From Organization to Viability: A Multi-Level Analysis of Gait Dynamics Under Occlusal Constraint

Clinical interpretation often assumes that observable performance sufficiently reflects the organization of an adaptive system.

However, similar observable performance may arise from distinct latent organizations.

This study extends a previous multi-level framework by introducing Level 4, centered on observed longitudinal viability.

Using an exploratory single-case design, gait data from a Parkinsonian participant were recorded with instrumented insoles under three occlusal conditions: neutral natural occlusion (ONL), a 2.5-degree increase in vertical dimension of occlusion (OC2.5), and a 3-degree increase (OC3).

Two sessions, separated by eleven weeks and a structured sensorimotor intervention, were analyzed.

The vertical dimension of occlusion was treated as an experimentally varied constraint applied to an adaptive neuromechanical system.

Although observable performance remained broadly comparable across conditions, PCA-based latent-space analysis revealed distinct longitudinal centroid displacements: OC3 showed the smallest displacement, ONL an intermediate displacement, and OC2.5 the largest.

Bootstrap resampling confirmed the stability of this Euclidean ordering within the dataset.

A complementary Mahalanobis analysis, accounting for within-condition covariance, produced a different ranking, indicating that covariance contributes to the observed displacement.

Rather than invalidating Level 4, this finding shows that the proposed viability measure should be interpreted as an exploratory observational proxy rather than as a covariance-independent metric or validated biomarker.

These within-subject, exploratory, and non-causal results suggest that clinical relevance depends not only on instantaneous performance or latent organization, but also on the capacity of a configuration to maintain coherent longitudinal behavior over time.