PuzzleAide: Comparing Audio and Embodied Assistants for MR Puzzle-Solving
Conversational Virtual Agents (CVA) have great potential to assist users in their task performance. Whether these agents need to be embodied or not was the main research question of the present pilot study. To answer this question, we designed and developed a Mixed Reality (MR) application for solving a physical puzzle while interacting with a CVA. Eleven participants took part in our between-subject pilot study and interacted with two different representations of a CVA (i.e., voice-only and embodied agent). In this short paper, we descriptively report on the participants’ problem-solving time, the number of assistance requests they made to the CVAs, and the social presence they perceived while doing so.
Authors
- Gustavo Marfia (ORCID: https://orcid.org/0000-0003-3058-8004)
- A. Lux (ORCID: https://orcid.org/0009-0005-9251-4985)
- Fariba Mostajeran (ORCID: https://orcid.org/0000-0002-7424-9031)
- Frank Steinicke (ORCID: https://orcid.org/0000-0001-9879-7414)
- Pasquale Cascarano (ORCID: https://orcid.org/0000-0002-1475-2751)
- Shirin Hajahmadi (ORCID: https://orcid.org/0000-0001-8494-9456)
- Kevin Heuer (ORCID: https://orcid.org/0009-0001-7130-2888)
- Gil Otis Mends-Cole
Institutions
- Universität Hamburg (DE)
- University of Bologna (IT)
Publication Details
- Published
- 2024-09-13
- DOI
- https://doi.org/10.1145/3677386.3688899
- Primary Topic
- Intelligent Tutoring Systems and Adaptive Learning
- Type
- article
- Field-Weighted Citation Impact
- 0.00