Evaluating Explainable AI with Humans: Trust, Reliance, Understanding, and Performance
Explainable AI (XAI) is often evaluated through claims about trust, reliance, understanding, and performance, yet studies measure these constructs differently and frequently overlook confounding factors. This PRISMA-guided systematic review with narrative synthesis examines human-centered XAI evaluation across these fo...