Conference Paper

Towards Automatic Personalized Content Generation for Platform Games.

In proceeding of: Proceedings of the Sixth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, AIIDE 2010, October 11-13, 2010, Stanford, California, USA
Source: DBLP
0 Bookmarks
 · 
57 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Traditional personality assessment methods are based on be-havioral, observational, and self-report measures [8], each of which suffers from weaknesses that stem from ambiguity (be-havioral measures), cost-payoff ratio (professional observa-tion), and reliability (self-report). Assessment through video game play offers a way of quantifying behavior, automating observations, and side-stepping self-report. To determine whether video games are a valuable addition to the arse-nal of personality assessment methods, we set out to answer the question: Does the statistically trackable play style of a player significantly correlate to his personality? To find the answer, we conducted a survey among Battlefield 3 players. Through the use of a promotional campaign, dubbed 'Psy-Ops', the response to the survey ran up to 13,376 individuals. Each participant was asked to fill out a 100-item IPIP (Inter-national Personality Item Pool) Big Five personality ques-tionnaire, and requested for permission to draw their game statistics from a public database. All in all, 175 game vari-ables, 100 personality scores, and 5 personality dimensions were correlated for the total sample, and 11 demographic subsamples. We found that play style and personality do correlate significantly, showing three key themes. (1) Con-scientiousness is negatively correlated with speed of action. (2) The game variable Unlock Score per Second correlates most often and most strongly with personality, especially with Conscientiousness and Extraversion. (3) Work ethic correlates negatively with performance in the game. Apart from these three themes, subsamples differ in correlational patterns. Two additional results were found when performing a post-hoc analysis on age. First, correlations between age and play style were greater than those between play style and person-ality. While themes (1), (2) and (3) showed effect sizes up to the 0.2 range, age offered effect sizes in the 0.3 range Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. FDG 2013 Chania, Krete Greece Copyright 20XX ACM X-XXXXX-XX-X/XX/XX ...$15.00. for game performance and game length preference, as well as a correlation of r = 0.42 with Unlock Score per Second. Secondly, age and personality correlate with a similar effect size as play style and personality. Therefore, age correlates strongly to play style, while age and play style offer compli-mentary correlations to personality.
    FDG2013; 01/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We argue for the use of active learning methods for player modelling. In active learning, the learning algorithm chooses where to sample the search space so as to optimise learning progress. We hypothesise that player modelling based on active learning could result in vastly more efficient learning, but will require big changes in how data is collected. Some example active player modelling scenarios are described. A particular form of active learning is also equivalent to an influential formalisation of (human and machine) curiosity, and games with active learning could therefore be seen as being curious about the player. We further hypothesise that this form of curiosity is symmetric, and therefore that games that explore their players based on the principles of active learning will turn out to select game configurations that are interesting to the player that is being explored.
    12/2013;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Estimating affective and cognitive states in conditions of rich human-computer interaction, such as in games, is a field of growing academic and commercial interest. Entertainment and serious games can benefit from recent advances in the field as, having access to predictors of the current state of the player (or learner) can provide useful information for feeding adaptation mechanisms that aim to maximize engagement or learning effects. In this paper, we introduce a large data corpus derived from 58 participants that play the popular Super Mario Bros platform game and attempt to create accurate models of player experience for this game genre. Within the view of the current research, features extracted both from player gameplay behavior and game levels, and player visual characteristics have been used as potential indicators of reported affect expressed as pairwise preferences between different game sessions. Using neuroevolutionary preference learning and automatic feature selection, highly accurate models of reported engagement, frustration, and challenge are constructed (model accuracies reach 91%, 92%, and 88% for engagement, frustration, and challenge, respectively). As a step further, the derived player experience models can be used to personalize the game level to desired levels of engagement, frustration, and challenge as game content is mapped to player experience through the behavioral and expressivity patterns of each player.
    IEEE transactions on cybernetics. 12/2013; 43(6):1519-1531.

Full-text (2 Sources)

View
18 Downloads
Available from
Jun 5, 2014