User Preferences for Interaction Modalities: The Influence of Task, Context, and User Characteristics when Interacting with Conversational Agents

Files
0391.pdf (521.95 KB)
Date
2024-01-03
Authors
Riefle, Lara
Benz, Carina
Contributor
Advisor
Department
Instructor
Depositor
Speaker
Researcher
Consultant
Interviewer
Annotator
Journal Title
Journal ISSN
Volume Title
Publisher
Volume
Number/Issue
Starting Page
3960
Ending Page
Alternative Title
Abstract
When using conversational agents (CAs), the interaction is typically either text- or speech-based. Existing research focuses on the effects of these interaction modalities or the general adoption of either text- or speech-based interaction, leaving an important research gap regarding users’ underlying preferences for interaction modalities. Therefore, this study investigates the influence of task, context, and individual user characteristics on user preferences for interaction modalities. We use a two-step approach consisting of exploratory interviews to identify 14 influencing factors, followed by a scenario-based experiment to quantitatively assess the impact of the identified task, context, and user characteristics. The results provide insights into the drivers for users’ preferences for interaction modalities when interacting with CAs. Thereby, we contribute to a more holistic understanding of human-CA interaction and provide a starting point for future research. The findings can further guide practitioners regarding which factors to consider in their decisions when investing in CAs.
Description
Keywords
Artificial Intelligence-based Assistants and Platforms, conversational agents, human-computer-interaction, interaction modality, user preferences, user study
Citation
URI
https://hdl.handle.net/10125/106864
Extent
10 pages
Format
Geographic Location
Time Period
Related To
Proceedings of the 57th Hawaii International Conference on System Sciences
Table of Contents
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International
Rights Holder
Local Contexts
Collections
Artificial Intelligence-based Assistants and Platforms
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.