The Effectiveness of Artificial Intelligence-Assisted Learning Stations for Differentiated Learning Based on Students' Learning Styles

Rullyana, Gema
Ardiansah, Ardiansah

Cetta: Jurnal Ilmu Pendidikan
Vol 9 No 1 (2026)

The integration of learning stations in higher education continues to face challenges in effectively addressing diverse student learning styles. Although differentiated instruction offers substantial promise, its implementation is often hindered by limited resources and the complexity of delivering personalized learning experiences at scale. This study explores the effectiveness of artificial intelligence (AI)-enhanced learning stations in supporting differentiated instruction aligned with individual learning preferences. Using a mixed-methods explanatory design, the research involved 82 students from an Educational Technology Study Program, divided into an experimental group (utilizing AI-supported learning stations) and a control group (traditional stations without AI). Data collection methods included pre- and post-tests, structured observations, VARK learning style inventories, and semi-structured interviews. Quantitative results indicated statistically significant improvements in learning outcomes for the experimental group, reflected in higher post-test scores and greater normalized gains. T-test and ANOVA analyses confirmed the intervention’s overall effectiveness, with no significant variation in learning gains across learning style categories within the experimental group. Qualitative findings supported these outcomes, with participants reporting that the AI-assisted environment fostered more personalized, relevant, and reflective learning experiences. Moreover, the integration of AI was associated with increased learner engagement, heightened motivation, and improved metacognitive awareness of learning preferences. This study contributes empirical evidence supporting the role of AI in enabling differentiated instruction within higher education contexts, highlighting its potential to provide scalable, personalized learning experiences. The findings suggest that AI-driven solutions may address key limitations in traditional instructional design by offering inclusive and adaptive strategies responsive to individual learner needs.