An evaluation of mental workload with frontal EEG

Area of Science:

• Neuroscience
• Cognitive Science
• Human-Computer Interaction

Background:

• Assessing mental workload is crucial for optimizing performance and preventing errors.
• Dynamic evaluation of mental workload often requires non-invasive and real-time methods.
• Electroencephalography (EEG) offers a promising avenue for monitoring brain activity related to cognitive load.

Purpose of the Study:

• To investigate the feasibility of using short-term frontal electroencephalography (EEG) to evaluate dynamic changes in mental workload.
• To identify common EEG features indicative of increasing task difficulty.
• To assess the accuracy of classifying mental workload levels from EEG data.

Main Methods:

• Recorded frontal EEG signals from twenty healthy subjects using a wireless, single-channel EEG device.
• Subjects performed four distinct cognitive and motor tasks: arithmetic operation, finger tapping, mental rotation, and lexical decision task.
• Utilized a Support Vector Machine (SVM) model for classification of mental workload levels based on EEG features.

Main Results:

• Theta activity in the frontal EEG consistently increased with task difficulty across all four tasks.
• The level of mental workload could be classified from EEG features with 65%-75% accuracy.
• Short-time analysis windows were effective for capturing dynamic changes in mental workload.

Conclusions:

• Frontal EEG, particularly theta activity, serves as a reliable indicator of mental workload.
• Short-term frontal EEG analysis is a feasible method for evaluating dynamic changes in mental workload.
• This approach holds potential for real-time workload monitoring in various applications.