Q&A or Document-Based? The Effects of Interface Type on How Screen Reader Users Access Interconnected Documents

Published in The 28th International ACM SIGAC-CESS Conference on Computers and Accessibility (ASSETS ’26), 2026

Blind and low-vision people are using AI chatbots more and more to read and understand documents. But do these chatbots actually help them learn better than reading a normal, structured document?

To find out, we asked 16 blind and low-vision people to explore imaginary worlds using two different tools: an AI chatbot (where you ask questions and get answers) and a traditional document format (like a standard webpage or text file).

When people used the traditional documents, they actually learned more, understood the big picture better, and were better at using what they learned. But interestingly, most people liked the AI chatbot more. They even thought they had learned more with the chatbot, even though the tests showed the exact opposite.

This shows that while AI chatbots feel easy and helpful to use, they might actually make it harder for people to truly connect the dots and get a full understanding of a new subject.

Recommended citation: Colleen F. Cipriano, Yichun Zhao, Miguel A. Nacenta, Kotaro Hara, and JayleeSoh. 2026. Q&A or Document-Based? The Effects of Interface Type on HowScreen Reader Users Access Interconnected Documents. In The 28th International ACM SIGAC-CESS Conference on Computers and Accessibility (ASSETS ’26), October 25–28,2026, Vila Nova de Gaia, Portugal. ACM, New York, NY, USA, 17 pages.https://doi.org/10.1145/3797890 (ASSETS, paper presentation) link