This arXiv paper presents a video-based scaffolding protocol for eliciting explainable AI requirements in stroke rehabilitation. The formative study involved three stroke survivors, two with moderate-to-severe aphasia, and three caregivers. Facilitators used analogical bridging, projective personas, binary forcing, and extended response time to reduce communication and cognitive barriers. The protocol surfaced heterogeneous and sometimes conflicting transparency needs. Reflexive analysis also identified three facilitation risks: normative bias, hypothesis-confirmation bias, and presence effect. The authors frame the protocol and risk guidelines as reusable methods for patient-facing XAI design.
No heat snapshots are available in the last 24 hours.