Date: 2016 - Present
Hardware and software: Oculus DK2, HTC Vive, Leap Motion, MYO, Polar H7 , Unity, Accord.NET, MATLAB.
Description: In my PhD research at Queen Mary University of London, I have developed a video game in Virtual Reality (VR) called 'Memory Break'. The game is designed to train the player's working memory (WM) performance keeping them engaged and immersed while challenging their WM (Gabana et al, 2017). I have trained a Machine Learning algorithm (SVM) that detects the affective states of the player and adapts the game accordingly (i.e.: if the player is frustrated, the game would reduce the difficulty) in order to keep him/her in an optimal affective state. The detection of affective states is realised analysing the heart rate (Polar H7), hand's pressure (MYO) and head motion (HTC Vive headset). The aim of 'Memory Break' is to help people diagnosed with Attention Deficit and Hyperactivity Disorder (ADHD) who have low WM capacity.
The game consists of throwing balls at different stationary or moving obstacles to successfully pass through without crashing into them. If the player crashes, five points are deducted from the score; In order to get points, the participants have to throw and hit the gems found on their way. Every 30s of game play, the game stops at a door where a random sequence of letters appears that has to be remembered. After the sequence is displayed, the doors open and the game continues for another 7s. The game then stops again at another door where the player has to recall and input the letters previously shown in the exact same order.
'Memory Break' has been developed and tested also for various interaction modes (VR, Desktop and iPad) for research purposes (see Gabana et al, 2017).
References: Gabana, D., Tokarchuk, L., Hannon, E., and Gunes, H. (2017). “Effects of Valence and Arousal on Working Memory Performance in Virtual Reality Gaming”. In Affective Computing and Intelligent Interaction (ACII) 2017.