Wilson Chen

Research work.

Research experience

AR/VR visual comfort and health

Research Assistant · Meta via TEKsystems · 2025–present

I have designed, moderated, and analyzed more than ten human factors studies. Their findings informed constraints, trade-offs, and product decisions on visual comfort and safety in novel AR/VR devices. I run statistical analyses on high-frequency, high-dimensional data with AI-assisted workflows that keep a person in the loop, and build pipelines and scripts for storing and accessing human subject data securely.

Visual misinformation

Graduate Research Assistant · University of Washington · 2024–2025

I coded political imagery, developed image-analysis pipelines, and led research on media provenance presented at the Trust and Safety Research Conference at Stanford.

Case studies

  1. DubsBooks

    I led research with the University Book Store. High student satisfaction and staff priorities shifted our focus toward comparing prices and buying options.

    The store’s leadership wanted students to find the best price, even elsewhere. We designed a portal to compare options and showed which parts depended on the store’s inventory and instructor data.

    Research + Design · 2025

    View case study

  2. Misinformation interventions

    I built the Community Notes prototype and shared research planning and analysis for a think-aloud evaluation of three moderation concepts.

    Participants accepted added context more readily than restriction. We recommended a narrower notification rule, holds only moderators can start, and criteria for choosing journalists. The prototypes were not revised or retested.

    Concept evaluation · 2024

    View case study

  3. Duo account linking

    In a usability study sponsored by Duo, every participant logged in easily, but few linked a personal account, and none without help.

    Linking stalled at awareness, trust in Duo, and the handoff to another app, so our recommendations covered how Duo explains itself as well as the setup screens.

    Usability study · 2024

    View case study

  4. A Digital Ethnography of Horizon Worlds

    In nine weeks of fieldwork, we found that official tools protected individuals, while world creators scripted their own tools against trolls.

    A course study combining about 50 hours of participant observation with semi-structured interviews. I ran half the interviews and observed community events.

    Ethnographic study · 2022

    View case study

Garage

Open the garage