/UnmuteAll: Modeling Verbal Communication Patterns of Collaborative Contexts in MOBA Games

TL;DR

This paper introduces a framework for analyzing verbal communication in MOBA games, revealing how team interactions vary with training and game situations, and providing a tool for assessing communication effectiveness.

Contribution

It presents a novel method for automatically assessing verbal communication patterns in esports, focusing on voice-based interactions in MOBA games.

Findings

Players show broader participation in collaborative situations.

Utterances increase over time, especially during decision making.

Team communication varies based on professional training.

Abstract

Team communication plays a vital role in supporting collaboration in multiplayer online games. Therefore, numerous studies were conducted to examine communication patterns in esports teams. While non-verbal communication has been extensively investigated, research on assessing voice-based verbal communication patterns remains relatively understudied. In this study, we propose a framework that automatically assesses verbal communication patterns by constructing networks with utterances transcribed from voice recordings. Through a data collection study, we obtained 84 game sessions from five League of Legends teams and subsequently investigated how verbal communication patterns varied across different conditions. As a result, we revealed that esports players exhibited broader and more balanced participation in collaborative situations, increased utterances over time with the largest rise in decision making, and team-level differences that were contingent on effective professional training. Building upon these findings, this study provides a generalizable tool for analyzing effective team communication.