Neuromechanical factors shaping upper limb control

Abstract

Humans have several, potentially infinite, ways to complete any single task. The purpose of this thesis was to better understand how and why individuals select specific ways to perform a given task. In Chapter 3, sampling methods were combined with a biomechanical model of the shoulder to identify the landscape of feasible muscle activity patterns for static shoulder exertions. I demonstrated how muscle activity patterns are successively shaped by each joint of the shoulder complex, the degree of musculoskeletal redundancy afforded to different muscle groups, and the pool of solutions available to our nervous system should it choose to converge on solutions that require little effort. In Chapter 4, this modelling approach was combined with empirical data to identify where real-world muscle activity patterns may be located in the landscape of all feasible solutions. The findings revealed that although effort-based criteria may help shape muscle coordination, effort alone does not sufficiently capture variation in real-world muscle activity patterns. The models used in Chapters 3 and 4 assume that muscles act independently. However, neural and mechanical dependencies can constrain muscle coordination, which are particularly prominent at the hand. In Chapter 5, I used opensource tools to develop and test methods for markerless tracking of 3D finger movements. The markerless tracking was then applied in Chapter 6 to study constraints to finger independence during isometric and movement tasks following a fatigue protocol aimed at reducing force transfers due to neural constraints. Fatigue reduced involuntary finger forces but increased involuntary finger movement, highlighting that neural factors play a larger role in constraining finger independence during isometric tasks while mechanical factors impose larger constraints during movement tasks. Overall, by combining theoretical frameworks, computational modelling, technological advancements, and experimental data, this thesis provides fundamental understanding on neuromuscular control of the upper extremity.