This paper explores the difference between reactive and proactive systems in robot failure explanations. As robots perform increasingly complex tasks, providing clear explanations is essential for building trust and acceptance. The study compares reactive systems, which explain failures after they occur, with proactive systems, which predict and explain potential issues in advance. The results show that proactive systems are perceived as more intelligent and trustworthy, with explanations that are more understandable and timely. The study also finds that proactive systems lead to better human-robot collaboration by enabling people to diagnose and assist with robot failures more effectively. The research involved a user study with 186 participants, who evaluated the performance of reactive and proactive systems in a scenario involving a robot assembling a gearbox kit. The robot was equipped with a behavior tree (BT) to generate explanations. In the reactive system, explanations were provided only after a failure occurred, while the proactive system used assumption checkers to detect and explain potential failures before they happened. The study found that proactive systems were perceived as more intelligent and trustworthy than reactive systems. Participants also rated the explanations from proactive systems as more understandable and timely. Additionally, participants preferred the explanations from proactive systems, indicating that they were more effective in helping users understand and resolve robot failures. The findings suggest that proactive systems are more effective in improving human-robot collaboration by providing timely and understandable explanations of robot failures. This can lead to better task performance and increased trust in robot systems. The study also highlights the importance of providing clear and timely explanations to help users understand and resolve robot failures, which is essential for the successful deployment of robots in shared human-robot environments.This paper explores the difference between reactive and proactive systems in robot failure explanations. As robots perform increasingly complex tasks, providing clear explanations is essential for building trust and acceptance. The study compares reactive systems, which explain failures after they occur, with proactive systems, which predict and explain potential issues in advance. The results show that proactive systems are perceived as more intelligent and trustworthy, with explanations that are more understandable and timely. The study also finds that proactive systems lead to better human-robot collaboration by enabling people to diagnose and assist with robot failures more effectively. The research involved a user study with 186 participants, who evaluated the performance of reactive and proactive systems in a scenario involving a robot assembling a gearbox kit. The robot was equipped with a behavior tree (BT) to generate explanations. In the reactive system, explanations were provided only after a failure occurred, while the proactive system used assumption checkers to detect and explain potential failures before they happened. The study found that proactive systems were perceived as more intelligent and trustworthy than reactive systems. Participants also rated the explanations from proactive systems as more understandable and timely. Additionally, participants preferred the explanations from proactive systems, indicating that they were more effective in helping users understand and resolve robot failures. The findings suggest that proactive systems are more effective in improving human-robot collaboration by providing timely and understandable explanations of robot failures. This can lead to better task performance and increased trust in robot systems. The study also highlights the importance of providing clear and timely explanations to help users understand and resolve robot failures, which is essential for the successful deployment of robots in shared human-robot environments.