Avatar Customisation and Empowerment in VR Collaboration – A New Study on Self-Representation and Self-Perception

In a newly published article in New Media & Society, we investigated how the ability to customise avatars in social immersive virtual reality (VR) affects users’ self-perceptions during collaborative tasks. The study, titled “Virtually Better: Multi-user Experiment on Avatar Self-representation, Self-discrepancies, Avatar Style and Self-perceptions in a VR Collaboration” (Bujić et al., 2025), addresses a timely set of questions at the intersection of digital identity, avatar representation and embodiment, and self-perceptions in shared virtual spaces.

Drawing on media and social psychology and human-computer interaction we designed a controlled multi-user lab experiment to compare the psychological impact of two conditions: one where participants were free to create their own avatars, and another where avatars were pre-assigned and visually neutral. Sixty-six participants took part in the study at Tampere University, completing a collaborative puzzle-solving task in a virtual reality environment that simulated social VR interactions. Pre- and post-task questionnaires captured self-evaluation and feelings of external shame (conceptualised as ‘other-as-shamer’), while participants also provided qualitative reflections on their avatar design choices.

The study’s findings offer a nuanced contribution to a gap in existing literature on avatar customisation and its implications in social VR. Participants who were able to create their own avatars reported a small but consistent increase in self-evaluation after the task, suggesting that agency over one’s digital self-representation may foster a sense of personal alignment or empowerment. Interestingly, this effect was not mirrored in changes to external shame; participants did not report significant shifts in how they thought others perceived them based on the avatar they used.

One of the original aspects of the study lies in how avatar design preferences were linked to individual self-discrepancy profiles in an experiment in social VR. Participants’ motivations for avatar creation were categorised according to whether they aimed to represent their actual self, their ideal self, how they believed they ought to be perceived, or a more playful or a symbolic version of themselves. Those who chose idealised or playful avatars were more likely to show improvements in self-evaluation, hinting at the expressive, playful, and potentially therapeutic value of customisation in VR. Ultimately, this study contributes to a growing recognition that the psychological dynamics of self-representation in immersive environments deserve closer attention.


While prior work on the Proteus effect has shown that avatars can influence user behaviour and perception, this study refines the conversation by foregrounding the role of user agency in customisation and the psychological mechanisms that underlie self-representation choices. It adds important context to how users bring their own identities, aspirations, and social considerations into digital embodiment—especially in collaborative or socially meaningful VR settings.

The implications extend to several fields. For researchers and designers working in human–computer interaction (HCI), immersive media, or educational technology, the findings suggest that avatar customisation is not merely a visual preference but a design feature with potential cognitive and affective consequences. In areas such virtual group work, social virtual reality platforms, and even mental health contexts, allowing users to personalise their avatars may offer subtle yet meaningful benefits for confidence and engagement.


Bujić, M., Macey, AL., Kerous, B., Buruk, O., & Hamari, J. (2025). Virtually Better: Multi-user Experiment on Avatar Self-representation, Self-discrepancies, Avatar Style and Self-perceptions in a VR Collaboration. New Media & Society (First published: April 22, 2025).

DOI: https://doi.org/10.1177/14614448251323904

Authors: Mila Bujić. Anna-Leena Macey, Bojan Kerous, Oğuz Buruk, and Juho Hamari

Funding: This research was funded by the Research Council of Finland (342144; ‘POSTEMOTION’), Foundation for Economic Education (16-9394) and Finnish Cultural Foundation (00230774).

#SocialVR #Embodiment #AvatarResearch #Embodiment #SelfDiscrepancy #VirtualIdentity #Gamification #HCI #Metaverse

No Comments

Sorry, the comment form is closed at this time.