Virtual Coach Application in Serious Games: Effects on Player Experience and Learning Outcomes

Abstract

In an effort to improve serious games, developers have increasingly started to implement virtual coaches. Many studies examining the impact of these coaches have found a positive effect on learning outcomes, but some studies failed to observe such an improvement. Research on whether virtual coaches beneficially affect the motivation and engagement of players is scarce. This research aims to contribute to the existing body of literature by assessing the effects of a virtual coach on both learning outcomes and player experience (i.e. motivation and engagement). The hypothesis is that the addition of a virtual coach improves these outcomes in a serious game aimed at teaching people how to effectively give feedback to others. A between-subjects research design was used. Participants were randomly allocated to one of two conditions: one group (n = 35) playing the serious game with a virtual coach and the other group (n = 35) playing without such a coach. Independent samples t-tests showed no significant effect of the coach on learning outcomes. In addition, no significant effect on motivation or engagement was observed. In accordance with previous studies, a significant negative correlation between age and learning outcomes was found, whereas in contrast to earlier research, females did not achieve better learning outcomes compared to males. This study further illustrates the well-known problem of the lack of generalisability regarding the findings of research examining virtual coaches. The limitations of this study concerning the application of the virtual coach are discussed extensively, culminating in recommendations of tangible steps that can be taken to improve the validity of similar research in the future.