Hallucinatory Domestication Theory (HDT): From In Silico Mechanism to In Vivo EEG Validation — Reproducible Package

This release documents the first in vivo empirical validation of the Hallucinatory Domestication Theory (HDT), previously established through a biologically constrained in silico computational framework. The objective of this publication is to demonstrate that HDT’s mechanistic predictions are not confined to simulation, but are operationally testable and observable in real human neural recordings.

HDT proposes that meaningful neural activity emerges through a structured dynamical cascade: generative high-entropy neural dynamics are stabilized through inhibitory domestication, crystallize into low-dimensional attractor states, become registered at the level of operational existence (rate coding), and are subsequently organized and coordinated via oscillatory framing and synchrony. A core quantitative prediction of this framework is that meaningful processing is marked not by sustained high complexity, but by a directed entropy collapse following structural stabilization.

To test these predictions, we applied HDT-consistent metrics to a public auditory oddball EEG dataset (OpenNeuro ds003061). The validation pipeline operationalizes dynamic entropy using normalized Lempel–Ziv complexity (H_dyn), entropy collapse rate (D = dH_dyn/dt), phase-locking value (synchrony, S), phase–amplitude coupling (cross-frequency coupling, CFC), and a correlation-based proxy for attractor stability (A). All metrics were implemented in direct correspondence with the in silico model to enable quantitative cross-level comparison.

The empirical results confirm HDT’s central dynamical predictions:

• Target stimuli exhibit approximately 21% entropy reduction (H_dyn decreasing from ~2.36 to ~1.86).
• Collapse rates reach peak negative values near −2.0, indicating active compression rather than static entropy differences.
• Temporal cascade analyses support the predicted ordering: cross-frequency coupling and synchrony precede or co-emerge with attractor stabilization, which in turn precedes entropy collapse.
• The lag between attractor stabilization and collapse (~0.20 s, r ≈ 0.90) demonstrates strong temporal alignment with the mechanistic model.
• Non-target trials do not exhibit the same structured cascade.

This convergence between in silico predictions and in vivo observations constitutes a cross-level mechanistic alignment rather than a post hoc interpretive match. While the present release represents a proof-of-principle single-subject validation, it establishes quantitative benchmarks and a fully reproducible operational pipeline for future multi-subject and comparative investigations.

The package includes:

• Full preprint manuscript detailing theoretical framework and empirical findings
• In Vivo Operational Guide with step-by-step reproducibility instructions
• Executable notebook implementing the full validation pipeline
• Exported CSV artifacts and figures
• Environment specifications for deterministic replication

The purpose of this release is to establish that HDT is an empirically testable and quantitatively grounded theory of neural meaning formation, bridging computational neuroscience and human electrophysiology. The results do not claim population-level generalization, clinical validity, or theoretical exclusivity; rather, they demonstrate that HDT generates precise, falsifiable, and reproducible predictions that manifest in real neural data.

This work marks the transition of HDT from a purely computational proposal to a cross-domain validated framework and provides a structured foundation for future experimental, comparative, and large-scale investigations.