Monitoring task loading with multivariate EEG measures during complex forms of human-computer interaction

Area of Science:

• Neuroscience
• Cognitive Psychology
• Human-Computer Interaction

Background:

• Cognitive load assessment is crucial for optimizing performance in complex tasks.
• Electroencephalography (EEG) offers a non-invasive method for measuring brain activity.
• Existing methods for quantifying cognitive load from EEG may lack participant-specific calibration.

Purpose of the Study:

• To investigate the relationship between task difficulty and EEG activity.
• To develop and validate a participant-specific EEG-based index for measuring cognitive load.
• To explore the potential of EEG for monitoring workload in computer-based tasks.

Main Methods:

• 16 participants performed a flight simulation task at varying difficulty levels (low, moderate, high).
• Electroencephalographic (EEG) data, specifically frontal midline theta and alpha band activity, were recorded.
• A personalized algorithm combined multiple EEG features to create a task load index, validated against new task data.

Main Results:

• Increased task difficulty correlated with elevated frontal midline theta EEG activity.
• Higher task difficulty was associated with decreased alpha band EEG activity.
• The participant-specific task load index demonstrated a significant, systematic increase with task difficulty.

Conclusions:

• Frontal midline theta and alpha band activity are sensitive indicators of cognitive load during computer-based tasks.
• A multivariate, participant-specific EEG index can reliably quantify cognitive load.
• This research supports the use of EEG for real-time workload monitoring in naturalistic computing environments.