From Feedback to Neural Dynamics: Network-Based Modeling of EEG in Interactive Tasks

Abstract

Feedback influences how people learn and make decisions, yet how it reshapes neural activity during interaction remains unclear. In this study, we examine feedback-driven neural dynamics during the Reading the Mind in the Eyes Test using EEG in a within-subject comparison of Feedback and No-Feedback conditions. We introduce two network-based frameworks, Neuro-Epistemic Network Analysis (NENA) and Neuro-Ordered Network Analysis (NONA), to model interactions between EEG frequency bands and behavioral outcomes as neural–behavioral states within an interactive task setting. NENA results show that feedback strengthens associations between higher-frequency activity (Beta, Gamma) and correct responses, while the absence of feedback is linked to lower-frequency activity (Theta, Alpha) and incorrect responses. NONA further shows that, temporally, under feedback, higher-frequency activity more often precedes correct responses (p = 0.01, Cohen’s d > 2). These findings show how feedback reorganizes neural–behavioral dynamics during interaction and suggest opportunities for designing adaptive, feedback-aware interactive systems using wearable neurophysiological sensing. The code and setup to reproduce the experiments are publicly available here.