Objective: Traffic accidents are mainly caused by driver-to-pedestrian collisions or driver-to-driver collisions. Prosocial driving behavior indicates that drivers exhibit altruistic behavior toward other drivers on roads. Yielding behavior demonstrates that drivers grant the right of passage to pedestrians at unsignalized crossings, while yielding attitude presents the subjective emotional and cognitive inclination to yield to pedestrians at unsignalized crossings. This study aims to explore the effect of altruism and drivers' perceived stress on prosocial driving behavior, yielding behavior, and yielding attitude. In addition, we endeavor to explore the effect of stress on prosocial driving behavior exhibiting an inverted "U-type" curve as Yerkes-Dodson's law suggests and test the moderating role of perceived stress on altruism and prosocial driving behavior/yielding behavior/yielding attitude.Methods: Using a survey method, we asked 454 participants to complete an altruism scale from the IPIT measuring altruism, a Perceived Stress Scale-10 measuring drivers' perceived stress, a prosocial driving scale from the PADI measuring prosocial driving behavior, and items on yielding behavior and yielding attitude. Then, a correlational matrix, a hierarchical multiple nonparametric regression analysis, and a moderating analysis of perceived stress were employed in sequence to reach our objective.Results: The hierarchical multiple nonparametric regression analysis showed that altruism positively predicts yielding attitude (F = 41.56, p < 0.001), yielding behavior (z = 8.46, p < 0.001, odds ratio = 4.90) and prosocial driving behavior (F = 110.66, p < 0.001), but stress predicts only prosocial driving behavior (F = 7.63, p < 0.001), not yielding attitude (F = 0.51, p > 0.05) or yielding behavior (z = 0.12, p > 0.05), which exhibits an inverted "U-type" curve. Moderating analyses showed that stress only moderates the relationship between altruism and yielding attitude (B = -0.24, t = -2.62, p < 0.01).Conclusions: Altruism is positively related to prosocial driving behavior, yielding behavior, and yielding attitude. Stress influences prosocial driving behavior only and exhibits an inverted "U-type" curve. Stress does not directly influence the yielding behavior. Instead, stress moderates the relationship between altruism and yielding attitude only and may further increase the possibility of yielding behavior.