Assessment of mental workload across cognitive tasks using a passive brain-computer interface based on mean negative...

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Signal Processing

Background:

• Brain-computer interfaces (BCI) offer real-time mental workload assessment for optimizing human-computer interaction.
• Classifying mental workload is challenging due to task-dependent neural signals, limiting classifier generalization across tasks.
• Previous methods for cross-task mental workload classification have shown limited success.

Purpose of the Study:

• To introduce and evaluate a novel algorithm for extracting frontal theta oscillations from EEG data.
• To demonstrate the algorithm's capability in detecting mental workload across different cognitive tasks.
• To compare the novel algorithm's performance against existing methods for mental workload classification.

Main Methods:

• Utilized a published dataset investigating subject-dependent task transfer.
• Applied a novel algorithm to extract frontal theta oscillations from electroencephalographic (EEG) recordings.
• Employed Filter Bank Common Spatial Patterns for data analysis and classification.

Main Results:

• The novel algorithm achieved high binary classification performance (92.00% and 92.35%) for workload vs. no workload conditions.
• Performance did not exceed chance level when comparing high vs. low workload conditions.
• While improving stability across tasks, independent component analysis preprocessing did not enhance performance over previous methods.

Conclusions:

• The proposed frontal theta oscillation algorithm shows potential for specific mental workload comparisons within BCI applications.
• It may outperform existing methods in certain workload detection scenarios, particularly when distinguishing from a no-workload state.
• The algorithm's generalizability across diverse cognitive tasks requires further investigation and refinement.