EEG Functional Connectivity Predicts Individual Behavioural Impairment During Mental Fatigue

Area of Science:

• Neuroscience
• Cognitive Science
• Human-Computer Interaction

Background:

• Mental fatigue, a common issue, impairs daily activities via the time-on-task (TOT) effect.
• Individual differences in behavior and brain activity hinder accurate fatigue detection.
• Existing methods struggle with personalized fatigue monitoring.

Purpose of the Study:

• To explore the potential of functional connectivity (FC) for predicting individual mental fatigue.
• To develop a data-driven framework for individualized fatigue prediction.
• To advance quantitative fatigue monitoring techniques.

Main Methods:

• Recorded EEG from 40 adults during a sustained attention task.
• Calculated FC using coherence, phase log index (PLI), and partial directed coherence (PDC) across frequency bands.
• Used differences in FC (diff(FC)) to predict time-on-task (TOT) slope and reaction time differences (∆ RT).

Main Results:

• Confirmed significant inter-subject variability in behavioral impairments.
• Achieved high accuracy in predicting individual behavioral impairments using diff(PDC).
• Identified theta and alpha bands as key contributors to prediction models, showing distinct patterns for TOT slope and ∆ RT.

Conclusions:

• Functional connectivity, particularly diff(PDC) in theta and alpha bands, can predict individual mental fatigue.
• This approach moves beyond correlational analysis towards individualized prediction of fatigue-related impairments.
• Findings support the development of practical tools for real-world fatigue monitoring.