Home // ACHI 2020, The Thirteenth International Conference on Advances in Computer-Human Interactions // View article
Authors:
Masato Kuno
Yoshimasa Ohmoto
Toyoaki Nishida
Keywords: human-agent interaction; cooperative problem.
Abstract:
We focus on problem-solving situations in which people cooperate with virtual interactive agents. The final goal is to achieve high-quality problem solving through a problem-solving process in which people recognise agents as effective collaborators and actively cooperate with the agents. It is known that trust, which is the basis of cooperation, has both affective and competent aspects. However, because the impression of agents tends to focus on the ability side, it is necessary to make the people recognise that the ability and emotional sides of agents are not separate but are integrated. In the proposed method, we apply the Alternate Estimation by representing Global and Local (AEGL) goal-oriented behaviour model, which adjusts the behaviour of an agent that show the agent's ability side and the emotional side by estimating the causes of the human behaviour. We demonstrated how both behaviours change consistently through interaction with people. In this study, we designed an experimental agent model to realise the proposed model. The people and agents are asked to perform cooperative decision-making tasks by exchanging opinions and adapting to each other's behaviour. The results suggest that the relationship between the ability and affective functioning of the agents eases tension and that people feel more comfortable talking to the agents.
Pages: 419 to 427
Copyright: Copyright (c) IARIA, 2020
Publication date: March 22, 2020
Published in: conference
ISSN: 2308-4138
ISBN: 978-1-61208-761-0
Location: Valencia, Spain
Dates: from November 21, 2020 to November 25, 2020