An Eye-Fixation Related Electroencephalography Technique for Predicting Situation Awareness: Implications for Driver...

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Automotive Engineering

Background:

• Maintaining situation awareness (SA) is critical for driver performance during transitions in Level 3 automated vehicles.
• Multimodal analysis of visual, behavioral, and cognitive processes aids real-time SA assessment in driver monitoring systems.

Purpose of the Study:

• To develop and validate a fixation-related electroencephalography band power (FRBP) approach for assessing situation awareness (SA) in automated driving.
• To investigate the relationship between electroencephalography (EEG) oscillations and SA during simulated driving tasks.

Main Methods:

• Thirty participants engaged in simulated automated driving tasks, with eye movements and EEG recorded.
• Situation Awareness Global Assessment Technique (SAGAT) assessed SA post-task.
• Mixed-effects models identified brain regions indicative of SA; machine learning models were developed for SA assessment.

Main Results:

• Alpha and theta oscillations in frontal and temporal brain areas correlated with SA.
• The FRBP technique achieved 88% accuracy in predicting driver SA using a neural network model.

Conclusions:

• The FRBP technique offers a comprehensive evaluation of SA by integrating eye movements and brain activity.
• FRBP holds significant potential for real-time monitoring of driver SA in automated driving systems.