Towards Human-centered Explainable AI: A Survey of User Studies for Model Explanations
This survey addresses the need for better human-centered evaluations in XAI to improve usability and collaboration for researchers and practitioners, but it is incremental as it synthesizes existing studies without introducing new methods.
The paper conducted a systematic review of 97 user studies in explainable AI (XAI) to analyze how human-centered evaluations are performed, categorizing them by trust, understanding, usability, and collaboration, and found that user evaluations are sparse and lack insights from cognitive or social sciences.
Explainable AI (XAI) is widely viewed as a sine qua non for ever-expanding AI research. A better understanding of the needs of XAI users, as well as human-centered evaluations of explainable models are both a necessity and a challenge. In this paper, we explore how HCI and AI researchers conduct user studies in XAI applications based on a systematic literature review. After identifying and thoroughly analyzing 97core papers with human-based XAI evaluations over the past five years, we categorize them along the measured characteristics of explanatory methods, namely trust, understanding, usability, and human-AI collaboration performance. Our research shows that XAI is spreading more rapidly in certain application domains, such as recommender systems than in others, but that user evaluations are still rather sparse and incorporate hardly any insights from cognitive or social sciences. Based on a comprehensive discussion of best practices, i.e., common models, design choices, and measures in user studies, we propose practical guidelines on designing and conducting user studies for XAI researchers and practitioners. Lastly, this survey also highlights several open research directions, particularly linking psychological science and human-centered XAI.