Investigating the effects of appearance style, familiarity, and facial animation source on the perception of digital doubles in desktop and VR
Year: 2026
Type: article
Abstract: Creating human digital doubles is becoming easier and more accessible through consumer-grade devices. In this work, we investigate the perception of 3D digital doubles through two experiments with a focus on appearance style, familiarity, and facial animation source. The desktop-based offline Experiment I focuses on how appearance style (realistic versus cartoonish) and familiarity (self, acquaintance, unknown person) affect self/other-identification, perceived realism, affinity, and social presence. Realistic digital double configurations received higher identification, perceived realism, and social presence ratings than cartoonish configurations. Participants rated self-like digital doubles as less realistic in appearance and overall, and as eerier, than unknown-person digital doubles. Both cartoonish and realistic digital doubles were animated using performance-driven facial animation in this experiment. In the offline VR-based Experiment II, we compare audio-driven and performance-driven facial animation. Audio-driven animation pipelines are becoming part of game engines, but there is not yet a clear understanding of their effects on the perception of digital doubles. In Experiment II, the evaluated performance-driven configuration received higher perceived realism, affinity, and social presence ratings than the audio-driven configuration. On the other hand, no statistically significant main effects of familiarity or interactions between familiarity and facial animation source were found in Experiment II. Together, the experiments provide complementary evidence on how appearance style, facial animation source, and familiarity relate to digital double perception in distinct desktop and VR contexts.
Source: Computers & Graphics
Institution: Utrecht University
Language: English
Subfield: Social Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: hybrid
Estimated APC paid: $2,770
OpenAlex ID: W7163580404
