Using physiological signal analysis to design affective VR games

One of the most important factors in human-computer interaction is the user emotional reaction. Interactive environments including serious games that are responsive to user emotions improve their effectiveness and user acceptances. Testing and training for user emotional competence is meaningful in health care, which has motivated us to analyze immersive affective games using feedback. In this article, a systematic model of designing interactive environment is presented, which consists three modules as affect modeling, affect recognition, and affect control. For affect control module, a graph-based structure is described to control the game scenarios dynamically by analyzing user emotional states. The affect recognition module makes use of physiological signal analysis such as autonomic nervous system variables to evaluate user emotional reaction to different stimuli, then send it as biofeedback to affect control module. The analysis should be used as a guiding principle for designing affective serious games. Oculus Rift DK2 is used in experiments to provide immersive virtual reality with affective scenarios. The experiments demonstrate the ability to induce measurable and distinguishable emotions from physiological signals. The measurement of effectiveness is discussed. Possible applications include emotion competence training for children with autism spectrum disorder.