Non-Invasive Brain-Computer Interface for Virtual Object Control

A brain-computer interface (BCI) is a system that provides a separate output pathway for neurological signals whereby they can be interpreted to determine the user's intended cognitive action. Utilizing EEG-based sensorimotor rhythms (SMRs) generated in the motor cortex has allowed subjects to control virtual computer cursors in up to three dimensions by simply imagining the movement of a specified body part. Nevertheless, the scalp EEG signals are smeared by the volume conduction effect and measurement noise. The overall hypothesis of this study is that EEG-based virtual object control may help reveal optimal motor imagination tasks best used in a BCI. The PI's hypotheses include: (1) The use of advanced signal processing techniques will better reveal characteristics of EEG signals that represent the underlying motor cognitive function of the subject; (2) BCI systems based on SMR generated using motor imaginations will allow effective control of a virtual object in real time; (3) EEG imaging techniques will provide insight into the areas of cortical activation during a motor imagery task that can be utilized to increase the spatial resolution of non-invasive BCI's.