Affect-based adaptation of an applied video game for educational purposes

Abstract

Purpose This paper aims to clarify how affect-based adaptation can improve implicit recognition of playing style of individuals during game sessions. This study presents the “Rush for Gold” game using dynamic difficulty adjustment of tasks based on both player performance and affectation inferred through electrodermal activity and facial expressions of the player. The game applies linear regression for calculating playing styles to be applied for achieving a style-based adaptation in other educational video games. Design/methodology/approach The experimental procedure included subject selection, demonstration, informed consent procedure, two game sessions in random order – one without and another with affective adaptation control – and post-game self-report. The experiment was conducted with participation of 30 master students and university lecturers in informatics. Findings This study presents experimental results concerning the impact of affective adaptation over playing style recognition, game session time, task’s effectiveness, efficiency and difficulty and, as well, player’s assessment of affectively adaptive gameplay obtained by an adaptation control panel embedded into the game and by post-game self-report. Research limitations/implications The proposed adaptive game limits recognised styles to such based on the Kolb’s Learning Style Inventory model. Another limitation of the study is the relatively small number of participants constrained by the extended experimental procedure and the desktop game version. Originality/value The paper presents an original research on the effect of affect-based adaptation on a novel approach for implicit recognition of playing styles.

Detection of Player Learning Curve in a Car Driving Game

“…Our purpose of runtime detection of learning curve patterns was and remains realization of an efficient and dynamic player-centric adaptation and, hence, improving the overall game playability. Our plan for future works includes dynamic adaptation of various game features, which fall into three main groups [15], namely:…”

A generic and efficient emotion-driven approach toward personalized assessment and adaptation in serious games

“…Thus, playing styles make it possible to play the same level in different ways. A playing style presents persistent traits of the play (Bontchev & Vassileva, 2017). Several models of playing styles, also known as player types, are proposed in the literature, like Bartle's model (Bartle, 1996), Lazzaro's model (Lazzaro, 2004(Lazzaro, , 2008), Yee's Model (Yee, 2006) and augmented Yee Model (Arelíusarson, 2013).…”

Towards Automatic Generation of Serious Maze Games for Education

“…Mazes are broadly used in entertainment games and that makes them very appropriate for interactive representation of content, where the player chooses one of the several options for an action. Furthermore, mazes can be combined with quizzes, puzzles and other mini-games situated at appropriated places into the maze that makes them suitable for GBL in any learning domain [19]. The player navigates between the rooms of the maze as specified by the connecting graph designed by the educator for a given learning curriculum.…”