Interacting with multiple game genres using Motion Onset Visual Evoked Potentials

Abstract

Motion Onset Visually Evoked Potentials (mVEPs) allow users to interact with technology using non-visually fatiguing stimuli in a Brain Computer Interface (BCI). This study employs mVEP in an onscreen controller and evaluates players' ability to use mVEP for online gameplay with games from three different genres namely action, puzzle and sports. The onscreen controller consists of five mVEP stimuli that are presented as buttons to allow the participant to choose from five different actions in each game. The performance was assessed based on online BCI accuracy and game score for each game. Results indicate that the players could control the games with an average online accuracy of 71% (5 class classification chance accuracy is 20%). The results also suggest that the use of the mVEP controller with a detailed environment and stimulating feedback in the form of an action game helped to attain the highest online accuracy (75%).

Can teenagers control a 3D racing game using motion-onset visual evoked potentials?

“…Our findings show that when using leftwards motion of the stimuli, the three most commonly used EEG channels (Cz, P7 and O1) are located on the left visual hemisphere. This finding is consistent with previous studies where leftwards motion of the stimuli was also used [37][36] [39][40] [30]. These locations also correspond to the middle temporal (MT) and medial superior temporal (MST) areas of the brain which is specialised for the processing of motion.…”

“…mVEPs were first investigated in BCI in [30] where an mVEP-based BCI control scheme was implemented to investigate the users' responses to moving stimuli. In [36] mVEP stimuli were employed as a control modality for three online BCI video games of different genres and in [37], a follow-up study tested the same three BCI games using a heads-up display (HUD) in which the stimuli were presented on a white background as opposed to overlaying mVEP stimuli onto the games graphics. The results indicated a HUD with white background would likely enable better mVEP detection accuracy in such games.…”

Neurogaming With Motion-Onset Visual Evoked Potentials (mVEPs): Adults Versus Teenagers

“…During the calibration run, the user viewed gaming scenes similar to those in the online game, with the exception of some online-specific gaming elements such as the car model, checkpoints, and on-road visual cues. In previous studies, it was shown that maintaining consistency between the calibration training and online visuals could have an impact on gaming with mVEPs, compared to using white screen background during the calibration runs [24] [25]. During the calibration run, each of the five stimuli was a target 60 times yielding data from 300 trials (5 stimuli × 60 activations).…”

“…Motion-onset VEPs (mVEP) are an alternative to flash-based BCIs and evoked using motion-based stimuli. mVEPs have been used in BCI spelling applications [21], user interfaces [22] and neurogaming [23] [24] [25]. Typically, a mVEP stimulus comprises a rectangular white box with a black border with a total length of 1.24° and height of 0.76°.…”

“…Average offline accuracy across all participants for all games was >70% while online accuracy yielded accuracies of >60%. In a follow-up study [25] the approach was improved i.e., participants were trained and tested within the same environment, stimuli were delineated within a dedicated controller area and more control instances were added. All scenarios achieved average offline and online accuracies of >77%.…”