Högskolan i Skövde

Looking ahead, game developers are building virtual worlds that are more and more realistic and immersive to players, making immersion evaluation an area of increasing importance for the game user experience (GUX) field. This study applies an approach to quantify immersion from the perspective of human cognition.

Immersion is a player’s level of engagement with a video game (Brown and Cairns, 2004). To measure it, much previous research has focused on questionnaires and psychophysiological methods. Questionnaires are dependable and convenient, but their use in real-time scenariosis limited. As gaming is a dynamic flow, we wish to assess player immersion in real-time and study how it changes according to game content. Psychophysiological methods, such as electroencephalogram (EEG, a recording of brain activity), measure immersion in real-time. However, data interpretation could be challenging since one recordable signal can be a marker for multiple psychological factors. For instance, when one brain signal increases its value, is it because the game is immersive or because the player is stressed? This study attempts to explore a new method to find additional solutions for measuring immersion in real-time. 

There are two dynamics that exist in the human cognitive system: component dominant dynamics (neural signals are processed as a single component, e.g., the interpreting of brain signals) and interactive dominant dynamics (multiple processes interact with each other simultaneously, known as the complex systems theory). Psychophysiological measurements belong to the component dominant dynamics, and scientists have been devoted to this field for years. But little has been done with interactive dominant dynamics. This research aims to fill a void and shed light on quantitative immersion measurements based on interactive dominant dynamics.

Interactive dominant dynamics can be observed via pink noise, a fractal signal that exhibitsinteraction dominant dynamics, also called 1/f scaling. This study aims to observe if pink noise exists in gaming activities through eye-tracking data (processed with detrended fluctuationanalysis) and to determine if pink noise data is correlated to immersion (questionnaire data). In the experiment, participants (n = 22) played two games in random order: the open-world game Elden Ring and the first-person shooter Metro Exodus, while their eye data is recorded and collected in real-time. By comparing the statistical properties, the main findings are:i. Pink noise exists in both Elden Ring and Metro Exodus (0.8 < h > 1.2)ii. The amount of data with pink noise may be correlated with immersive questionnaire results for Elden Ring (r = 0.567*)

The study investigated a quantitative way of applying complex systems theory to the evaluation of game immersion, observed pink noise in gaming behaviour and demonstratedthat pink noise might be correlated to immersion in open-world games. Future research could conduct the same experiment in other open-world games to determine if a positive association exists there as well. Furthermore, the study enables cognitive scientists to collaborate with game researchers and delve into the meaning behind the data.