Authors:
Gregor Geršak
1
;
Sean M. McCrea
2
and
Domen Novak
2
Affiliations:
1
University of Ljubljana, Slovenia
;
2
University of Wyoming, United States
Keyword(s):
Affective Computing, Classification, Computer Games, Physiological Computing, User Experience.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Collaboration and e-Services
;
Data Manipulation
;
e-Business
;
Enterprise Information Systems
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Methodologies and Methods
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Soft Computing
;
Usability
;
Usability and Ergonomics
;
Web Information Systems and Technologies
;
Web Interfaces and Applications
Abstract:
Physiological games use classification algorithms to extract information about the player from physiological measurements and adapt game difficulty accordingly. However, little is known about how the classification accuracy affects the overall user experience and how to measure this effect. Following up on a previous study, we artificially predefined classification accuracy in a game of Snake where difficulty increases or decreases after each round. The game was played in a laboratory setting by 110 participants at different classification accuracies. The participants reported their satisfaction with the difficulty adaptation algorithm as well as their in-game fun, with 85 participants using electronic questionnaires and 25 using paper questionnaires. We observed that the classification accuracy must be at least 80% for the physiological game to be accepted by users and that there are notable differences between different methods of measuring the effect of classification accuracy. Th
e results also show that laboratory settings are more effective than online settings, and paper questionnaires exhibit higher correlations between classification accuracy and user experience than electronic questionnaires. Implications for the design and evaluation of physiological games are presented.
(More)