Decoding the Flow Experience in Video Games: An Analysis of Physiological and Performance Metrics

We explore the application of psychophysiological measures to gain real-time and objective insights into the flow experience of gamers, aiming to overcome the limitations of traditional self-report surveys, which can be intrusive and lack real-time capabilities. In this study, a self-developed game platform is employed to capture users’ real-time performance data and physiological responses. The recorded data includes in-game process logs, eye movement patterns, and galvanic skin response (GSR). To objectively quantify gaming performance, a novel method employing a gradient boosting model is introduced, offering a comprehensive understanding of the gaming experience. The experimental results indicate that gaming performance tends to improve during flow states compared to non-flow states. Flow states are characterized by specific physiological markers, including saccade number in the area of interest (AOI), average peak velocity of saccades, pupil diameter, blink rate, and fixation number. Additionally, GSR signals exhibit significant differences between flow and non-flow states, with an increased peak frequency during the flow state. The integration of real-time multi-modal measurements offers a novel approach to investigating the flow experience in games, providing valuable insights for future research in immersive game design and user experience evaluation.