DUAL-TRACK: Pairing a Reserve Biomarker with HRV Function in n=1 Self-Tracking

DUAL-TRACK is a general method paper. It describes a way of tracking one person with consumer devices. A slow-changing "reserve" biomarker is paired with a fast-changing heart rate variability "function" biomarker. The relationship between the two is analysed separately depending on whether the tracked system is stable or actively changing, and is not pooled. The method is demonstrated on one participant's daily data (May 18 to September 21, 2026). The reserve marker is the skin carotenoid score from the Prysm iO device (Nu Skin). The function marker is HRV Total Power, defined here as LF+HF power with VLF excluded (explained in the text), with RMSSD and Baevsky's Stress Index as additional variables. In this participant, the relationship between the two looked different in the stable and perturbation windows. A curved relationship was seen in the original Stable window. In a later Stable window (September 22 to October 14, 2026; called Phase 3 in the companion TRACE-CoQ) it was not reproduced. Results are descriptive and hypothesis-generating, from a single unblinded, non-randomized self-tracking record, and they support no general conclusions. No biological mechanism is proposed. The paper also states where the method does not fit as written. If the perturbing input directly raises the reserve marker (for example, a carotenoid supplement and a skin carotenoid score), a steady dose alone does not define a stable period; this has not been tested here. The CoQ10 application of the method, with its dosing details and results, is reported in the companion document TRACE-CoQ (short paper and appendix, same author; concept DOI https://doi.org/10.5281/zenodo.23308049 This is an ongoing project, not a finished study. Tracking continues, and later versions will report further results. The author is an independent investigator. AI language models (Anthropic Claude) were used as interactive tools for statistical discussion, language refinement and version integration, as described in the manuscript. Single subject; not peer reviewed; no causal or mechanistic claims. Correspondence: [email protected]