An oculometrics-based biofeedback system to impede fatigue development during computer work: A proof-of-concept study

Area of Science:

• Human-Computer Interaction
• Cognitive Science
• Ergonomics

Background:

• Fatigue significantly impacts performance during prolonged computer use.
• Objective identification of early-stage fatigue is crucial for effective intervention.
• Current micro-break strategies often lack personalization and objective triggers.

Purpose of the Study:

• To develop and evaluate a biofeedback system using oculometrics to detect fatigue.
• To compare the effectiveness of biofeedback-triggered micro-breaks versus self-triggered breaks.
• To assess the impact of the biofeedback system on perceived workload and fatigue.

Main Methods:

• Developed a biofeedback system using eye movements and pupil dilation (oculometrics).
• Implemented a Decision Tree Ensemble model for fatigue classification based on oculometrics.
• Conducted a study with 20 participants comparing manual vs. biofeedback-triggered micro-breaks during a computer task.
• Measured perceived fatigue using Karolinska Sleepiness Scale (KSS) and workload using NASA-TLX.

Main Results:

• The biofeedback system achieved approximately 70% accuracy in identifying fatigue.
• Participants reported significantly lower perceived workload in biofeedback-triggered sessions compared to manual sessions (p = 0.01).
• Oculometrics-based biofeedback effectively moderated fatigue development.

Conclusions:

• Integrating oculometrics-based biofeedback into micro-break timing plans can effectively impede fatigue during computer work.
• Biofeedback-triggered micro-breaks are more effective in reducing perceived workload than self-triggered breaks.
• This system offers a promising approach for objective fatigue management in computer-intensive tasks.