Concept mapping has been widely used in various fields to facilitate students' organization of knowledge. Previous studies have, however, pointed out that it is difficult for students to construct concept maps from the abundant searched data without appropriate scaffolding. Thus, researchers have suggested that students could produce high quality concept maps when they construct them with appropriate guidance. Accordingly, this study proposes a question prompt-based concept mapping approach, which was conducted in a computer course with a duration of 14 weeks on the research issue of global warming. There were 176 senior high school students from four classes who participated in the experiment. The study adopted a quasi-experimental design for the investigation of the effectiveness of the proposed approach; two classes were randomly assigned to be the experimental group, learning with the question prompt-based concept mapping approach, while the other two groups were the control group, who used the conventional concept mapping approach. The results indicated that the students in the experimental group had better performance than those in the control group in terms of their learning achievement, 5C competences, and learning attitudes. In addition, the results also showed that students with a reflective style had better learning achievement than those with an active style. Thus, it is worth investigating, for such an issue, how to propose more effective learning models for diverse student learning styles in the future. This can provide useful guidance for educators, practitioners and researchers in the areas of science education and computer-assisted education.