The role of shared mental models in human-AI teams: a theoretical review

Abstract

Mental models are knowledge structures employed by humans to describe, explain, and predict the world around them. Shared Mental Models (SMMs) occur in teams whose members have similar mental models of their task and of the team itself. Research on human teaming has linked SMM quality to improved team performance. Applied understanding of SMMs should lead to improvements in human-AI teaming. Yet, it remains unclear how the SMM construct may differ in teams of human and AI agents, how and under what conditions such SMMs form, and how they should be quantified. This paper presents a review of SMMs and the associated literature, including their definition, measurement, and relation to other concepts. A synthesized conceptual model is proposed for the application of SMM literature to the human-AI setting. Several areas of AI research are identified and reviewed that are highly relevant to SMMs in human-AI teaming but which have not been discussed via a common vernacular. A summary of design considerations to support future experiments regarding Human-AI SMMs is presented. We find that while current research has made significant progress, a lack of consistency in terms and of effective means for measuring Human-AI SMMs currently impedes realization of the concept.

Keywords:

• mental models
• shared mental models
• explainable AI
• human-machine teaming
• human-AI teaming

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Here we use the term AI to represent the full range of possible automation and autonomy, whether enabled by learning agent such as ML, deep learning, artificial intelligence, or a more static reasoning system.

2 A further model emphasizing the information content, rather than the formation, of this relationship can be found in (of Sciences Engineering and Medicine 2022, 28); see our addendum in §8.

Additional information

Notes on contributors

Robert W. Andrews

Robert "Bowe" Andrews is a Lieutenant in the United States Air Force and student pilot in the Euro-NATO Joint Jet Pilot Training (ENJJPT) program at Sheppard AFB. He received undergraduate degrees from Georgia Tech in 2020 in Russian and Mechanical Engineering, and a Master of Science in Aerospace Engineering in 2021. During graduate school, Bowe conducted research with the Cognitive Engineering Center studying shared mental models in human-AI teams. He is particularly passionate about the optimization of next generation human-machine aerospace systems.

J. Mason Lilly

J. Mason Lilly is an M.S. Computer Science student and research assistant in the Cognitive Engineering Center at Georgia Tech. His chief interests are deep learning applications in robotics and human-AI interaction. He earned his B.A. in Computer Science & Applied Music from Transylvania University in 2016. Following graduation, he will be joining Microsoft in Atlanta in a robotics role. In his spare time, Mason plays clarinet and mandolin with his family’s band and practices photography.

Divya Srivastava

Divya Srivastava is a fourth-year Mechanical Engineering Ph.D. student in the Cognitive Engineering Center at Georgia Tech. She is researching how shared mental models can be developed and maintained for increased performance in human-autonomy teams. Her research interests are in HRI and user-centric product design. Previously, Divya has worked for Sandia National Laboratories, Amazon Robotics, the Office of Naval Research, and NASA’s Jet Propulsion Laboratory, all in the general field of human-robot interaction or autonomous vehicles. Divya received her M.S. degree in Mechanical Engineering from Georgia Tech. She received her B.S. degree in Mechanical Engineering (Honors) with thesis option and concentration in Aerospace, along with a minor degree in Computer Science from Rutgers University. In her spare time, Divya enjoys cooking, reading fiction novels, and singing off-key to the radio.

Karen M. Feigh

Karen M. Feigh is a professor of cognitive engineering at Georgia Institute of Technology. She earned her PhD in industrial and systems engineering from Georgia Institute of Technology and her MPhil in aeronautics from Cranfield University in the United Kingdom. Her current research interests include human-automation interaction and using work domain analysis in systems engineering.