Educational game to model learner's personality

Abstract

Educational games have become an integral part of the educational process at various levels, with their usage rapidly increasing as innovative e-learning methods. These games have proven to be highly engaging and effective in knowledge retention. Traditionally personality assessment relies on psychometric questionnaires, with the Big Five Inventory (BFI) being a widely used tool. However, these approaches often have certain drawbacks as respondents tend to carefully consider their answers, prioritizing correctness over authenticity. To address these limitations, novel approaches are being developed that incorporate gaming elements to indirectly measure personality. Therefore, an intriguing question arises: Can the subconscious moves, choices, and behaviours exhibited during gameplay serve as indicators of players’ personality? In this thesis, an educational game focused on Databases courses for university students was developed. The game aims to capture everyday life experiences at the university such as social connections and curiosity or willingness to try new things, based on the Five-Factor Model (OCEAN). The objectives were to strengthen the knowledge and comprehension of Databases subject and also to gather information about players’ gaming behaviour and thus predict their scores on two personality traits: Extraversion and Openness to Experience, based on the Five- Factor Model. A total of 149 computer science students of the University of Macedonia participated in the study by playing the game and completing the BFI questionnaire. We utilized classification algorithms (Naive Bayes Classifier, Decision Trees, k-Nearest Neighbour (k-NN) and Logistic Regression) to develop a model to predict students’ personality. The goodness of the model was assessed using different metrics and the results showed that it is effective to model both the Extraversion and Openness personality dimensions using serious games instead of questionnaires. These findings can be used by educators and game designers to develop personalized educational games considering the learner’s personality and thus provide valuable insights for future research in this domain.