BCIs are driven by Electroencephalogram (EEG) signals recorded from the scalp by an array of electrodes mounted in a cap. EEG analysis and BCI have many promising applications for remote control, enhancing human capabilities, and rehabilitation. The success of BCIs in interpreting human intentions depends on efficient preprocessing and machine learning techniques for both feature detection and classification. Typical error rates are currently around 30%, and dealing with real-time error detection and reduction is one of the challenging research areas.
The novel paradigm that we are developing is based on detecting and decoding Error-related Potentials (ErrP), which are evoked in response to errors or conflicts between expected and actual feedback. Such errors may occur not only due to interface errors, but also due to any unexpected feedback or unexpected features in the environment. We have already characterized ErrPs in response to visuo-motor disturbances during reaching movements, and demonstrated that their peak-to-peak amplitude increases with the magnitude of the disturbance. The information available in error-related potentials is expected to enhance and improve BCIs, remote control, simulator training, decision making, human performance monitoring, and robot-assisted rehabilitation.
Our current research is targeted at investigating ErrP, and their applications for enhancing BCIs:
- Characterizing ErrPs in response to different disturbances.
- Enhancing BCIs with ErrPs.
- Enhancing rehabilitation with ErrPs.
- BCI control of wheel-chair
Future projects:
- CPG and its’ EEG correlates
- Balance and its’ EEG correlates
- Palm rehabilitation