Demonstrating Brain-Level Interactions Between Visuospatial Attentional Demands and Working Memory Load While Driving...

Area of Science:

• Neuroscience
• Cognitive Psychology
• Human Factors Engineering

Background:

• Driving is a complex cognitive task requiring multiple resources.
• Limited research exists on brain-level interactions during dual-task driving scenarios.
• Understanding these interactions is crucial for driver safety and advanced driver-assistance systems.

Purpose of the Study:

• To investigate the brain-level interactions between visuospatial attentional demands and working memory load (WML) during driving.
• To predict driving difficulty using functional near-infrared spectroscopy (fNIRS) data under varying WML conditions.
• To identify brain regions involved in multitasking during driving.

Main Methods:

• Utilized high-density fNIRS to measure brain activation in 15 participants driving in a virtual reality simulator.
• Manipulated driving difficulty (standard vs. construction zone) and WML (0-back to 4-back task).
• Employed multivariate logistic ridge regression for predictive modeling of driving difficulty.

Main Results:

• Achieved 75.0% accuracy in predicting driving difficulty across all WML levels using separate classifiers.
• Prediction accuracy varied significantly with WML levels (62.2-87.1%), peaking at intermediate loads.
• Combined models across WML levels reduced prediction accuracy to 46.8%.

Conclusions:

• Visuospatial attention and WML involve interacting brain processes, evidenced by diminishing predictive patterns at higher WML.
• Bilateral dorsal frontal, inferior parietal, and left superior parietal areas were key predictors of driving difficulty.
• Multitasking strategies may shift with increasing WML, aligning with multiple resource theory.