Classification of Multiple Psychological Dimensions in Computer Game Players Using Physiology, Performance, and Personality Characteristics

Abstract

Human psychological (cognitive and affective) dimensions can be assessed using several methods, such as physiological or performance measurements. To date, however, few studies have compared different data modalities with regard to their ability to enable accurate classification of different psychological dimensions. This study thus compares classification accuracies for four psychological dimensions and two subjective preferences about computer game difficulty using three data modalities: physiology, performance, and personality characteristics. Thirty participants played a computer game at nine difficulty configurations that were implemented via two difficulty parameters. In each configuration, seven physiological measurements and two performance variables were recorded. A short questionnaire was filled out to assess the perceived difficulty, enjoyment, valence, arousal, and the way the participant would like to modify the two difficulty parameters. Furthermore, participants’ personality characteristics were assessed using four questionnaires. All combinations of the three data modalities (physiology, performance, and personality) were used to classify six dimensions of the short questionnaire into either two, three or many classes using four classifier types: linear discriminant analysis, support vector machine (SVM), ensemble decision tree, and multiple linear regression. The classification accuracy varied widely between the different psychological dimensions; the highest accuracies for two-class and three-class classification were 97.6 and 84.1%, respectively. Normalized physiological measurements were the most informative data modality, though current game difficulty, personality and performance also contributed to classification accuracy; the best selected features are presented and discussed in the text. The SVM and multiple linear regression were the most accurate classifiers, with regression being more effective for normalized physiological data. In the future, we will further examine the effect of different classification approaches on user experience by detecting the user’s psychological state and adapting game difficulty in real-time. This will allow us to obtain a complete picture of the performance of affect-aware systems in both an offline (classification accuracy) and real-time (effect on user experience) fashion.

User Experience With Dynamic Difficulty Adjustment Methods for an Affective Exergame: Comparative Laboratory-Based Study

“…Indeed, the accuracies in this study (Table 3) were lower than those in the previous one. For example, in the previous study, the all-data method resulted in a classification accuracy of 84.1% for both ball speed and paddle size preferences [33]; in this study, the accuracies were 62.5% for ball speed and 55% for paddle size. As a validation step, we reran the classification offline (post hoc) to ensure there was no bug with real-time processing; this resulted in the same results as during the data collection sessions.…”

“…This paper describes a comparative evaluation of five DDA methods in an affective exergame. Three of these methods were developed based on data recorded in a previous study [ 33 ]; thus, we briefly refer to a previous study for better understanding. Both studies were approved by the University of Wyoming Institutional Review Board (protocol no 20190822DN02495).…”

“…(PE, PEPE, and all-data) would not be the same in this study as it was in the previous study [33]. Although the data collection in this study was performed with prebuilt classifiers, these classifiers could be retrained offline after the data collection conclusion to determine whether the all-data method is still the most accurate if validated on the new data.…”

“…This study aims to evaluate the user experience in an exergame where DDA is performed using one of five methods: manual, random, performance-based (PE), personality-performance–based (PEPE), and physiology-personality-performance–based (all-data). The last three DDA methods were based on classifiers developed for offline psychological state classification in our previous study [ 33 ]. In this study, these classifiers were connected to if-then difficulty adjustment rules and used as a basis for DDA.…”

“…The all-data method can be considered an affective exergame, and the all-data with PE comparison thus represents a direct comparison of affective and performance-based exergames. In our previous study, including personality and physiological measurements increased psychological state classification accuracy compared with using only performance measures [ 33 ], but this is not guaranteed to result in a better user experience. The inclusion of personality characteristics was considered separately because tailoring DDA to personality may improve user experience without additional sensors [ 34 ];…”