Effects of task and EEG-based reference signal on performance of on-line ocular artifact removal from real EEG

The effects of two factors on the performance of online ocular artifact (OA) removal methods on real electroencephalogram (EEG) data are evaluated. A new metric is proposed that captures both the amount of artifact removed and the likelihood that a given method would distort the underlying EEG. The metric is then used to measure the performance of an existing on-line OA removal algorithm during periods of (i) eye movements and blinks (EM), (ii) a motor related potential (MRP), and (iii) neither EM nor MRP. The results show that the performance of the algorithm is significantly (p<.05) affected by both EM and MRP. Also, the algorithm-like many others-requires a reference electro-oculogram (EOG) signal, yet for certain applications (e.g., a brain computer interface or BCI) it is preferable or necessary to avoid attaching electrodes around the eyes. Thus, the new metric is also used to investigate the effects of using 3 frontal EEG channels as the reference signal instead of EOG. The results show that replacing the EOG with the selected 3 EEG channels does not significantly (p<.05) affect the performance of the algorithm, making it a viable option for online applications where the use of EOG is not suitable (e.g., BCI).