EEG-based detection of modality-specific visual and auditory sensory processing

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Cognitive Science

Background:

• Passive brain-computer interfaces (pBCI) monitor user mental states to adapt human-machine interactions.
• Detecting mental workload level is common, but identifying the specific type of cognitive resources (e.g., visual vs. auditory) is crucial for more effective adaptations.
• Current pBCI research primarily focuses on workload levels, overlooking the potential of differentiating resource types.

Purpose of the Study:

• To investigate if electroencephalography (EEG) can differentiate between visual and auditory processing tasks.
• To determine the impact of sensory processing demand levels on the accuracy of distinguishing task types.
• To explore the feasibility of a pBCI that detects both the level and type of attentional resources.

Main Methods:

• 15 participants performed designed visual and auditory tasks under varying demand levels.
• Electroencephalography (EEG) data was recorded during task performance.
• Traditional machine learning algorithms were employed to classify task types based on EEG signals.

Main Results:

• Auditory and visual processing tasks were distinguished with 77.1% accuracy under high demand conditions.
• Classification accuracy did not exceed chance levels in the low demand condition.
• EEG signals show potential for differentiating cognitive resource types.

Conclusions:

• The study supports the feasibility of developing pBCIs that can identify the type of attentional resources required.
• High demand scenarios showed promising accuracy in distinguishing between visual and auditory tasks.
• Further research is needed to establish demand thresholds for accurate type detection, but results are promising for safety-critical applications.