Follow the Yellow or Red Brick Road? Investigating the Impact of Narratives in a Guided Navigation Task

Bruce W. Wilson, Shivaanee Eswaran, Omar Riyaz, Javier Chiyah-Garcia, Matthew Peter Aylett

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Sharing casual stories in conversations is a natural human behaviour that adds personal and unique details, while also offering measurable benefits such as improved memory recall and more positive social human-human interactions. However, as social robots become more common in everyday environments, it is unclear whether these stories, or narratives, have similar effects on human-robot interactions with embodied agents.

In this paper, we develop and run a pilot evaluation with a direction-giver agent to study the effects of narratives in the interaction. We use a Furhat robot with an LLM to guide participants through a virtual environment in Minecraft, providing instructions and answering follow-up questions throughout the task. We present preliminary results that show participants made fewer navigational mistakes and preferred narratives, and assess using virtual environments for tasks that typically require a physical one.
Original languageEnglish
Title of host publicationHAI '24: Proceedings of the 12th International Conference on Human-Agent Interaction
PublisherAssociation for Computing Machinery
Pages411-413
Number of pages3
ISBN (Print)9798400711787
DOIs
Publication statusPublished - 24 Nov 2024
Event12th International Conference on Human-Agent Interaction 2024
- Swansea University, Swansea, United Kingdom
Duration: 24 Nov 202427 Nov 2024
https://hai-conference.net/hai2024/

Conference

Conference12th International Conference on Human-Agent Interaction 2024
Country/TerritoryUnited Kingdom
CitySwansea
Period24/11/2427/11/24
Internet address

Fingerprint

Dive into the research topics of 'Follow the Yellow or Red Brick Road? Investigating the Impact of Narratives in a Guided Navigation Task'. Together they form a unique fingerprint.

Cite this