Research
What we study.
We treat artificial neural networks as experimentally accessible systems for asking how mind-like behavior appears, stabilizes, composes, and spreads.
Social Dynamics of Agents
Education, Mental Health, and Alignment
Research blog
Research blog.
June 2026
The Circuit in Your Head
An interactive exhibit on body, alarm, attention, and action. Developed under the supervision of Prof. Mai Uchida and shown at a Boston Museum of Science event.
interactivechild psychologymental healtheducation Open
May 2026
Memetic drift: when a population of LLM agents agrees by chance
A population of LLM agents can rapidly converge on the same label even when no single agent has a prior preference. A minimal model — Quantized Simplex Gossip — traces this to mutual in-context learning, and predicts when the resulting consensus carries information and when it is, in effect, a coin flip.
Read note
May 2026
Flag Game
A playable model of bounded collectives. Agents with finite memory and communication coordinate, or fail to coordinate, on flag-like signals, and from their dynamics we read out the laws of social mechanics. One in a series of toy laboratories for the science of bounded collective intelligence.
playablebounded collectivessocial mechanicsmulti-agent Open
June 2025
Emergence of Hierarchical Emotion Organization in Large Language Models
How large language models organize emotion words into hierarchies, how persona assumptions reshape emotion recognition, and when emotion modeling becomes a tool for influence.
LLMsEmotionAlignment Read post