Multimodal Physiological and Behavioral Measures to Estimate Human States and Decisions for Improved Human Autonomy Teaming
Abstract
For effective human autonomy teaming to occur, it is essential to manage the interactions between humans and autonomous agents. This leads to optimized performance, ensures resilient and stable team member states, and supports the capability to appropriately build, calibrate, and maintain trust. Toward this aim, it is critical to predict human teammate decisions, and the underlying mental states that drive those decisions. By predicting these decisions, we will be able to design new intervention strategies and technologies, such as display designs, agent feedback, or adaptive behavior, to improve teaming and mitigate possible negative interactions such as performance degradations and miscalibrated trust. In this report, we motivate the importance of estimating the psychological states that impact these decisions and summarize the known relationships they have with physiological and behavioral measures that can be captured in real time with noninvasive or wearable technologies. This provides a foundation to employ a priori constraints on models that utilize multiple physiological and behavioral signals to infer mental or psychological states (stress, fatigue, workload, trust, etc.), to improve prediction of human decisions when interacting with autonomous agents in military-relevant environments.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 02, 2020
- Accession Number
- AD1111968
Entities
People
- Ashley H. Oiknine
- Benjamin Files
- Catherine Neubauer
- Derek Spangler
- Gregory Gremillion
- J. Cortney Bradford
- Kristin E. Schaefer
- Stephen M. Gordon
- Steven M Thurman
Organizations
- United States Army Research Laboratory