I conducted research with the University of Washington in the department of Human Centered Design & Engineering and the Information School, working with faculty and research scientists.
We wanted to examine how people interact with body tracking in VR and whether they collaborate better through VR avatars or real-life interactions. With more teams using avatars in online meetings and people displaying Apple Memoji avatars on FaceTime, we wanted to explore how this impacts communication and teamwork. To measure success rates, we had users complete tasks and play games both in VR with avatars and in real life. We analyzed their experiences, finding challenges with motion capture accuracy, avatar responsiveness, and adapting to VR controls. Some users felt disoriented or experienced physical discomfort, while others found body tracking enhanced immersion. Our insights help improve VR interface design to make virtual interactions feel more natural and effective. Check out our final presented findings.
Working cross-functionally with a team across the University of Seoul, we explored how UX professionals used ChatGPT in their work. We interviewed 30 practitioners from South Korea and the U.S. to understand their experiences. We found key differences: Korean designers struggled with prompt generation, while U.S. designers were more concerned about data privacy. Based on these insights, we proposed design solutions to improve AI tools for UX work. Check out the full published paper submitted to the CHI Conference on Human Factors in Computing Systems.
Conducting research with the Information School, we wanted to study the limitations of Google Translate and how its inaccuracies impact different cultural and ethnic communities. With translation tools playing a key role in communication, we wanted to explore the risks of misinformation, especially in critical areas like healthcare. We conducted surveys and interviews with multilingual users to understand their experiences. We found that translation errors often distort meaning, lack cultural nuance, and can even mislead people in serious situations. Our research highlights the need for more ethical and accurate AI translation tools that respect linguistic and cultural diversity. View our research paper here.