Empowering User Agency in Smart Indoor Workspaces

Office workers increasingly manage their own lighting environments using mobile devices, yet existing control interfaces—typically slider-based panels—impose high cognitive load and fail to support divided attention in dynamic work settings. This research investigates how a mobile, context-aware recommendation interface can reduce this burden while preserving user agency. This study designed a tablet-based prototype that presents a single task-specific lighting recommendation with three symmetrically weighted actions—accept, adjust, and dismiss—accompanied by a real-time illuminance preview. A mixed-design user study with 36 office workers compared the adaptive interface against a manual control baseline. The adaptive interface reduced perceived workload by 28.6% on the NASA-TLX, improved visual comfort by 41.2%, and achieved significantly higher System Usability Scale (SUS) scores. Interaction logs revealed a 70.4% acceptance rate alongside frequent fine-grained adjustments. The findings offer design implications for mobile interfaces that balance system proactivity with user agency.