QUANTA: Quantitative Network-based Models of Adaptive Team Behavior

Abstract

This Multi-University Research Initiative will advance the development of the Network Science of Teams by creating Quantitative Network-based Models of Adaptive Team Behavior. The overarching contribution of this research is to build quantifiable infonnative models of team behavior as dynamical systems interacting over multiple networks, develop rigorous models that relate interaction patterns and network evolution to task performance, break new ground in the learning of optimal design of teams, scaling teams to complex tasks, and advancing social science theories of team performance. This research will produce methods to optimize team performance under different contexts and resource constraints. The research team will identify bottom-up approaches for improving team performance. They will identify meso-level approaches to get teams to adapt and increase their robustness. And finally, they will develop top-down strategies for effectively assigning individuals to teams, and teams to tasks. The proposed effort brings together seven academics from five universities and pulls together an excellent balance of multidisciplinary scholars from sociology, cognitive and social psychology, health and behavioral sciences, computer science, statistics, controls and dynamical systems, and network science. Thrust 1 (teams as networks of interacting entities) will model teams as semantically-labeled multilevel networks with attributes on nodes, edges, and subgroups and these conceptual models will be experimentally validated; develop statistical models of team performance as information flows; and examine the relationship between influence networks, collective intelligence, task perfonnance, and robustness and optimization. Thrust 2 (analysis and models of dynamic team behavior) will examine the perfonnance of teams on a sequence of tasks. They will examine changes in social influence networks resulting from repeated tasks and examine team adaptability using methods from computational learning and data-driven optimization. Thrust 3 (network science of teams-of-teams) will examine the scalability of teams and tasks. This thrust will consider how to decompose a complex task into subtasks, the effect of inter-task dependencies, and how to assign teams to such tasks.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 12, 2017
Source ID
W911NF1510577

Entities

People

  • Ambuj Singh

Organizations

  • Army Contracting Command
  • United States Army
  • University of California, Santa Barbara

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Neural Network Machine Learning.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.