Exploring Narrative Structures and Elements in Children's Collaborative Storytelling With AI and Humans

• In Event: AERA Roundtable Session Friday 1:30 pm Four Seasons Ballroom 4
  In Roundtable Session: AI and Literacy in 21st-Century Classrooms (Table 3)

Abstract

Collaborative storytelling provides a dynamic context for children’s language development. Recent advances in artificial intelligence (AI) have enabled child-AI collaborative storytelling. However, the narrative features of child-AI interactions and how they differ from child-human interactions remain underexplored. This study investigates the similarities and differences in narrative elements and structures in children’s collaborative storytelling with AI and human partners. Our findings reveal that while the narrative structures differ significantly between stories co-created with AI and those with human storytellers, the contributions to narrative elements are similar among children in both conditions. Additionally, both AI and human storytellers contribute comparably to most narrative elements. This study enhances our understanding of narrative interactions in child-AI versus child-human storytelling.

Authors

• Xuechen Liu, University of Michigan

• Echo Zexuan Pan, Harvard University

• Yuqing Xing, University of Southern California

• Jinshil Choi, University of Michigan

• Hongyi Meng, University of Michigan

• Youran Chen, University of Hong Kong

• Wenfei Pei, University of Michigan

• Ying Xu, Harvard University