The use of generative AI for multimodal learning, teaching and assessment in higher education: A scoping review
Generative Artificial Intelligence (GenAI) presents opportunities for multimodal learning, but this remains under-explored. This scoping review identifies trends across 19 multinational and multidisciplinary peer-reviewed empirical studies. Findings reveal that image-based modalities dominate multimodal GenAI applications in teaching/learning, while assessment examples are scarce. Studies also focus either on the technicality or the pedagogy in multimodal learning designs. Application within pedagogic contexts remains under-evaluated. Challenges related to ethics, technological sustainability, and inclusivity also persist.
Authors
- Sue Beckingham (ORCID: https://orcid.org/0000-0001-5660-125X)
- Samuel S. Saunders (ORCID: https://orcid.org/0000-0002-7515-9214)
- Run Wen (ORCID: https://orcid.org/0000-0001-5536-3907)
- Tünde Varga‐Atkins (ORCID: https://orcid.org/0000-0002-9492-4284)
- Nataša Lacković (ORCID: https://orcid.org/0000-0001-7086-4822)
- Na Li (ORCID: https://orcid.org/0000-0003-2395-3499)
Institutions
- University of Liverpool (GB)
- Lancaster University (GB)
- Xi’an Jiaotong-Liverpool University (CN)
- Sheffield Hallam University (GB)
Publication Details
- Journal
- Innovations in Education and Teaching International
- Published
- 2026-11-12
- DOI
- https://doi.org/10.1080/14703297.2026.2763524
- Primary Topic
- Artificial Intelligence in Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Staff and Educational Development Association
- National Natural Science Foundation of China
- Xi’an Jiaotong-Liverpool University