A particular element of computer games which significantly affects the gaming experience is avatars, i.e., representations of virtual players. In particular player-avatar similarity is related to the game experience, however, the results of previous studies are not conclusive. There is also a lack of research in which participants actually create their own avatars. The first aim of this study is to check whether there is a relationship between the player-avatar similarity and the game experience: game efficacy, game enjoyment and immersion. As a second goal the study examines gender as the mechanism responsible for the relationship between avatar player-similarity and game experience: game enjoyment, game efficacy and immersion. In the study (N = 130) participants created their avatars, played designed computer game Characterium and assessed their experience. The results demonstrate that player-avatar similarity was positively related to immersion level (but not to game efficacy and game enjoyment). Gender was not a significant moderator of the relationships between avatar similarity and game efficacy, game enjoyment and immersion. We consider this study to be a step forward in better understanding the mechanisms underlying the relationship between player-avatar similarity and game experience. For the first time, the relationship between player-avatar similarity and immersion was demonstrated in the condition of creating an avatar by the subjects and controlling it in the game. We also identify potential reasons why our hypotheses have not been confirmed.

Avatars are virtual representations of virtual world users. Creating a virtual representation can be considered an important element of the self-esteem formation process in modern adolescents. As previous research suggests, the similarity level between an avatar and its creator is related to the latter’s self-esteem. However, previous studies were limited by small and unrepresentative samples, which creates difficulties in generalizing the results. Moreover the studies usually did not include adolescents. This study aims to investigate possible predictors of adolescent–avatar similarity (AAS), in particular global self-esteem, which develops in adolescence. In addition to self-esteem, the importance of personality and gender was also examined. 130 high school students played the Characterium computer game, which was designed for this project. The participants also completed measures of global self-esteem and personality and explained why they had created such avatars. In order to test the research hypotheses, hierarchical regression analyses were performed. No relationship was found between adolescents’ global self-esteem and AAS; however extraversion and gender were significant predictors of adolescent–avatar similarity. Due to the limited number of studies in adolescents, we discuss our results based on the results of adults. The lack of verification of the first hypothesis may be due to the fact that adolescents use avatars to experiment with their own identity, regardless of their self-esteem. This suggests that they use avatars for a different purpose than adults (who want to boost their self-esteem). We discuss gender and personality results in terms of gender and personalty differences in (adult) players. Additionally we propose a hypothesis of cognitive overload (for adolescent extroverts).