Decoding peripheral neural correlates of dexterous movements

Abstract

In order for humans to move, a complex interplay of neural signals needs to transpire. A simple activity of the finger pointing requires coordinating the volition and motor control centers in our brains, inter-connectivity of the forearm and hand’s tendons, and finally, the firing of spinal motor neurons, leading up to the contraction of muscles, leading the movement directly. Due to this high level of complexity, cracking the neural code of volitional, dexterous movement remains one of the great unsolved riddles of modern neuroscience until today. In this thesis, I focussed on developing algorithmic approaches to decoding the peripheral part of the neuromuscular and neural activities during dexterous finger movements. The algorithms are human-centered and suited to the electromyographic and peripheral-neural data I collected during the course of my doctoral degree. Throughout the chapters of the thesis, I show the deployment of a non- invasive experimental setup which allowed me to collect high quality peripheral neural signals from humans. Additionally, I present the development of a biologically-inspired analytical framework, which led me to the identification and tracking of individual finger movement correlates. Finally, I introduce a novel peripheral neural connectivity analysis, and show the robustness of the developed algorithmic approaches to extend beyond a single finger movement decoding.