Improving Our Knowledge of Driver Qualities by Measuring Their Cognitive Processes

Abstract

In this study, we provide a new method for interpreting EEG data collected from drivers during a driving simulator. As indicators of brain nonlinear dynamics, we zeroed in on the Hurst exponent, Shannon entropy, and fractal dimension. While the Hurst exponent shows learning patterns in memory retention and habit development, the Shannon entropy and fractal dimension change during driving condition changes. These trends are statistically significant. These results provide new opportunities for evaluating driver performance, detecting safety hazards, and expanding our knowledge of the non-linear dynamics of human cognition in relation to driving and beyond, and they imply that the tools of Non-linear Dynamical (NLD) Theory can be used as indicators of cognitive state and changes in driving memory. Our research shows that NLD techniques have the ability to shed light on brain states and system variations, which opens the door for their incorporation into existing ML and Deep Learning models. Beyond its use in driving apps, this integration has the potential to enhance cognitive learning, which in turn boosts productivity and accuracy.