Toward a Subject-Independent EEG-Based Neural Indicator of Task Proficiency During Training

Area of Science:

• Neuroscience
• Cognitive Science
• Human-Computer Interaction

Background:

• Developing objective measures of cognitive states is crucial for optimizing learning.
• Passive brain-computer interfaces (BCIs) offer potential for real-time feedback and enhanced training.
• Subject-independent classification of mental states from EEG is a key challenge.

Purpose of the Study:

• To investigate the feasibility of an electroencephalography (EEG)-based neural indicator (NI) for tracking task proficiency.
• To assess the subject-independent classification of mental states for proficiency assessment.
• To explore the potential of this NI in passive BCIs for training enhancement.

Main Methods:

• A spatial knowledge acquisition task was performed by 15 participants in a virtual environment.
• Electroencephalography (EEG) data, task performance, and perceived certainty were recorded.
• A support vector machine classifier was trained on subject-independent EEG data to classify proficiency states (low/high).
• A neural indicator (NI) was derived from epoch classifications to quantify proficiency progression.

Main Results:

• The EEG-based neural indicator effectively tracked task proficiency progression in 9 out of 15 participants.
• A significant increase in NI was observed from the first to the last training block (NI=0.15 to NI=0.81).
• The neural indicator reached a plateau after the 7th training block for successful participants.

Conclusions:

• A subject-independent EEG-based neural indicator shows potential for real-time assessment of task proficiency.
• This approach could facilitate the development of passive BCIs to improve training outcomes.
• Further research is warranted to refine the neural indicator and its application in diverse training scenarios.