VTuber self-presentation is built with a character and a crowd

Nine interviews and seven observed channels

Qian Wan and Zhicong Lu studied Chinese-speaking VTubers through semi-structured interviews with nine creators and observations of archived videos and livestreams from seven additional channels. The researchers examined identity construction, avatar-mediated performance and gender expression. They describe creators combining elements of their offline selves with invented identities, performing in exaggerated ways, and negotiating sexualized gender presentation in misogynistic settings. All participants and observed channels belong to one linguistic and cultural context, and the authors flag limits around culture, affiliation and disclosure.

The model is part of the social script

The paper is useful because it does not reduce a VTuber model to a privacy screen. Character lore, clothing, expressions and voice give a creator material to perform with, while chat and fan suggestions can feed back into the character. That makes self-presentation partly collaborative. A redesign can therefore alter more than appearance: it may change the jokes an audience recognizes, the gestures a performer can deliver and the boundary between character continuity and personal change.

For creators, the practical question is not whether an avatar is authentic in the abstract. It is which parts of the presentation are chosen, negotiated or imposed. A useful model brief can record who controls lore changes, how audience ideas are accepted, and which expressions the performer does not want to repeat. Those decisions make authorship visible without demanding disclosure of the person behind the rig.

Do not flatten a situated account

A sixteen-channel qualitative study can reveal mechanisms and tensions; it cannot estimate how common they are across languages, agencies or independent scenes. Two interviewees also appeared through virtual identities during interviews, which the authors note may have shaped disclosure. The result supports careful questions about pseudonymity, gender and co-created identity. It does not establish that VTubing generally prevents harassment or that exaggerated performance affects every creator in the same way.