Date of Award

12-2024

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical Engineering and Computer Science

First Advisor

Thomas C. Eskridge

Second Advisor

Siddhartha Bhattacharyya

Third Advisor

Moti Mizrahi

Fourth Advisor

Troy R. Weekes

Abstract

Increasingly capable machines, including Artificial Intelligence (AI) agents are playing a more important role in a wide range of applications, including human daily activities and safety-critical systems. They can benefit even more when humans and such machines agents work together as a team by leveraging each other's strengths and complementing each other to enhance overall performance. To design high-performing teams, it is critical to analyze the team dynamics and understand how humans and machines interact with each other. Collaboration, Coordination, and Cooperation (3Cs) are terms typically used to describe the behavior of teams. However, these terms tend to be used interchangeably as well as to be defined in different ways, introducing an inconsistency across studies. The primary contributions of this dissertation are fourfold. First, we propose a new approach to classifying Human-Machine teams based on interdependence and input compositionality (i.e., how inputs from team members are used), enabling a more informative, consistent use of 3Cs. Second, we present a series of experiments that investigated patterns of interactions to study if there is a ``sweet spot'' to achieve higher team performance; we got inspiration of the notion of a sweet spot from social physics. The experimental results found different patterns of interactions between high- and low-performing teams and suggested that User Interface (UI) design could help humans better interact with agents to achieve higher team performance. Third, based on our findings from the experiments, we offer preliminary design guidelines aimed at better analyzing and enhancing human-machine team performance by associating them with our proposed team structures. These guidelines are expected to be applicable to any teams that follow the outlined team structures. Fourth, we discuss the implications of our proposed approach to classifying team structures and highlight potential areas for future research such as investigating patterns of transitions between team structures and exploring differences in trust and reliance models across team structures. Altogether, this dissertation serves as a foundation for creating high-performing and high-functioning human-machine teams (i.e., not only human-machine teams exhibit high team performance, but humans maintain a high level of Quality of Life).

Download

Share

COinS