Cortical Control Using Multiple Signal Modalities

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Cortical control is emerging as an important area in neuroprosthetic research. The different types of neural signals that are being investigated as control signals depend on the volume of brain tissue that is sampled and the invasiveness of the recording electrode. In particular, multi-channel neuronal spike recordings and local field potentials have been separately considered for real-time cortical control applications. While these signals are both intracortical extracellular potentials that are recorded simultaneously using implantable microelectrode arrays, the relationship between these signals in the context of providing optimized cortical control has not yet been explored. The objective of this project is to investigate the information content of multi-channel spike recordings and local field potentials used either in isolation or in concert in a real-time cortical control paradigm. This integrated look at the control aspects of spike recordings and local field potentials seeks to understand the degree to which combining these signals can be used to optimize the cortical control signal. This project has two specific aims. Specific aim 1 is directed at quantitatively determining the shared information between multi-channel local field potentials and spike recordings. Animals will be operantly conditioned to modulate either multiple local field potential signals only or multiple spike recordings only, with both types of signals being recorded. Offline analysis will investigate the degree to which information about the target states is distributed between and within the two types of signals. Specific aim 2 is directed at investigating the relative contribution of local field potentials and spike recordings when subjects perform an operant conditioning task that utilizes both types of signals. Subjects will be provided the opportunity to use both (or either) local field potentials and ensemble spike activity to perform a behavioral task. The relative decoding filters for both signals will be compared to determine subject preference. Task difficulty will then be varied in order to study the interplay between task difficulty and modality preference. This project investigates a novel use of intracortical signal combinations for neuroprosthetic control and it is likely to contribute to the development of future clinical neuroprostheses.