Search and Filter

Repository Feedback

Your feedback helps us improve the repository's content relevance and usability. Please share your thoughts to help us better serve researchers and practitioners.

Submit feedback

Submit a research study

Contribute to the repository:

Add a paper

How University Disability Services Professionals Write Image Descriptions For Hci Figures Using Generative AI

Date
Publisher
arXiv
Disability Services Office (DSO) professionals at higher education institutions write alt text for visual content. However, due to the complexity of visual content, such as HCI figures in research publications, DSO professionals can struggle to write high-quality alt text if they lack subject expertise. Generative AI has shown potential for understanding figures and writing descriptions, yet its support for DSO professionals remains underexplored, and few studies evaluate the quality of AI-assisted alt text. In this work, we conducted two studies: first, we investigated generative AI support for writing alt text for HCI figures with 12 DSO professionals. Second, we recruited 11 HCI experts to evaluate the alt text written by DSO professionals. Findings show that alt text written solely by DSO professionals has lower quality than alt text written with AI assistance. AI assistance also helped DSO professionals write alt text more quickly and with greater confidence; however, they reported inefficacy in interactions with the AI. Our work contributes to research on AI support for non-subject-expert accessibility professionals.
Study at a glance
  • Question: How does generative AI assistance affect how disability services professionals write image descriptions for complex scientific figures?
  • Setting and data: Higher education institutions in the US, mid-2025; 12 disability services professionals and 11 HCI experts evaluating descriptions for 16 figures.
  • Method: Two-phase mixed methods user study (user testing, think-aloud protocols, interviews, rubrics, and expert evaluations).
Why it matters

University disability services professionals struggle to write effective alternative text for complex scientific figures outside their expertise. Without adequate descriptions, students with accessibility needs face substantial barriers when accessing visual academic content.

Key findings
  • HCI experts rated AI-assisted figure descriptions significantly higher than human-only descriptions across accuracy, completeness, clarity, effectiveness, and descriptiveness.
  • AI assistance helped disability professionals complete descriptions more quickly and with higher confidence, though they rarely engaged in iterative prompting to refine or verify outputs.
  • Experts frequently misidentified text generation sources, mistaking human-written descriptions for AI 42% of the time and AI-assisted descriptions for human-written 45% of the time.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

Who is the user?
Who age?