Physiological Signals as Predictors of Cognitive Load Induced by the Type of Automotive Head-Up Display

Abstract

The visual information complexity of automotive head-up displays (HUDs) may affect cognitive load and reduce driver performance in critical situations. This study investigated whether physiological indicators of cognitive load can predict the type of HUD while driving. Physiological signals of heart rate variability (HRV), electrodermal activity (EDA), skin temperature, and pupil dilation were recorded from 28 participants using a motion-based driving simulator. Two types of HUD with different information complexities were compared: baseline HUD and augmented reality HUD. Heart rate and EDA were processed to create standardized biomedical features. Time-series analysis and basic statistics generated two sets of features for pupil dilation and skin temperature. The effect of signal combinations on classification performance was tested using signal fusion. Three gradient boosting classifiers (LGBM, HGBC, and XGB) were trained on physiological signals to predict HUD type. The fusion of HRV, EDA, and time-series features for skin temperature and pupil dilation yielded moderate performance, with average AUC ROC scores of XGB = 0.67, LGBM = 0.69, and HGBC = 0.70. Combining HRV, EDA, and basic statistical features for skin temperature and pupil dilation, the classifiers achieved an improved average AUC ROC score of 0.76. The best scores were 0.96 (LGBM and XGB) and 0.98 (HGBC). These results demonstrate the potential of physiological signals for modeling HUD-induced cognitive load and dynamically regulating its effects in real-time.

Keywords

• Advanced driver assistance system (ADAS)
• augmented reality
• automotive head-up display
• autonomous vehicles
• classification
• conditionally automated driving