In order to give players a good experience, the difficulty in a computer game needs to be adjusted appropriately. In their new work, Korean scientists have developed a new approach in which players’ emotions are estimated with the help of in-game data and the difficulty level is adjusted to maximize player satisfaction. The researchers’ work contributes to balancing the difficulty of the games and making the games more attractive to all types of players.
Difficulty level is a challenging aspect of video games. Some people prefer video games because they are difficult, while others prefer to enjoy an easy experience. Many developers use ‘dynamic difficulty adjustment (DDA)’ to make this process easier. DDA allows a game’s difficulty level to be adjusted in real time based on the player’s performance. For example, if the player’s performance exceeds the developer’s expectations for a certain difficulty level, the game’s DDA tool automatically increases the difficulty level. Although useful, this strategy only takes into account the player’s performance and does not look at how much fun the players are actually having.
A research team at the Gwangju Institute of Science and Technology decided to tweak this DDA approach a bit in a new study published in Expert Systems With Applications. Instead of focusing on player performance, scientists have developed DDA tools that adjust the game’s difficulty to maximize one of the four elements associated with player satisfaction (difficulty, skill, fluency, and attraction value). DDA tools are trained with a machine learning approach using data gathered from real human players who play various AI fighting games and then fill out a questionnaire about their experience.