Many studies support that children with autism spectrum disorder (ASD) show executive dysfunction. The applicant's NSC project report also indicated that in all areas in the behavior rating inventory of executive function, the second edition (BRIEF-2), including inhibit, self-monitor, shift, emotional control, initiate, working memory, plan/organize, task-monitor, and organization of materials, children in the typically developmental group showed better performance than children in the ASD group. The results also indicated that compared to the diagnosis of ASD, an executive function could be a better predictor of daily life adaptation. Similarly, more and more researchers put their efforts into executive training. However, the results are mixed. A meta-analysis showed that technology-based training can effectively improve the executive function of children with ASD. Children with ASD love game-like training as it provides friendly environments to children with ASD and induces less anxiety. The applicant, therefore, plans to develop a virtual reality executive function training program in the first year based on the applicant's previous experience with executive function training. The game will include the components of visual searching, reaction time, shift, planning, and working memory. At the end of the game, the game will provide visual feedback to the children to train in self-monitoring. In the second year, the applicant will conduct a randomized controlled trial to investigate the effectiveness of the training program. I plan to include 70 children with ASD and randomly assign them to either the executive training or control groups. Executive function and daily life adaptation will be the indexes of the effectiveness of the study.