Enrich story co-creation by adding visual stimuli and interaction turns, and prioritise interaction fluency¶

Strategy · All strategies
Evidence · no claims cited

Description¶

Based on user feedback and failure analysis, the authors recommend enriching the number of visual stimuli and increasing the interaction turns between robot and user to improve the richness and depth of co-created stories, since participants found three turns too few (on average 5 ± 0.45 desired). They further recommend addressing technical issues to make the interaction more fluent and quicker, adding longer chatting phases especially as ice-breaking at the beginning, and integrating visual or tactile controls so iCub can restart the cube handover in case of failure.

Design Implications¶

Context¶

Requirements¶

• Technical work on voice transcription robustness and latency introduced by external services such as ChatGPT

Constraints¶

• The recommendation arises from adult usability feedback; the article states experiments with adults are an intermediate step before clinical studies with young patients

Target Learners¶

• children under 10 in pediatric rehabilitation settings

Target Learning Goals¶

• more engaging and richer collaborative storytelling

Related Strategies¶

• Interactive Storytelling

Examples¶

-

Key Sources¶

• Maria Lombardi, Carmela Calabrese, Davide Ghiglino, Caterina Foglino, Davide De Tommaso, Giulia Da Lisca, Lorenzo Natale, Agnieszka Wykowska. (2025). Would you let a humanoid play storytelling with your child? A usability study on LLM-powered narrative Human-Robot Interaction. https://arxiv.org/abs/2508.02505