Predicting Team Performance from Communications in Simulated Search-and-Rescue

2025-03-05

AI Summary

This study addresses the challenge of predicting team performance and uncovering how unobservable individual traits influence collaborative effectiveness, using conversational data from simulated search-and-rescue tasks. Leveraging a Minecraft-based rescue environment, we extract multidimensional communication behavioral features and propose a novel method integrating LDA topic modeling with unsupervised clustering to infer latent team traits—such as leadership and coordination propensity—in real time. Subsequently, multilevel correlation analysis quantifies the differential contributions of these traits to task success. Results identify three distinct communication patterns significantly predictive of team success; moreover, jointly modeling individual traits and team dynamics improves prediction accuracy by 27% over baseline models. This demonstrates that dialogue data robustly encode implicit collaborative competencies. The work establishes an interpretable, computationally tractable paradigm for team effectiveness assessment grounded in behavioral linguistics and computational social science.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Multiagent Systems: Teamwork
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Topic discovery and tracking
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

Understanding how individual traits influence team performance is valuable, but these traits are not always directly observable. Prior research has inferred traits like trust from behavioral data. We analyze conversational data to identify team traits and their correlation with teaming outcomes. Using transcripts from a Minecraft-based search-and-rescue experiment, we apply topic modeling and clustering to uncover key interaction patterns. Our findings show that variations in teaming outcomes can be explained through these inferences, with different levels of predictive power derived from individual traits and team dynamics.

Problem

• Analyze conversational data to identify team traits.
• Correlate team traits with teaming outcomes using topic modeling.
• Predict team performance from communication patterns in simulations.

Innovation

• Analyze conversational data for team traits
• Apply topic modeling and clustering techniques
• Predict team performance using interaction patterns