Dynamics of human learning of a brain-computer interface

Methods of using human brain signals to control devices have evolvedin the past decade. These brain-machine interfaces (BMI) have obviouspotential to restore function across a broad range of neurologic diseases. Though mathematical tools are critical in optimizing signalextraction, the capacity of human learning mechanisms may allow forrapid modification of control signals, once feedback is given. This discussion will look at the possible signals used for human BMIapplications as explored in human electrocorticography. Theinteraction between motor, sensory and feedback will be explored. The human capacity to rapidly adapt to feedback is of particular interest,thus, the psychophysics of BMI acquisition, and the networks involvedin learning of this highly novel output are examined.