Abstract The rise of videoconferencing (VC) technologies has transformed how individuals collaborate and interact across professional and personal contexts. However, empirical studies comparing VC and face-to-face (FtF) interactions remain fragmented, partly due to a lack of open, multimodal datasets capturing both modalities with rich behavioural, physiological, and self-report measures. To address this gap, we introduce the dataset, comprising approximately 180 hours of audio-visual recordings of unacquainted dyads engaging in structured and creative collaborative tasks under controlled laboratory conditions. Participants were randomly assigned to VC or FtF interaction. The dataset includes six salivary oxytocin measurements, self-reports on affect, personality traits, relevant attitudes, communication outcomes, and a repeated sustained attention task. Behavioural recordings from frontal and side camera views are available for most dyads, with individual-level data for 127–131 participants. The dataset enables research into social bonding, cooperation, behavioural and physiological synchrony, and broader communication dynamics, filling a critical gap in resources for comparative communication studies.
Tsigeman et al. (Thu,) studied this question.