Electrical Stimulation of Rat Primary Motor Cortex for Neurorehabilitation and Neuroprosthetic Applications.

Abstract

Cortical electrical stimulation (CES) has been used extensively in experimental neuroscience to modulate neuronal or behavioral activity, which has led this technique to be considered in neurorehabilitation. Because the cortex and the surrounding anatomy have irregular geometries as well as inhomogeneous and anisotropic electrical properties, the mechanisms by which CES has therapeutic effects are poorly understood. Therapeutic effects of CES can be improved by optimizing the stimulation parameters for targeted brain regions. In this dissertation, the effects of CES pulse polarity on neural signals such as unit activity (spikes), local field potentials (LFP), and electrocorticograms (ECoG) recorded from rat primary motor cortex were investigated. The results showed that units located in lower cortical layers are preferentially excited by anodic stimulation, while cathodic stimulation excites those located in upper cortical layers. These opposing effects were also frequency- and amplitude-dependent. Time-frequency analysis of LFPs showed high correlation of gamma (30-120Hz) power with unit activity in corresponding layers. On the other hand, high gamma (60-120Hz) power of ECoG signals only showed high correlation with the unit activity in lower layers. Time-frequency correlations, which were found between LFPs, ECoGs and unit activity were also frequency- and amplitude-dependent. In addition the intracortical microstimulation study showed that lower motor thresholds can be obtained by anodic stimulation in upper layers of motor cortex compared to cathodic and vice versa in lower layers (V/VI). The data demonstrates that the poststimulus effects in neural activity after manipulation of CES parameters changes according to the location (depth) of the recorded neural activity in motor cortex. The signature of the neural activity observed in LFP and ECoG signals provides a better understanding of the effects of stimulation on the affected network and has a promising potential to be used in closed-loop control stimulation systems. These results demonstrate that the neurorehabilitation and neuroprosthetic applications of CES can be further improved by optimizing CES parameters.