<p>Beyond ensuring safe driving, intelligent vehicles must also provide drivers with a sense of security that meets their psychological expectations. To this end, this study utilizes multi-source natural driving data to analyze the interrelationships among vehicle dynamics, driver behaviors, and physiological activation in car-following scenarios. First, multi-source “Human—Vehicle—Environment” data are collected in natural driving conditions, from which skin conductance activities, driver operations, and inter-vehicle state data are extracted for car-following events. Next, based on the drivers’ skin conductance indicators, physiological activation states are distinguished from stable states, laying the groundwork for examining car-following behavior characteristics under physiological activation. This systematic analysis reveals the distinct differences in driving behaviors and physiological responses among various drivers during car-following. Finally, real-world driving data are used to validate the proposed analysis method, confirming its applicability and demonstrating that different drivers exhibit individualized physiological responses and driving habits during car-following. These findings provide a theoretical basis for developing personalized car-following control strategies that incorporate drivers’ physiological states.</p>