Botune: an AI-supported scaffolding system for children’s drawing-based narrative expression in nature learning
Nature exploration fosters children’s cognitive development and environmental awareness; however, kindergarten curricula are often constrained by limited resources and repetitive interaction, leaving children’s observations only weakly integrated into structured understanding and expression. We present Botune, an AI-supported creative-expression tool for kindergarten planting activities that integrates plant observation, drawing, spoken narration, and animation-based sharing. We conducted an exploratory, fixed-order within-participant study in kindergarten planting classes with 27 children and three educators; 24 complete cases were analysed after excluding data from three children who did not complete the drawing task. The traditional activity preceded the Botune-supported activity by seven days. The proportion of activity time spent observing and teacher-rated involvement scores during sharing were higher in the later session. The later narrations contained significantly more actions and character interactions; character counts were descriptively higher but did not remain statistically significant after Holm correction, and state- or emotion-related expressions did not differ significantly. These semantic counts index narrative elaboration rather than overall narrative quality or general narrative ability. These differences cannot be attributed to Botune alone. Scientific knowledge and conceptual understanding were not directly assessed by these measures. The study offers exploratory observations and design considerations for AI-supported narrative expression in kindergarten nature learning.
Authors
- 陈忆帆
- 季书战
- Yao Hu (ORCID: https://orcid.org/0009-0002-0651-8487)
- Yuyan Ma
- Yi Li
Institutions
- Hunan University (CN)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-10-16
- DOI
- https://doi.org/10.1038/s41598-026-75865-7
- Primary Topic
- Early Childhood Education and Development
- Type
- article
- Field-Weighted Citation Impact
- 0.00