Designing for Effective Human-XAI Interaction: User Experience Research Plays and Cards

Authors: Naiseh, M., Dogan, H., Giff, S., Malhi, A. and Jiang, N.

Journal: Lecture Notes in Computer Science

Volume: 15936 LNCS

Pages: 229-241

eISSN: 1611-3349

ISSN: 0302-9743

DOI: 10.1007/978-3-032-01399-6_13

Abstract:

Explainable Artificial Intelligence (XAI) has emerged as a critical field for fostering trust, transparency, and comprehension in human-AI interactions. However, existing XAI systems often fall short of addressing real-world usability challenges, resulting in suboptimal adoption and engagement. This paper applies the User Experience Research Point of View (UXR PoV) playbook to Human-XAI interactions as a case study, i.e., a structured framework designed to guide multidisciplinary teams in creating effective human-centered XAI systems. The playbook consists of actionable play cards, organised into three dimensions: Usability Enhancement, Human-Like Enhancement, and Learning Enhancement. Our proposed Human-XAI plays and cards aim to improve the usability and long-term impact of XAI systems by leveraging iterative design principles, interdisciplinary collaboration, and evidence-based practices.

Source: Scopus

Designing for Effective Human-XAI Inter-action: User Experience Research Plays and Cards

Authors: Naiseh, M., Dogan, H., Giff, S., Malhi, A. and Jiang, N.

Conference: Explainable, Trustworthy, and Responsible AI and Multi-Agent Systems (EXTRAAMAS 2025)

Dates: 19-20 May 2025

Source: Manual