Kinematics-dependent brake response in naturalistic rear-end emergencies emerges from cognitive behavior model

Understanding and modeling driver behavior during critical traffic situations is crucial for improving traffic safety and developing effective driver assistance systems. Previous research has shown that driver responses are strongly influenced by kinematic urgency.In this study, we evaluate whether our driveBOT cognitive behavior model can replicate human-like braking behavior under varying kinematic conditions. Using simulated surprising rear-end scenarios, we compare driveBOT's performance against naturalistic human data, particularly focusing on deceleration onset timing, ramp-up (jerk), and maximum deceleration. Our results demonstrate that driveBOT's cognitive architecture captures the key kinematics-dependent responses observed in human behavior of threshold-like responses to visual looming and rapid reactions in high-urgency situations. These findings validate driveBOT as a practical tool for modeling driver responses and contribute to bridging the gap between human behavior models and real-world applications.