“Interaction Modality and Trust: Investigating System-Driven Conversational Recommendation and Faceted Search”
iCAIML PhD student Ahmadou Wagne has a paper accepted at ACM RecSys 2026.
Ahmadou Wagne co-authored the short paper Interaction Modality and Trust: Investigating System-Driven Conversational Recommendation and Faceted Search accepted for publication at the 20th ACM Conference on Recommender Systems (RecSys 2026), taking place in Minneapolis, Minnesota, USA, from September 28–October 2, 2026. The paper investigates how interaction modality influences user trust in recommender systems by comparing a system-driven Conversational Recommender System (CRS) with a functionally equivalent faceted search interface in a between-participants user study.
The findings show that while explainability and anthropomorphism were positively associated with trust across both interaction modalities, no significant differences in overall trust emerged. However, users of the conversational system reported significantly lower perceived control, highlighting an important design trade-off between providing structured, safe conversational guidance and preserving user agency. The work contributes practical insights for the design of trustworthy conversational recommender systems.
The accepted contribution is listed at: https://recsys.acm.org/recsys26/contributions/#content-tab-1-1-tab