BME PhD Defense: Steven M. Peterson
Effects of Sensorimotor Perturbations on Balance Performance and Electrocortical Dynamics
WHEN: August 17, 2018 10:30 am-11:30 amADD TO CALENDAR
Humans must frequently adapt their posture to prevent loss of balance. Such balance control requires complex, precisely-timed coordination among sensory input, neural processing, and motor output. Despite its importance, our current understanding of cortical involvement during balance control remains limited by traditional neuroimaging methods, which are stationary and have poor time resolution. High-density electroencephalography (EEG), combined with independent component analysis, has become a promising tool for recording cortical dynamics during balance perturbations due to its portability and high temporal resolution. Additionally, recent improvements in immersive virtual reality headsets may provide new rehabilitative paradigms, but the effects of virtual reality on balance and cortical function remain poorly understood.
In my first study, I recorded high-density EEG from healthy, young adult subjects as they walked along a beam with and without virtual reality high heights exposure. While virtual high heights did induce stress, the use of virtual reality during the task increased performance errors and EEG measures of cognitive loading compared to real-world viewing without a headset. In my second study, I collected high-density EEG from healthy young adults as they walked along a treadmill-mounted balance beam to determine the effect of a transient visual perturbation on training in virtual reality. Subjects in the perturbations group improved comparably to those that trained without virtual reality, indicating that the perturbation helped subjects overcome the negative effects of virtual reality on motor learning. The perturbation primarily elicited a cognitive change. In my third study, healthy, young adult EEG was recorded during physical pull and visual rotation perturbations to tandem walking and tandem standing. I found similar electrocortical patterns for both perturbation types, but different cortical areas were involved for each. In my fourth study, I used a phantom head to validate EEG connectivity methods based on Granger causality in a real-world environment. In general, connectivity measures could determine the underlying connections, but many were susceptible to high-frequency false positives. Using data from my third study, my fifth study analyzed corticomuscular connectivity patterns following sensorimotor balance perturbations. I found strong occipito-parietal connections regardless of perturbation type, along with evidence of direct muscular control from the supplementary motor area during the standing perturbation response.
Taken together, the work presented in this dissertation greatly expands upon the current knowledge of cortical processing during sensorimotor balance perturbations and the effect of such perturbations on short-term motor learning, providing multiple avenues for future exploration.
CO-CHAIRS: Dr. Cynthia Chestek and Dr. Daniel P. Ferris