Dynamic Difficulty Adjustment in Games Based on Cognitive Workload

Abstract: Games should provide adequate challenge to players in order to provide an engaging and fun player experience. If a game is too easy or too hard, the players might get bored or frustrated. Building on the flow theory, the use of dynamic difficulty adjustment (DDA) systems have become popular for adjusting the game difficulty in real time according to the performance of the player. An alternative to using player performance as input to the DDA system is to use measurements of the player, such as heart rate or brain metrics. In this study, we examined Detection-response task (DRT), which is a method for measuring cognitive workload, in a gaming context. The study aimed to determine if the DRT is sensitive to real-time changes in player workload and to study the challenges with integrating the DRT into gameplay and considering these findings determine if the DRT is a suitable method for DDA. We created a game and integrated the DRT into the gameplay, and 26 participants played the game on different difficulty levels. The results show that the DRT is not sensitive to real-time changes in workload in our game, and there are some difficulties with integrating the DRT into gameplay. These results indicate that it is difficult to implement the DRT in games and that the DRT might not be a suitable method to use in DDA systems. More research is needed to assess the method further.