Behavioral research for platforms that want to understand their users, not just measure them.
Tenured researcher and practicing consultant. My work spans social platforms, online gaming, VR/XR environments, and trust & safety, with formal research partnerships with Riot Games and VRChat ongoing. Co-PI on a completed $1.6M NSF initiative spanning five studies and 13 peer-reviewed papers. Through SIM DAD LLC, I ship the research tools the work depends on.
02 · Research
How people form relationships, experience presence, and behave in virtual spaces, including social VR platforms, virtual classrooms, and immersive games. Foundational for platform design, onboarding systems, and safety policy in emerging spatial computing environments.
Harassment typologies, hate-speech detection, moderation effectiveness, and the behavioral mechanics of toxic and prosocial communities. Relevant to trust & safety teams, platform policy groups, and anyone auditing community governance systems.
How audiences form emotional bonds with streamers, creators, and characters, and what platform features drive or disrupt those bonds. Built on large-scale behavioral datasets including 321M+ Twitch chat messages and 5M+ proprietary player records from Riot Games.
Surveys, experiments, semi-structured interviews, NLP corpus analysis, and structural topic modeling, often combined in a single study. For anyone who needs findings that hold up to scrutiny from product teams, program officers, and peer reviewers alike.
03 · Track Record
Five-study mixed-methods program on virtual-meeting equity for game developers (2021–2025). Formal data-access partnership with VRChat; produced 13 peer-reviewed publications and 7 conference proceedings across an international team of 30+ researchers.
Formal research data agreement with Riot Games providing access to 5M+ proprietary player behavioral records. Findings on gaming behavior and identity published in a peer-reviewed journal.
Distributed Raspberry Pi cluster collecting 321M chat messages across 117,943 channels. Developed an original NLP verbal-immediacy scoring methodology; published findings on parasocial relationships at scale in Communication Monographs.
Formal data-access partnership with VRChat supporting the NSF program's virtual-meeting studies, enabling behavioral research inside a live social-VR platform at scale.
04 · Engage
Applied behavioral research for product, policy, or strategy teams. Study design through final deliverable: methods, analysis, and findings you can use.
Seeking Co-PI and co-investigator relationships for NSF, NIH, or private-foundation proposals. Particularly interested in human-computer interaction, platform behavior, and AI-augmented environments.
Conference keynotes, executive briefings, and workshops on social platforms, trust & safety, VR/XR behavior, and computational research methods.
Research tools built at SIM DAD LLC, including TASS (text analysis), Ibis (transcription), Lector (read-along TTS reader), and Resero (vault recovery), available for institutional and commercial licensing. See simdadllc.com.
05 · Publications
Leith, A. P. (2021). Parasocial cues: The ubiquity of parasocial relationships on Twitch. Communication Monographs, 88, 111–129. doi.org/10.1080/03637751.2020.1868544
Demonstrates how Twitch's live platform features systematically drive parasocial bond formation at scale, establishing a corpus-based methodology for measuring viewer–streamer relationships across 321M+ messages.Foxman, M., Beyea, D., Leith, A. P., Ratan, R. A., Chen, V. H. H., & Klebig, B. (2022). Beyond genre: Classifying virtual reality experiences. IEEE Transactions on Games, 14(3), 466–477. doi.org/10.1109/TG.2021.3119521
Proposes a taxonomy that moves beyond gaming-genre labels to classify the full spectrum of VR experiences, giving platform designers and researchers a shared vocabulary for what immersive media actually is.Lim, C., Ratan, R., Foxman, M., Beyea, D., Jeong, D., & Leith, A. P. (2025). Examining attitudes about the virtual workplace: Associations between Zoom fatigue, impression management, and virtual-meeting adoption intent. PLOS One. doi.org/10.1371/journal.pone.0312354
Finds that Zoom fatigue and impression-management anxiety predict lower intent to adopt virtual meetings, with direct implications for how platforms design meeting tools and frame adoption to their users.06 · Projects
Local AI tutor · reflective journaling · instructor oversight dashboard
Research tools for institutional and commercial licensing