EUVIMEDEUVIMED entdecken
Mixed-Methods-StudieOpen Access · CC BY 4.0

Personalising Learning at Scale: A Mixed-Method Study of AI-Generated Bilingual Avatars and Adaptive Feedback

Jasmine Jain, Pei Lin Tay, Yee Ling Lee, Shin Yen Tan

Asia Pacific Journal of Educators and Education · 09.06.2026 · 41 (1) · 247-273

DOI: 10.21315/apjee2026.41.1.11 ↗

Kurz gefasst

Mixed-Methods-Studie in der Zeitschrift Asia Pacific Journal of Educators and Education, online erschienen am 09.06.2026. Verfasst von Jasmine Jain und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The increasing use of Artificial Intelligence (AI) in education has shown strong potential to transform personalised learning, yet the application of AI tools in flexible, bilingual learning environments remains underexamined. Guided by the Cognitive Load Theory (CLT) and Self-Regulated Learning (SRL) theory, the study examined how students navigated and engaged with an AI learning assistant and the use of AI-generated bilingual avatars, where lecturers provided only English scripts and AI, via Synthesia.io, automated the delivery of lecture content in students’ preferred language (English or Chinese). Additionally, Noodle Factory, an AI-powered personalised learning platform was used to personalise weekly quizzes by generating AI-based feedback in students’ chosen language and creating custom learning pathways responsive to individual performance. This exploratory mixed-method study was conducted over five academic weeks, involving 195 Bachelor of Education student teachers from a private university in Malaysia. Quantitative data were collected on AI-generated video view frequency, quiz participation in Noodle Factory, and engagement frequency with the AI learning assistant. Qualitative data were obtained from interviews with eight purposively selected students. The study found that students strategically engaged with bilingual AI video lectures and the AI learning assistant in ways that supported comprehension and self-regulated learning. Quantitative data showed increased quiz attempts during weeks when bilingual videos were made available, and language usage patterns indicated a preference for English, with selective reliance on Chinese for clarification. Qualitative findings revealed that bilingual access enhanced understanding, enabled flexible learning, and supported academic preparation. The AI learning assistant was used primarily for revision and feedback, although some students noted technical limitations and translation issues. These findings suggest that AI-driven bilingual tools, when aligned with students’ cognitive and linguistic needs, can enhance engagement and autonomy in self-paced learning environments. These findings have practical implications for institutions seeking to support bilingual and international student populations.

Zitieren

APA

Jain, J., Tay, P. L., Lee, Y. L., & Tan, S. Y. (2026). Personalising Learning at Scale: A Mixed-Method Study of AI-Generated Bilingual Avatars and Adaptive Feedback. Asia Pacific Journal of Educators and Education, 41(1), 247-273. https://doi.org/10.21315/apjee2026.41.1.11

Vancouver

Jain J, Tay PL, Lee YL, Tan SY. Personalising Learning at Scale: A Mixed-Method Study of AI-Generated Bilingual Avatars and Adaptive Feedback. Asia Pacific Journal of Educators and Education. 2026;41(1):247-273. doi:10.21315/apjee2026.41.1.11

BibTeX
@article{Jain2026,
  author = {Jain, Jasmine and Tay, Pei Lin and Lee, Yee Ling and Tan, Shin Yen},
  title = {Personalising Learning at Scale: A Mixed-Method Study of AI-Generated Bilingual Avatars and Adaptive Feedback},
  journal = {Asia Pacific Journal of Educators and Education},
  year = {2026},
  volume = {41},
  number = {1},
  pages = {247--273},
  doi = {10.21315/apjee2026.41.1.11}
}
Kostenlos starten →