Communication Adaptation in Virtual Teams: An Agent-Based Modelling Approach Using Synthetic Data

This study examines the evolution of communication preferences in virtual teams, focusing on language proficiency, social influence, and technological context. Utilizing Media Richness Theory (MRT) and Adaptive Structuration Theory (AST), it employs a novel computational framework incorporating synthetic data and agent-based modeling (ABM). A dataset of 5,000 synthetic team members simulates baseline communication, while the ABM assesses adaptation dynamics over 20-time steps. Findings indicate that communication preferences are not stable nor linear. Instead, medium-switching behavior arises from interactions among agents influenced by language proficiency, team structure, and peer norms. The results indicate that 42.6% prefer text-based communication, 26.2% voice, and 31.2% video, with a weak correlation (r = 0.08) between language proficiency and adaptation. The study redefines how communication is theorized and modelled in digitally complex team environments, offering a method to study adaptation in the context of immersive and distributed technologies like AI, blockchain, AR, VR, and the metaverse.