Can We Use a Brain-Computer Interface and Manipulate a Mouse at the Same Time?

Brain-Computer Interfaces (BCI) introduce a novel way of interact-ing with real and virtual environments by directly exploiting cere-bral activity. However in most setups using a BCI, the user is ex-plicitly asked to remain as motionless as possible, since muscular activity is commonly admitted to add noise and artifacts in brain electrical signals. Thus, as for today, people have been rarely let using other classical input devices such as mice or joysticks simul-taneously to a BCI-based interaction. In this paper, we present an experimental study on the influence of manipulating an input de-vice such as a standard computer mouse on the performance of a BCI system. We have designed a 2-class BCI which relies on Alpha brainwaves to discriminate between focused versus relaxed mental activities. The study uses a simple virtual environment inspired by the well-known Pac-Man videogame and based on BCI and mouse controls. The control of mental activity enables to eat pellets in a simple 2D virtual maze. Different levels of motor activity achieved with the mouse are progressively introduced in the gameplay: 1) no motor activity (control condition), 2) a semi-automatic motor activ-ity, and 3) a highly-demanding motor activity. As expected the BCI performance was found to slightly decrease in presence of motor activity. However, we found that the BCI could still be success-fully used in all conditions, and that relaxed versus focused mental activities could still be significantly discriminated even in presence of a highly-demanding mouse manipulation. These promising re-sults pave the way to future experimental studies with more com-plex mental and motor activities, but also to novel 3D interaction paradigms that could mix BCI and other input devices for virtual reality and videogame applications