Properly balancing the difficulty of a video game is essential to providing gamers with an enjoyable experience. In a recent study, Korean scientists developed a novel dynamic difficulty adjustment approach in which players’ emotions are estimated using game data, and the difficulty level is adjusted accordingly to maximize satisfaction. players. Their efforts could help balance the difficulty of games and make them more appealing to all types of gamers.
Difficulty is a difficult aspect to balance in video games. Some people prefer video games that present a challenge while others enjoy an easy experience. To make this process easier, most developers use “Dynamic Difficulty Adjustment (DDA)”. The idea of DDA is to adjust the difficulty of a game in real time based on player performance. For example, if the player’s performance exceeds the developer’s expectations for a given difficulty level, the game’s DDA may automatically increase the difficulty to increase the challenge presented to the player. While useful, this strategy is limited in that only player performance is considered, not how much fun they actually have.
In a recent study published in Expert systems with applications, a research team from the Gwangju Institute of Science and Technology in Korea decided to modify the DDA approach. Instead of focusing on player performance, they developed DDA agents that adjusted game difficulty to maximize one of four different aspects related to a player’s satisfaction: challenge, skill, flow, and valence. DDA agents were trained via machine learning using data collected from real human players, who played a fighting game against various artificial intelligences (AIs), and then answered a questionnaire about their experience.
Using an algorithm called Monte Carlo tree search, each DDA agent used real game data and simulated data to adjust the opposing AI’s fighting style to maximize a specific emotion, or “affective state”. “One of the advantages of our approach over other emotion-focused methods is that it does not rely on external sensors, such as electroencephalography,” comments Associate Professor Kyung-Joong Kim, who led the research. ‘study. “Once trained, our model can estimate player states using only game features.”
The team verified – through an experiment with 20 volunteers – that the proposed DDA agents could produce AIs that improved the overall experience for players, regardless of their preference. This is the first time that affective states have been incorporated directly into DDA agents, which could be useful for commercial games. “Commercial game companies already have massive amounts of player data. They can leverage this data to model players and solve various game balancing issues using our approach,” remarks Associate Professor Kim. It should be noted that this technique also has potential for other areas that can be “gamified”, such as health care, exercise and education.
This article was posted online on June 3, 2022 and will be published in volume 205 of the journal on November 1, 2022.
Source of the story:
Materials provided by GIST (Gwangju Institute of Science and Technology). Note: Content may be edited for style and length.