The Three Treasures as Structural Heuristics of an Informational Nucleus: Jing-rho, Qi-v_IC, Shen-Phi - One Mapping Sustained, One Weakened, One Demoted, and the Boundary It Reveals
Zenodo (CERN European Organization for Nuclear Research) September 9, 2026 DOI: 10.5281/zenodo.22680858 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Preprint Preregistered Peer reviewed |
|---|---|
| Sample size | 47 |
| Population | Expert contemplative practitioners from three traditions (N=47) and matched controls (N=16); additional expert sample (N=46) for the absorption task; N=15 for the audio-guided meditation task |
| Measures | spectral condensation of the cortical correlation matrix (dC), breathing-heart coherence in the baroreflex resonance band, Phi_vN, gamma trait |
| Key points | The authors report that a substrate-level marker (Jing-rho) is supported: expert meditators showed higher spectral condensation of cortical correlation matrices than controls (dC=+0.072, d=+0.85, p=0.0053), with no internal dose gradient. A process-level marker (Qi-v_IC) was weakened: audio-guided meditation raised breathing-heart coherence from 0.111 to 0.599, but only 7/15 reached the resonance band. An apex-level marker (Shen-Phi) was demoted: absorption did not raise Phi_vN, which the authors argue indexes sleep-stage transitions instead. |
Abstract
Contemplative traditions of China and India developed detailed cartographies of internal states - the Three Treasures (Jing/Qi/Shen) of Taoist neidan and the Kundalini circuit of tantric yoga - without measurable operational definition. We proposed these cartographies as structural heuristics of an informational nucleus (density matrix rho, velocity field v_IC, integrated differentiation Phi, structural binding O_info), each with an explicit demarcation and a falsifiable order-of-magnitude prediction (P1-P4), and we then executed the predictions against public datasets. Jing-rho (substrate-substrate) is sustained: expert contemplative practitioners (N=47, three traditions, 576-43,800 lifetime hours) show higher spectral condensation of the cortical correlation matrix than matched controls (N=16) in a common resting condition (dC=+0.072, 7x the pre-registered threshold; d=+0.85, p=0.0053), homogeneous across traditions, with no internal dose gradient - threshold acquisition, not accumulation. Qi-v_IC (process-process) is weakened, not refuted: audio-guided meditation without pacing raises breathing-heart coherence in the baroreflex resonance band from 0.111 to 0.599 (d_z=+1.616, p=0.00012, N=15), but only 7/15 subjects reach the resonance band. Shen-Phi (apex-apex) is demoted on the state branch: in 46 experts under self-induced absorption vs instructed mind-wandering, Phi_vN does not rise (d_z=-0.286, p=0.086; power 0.92), the dose gradient is null, and the pipeline was validated against the dataset's own published gamma trait (p=0.00089) - the boundary belongs to the construct: Phi is an observable of level transitions (wake-NREM-REM), not of the contemplative axis. A boundary theorem completes the demotion: the three spectral faces (Phi, C, O) are deterministic functions of one eigenvalue spectrum, so between-face coupling is inaccessible by construction. The cartography is structurally valid but partially translatable: the measurable return happens in the substrate, the process flows, and the apex is not where the tradition places it. Preprint of a manuscript under journal submission. All datasets are public; analysis code at Zenodo 22178708.