Platformer level design for player believability

Abstract

Abstract-Player believability is often defined as the ability of a game playing character to convince an observer that it is being controlled by a human. The agent's behavior is often assumed to be the main contributor to the character's believability. In this paper we reframe this core assumption and instead focus on the impact of the game environment and aspects of game design (such as level design) on the believability of the game character. To investigate the relationship between game content and believability we crowdsource rank-based annotations from subjects that view playthrough videos of various AI and human controlled agents in platformer levels of dissimilar characteristics. For this initial study we use a variant of the well-known Super Mario Bros game. We build support vector machine models of reported believability based on gameplay and level features which are extracted from the videos. The highest performing model predicts perceived player believability of a character with an accuracy of 73.31%, on average, and implies a direct relationship between level features and player believability.

Exploring Player Adaptivity through Level Design: A Platformer Case Study

“…Besides the presented works, there are other ways to create players' models and personalize game content. For instance, Camilleri, Yannakakis, and Dingli [8] crowdsourced rank-based annotations from users watching videos of characters controlled by AI and humans playing a 2D platform game to model believability, i.e., the ability to convince observers that a human player is controlling a character. Blom et al [7] personalized levels of Infinite Mario Bros [45] according to players' facial expressions, i.e., during the gameplay, the experience adapted to the players' emotions inferred from the facial expression recognition.…”

Badanie Wpływu Sterowania Grą Platformową Na Efektywność Rozgrywki

“…In [3] an experiment was carried out in which two players and two computers with artificial intelligence play a platform game similar to "Infinite Mario Bros" (a platform game). The following parameters were selected to determine the effectiveness of the players' gameplay: completion time, last life game play duration, percentage of time spent running left, percentage of time spent running, number of jumps, number of unsuccessful jumps, number of collected coins, number of kicked shells, number of deaths caused by falling into a hole, number of deaths caused by enemies, number of killed enemies, number of enemies killed by a kicked shell.…”

Trace It Like You Believe It: Time-Continuous Believability Prediction

“…Finding normative protocols for believability assessment is important; while the common belief is that an agent's believability depends only on the AI that controls it [32], this is far from being true. In particular, Camilleri et al showed that changing an agent's environment affects their perceived believability levels [33]. In their study the authors asked players to annotate the believability of a platform game play-through.…”