Home // CENTRIC 2019, The Twelfth International Conference on Advances in Human-oriented and Personalized Mechanisms, Technologies, and Services // View article

Difference in Attitudes Toward Suggestions Given by an Agent Using Impasse Estimation

Authors:
Yoshimasa Ohmoto
Hanako Sonobe
Toyoaki Nishida

Keywords: Human-agent interaction; Metacognitive suggestion; Insight problem solving.

Abstract:
When a human and a conversational agent cooperate to perform a task, the agent may need to provide advice to the human. In that case, the agent has to induce attitudes that ensure the human accepts the advice of the agent in advance or through the task. This study aimed to investigate whether the acceptance of advice provided by an agent could be encouraged by controlling the content and timing of advice presentation according to the state of the participant. We focused on metacognitive suggestions during insight problem solving as an example of advice that was effective even if an agent performs it. We conducted an experiment to investigate whether participants would be likely to accept the suggestion based on the estimation of the inner state of them even when the agent provided the suggestion. Based on results from the analysis of operation history log, the acceptance rate of suggestions in the participants interacting with the state-considering-suggestion agent was significantly higher than that in the participants interacting with fixed-interval-suggestion agent, but the task performance was not high enough.

Pages: 11 to 16

Copyright: Copyright (c) IARIA, 2019

Publication date: November 24, 2019

Published in: conference

ISSN: 2308-3492

ISBN: 978-1-61208-754-2

Location: Valencia, Spain

Dates: from November 24, 2019 to November 28, 2019