JOURNAL ARTICLE · EDITORIAL BRIEF
Human-centered evaluation of data visualization understanding in vision-language models
Arnav Verma; Kushin Mukherjee; Christopher Potts; Elisa Kreiss; Judith E. Fan
Scientific Reports · 19 October 2026
The finding or development
The tested models scored below people on average and made reliably different patterns of errors. A good overall score alone therefore does not establish that a model interprets charts as people do.
What supports it
Peer-reviewed early publication comparing eight vision-language models with human responses across six visualisation assessments.
Reviewed: Publisher abstract and publication metadata only; methods, model versions and participant sampling not reviewed.
Why it matters for design
For designers evaluating dashboards or AI chart assistants, compare specific misunderstandings as well as accuracy. This is an editorial implication, not a tested design intervention.
Read with this in mind
Abstract-only review. Results concern the tested models and tasks, not every current AI system or real-world dashboard. The publisher identifies this as an accepted early version awaiting final editing.
Original source
Opens the publisher outside this reading room. If nothing opens, copy the source address below.
Reference tools · Paperpile & more
Using Paperpile? Open the original source above, then use the Paperpile extension to save it to your library. Its inline buttons appear on supported websites.
Paperpile’s browser extension guide (opens an external website)Or download the original reference and import the file into Paperpile, Zotero or EndNote. Check the details against the source before citing.
Citation details are requested from Crossref when you download. The file contains a reference, not the full paper.
DOI: 10.1038/s41598-026-68119-z
This is the complete UX Observatory brief, not the publisher’s full article. Follow the original source to read the work.