Virtual characters, real emotions: Avatar identification and psychological transformation across esports and K-pop fandoms

As virtual idols become central actors in transmedia entertainment ecosystems, the psychological connections between fans and virtual characters have grown more diverse and complex. K/DA, a virtual girl group that blends the League of Legends game universe with K-pop idol imagery, serves as a representative case of transmedia character design. Guided by social cognitive theory and drawing on self-expansion theory, this study examines how fans form avatar identification through engagement with virtual characters. We distinguish three fan groups—esports players, K-pop fans, and dual fans—and construct a psychological transformation model based on four analytical categories: avatar characteristics, celebrities’ real-world influence, media richness, and self-image congruence. This study collected survey data from 800 valid respondents and analyzed the data using partial least squares structural equation modeling. The results indicate that narrative engagement, ideal self-image congruence, and ideal others-image congruence are stable predictors of avatar identification across fan groups, whereas physical attractiveness, worship, and vividness exhibit group-specific effects. These findings deepen the understanding of the psychological processes underlying avatar identification in transmedia contexts and provide empirical implications for virtual idol management and fan engagement strategies.