Rest Assured: Detecting Mental Fatigue and Recovery with EEG Headphones

Paper Title

Rest Assured: Detecting Mental Fatigue and Recovery with EEG Headphones

Publication Info

• Topic area: Mental fatigue detection and recovery using wearable EEG technology.
• Keywords: Mental fatigue, EEG headphones, recovery, digital breaks, alpha power, social media, video games, mindful rest, cognitive performance, HCI.

Background and Problem

• Problem / challenge: Traditional EEG setups for detecting mental fatigue are cumbersome and impractical for everyday use, hindering naturalistic monitoring of fatigue and recovery. Existing methods like questionnaires are subjective and interruptive.
• Significance: Mental fatigue affects concentration, well-being, and long-term health, making its detection and management critical in modern, cognitively demanding environments.
• Motivation and related work: Prior research has established EEG as a reliable marker of mental fatigue, with alpha and theta power changes linked to fatigue states. Advances in wearable EEG (e.g., headphone-EEG) offer promise for practical applications, but their potential for fatigue and recovery monitoring remains untested.

Solution

• Proposed approach: Use headphone-EEG to detect mental fatigue and recovery during work–break sequences, comparing three break types: video gaming, social media browsing, and mindful rest.
• Novelty:
  1. Demonstrates the feasibility of headphone-EEG for detecting mental fatigue and recovery.
  2. Differentiates recovery effects of digital break activities using EEG, self-reports, and task performance.
  3. Establishes alpha power as a key biomarker for fatigue and recovery in naturalistic settings.
  4. Validates the use of around-the-ear electrodes for practical fatigue monitoring.
• Procedure and key techniques:
  1. Participants completed two 16-minute TloadDback task blocks separated by a 10-minute break (video game, social media, or mindful rest).
  2. EEG data (alpha and theta power) were recorded using an 8-channel headphone-EEG system.
  3. Self-reports (F-ISA, NASA-TLX) and task performance (accuracy, reaction time) were collected to validate EEG findings.
  4. EEG data were preprocessed to remove noise and artifacts, and relative alpha and theta power were analyzed.

Results

• Concrete findings:
  • Alpha power increased during the first task block (indicating fatigue) and decreased during the social media break but remained elevated after the video game and mindful-rest breaks.
  • Social media and mindful-rest breaks showed significant reductions in self-reported fatigue, while the video game break did not.
  • Task performance (accuracy and reaction time) improved after all breaks but deteriorated more rapidly in the video game condition during the second task block.
• Advantage over baselines: Headphone-EEG successfully detected fatigue and recovery dynamics, with alpha power changes aligning with subjective and performance-based measures. Around-the-ear electrodes provided comparable results to full setups.
• Experiments / evaluation:
  • 97 participants (72 valid datasets) completed the experiment in a controlled lab setting.
  • EEG, self-reports, and performance metrics were analyzed across four stages: pre-task, post-task, post-break, and post-second task.
  • Break types were compared for recovery experience, mental workload, and familiarity.
• Limitations and future work:
  • Short experimental duration (60–90 minutes) may not fully capture real-world fatigue dynamics.
  • Controlled lab setting limits generalizability to naturalistic environments.
  • Future studies should explore longer-term effects, real-world applications, and the impact of break familiarity and duration.

Summary

This study demonstrates the feasibility of using headphone-EEG to detect mental fatigue and recovery during work–break sequences. Alpha power emerged as a reliable biomarker, with social media and mindful-rest breaks showing stronger recovery effects than video games. The findings align with self-reports and task performance, highlighting the potential of headphone-EEG for naturalistic fatigue monitoring. Future work should extend these findings to real-world settings and explore personalized fatigue management strategies.