EEG band powers for characterizing user engagement in P300-BCI

An asynchronous P300-based brain computer interface (BCI) allows users to operate the BCI at their own pace by being able to detect a user's engagement. In our previous work, band powers has been shown to be able to provide additional information for characterizing user engagement and yielded better performance compared to the use of only the amplitudes of event-related potentials. In this follow up study, 19 subjects participated in an experiment which was designed to further evaluate additional predictors of user engagement using band powers. In addition to the regular P300 attended condition, two not-engaged conditions were considered: one with the P300 stimulus matrix still shown (control 1) and the other with stimulus covered by a blank screen (control 2). Alpha and beta band activities decreased in the order of control 2, control 1 and attended. Furthermore, the attended condition had lower delta activity compared to the control conditions. Classification results indicated that band powers were better at differentiating attended and control 2 conditions. Using band powers as additional features resulted in a moderate to moderately large (dz= 0.52 to 0.74) improvement over the classification of the two conditions.

[1]  Hasan Ayaz,et al.  Detection of attention shift for asynchronous P300-based BCI , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[2]  Ying Sun,et al.  Asynchronous P300 BCI: SSVEP-based control state detection , 2010, 2010 18th European Signal Processing Conference.

[3]  E. Donchin,et al.  Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials. , 1988, Electroencephalography and clinical neurophysiology.

[4]  T. Sejnowski,et al.  Estimating alertness from the EEG power spectrum , 1997, IEEE Transactions on Biomedical Engineering.

[5]  F Babiloni,et al.  P300-based brain–computer interface for environmental control: an asynchronous approach , 2011, Journal of neural engineering.

[6]  Hasan Ayaz,et al.  A P300-based EEG-BCI for spatial navigation control , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[7]  Helge J. Ritter,et al.  An Augmented-Reality Based Brain-Computer Interface for Robot Control , 2010, ICONIP.

[8]  W. Klimesch,et al.  EEG alpha oscillations: The inhibition–timing hypothesis , 2007, Brain Research Reviews.

[9]  Cuntai Guan,et al.  Asynchronous P300-Based Brain--Computer Interfaces: A Computational Approach With Statistical Models , 2008, IEEE Transactions on Biomedical Engineering.

[10]  Edgar Erdfelder,et al.  G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences , 2007, Behavior research methods.

[11]  A. Kübler,et al.  ERPs contributing to classification in the ”P300” BCI , 2011 .

[12]  G. Pfurtscheller,et al.  Brain-Computer Interfaces for Communication and Control. , 2011, Communications of the ACM.

[13]  Michael E. Smith,et al.  Monitoring Working Memory Load during Computer-Based Tasks with EEG Pattern Recognition Methods , 1998, Hum. Factors.

[14]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .