BCIwiki (bciwiki.com) — Researchers report that a formative study eliciting explainable AI (XAI) requirements from three stroke survivors and three caregivers revealed three systematic facilitation biases, according to a preprint posted to arXiv on July 28, 2026 that has not yet undergone peer review.
Eliciting XAI requirements from stroke survivors presents a methodological challenge with direct implications for designing AI-driven adaptive rehabilitation systems, but existing protocols typically conduct dyadic interviews and overlook facilitation dynamics. Moving from dyad to triad, facilitators in the formative study employed four scaffolding approaches, analogical bridging, projective personas, binary forcing, and extended response time. Reflexive analysis revealed three systematic facilitation biases, normative bias, hypothesis confirmation bias, and presence effect, which the authors present as protocol risk guidelines for practitioners.