Impact of Generative AI-assisted programming on the computational thinking of high school students
Rong Guo, Guian Li, Haifei Miao, Zhongling Pi, LiHua Xie · Frontiers in Psychology · 2026
Introduction Generative AI has demonstrated remarkable performance in the field of education due to its powerful text-generation capabilities. However, its impact on programming education is still in the early exploration stage, with relatively few related empirical studies. Methods This research explored the impact of Generative AI-assisted programming on high-school students’ computational thinking. It recruited 83 high-school students. This study adopted a convergent parallel mixed-methods design. First, quantitative data was analyzed to measure changes in computational thinking. Subsequently, follow-up interviews were conducted with selected participants to explain and elaborate on the quantitative research findings. Results The computational thinking ( p < 0.01) of the experimental group has demonstrated a significant improvement. The analysis revealed significant, differentiated gains: in computational thinking, algorithmic thinking and abstraction improved substantially ( p < 0.01). These quantitative outcomes were further illuminated by qualitative findings, which highlighted Generative AI’s role in providing real-time, personalized scaffolding-a key mechanism driving student cognitive skill development. Discussion Generative AI-assisted programming learning strategy promotes personalized learning and stimulates students’ enthusiasm for learning programming. The findings suggest the potential for applying Generative AI-assisted programming in similar educational settings.