AFH* Model: Structural Consciousness Theory with ECLIPSE Validation

Research repository accompanying:
Sjöberg Tala, C.A. (2025). ECLIPSE: A systematic falsification framework for consciousness science. Submitted to Scientific Reports, under review.

---

Abstract

The AFH* (Autopsychic Fold + H* Horizon) model proposes a structural framework for consciousness emergence, integrating:

• Topological organization (κ_topo)
• Multiscale temporal coherence (Σ_stability)
• Causal information integration (Φ_integration)

This repository includes the full implementation of ECLIPSE, a falsification pipeline developed to validate AFH* across population-scale EEG data. Unlike post-hoc rationalizations, ECLIPSE enforces pre-registered, irreversible, and binary-outcome criteria.

---

Theoretical Framework

Mathematical Formulation

Consciousness ⇔ (κ_topo ≥ θ_κ) ∧ (Σ_stability ≥ θ_Σ) ∧ (Φ_integration ≥ θ_Φ)

Where: - κtopo: Ollivier-Ricci curvature of functional brain networks - Σstability: Multiscale temporal coherence across frequency bands
- Φ_integration: Transfer entropy-based directional causal integration

H* Horizon Hypothesis

Consciousness arises when neural systems cross a critical threshold in structural space, forming an Autopsychic Fold—a self-referential configuration capable of sustaining subjective experience.

---

ECLIPSE Validation Protocol

Five-Stage Falsification Framework

1. Irreversible Data Partitioning – Cryptographic split ensures zero optimization leakage
2. Pre-Registered Thresholds – Targets: F1 ≥ 0.60, Precision ≥ 0.70, Recall ≥ 0.50
3. Clean Development – Separation of exploratory and confirmatory phases
4. Single-Shot Validation – One-time test on holdout; no iterative tuning
5. Binary Assessment – Pass/fail outcome removes interpretative ambiguity

Dataset: Sleep-EDF Expanded (PhysioNet)

| Metric | Value | |--------|-------| | Subjects | 153 full-night EEG recordings | | Epochs | 126,160 × 30-second segments | | States | Wake/REM (conscious) vs N1/N2/N3 (unconscious) |

---

Empirical Results

| Version | Scope | F1 Score | Precision | Recall | Status | |---------|-------|----------|-----------|--------|---------| | v3.7 | Basic AFH* prototype | 0.031 | 0.032 | 0.030 | ❌ Falsified | | v4.1 | Refined topological operations | 0.037 | 0.025 | 0.069 | ❌ Falsified |

Interpretation

Systematic falsification confirms empirical limits of partial implementations and validates the robustness of the ECLIPSE methodology.

---

Repository Structure

AFH-MODEL/ ├── README.md → Main project overview ├── requirements.txt → Python dependencies ├── LICENSE → MIT license ├── .gitignore → Version control exclusions ├── CITATION.cff → Citation metadata ├── NISA'S CODE → Alternative implementations │ ├── 1. Docs/ → Presentations & conceptual material │ ├── figures/ → Scientific illustrations │ ├── 1.1 presentation.md → Lab presentation script │ ├── 1.2 glossary.md → Terms and variables │ ├── 1.3 ethical_manifesto.md → Ethical commitments │ └── 1.4 acknowledgments.md → Scientific credits │ ├── 2. Theory/ → Core theoretical files │ ├── 2.1 model_architecture.md → AFH* formalism │ ├── 2.2 horizon_thresholds.md → H* criticality │ ├── 2.3 resonant_residue.md → Symbolic divergence │ └── 2.4 falsifiability_criteria.md → Empirical test design │ ├── 3. Experimentation/ → ECLIPSE pipeline │ ├── 3.1 COMPUTATIONAL STUFF/ → Main implementations │ │ ├── legacy/ → Older versions │ │ ├── results 3.7/ → Output v3.7 │ │ ├── results 4.1/ → Output v4.1 │ │ ├── ECLIPSE 3.7.py → Code v3.7 │ │ └── ECLIPSE 4.1.py → Code v4.1 │ └── README.txt → Pipeline description │ └── 4. Annexes/ └── 4_open_questions.md → Theoretical frontiers

---

Quick Start

Setup

bash pip install -r requirements.txt

Run Validation

bash cd "3.Experimentation/3.1 COMPUTATIONAL STUFF/" python "ECLIPSE 4.1.py" --model afh --validation complete

Outputs Include:

• Holdout predictions
• Cryptographic SHA256 checksum
• Automatic threshold comparison
• Reproducibility seed log (Seed 2025)

---

Scientific Contributions

AFH Model*

• Structural theory of consciousness grounded in topology, causality, and temporal dynamics
• Explicit, falsifiable predictions across neurotypical populations

ECLIPSE Pipeline

• First systematic falsification tool for consciousness science
• Avoids statistical artifacts and overfitting
• Theory-agnostic: applicable to IIT, GWT, QSP, and beyond

---

Clinical & Technological Applications

| Domain | Application | |--------|-------------| | Anesthesia | Structural detection of loss-of-consciousness transitions | | Disorders of Consciousness | Diagnosis beyond behavioral inference | | BCI | Integration into neural decoding pipelines | | Pharmacology | Mechanistic assessment of consciousness-modulating compounds |

---

Citation

bibtex @article{sjoberg2025eclipse, title={ECLIPSE: A systematic falsification framework for consciousness science}, author={Sjöberg Tala, Camilo Alejandro}, journal={Scientific Reports}, year={2025}, note={Submitted – under review}, doi={10.5281/zenodo.15541550} }

---

Author

Dr. Camilo Alejandro Sjöberg Tala, M.D.
Independent Researcher – Viña del Mar, Chile
Email: cst@afhmodel.org
ORCID: https://orcid.org/0009-0009-6052-0212

---

Future Directions

• Implementation of fold detection (ψ) and symbolic resonance (∇Φ)
• Multimodal datasets (fMRI, MEG) for structural convergence
• Experimental paradigms targeting causal topology transitions
• Cross-cultural validation and developmental neuroscience applications

---

"The systematic falsification of AFH v3.7 and v4.1 represents methodological success, not theoretical failure. ECLIPSE ensures consciousness science enters its post-hoc-free era."

"Cogito ergo sum, quamodo sum?" — AFH-R Heuristic