Why Is Neuromechanical Modeling of Balance and Locomotion So Hard?

Abstract

A goal and challenge in neuromechanical modeling is to develop validated simulations to predict the effects of neuromotor deficits and therapies on movements. This has been particularly challenging in balance and locomotion because they are inherently unstable, making it difficult to explore model parameters in a way that still coordinates the body in a functional way. Integrating realistic and validated musculoskeletal models with neural control mechanisms is critical to our ability to predict how human robustly move in the environment. Here we briefly review both human locomotion models, which generally focus on modeling the physical dynamics of movement with simplified models of neural control, as well as balance models, which model sensorimotor dynamics and processing with simplified biomechanical models. Combining complex neural and musculoskeletal models increases the redundancy in a model and allows us to study how motor variability and robustness are exploited to produce movements in both healthy and impaired individuals. To advance, the integration of neuromechanical modeling and experimental approaches will be critical in testing specific hypotheses concerning how and why neuromechanical flexibility is both exploited and constrained under various movement contexts. We give a few examples of how the close interplay between models and experiments can reveal neuromechanical principles of movement.