Related Experiment Video
Updated: Jul 15, 2025
07:48
Eye Tracking During A Complex Aviation Task For Insights Into Information Processing
Published on: April 4, 2025
358
Eye-Tracking in Physical Human-Robot Interaction: Mental Workload and Performance Prediction
Satyajit Upasani1, Divya Srinivasan2, Qi Zhu3
1Virginia Tech, Blacksburg, VA, USA.
Human Factors
|October 4, 2023
Summary
Eye-tracking measures reliably assess cognitive load in physical human-robot interaction (pHRI). These metrics can predict task success and adapt robots to human skill levels during collaboration.
Area of Science:
- Human-Robot Interaction
- Cognitive Science
- Robotics
Background:
- Physical Human-Robot Interaction (pHRI) requires learning robot dynamics, increasing cognitive load.
- Eye-tracking metrics offer insights into mental workload fluctuations during learning.
Purpose of the Study:
- Assess sensitivity and reliability of eye-tracking for task difficulty variations in pHRI.
- Evaluate eye-tracking's capability to predict performance in human-robot collaboration.
Main Methods:
- Participants performed a virtual pick-and-place task with a bimanual robot.
- Robot joint stiffness was manipulated to alter motor-coordination demands.
- Investigated psychometric properties of eye-tracking measures and performance prediction.
Main Results:
- Stationary Gaze Entropy and pupil diameter showed high reliability and sensitivity to workload changes.
- Increased task difficulty led to more robot-monitoring strategies.
- Eye-tracking predicted trial success/failure with 70% sensitivity and 71% accuracy.
Conclusions:
- Eye-tracking measures demonstrate acceptable sensitivity and reliability for pHRI workload.
- Gaze behaviors indicating visual monitoring are sensitive to task difficulty.
- Further exploration of eye-tracking in pHRI is warranted for understanding workload and internal-model formation.
