Authors: Ryan Beveridge, David Marshall, Shane Wilson and Damien Coyle
Motion Onset Visually Evoked Potentials (mVEPs) have the advantage of being an elegant and less visual fatiguing stimuli than that of other VEPs such as the steady state VEP (SSVEP) or P300 and therefore may be apposite for use in movement-free brain-controlled computer games using brain-computer interface (BCI) technology. To investigate the effects of 3-Dimensional (3D) graphic variations on mVEPs, we present in this pilot study a set of five hypothetical game levels with differing graphics, each with increasing visual complexity, in which the user/player must attend to one of five mVEP inducing stimuli. The mVEP based on-screen virtual buttons involves a leftward motion lasting 140 milliseconds to elicit a response from the dorsal pathway. This pilot study focused on offline classification results. BCI classification accuracy results for separating target vs. non target mVEP stimuli (2 class) as well as classifying target stimuli among the five stimuli (5 class) are compared for each variation in graphic complexity. The results of the study show a trend indicating the classification accuracy is inversely proportional to graphic complexity however the difference in BCI classification accuracy for each level of complexity are not significant (>70%, p>0.05). The results are encouraging, suggesting that the use of 3D graphics of varied complexity is possible when using mVEP based BCI as a control strategy.
Keywords: Brain-Computer Interface (BCI), Motion Onset Visually Evoked Potentials (mVEP), Electroencephalography (EEG), Gaming, 3D, Graphics, Visual