An RSVP-Based BCI Conducted in Motional Workspace: A Pilot Study

Current rapid serial visual presentation (RSVP) paradigm-based brain-computer interface (BCI) systems have primarily been conducted in static workspace. Investigating these systems in motional workspaces is also valuable since users often perform tasks in motional workspaces. In this paper, we conducted an RSVP BCI-based pedestrian detection task in the meantime with and without roll motion of the workspace. We defined three different conditions related to roll motion with a sinusoidal vibration amplitude of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$5^{\circ}, 10^{\circ}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$15^{\circ}$</tex>, and a control group with a vibration amplitude of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0^{\circ}$</tex>, i.e. static state. Neural signatures of electroencephalogram (EEG) signals and detection performance were investigated and compared between static condition and motional conditions. Results from ten subjects showed that the detection performance of the RSVP-based BCI significantly decreased by nearly 10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> compared with the static workspace. Our findings reveal the differences in neural signature and detection performance of the RSVP-based BCI between static workspace and motional workspace. This work contributes to broadening the application scenarios of brain-computer interfaces for human-machine interaction.