What If Conversational Agents Became Invisible? Comparing Users’ Mental Models According to ...
summary
Study compares visible versus hidden conversational agents using a drawing-and-interview study with 30 novice users interacting with 3 AI speakers placed in three locations or hidden from view. Visible devices are drawn as three distinct agents with location-specific roles and elicit users to face and maintain “eye contact,” aiding learnability and social cues. Invisible agents are perceived as a single whole-home controller, enable seamless multitasking and ambient support, and produce no significant difference in overall user experience across conditions.
AI-written summary · about
chapters (5)
related talks
22mIs the Invisible Interface Really Invisible?
36mEmbodied AI: A Cognitive Shift, Not a Cosmetic Change
55mDon’t Make Me Prompt: Intelligent Interfaces Users Actually Like
28mHuman Computer Interaction: The UX of AI
57mDesigning Human-AI Experience
34mAI-first Without UI-first: Moving from Graphical Interfaces to Cognitive Systems
16mWeak signals of a vocal revolution in UIs
49mThe Shape of AI