Nykko Vitali
I am a PhD student in Psychology at Harvard University. I work primarily with Jason Mitchell and Mahzarin R. Banaji.
My research program is bi-directional. On one end, I combine behavioral experiments and computational modeling to study how people metacognitively evaluate information, update their beliefs, and what cues they focus on in digital environments. On the other, I am quite interested in exploring what aspects of human cognition have seeped into the weights of LLMs, and how aligned they are with human behavior.
I also love stand-up comedy, building tools for various applications, and the manga One Piece :)
Current Focus
Human-AI interaction and belief updating during uncertainty & Reasoning inconsistencies in LLMs
Publication Spotlight
Lehr, S. A., Saichandran, K. S., Harmon-Jones, E., Vitali, N., & Banaji, M. R. (2025).
Kernels of selfhood: GPT-4o shows humanlike patterns of cognitive dissonance moderated by free choice.
Proceedings of the National Academy of Sciences, 122(20), e2501823122.
Read the paper
Recent Poster Presentation
See past posters →Side Projects
ClaudeR
R MCP integration for interactive, agent-assisted coding in RStudio. View repository
Collective Anthropomorphization
My exploration into how viewers on a Twitch stream watched a reasoning trace visible LLM play Pokemon Red for three days straight including regression and topic models. View write-up
A Quantitative Comaprison of Wikipedia vs. Grokipedia
A comparative project exploring information quality and narrative framing across model-generated and encyclopedia-style content. View repository