A Large-Scale Mixed-Methods Analysis of Blind and Low-vision Research in ACM and IEEE
Thoo, Yong-Joon, Jeanneret Medina, Maximiliano, Froehlich, Jon E., Ruffieux, Nicolas, Lalanne, Denis · 2023 · Proceedings of the 25th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS) · doi:10.1145/3597638.3608412
Summary
This paper provides the most comprehensive mapping to date of research on technology for blind and low-vision (BLV) people published across ACM and IEEE venues between 2010 and 2022. The authors combined quantitative bibliometric methods — specifically documents bibliographic coupling analysis (DBCA) and text mining of titles, abstracts, and keywords — with qualitative coding of the 100 most-cited papers. Starting from a Scopus query that yielded 3,378 results, the researchers applied systematic screening, filtering for relevance and citation impact using Z-score normalization, arriving at a final dataset of 880 papers published across 240 different venues. CHI (17.7%) and ASSETS (16%) were the most prevalent venues. The 880 papers were written by 2,458 unique authors with a mean of 4.2 authors per paper. The research team developed an eleven-category conceptual framework for qualitative coding that captured research area, issue addressed, contribution type, community of focus, age category, technology, device, vision use strategy, and input/output modalities. Using VOSviewer for science mapping with the Leiden clustering algorithm, the authors organized the field into five bibliometric clusters, which they then interpreted into four primary research areas: Accessibility at Home and on the Go (N=280, 31.8%), Non-Visual Interaction (N=195, 22.2%), Orientation and Mobility (N=331, 37.6%), and Education (N=54, 6.1%).
Key findings
The largest research area is Orientation and Mobility, encompassing navigation assistance systems that localize users, detect obstacles, plan paths, and provide audio or haptic feedback — often leveraging computer vision and deep learning. Accessibility at Home and on the Go focuses on helping BLV people with daily visual tasks including image description, visual question answering (notably the VizWiz project), and personal object recognition. Non-Visual Interaction research centers on speech and touch modalities, with voice assistants and touchscreen accessibility being major topics. Education research emphasizes tangible technologies and collaborative learning for STEM subjects. Technologically, mobile devices dominate (25.5% of papers), followed by computer vision (15.8%), wearables (12%), and navigation assistance (12%). AI and mixed reality have shown notable growth since 2015-2016. A striking finding is the diversity of terminology: the authors identified seven distinct denomination categories used to refer to BLV communities, with person-first and identity-first language varying significantly across papers. Most solutions (N=50) target both blind and low-vision users, though 36 papers focus solely on blind users. Audio is the dominant output modality (N=46), while 31 systems require no user input at all, operating automatically via computer vision or sensors.
Relevance
This review is an essential resource for anyone working in BLV accessibility research or practice, providing a bird's-eye view of where the field has been and where it is heading. For practitioners, the identification of four major research streams helps contextualize individual technologies within the broader landscape. The paper raises important methodological concerns: the lack of a standard for reporting participants' visual abilities makes it difficult to compare studies or replicate findings. The authors also note that solutions overwhelmingly favor visual substitution (replacing visual information with non-visual output) over visual enhancement strategies for people with residual vision — an underexplored opportunity. The analysis of terminology diversity is particularly valuable for organizations developing inclusive language guidelines. Future directions include greater attention to low-vision-specific solutions, better reporting standards for participant visual abilities, and investigation of how factors beyond visual acuity — such as cognitive-emotional state and personal preferences — affect technology adoption.
Tags: blind and low vision · systematic review · bibliometrics · visual impairment · literature review · research trends · assistive technology · HCI