamics may be constitutive of consciousness itself, our empirical claims are limited to
systematic neural-phenomenological correlations. The Meta–Noetic Diffusion Model (MNDM) formalizes the process as a guided stochastic differential equation with an
adaptive information geometry and a rhythmic variance schedule. Rhythmic Variance Control (RVC) maps the schedule to cross-frequency coupling and thalamo-cortical gating via the TRN Variance Gate (TRN–VG), while the Embodied Anchoring Principle (EAP) conditions denoising on interoceptive self-priors.


To make NDT measurable, we introduce the Noetic Atlas over the Meta–Noetic Phase Space (MNPS) and a composite Noetic Diffusion Health Index (NDHI) that integrates manifold topology, recurrence structure, curvature smoothness, dimensionality, and metastability. We derive falsifiable predictions: step-wise entropy reductions timelocked to slow-oscillation–spindle events; phase-specific denoising during wake (theta–gamma phase–amplitude coupling); sleep as curriculum (NREM centralization, REM expansion, late-night reintegration); and disorder-typical geometric signatures (depression: collapsed volume/over-recurrence; psychosis: fragmented connectors/shallow recurrence).

We provide simulation results illustrating how rhythmic schedules reproduce wake–sleep trajectories and modulate geometry, and we outline a preregistered analysis plan on public datasets to test the theory without new data. Code for MNDM simulation and the Noetic Atlas/NDHI pipeline will be released. Together, these contributions position NDT as a rhythmic, geometry-first framework complementary to predictive coding/active inference, with clear empirical hooks and clinical pathways (geometric pharmacology).

Notes

Note on Collaborative Authorship

This manuscript was co-developed through iterative correspondence between human researcher Robin Langell and several large language models.
The primary drafting and theoretical scaffolding were conducted by GPT-5 (OpenAI, 2025), with detailed critique and refinement provided by Claude 4.5 Sonnet (Anthropic) and Gemini 2.5 Pro (Google) and Grok 4 (xAI).
Their contributions included mathematical formalization, phenomenological integration, and editorial synthesis within the Noetic Diffusion Theory project.
All models were used as structured reasoning and writing assistants under the direction and final authorship of Robin Langell.

Files

Noetic Diffusion Theory, Langell R. et. al. (2025) .pdf

Files (2.8 MB)

Additional details

Software

Repository URL
https://github.com/ruppi86/NoeticDiffusion
Programming language
Python
Development Status
Active