Abstract
This study explores the use of a generative AI tool called ‘GAIscaffold’ as a cognitive scaffold in English as a foreign language (EFL) writing in a metaverse environment. Two undergraduate EFL writing classes participated in describing a historical place in the metaverse. Students were randomly assigned to either an experimental group, which used ‘GAIscaffold’ or a control group which did not use ‘GAIscaffold’. Data collection included student questionnaires on learning perceptions and engagement with ‘GAIscaffold’ as well as their final writing performance. An independent sample t-test was used to compare students’ perceptions in the two groups, and multivariate analysis of covariance (MANCOVA) was adopted to analyse engagement and writing performance. Results showed that the experimental group demonstrated statistically significant improvements in both engagement and writing performance. The group also reported a more positive perception of ‘GAIscaffold’. These findings suggest that integrating generative AI as a cognitive scaffold can enhance students’ learning experiences, learning engagement, and performance in EFL descriptive writing in the metaverse environment. Copyright © 2025 Informa UK Limited, trading as Taylor & Francis Group.
| Original language | English |
|---|---|
| Journal | Educational Psychology |
| Early online date | Sept 2025 |
| DOIs | |
| Publication status | E-pub ahead of print - Sept 2025 |
Citation
Ding, J., & Song, Y. (2025). Descriptive writing using generative AI as a cognitive scaffold in the metaverse environment: university students’ perceptions, learning engagement, and performance. Educational Psychology. Advance online publication. https://doi.org/10.1080/01443410.2025.2568189Keywords
- Generative AI
- Cognitive scaffold
- Metaverse environment
- Descriptive writing
- EFL students