A Simulator Study on the Driver Failure and Traffic Conflict in Lane Change Situations on a 2+1 Road

This study examined driver failure and traffic conflict using a driving simulator during lane-change scenarios on a 2+1 road, focusing on differences between day and night conditions. Data on driver failure status and driving behavior were captured through a survey. The study also evaluated driver performance and electrodermal activity. The findings revealed that nighttime lane changes had a shorter minimum time-to-collision (MTTC), indicating a higher traffic conflict severity compared to daytime (p = 0.046). It was also found that increased driver stress was correlated with decreased MTTC (p = 0.039). Drivers who were prone to making mistakes were closer to collisions (p = 0.005), whereas those prone to violations avoided collisions better (p = 0.027). Despite facing traffic conflicts, the drivers reported no perception, decision, or planning errors. Moreover, "procedure error" was the most common cause of failure. The study suggests that advanced human-machine interface systems are required to aid perception and decision-making and recommends future research with larger, diverse samples.