The impact of artificial intelligence-assisted teaching on medical students’ learning outcomes: an integrated model based on the ARCS model and constructivist theory

Artificial intelligence-assisted teaching, as an innovative model that combines intelligent technology and personalized education, is increasingly being emphasized in higher medical education. This study included 523 participants, with a valid response rate of 87.2%. An integrated model based on the ARCS motivation model and constructivist theory was developed to explore the factors influencing medical students' learning outcomes in the context of AI-assisted instruction. Descriptive statistics were conducted using SPSS 23.0, and a structural equation model was constructed and validated using Amos 23.0. Mediation analysis was performed with Process (version 3.3.1). The study confirmed that teaching quality had a positive effect on learning motivation (β = 0.645, P < 0.001) and learning outcomes (β = 0.128, P = 0.032). Learning motivation positively influenced learning attitude (β = 0.822, P < 0.001) and learning satisfaction (β = 0.350, P < 0.001). Learning attitude had a positive impact on both learning satisfaction (β = 0.530, P < 0.001) and learning outcomes (β = 0.232, P = 0.020). Learning satisfaction was also positively associated with learning outcomes (β = 0.415, P < 0.001). The external environment had a positive effect on learning motivation (β = 0.449, P < 0.001) and learning outcomes (β = 0.101, P = 0.033). Moreover, learning motivation played a significant mediating role in the relationships between teaching quality and learning outcomes (β_inmedia = 0.343, 95% CI [0.273, 0.414]), as well as between the external environment and learning outcomes (β_inmedia = 0.287, 95% CI [0.218, 0.355]). Teaching quality and external environment indirectly enhance medical learning outcomes by strengthening learning motivation. Learning motivation plays a key role in shaping learning attitude, satisfaction, and outcomes, confirming the positive value of AI-assisted teaching in optimizing learning mechanisms. This study contributes to the application of AI-assisted teaching in medical education and provides empirical support for improving medical students' learning performance.