Journal of Information Resources Management ›› 2025, Vol. 15 ›› Issue (4): 42-55.doi: 10.13365/j.jirm.2025.04.042

Previous Articles     Next Articles

Research on the Influence of Initiative in Conversational Search Systems on Human-AI Collaboration

Fu Shiting1　Jiang Tingting2,3　Song Defeng2   

  1. 1.School of Information Management, Central China Normal University, Wuhan,430079;　
    2.School of Information Management, Wuhan University, Wuhan,430072;　
    3.Center for Studies of Information Resources, Wuhan University, Wuhan,430072
  • Online:2025-07-26 Published:2025-08-31
  • About author:Fu Shiting, Ph.D, Lecturer, research interests include human-AI interaction, information behavior; Jiang Tingting(corresponding author), Ph.D, professor, research interests include human-AI interaction, information behavior, user experience measurement, human-centered data analysis, Email:tij@whu.edu.cn; Song Defeng, master candidate, research interests include human-AI interaction, human-centered data analysis.
  • Supported by:
    This work is supported by the Major Project of the National Social Science Fund of China titled "Research on the Reconstruction and Application of Information Service System Driven by Human-Centered Artificial Intelligence"(22&ZD325).

PDF

14

Abstract: This study examines the role of proactivity in conversational search systems (CSS), identifying three key proactive interaction strategies, i.e., clarification, suggestion and disclosure.A mixed-initiative CSS was developed based on these strategies and compared with a user-initiative CSS. Experimental results showed that proactive CSS significantly improved the intrinsic, contextual, and representation quality of search results, enhanced system usability and social presence, and increased user trust and task attraction. Team consensus consistently mediated these effects. The findings offer both theoretical and practical insights for designing proactive CSS.

Key words: Conversational search, Mixed-initiative, Proactive interaction, Human-AI collaboration, Human-AI interaction, Collaboration effectiveness

CLC Number: 

Copyright © Journal of Information Resources Management
Supported by:Beijing Magtech