Population level dynamics of rat hindlimb sensorimotor cortex cells during bipedal obstacle avoidance

Abstract

Restoration of locomotor function following spinal cord injury is both a highly desired outcome and critical health priority for paralyzed individuals (Anderson, 2004). Brain Machine Interfaces (BMIs) hold promise for restoration of voluntary locomotor function. BMIs decode information from simultaneously recorded populations of single neurons and use this information as a control signal to restore voluntary function (for review, see (Lebedev & Nicolelis, 2006)). During stereotypical flat stepping, the majority of cells in the motor cortex are modulated by a particular phase of the step cycle, and change their activity during complex, or visually guided, locomotion (Drew, Andujar, Lajoie, & Yakovenko, 2008). While the role of single cells during locomotion has been studied extensively, population level dynamics are not well understood. Development of a locomotor BMI to restore voluntary locomotor function requires better understanding of these population level dynamics. The long term goal of this work is the development of a BMI to restore voluntary locomotor function following a spinal cord injury. The central hypothesis of this thesis is that population-level changes in phase-modulated neural activity encode for voluntary changes to the step cycle. This thesis aims to identify the kinematic strategy adopted during a complex locomotor task (Aim 1), identify population level changes in neural activity during the same task (Aim 2), and incorporate findings into the design of a single trial BMI decoder for voluntary gait changes (Aim 3).%%%%Ph.D., Biomedical Engineering – Drexel University, 2015