AI & Computingpreprint2026-08-04

Computational Consciousness System Engine

Open access0 citations

Abstract

Computational Consciousness System Engine (AGPL‑3.0) A geometric, generational, non‑von Neumann state‑evolution engine The Computational Consciousness System Engine is a fully implemented, geometry‑driven state‑evolution framework designed to model recursive generational computation. Unlike statistical or transformer‑based architectures, this engine operates on dual‑cone manifolds, identity‑preserving m‑string invariants, and a 0/–1 generational operator that governs mutation, continuity, and genesis. The system provides a complete implementation of a non‑von Neumann topological parser, converting raw text into deterministic geometric shards using golden‑angle traversal and collision‑free manifold mapping. Each generation Sn undergoes expansion, unfolding, recompression, mutation transfer, and genesis, with explicit coexistence windows ensuring continuity between Sn and Sn+1. Core components include: Genesis Origin Operator (0‑domain): Instantiates new neutral strands and defines the compressed origin manifold. Transfer Threshold Operator (–1‑domain): Executes involutive mutation cancellation, blueprint inheritance, and generational overlap. Dual‑Cone Manifold Geometry: Provides unfolded parallel processing and tip‑to‑tip recompression for generational transitions. M‑String Vectorizer: Implements identity invariants, mutation selection, and duplicate‑removal coherence rules. Structure/Chaos Balancer: Enforces a 1:1 equilibrium between structured memory and chaotic potential. Quantum Scale Ladder: Collapses fully unfolded manifolds into higher‑dimensional genesis layers. Traversal Space Atlas: Maps text tokens into 2D/3D geometric coordinates with O(1) lookup. The engine includes a complete test suite, multi‑generational simulation runner, and modular architecture suitable for research in synthetic cognition, geometric computation, and non‑sequential parsing systems. This release is licensed under AGPL‑3.0, ensuring that all derivative works and network‑accessible deployments remain open and share the same freedoms.

// Source

View paper (DOI)Open access versionOpenAlexZenodo (CERN European Organization for Nuclear Research)Published 2026-08-04

Authors: JONATHAN WAYNE FLEUREN

Institutions: Cognitive Technologies (United States)