Exploring Player Adaptivity through Level Design: A Platformer Case Study

Abstract

Different people experience distinct feelings while playing the same game. For instance, some players like action-heavy games, while others prefer puzzles. Creating a game that engages everyone is not trivial and, ultimately, impossible. However, generating content adapted to the players could improve their experience, thus maximizing the potential player base.Therefore, a game designer's role can be challenging when dealing with games with static content. With adaptive games, these can be designed to act differently according to several actions without developing different versions. Although this approach seems to solve all the designers' problems, creating these games can be very time-consuming, increasing the development time and budget. Besides, creating a generic approach that adapts content for every game genre or even different games inside the same genre is not trivial.To tackle this problem, this work proposes a methodology to explore player adaptivity through level design that follows three phases: (i) the selection of a genre and a game, (ii) a first study to correlate players' profiles, preferences, and actions with single elements, and (iii) a second study, based on the previous, to evaluate the impact of generating and adapting game levels.Using a version of a well-known platform game, Super Mario Bros., the first results show it is not possible to correlate players' profiles with game elements. However, several preferences for levels with specific configurations were spotted and used for the second study's adaptation. Nevertheless, these adaptations did not sufficiently impact the players' overall experience in the second study.Although the results do not seem encouraging, other deeper analyses of the study's data should be performed in the future using distinct algorithms and machine learning. Moreover, the same methodology should be applied to other games to ensure similar results are obtained. Last but not least, a study should be conducted to explore correlations between players' profiles and groups of game elements, especially obstacles around gaps.