Operant Conditioning of the Sensori-Motor Rhythm in the Neurofeedback Paradigm for Endogenous Enhancement of Motor-Imagery Brain-Computer Interface

Motor Imagery Brain Computer Interface (BCI) is based on a principle that during real/imaginary movement of a certain part of the body we can observe desynchronization of the mu rhythm (8-12 Hz) in area of sensorimotor cortex, which corresponds to this body part representation. By detecting this desynchronization with one of the neuroimaging techniques, such as Electroencephalography (EEG), it is possible to predict what kind of movement the person is currently doing or thinking about. Such interfaces are attractive not only to the general public (e.g. new commercial products with easier control), but also to patients with severe motor dysfunction, who can use them for rehabilitation and operation of external devices (e.g. wheelchair). However, most people with motor impairments do not have a pronounced mu rhythm, which automatically makes such interfaces ineffective for them. In this thesis we propose to facilitate motor-imagery BCI with neurofeedback training, a methodology used to control the subject’s brain states by means of operant conditioning. We predict that increasing the power of mu-rhythm with Neurofeedback will allow the user to gain better control of the motor imagery BCI. We provide results of our pilot control experiment, in which 25 participants tested a joint NFB-BCI paradigm. We were able to show a significant increase of the mu-power in the experimental group. However, the associated increase in mu-desynchronization was not enough to improve the quality of motor-imagery BCI.