We have systematically moved from the data in Fig. 1 to the fit in Fig. 3A, and then from very simple well-understood physiological mechanisms to how healthy HR should behave and be controlled, reflected in Fig. 3 B and C. The nonlinear behavior of HR is explained by combining explicit constraints in the form (Pas, ?Odos) = f(H, W) due to well-understood physiology with constraints on homeostatic tradeoffs between rising Pas and ?O2 that change as W increases. The physiologic tradeoffs depicted in these models explain why a healthy neuroendocrine system would necessarily produce changes in HRV with stress, no matter how the remaining details are implemented. Taken together this could be called a “gray-box” model because it combines hard physiological constraints both in (Pas, ?O2) = f(H, W) and homeostatic tradeoffs to derive a resulting H = h(W). If new tradeoffs not considered here are found to be significant, they can be added directly to the model as additional constraints, and solutions recomputed. The ability to include such physiological constraints and tradeoffs is far more essential to our approach than what is specifically modeled (e.g., that primarily metabolic tradeoffs at low HR shift priority to limiting Pas as cerebral autoregulation saturates at higher HR). This extensibility of the methodology will be emphasized throughout.
The most obvious limit in using static models is that they omit important transient dynamics in HR, missing what is arguably the most striking manifestations of changing HRV seen in Fig. 1. Fortunately, our method of combining data fitting, first-principles modeling, and constrained optimization readily extends beyond static models. The tradeoffs in robust efficiency in Pas and ?O2 that explain changes in HRV at different workloads also extend directly to the dynamic case as demonstrated later.