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

Poster A in Poster Session A: Tuesday, August 4, 9:30 – 11:15 am, Kimmel Center, Shorin & Rosenthal Rooms

Signals vs. synchrony: decoding affective processing from voxel activation vs. functional connectivity in task-based fMRI

Cécile Abati¹, Michael Marxen², Dominik Grotegerd², Leonie Esser², Katharina Hierl², Udo Dannlowski³, Roland Zahn², Philipp Kanske², Ramona Leenings⁴, Katharina Förster²; ¹Technische Universität Dresden, ²Dresden University of Technology, ³Westfälische Wilhelms-Universität Münster, ⁴Johannes Kepler University Linz

Presenter: Cécile Abati

Emotions emerge from distributed patterns of brain activity, yet it remains debated whether voxel activation (VA) patterns or functional connectivity (FC) between brain regions provide a better basis for decoding affective states. While several fMRI studies have shown above-chance classification from VA patterns, recent work suggests that FC may capture additional emotion-discriminative information. However, well-powered direct comparisons are lacking. Here, we present a preregistered study comparing VA-based and FC-based decoding of affective processing in 389 healthy participants performing an emotional face-matching task (neutral, happy, fearful, angry faces; shapes). VA patterns and FC matrices serve as input features for a machine-learning pipeline using nested cross-validation with hyperparameter optimization. Classification follows a hierarchical strategy: face vs. shape, emotional vs. neutral face, and emotion categories, with FC and VA compared at each level. We hypothesize that FC will outperform VA, testing whether the FC advantage reported in small samples generalizes to a larger cohort with a more rigorous analytic pipeline.

Topic Area: Methods, Tools, Theory & Neural Coding