A review on polyadic chatbots: trends, challenges, and future research directions

Document Type

Article

Source of Publication

Knowledge and Information Systems

Publication Date

1-1-2024

Abstract

Polyadic chatbots, conversational agents created explicitly for multiparty interactions, are gaining appeal as game-changing technologies in human-AI engagement. Unlike dyadic chatbots, which are designed for one-on-one conversations, there is a lack of studies examining the recent research on polyadic chatbots. This research investigates the roles, design concepts, empirical data, and issues connected with polyadic chatbots. Based on a comprehensive review of 57 studies, this study concludes that chatbots are widely employed for general purposes, particularly to improve group dynamics, decision making, and brainstorming. Polyadic chatbots are used for education, tourism, and social media. While text-based polyadic chatbots are prevalent, there is an increasing trend toward embodied chatbots that use nonverbal cues. Polyadic chatbots serve primarily as discussion facilitators, mediators, collaborators, emotional support, and social companions. They help to address key communication challenges related to improving collaboration and balancing participation. Only a few studies have used theoretical frameworks to create polyadic chatbots, focusing on idea creation, conversation management, and social support. Experimentation is the key approach for testing these chatbots, demonstrating their efficiency in various tasks while also identifying drawbacks such as limited problem-solving ability, user acceptance, and technological limits. Moving ahead, research should focus on improving polyadic chatbots' adaptability in social contexts, understanding the importance of nonverbal clues, and matching their design with user experience theories and cultural sensitivities. Addressing ethical issues regarding transparency, prejudice, and privacy is critical to ensuring responsible and equitable deployment across various interaction settings.

ISSN

0219-1377

Publisher

Springer Science and Business Media LLC

Disciplines

Computer Sciences

Keywords

Chatbot, Conversational agent, Human–computer interaction, Multipart, Multiparty, Polyadic

Scopus ID

85212468255

Indexed in Scopus

yes

Open Access

no

Link to Full Text

Share

COinS