Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster E in Poster Session E: Thursday, August 6, 10:30 am – 12:15 pm, Kimmel Center, Shorin & Rosenthal Rooms

Behavior and Autonomic Physiology Characterize Brain State Dynamics During Naturalistic Gameplay

Yibei Chen¹, Yann Harel², Satrajit S Ghosh¹; ¹Massachusetts Institute of Technology, ²Centre de Recherche de l'IUGM

Presenter: Yibei Chen

How does the brain organize continuous, goal-directed behavior? Most brain state research relies on resting-state or passive paradigms where behavior is absent or impoverished. We analyzed fMRI data from five participants who each completed 18-30 sessions of Super Mario Bros gameplay, with concurrent behavioral event annotations and autonomic physiology (electrodermal activity, respiration). Fitting a Gaussian hidden Markov model per subject to parcellated, PCA-reduced brain data, we discovered 9-11 active brain states per individual. Action rate and time since last death ranked among the top three predictors of state identity for all five subjects, with autonomic signals contributing independent explanatory power beyond behavior alone. Death events produced the most consistent state reorganization. Despite fully independent modeling, 8 of 11 consensus states were shared across all five subjects (mean cosine similarity = 0.72, accounting for PCA sign ambiguity; p < 0.001). These results demonstrate that brain states during active naturalistic behavior are governed by a compact, shared neural repertoire whose transitions are jointly shaped by behavioral events and bodily arousal.

Topic Area: Methods, Tools, Theory & Neural Coding