Abstract
Museums are increasingly using chatbots to transform passive visits into interactive experiences, leveraging advancements in Large Language Models (LLMs) for more engaging interactions. However, design guidelines for chatbot roles and interactions tailored to user preferences in museum contexts remain underexplored. To address this, we conducted an online survey with 65 participants, examining preferred chatbot roles and their relationship to artifact characteristics. Participants strongly favored chatbots using a first-person narrative as artifact creators, appreciating their empathetic, immersive, and novel perspectives. The user perceptions of chatbot roles are also found to be influenced by artifact characteristics, including artifact category, its popularity, and whether it depicts human or animal figures. However, concerns about the authenticity and ethical representation of historical figures emerged. These findings provide valuable insights for designing engaging and culturally sensitive chatbot interactions in museums.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems |
| Place of Publication | Yokohama Japan |
| Publisher | Association for Computing Machinery (ACM) |
| Pages | 1-8 |
| Number of pages | 8 |
| ISBN (Print) | 9798400713958 |
| DOIs | |
| Publication status | Published - 25 Apr 2025 |