Trust Measurement using Multimodal Behavioral Analysis and Uncertainty Aware Trust Calibration
Abstract
This report summarizes our major research activities, study results and research accomplishments out of the trust measurement project in the past year. This is also the final report of the project. We have conducted different experiments on trust examination with varied system accuracy, and human trust in predictive decision making. From the study we have found that: 1) people can correctly perceive the accuracy of the system and adjust their trust accordingly; 2) there exists a strong link between human decisions, trust and perception, and trust can be inferred from a couple of decisions; 3) different uncertainty types (e.g. risk and ambiguity) affect human trust differently; 4)cognitive load levels also affect human trust differently because of cognitive resources available. These trust variations can be examined by physiological signals (e.g. GSR). Our future work will focus on investigating other physiological signals (e.g. BVP) as a means to quantify user trust, as well as identifying the trust patterns when humans play different roles in the human-machine collaboration.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 05, 2018
- Accession Number
- AD1046153
Entities
People
- Fang Chen
Organizations
- National ICT Australia