Dialogue Cues in Telecare Dialogue Agents: The Impact of Text Dialogue Methods and Dialogue Prompt Types on User Perceptions and Behaviors
Chien‐Hsiung Chen, Jiwei He · International Journal of Human-Computer Interaction · 2025
As patient-centered care grows, generative-AI conversational agents—unconstrained by time or location—offer novel telecare interactions. This study examined how text dialogue methods (task-led vs. social-led) and prompt cues (none, progress bar, text) affect Perceived Ease of Use (PEOU), Perceived Usefulness (PU), Confirmation (CON), Satisfaction (SAT), and Usage Intention (UI). Sixty participants took part in a 2 × 3 between-subjects experiment; data were analyzed via two-way ANOVA. Key findings: (1) Social-led text dialogues enhance PU and PEOU. (2) Progress bar cues boost expectation, SAT, and UI. (3) Text cues improve TAM (PEOU, PU) and ECM (CON) perceptions. (4) Interaction effects: without cues or with a progress bar, task-led dialogues yield higher PEOU and PU than social-led; when text prompts are present, social-led dialogues receive higher ratings. The results of this study provide a reference for the design and development of dialogue cues in telecare dialogue agents.