Dynamic facial trustworthiness perception in real-time social contexts

Haoming Qi¹Dongcheng He^2,3*

• ¹Chinese People's Public Security University, Beijing, Beijing, China
• ²School of Optometry, University of California, Berkeley, Berkeley, United States
• ³Hefei Comprehensive National Science Center Data Space Research Institute, Hefei, Anhui Province, China

Current understanding of the neural mechanisms underlying facial trustworthiness perception is primarily based on studies using static facial stimuli.However, real-life social interactions are dynamic and complex, and the neural processes involved in such naturalistic contexts remain largely unexplored. In this study, we analyzed EEG data collected by Chen et al. ( 2024) during a deception game involving two participants: a player and an observer engaged in real-time interaction. The player either followed instructions or made spontaneous decisions to lie or tell the truth, while the observer judged whether to trust the player based solely on his or her facial expressions. We examined observers' behavioral data, event-related potentials, and interhemispheric EEG asymmetries in both signal magnitude and instantaneous phase.The results revealed a significant effect of trustworthiness on hemispheric asymmetry in the observer's centroparietal phase activities especially after ~800 ms post-stimulus until the end of the trial at 3000 ms post-stimulus. Subsequent frequency-based analysis revealed that this asymmetry in phase progression was primarily driven by lateralized signal frequency. These findings suggest that the perception of facial trustworthiness involves dynamic hemispheric lateralization. Whereas previous studies using static face stimuli indicate that trustworthiness perception occurs rapidly, our findings suggest that trustworthiness perception can be modulated by persistent and dynamic affective processing in real-time social contexts.