An experimental study on enhancing learning motivation and academic performance through AI literacy training
Artificial intelligence (AI) is increasingly used in higher education, creating opportunities to support personalized learning. However, the educational value of AI depends not only on access to AI tools but also on students’ motivation, academic self-efficacy, and learning progress. This study presents the Motivation–Performance Adaptive AI Literacy Optimization Framework (MP-AILOF), a conceptual framework designed to integrate AI literacy instruction with learner-related indicators to inform adaptive instructional decisions. The framework combines learner profiling, motivational assessment, academic performance monitoring, and a proposed feedback-based instructional optimization process. Its empirical evaluation used a two-group pre-test–post-test dataset comprising 80 undergraduate students, with 40 students in the experimental group and 40 in the control group, from one higher-education institution. The experimental group received MP-AILOF-supported AI literacy instruction, whereas the control group received conventional classroom instruction. Academic self-efficacy, learning motivation, and grade point average (GPA) were assessed before and after the intervention. Paired-sample t-tests were used to examine within-group changes, and analysis of covariance (ANCOVA) was used to compare post-test outcomes while adjusting for baseline scores. In the supplied dataset, the experimental group showed statistically significant increases in all three outcomes. The ANCOVA analyses also indicated higher adjusted post-test scores for the experimental group than for the control group. These findings describe the observed results in the available sample and should be interpreted in light of the study design and the need to verify the recorded GPA values against the institutional grading scale. The implementation package reproduces the principal descriptive and inferential analyses; it does not independently validate the framework’s proposed GBR prediction and adaptive optimization components. Further research using verified data, documented allocation procedures, validated measurement instruments, and follow-up assessments is required to evaluate generalizability and the persistence of any observed changes.
Authors
- Guandong Song (ORCID: https://orcid.org/0000-0002-8036-5960)
- 甄洪芳
- Yuting Wang
Institutions
- Fushun Vocational Technology Institute (CN)
- Northeastern University (CN)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-15
- DOI
- https://doi.org/10.1038/s41598-026-79820-4
- Primary Topic
- Artificial Intelligence in Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00