Sander Bakkes

Generating Missions and Spaces for Adaptable Play Experiences

This paper investigates strategies to generate levels for action-adventure games. For this genre, level design is more critical than for rule-driven genres such as simulation or rogue-like role-playing games, for which procedural level generation has been successful in the past. The approach outlined by this article distinguishes between missions and spaces as two separate structures that need to be generated in two individual steps. It discusses the merits of different types of generative grammars for each individual step in the process. Notably, the approach acknowledges that the online generation of levels needs to be tailored strictly to the actual experience of a player. Therefore, the approach incorporates techniques to establish and exploit player models in actual play.

Challenge balancing for personalised game spaces

In this paper we propose an approach for personalising the space in which a game is played (i.e., levels) - to the end of tailoring the experienced challenge to the individual user during actual play of the game. Our approach specifically considers two design challenges, namely implicit user feedback and high risk of user abandonment. We contribute an approach that acknowledges that for effective online game personalisation, one needs to (1) offline learn a policy that is appropriate in expectation across users - to be used for initialising the online game, (2) offline learn a mapping from gameplay observations to the player experience - to be used for guiding the online game personalisation, and (3) rapidly converge to an appropriate policy for the individual user in online gameplay - employing the learned feedback model and a straightforward model of user abandonment. User studies that validate the approach to online game personalisation in the actual video game INFINITE MARIO BROS, indicate that it provides an effective basis for automatically balancing the games challenge level to the individual human player.

Inferring Player Experiences Using Facial Expressions Analysis

Understanding player experiences is central to game design. Video captures of players is a common practice for obtaining rich reviewable data for analysing these experiences. However, not enough has been done in investigating ways of preprocessing the video for a more efficient analysis process. This paper consolidates and extends prior work on validating the feasibility of using automated facial expressions analysis as a natural quantitative method for evaluating player experiences. A study was performed on participants playing a first-person puzzle shooter game (Portal 2) and a social drawing trivia game (Draw My Thing), and results were shown to exhibit rich details for inferring player experiences from facial expressions. Significant correlations were also observed between facial expression intensities and self reports from the Game Experience Questionnaire. In particular, the challenge dimension consistently showed positive correlations with anger and joy. This paper eventually presents a case for increasing the application of computer vision in video analyses of gameplay.

Correlation between Facial Expressions and the Game Experience Questionnaire

Quantitative methods in the domain of player experience evaluation provide continuous and real-time analyses of player experiences. However, current quantitative methods are mostly either too confined within in-game statistics, or require non-typical laboratory play setups to monitor real-life behavior. This paper presents preliminary results on the feasibility of using facial expressions analysis as a natural quantitative method for evaluating player experiences. Correlations were performed between the facial expressions intensities and self-reported Game Experience Questionnaire (GEQ) dimensions of players in two video games.

Circuit-adaptive challenge balancing in racing games

In this paper, we propose a novel approach to challenge balancing in racing games: circuit-adaptive challenge balancing. We propose to automatically adapt the actual racing circuit - while it is being played - such that the performed circuit adaptations intelligently balance the challenge for all players in parallel. Our approach to circuit-adaptive game balancing is submitted as an alternative to the traditional rubber banding method (not a replacement), that may contribute particularly to distinct design goals and gameplay events. Indeed, an interesting feature of the approach, is that each player will be targeted with distinct, player-appropriate circuit adaptations. Consequently, we test the hypothesis that a perceptively balanced game can be achieved via such player-appropriate adaptations. The approach itself is built around (A) a classifier that can assess a players in-game performance, and (B) an algorithm that employs the ability of targeted circuit adaptations, to the end of realising circuit-based challenge balancing. Experiments that validate the approach - by means of simulation studies and studies with actual human participants - suggest that the approach can automatically balance the challenge in actual racing games, by reducing the gap size between all players in parallel, while ensuring that it does not come at a cost in terms of the actual player experience.