Modeling Accessibility: Characterizing What We Mean by "Accessible"
Kelly Avery Mack, Jesse J Martinez, Aaleyah Lewis, Jennifer Mankoff, James Fogarty, Leah Findlater, Heather D. Evans, Cynthia L Bennett, Emma J McDonnell · 2025 · ASSETS 2025: 27th International ACM SIGACCESS Conference on Computers and Accessibility · doi:10.1145/3663547.3746344
Summary
This paper tackles a fundamental gap in accessibility research: the field lacks a clear, shared characterization of what "accessibility" actually means. While accessibility research aims to use technology to make the world more accessible to disabled people, the concept itself has remained undertheorized. The authors conducted hour-long semi-structured interviews with 25 disabled people who were intentionally recruited to represent diversity across disability types, racial identities, gender identities, and other intersecting characteristics. Participants included people with vision disabilities, motor disabilities, chronic illness, d/Deaf and Hard of Hearing conditions, mental health conditions, intellectual and developmental disabilities, neurodivergence, and addiction—many with multiple disabilities. The research is grounded in intersectionality theory from Black feminist scholarship (Kimberlé Crenshaw, Patricia Hill Collins), disability studies models of disability, and HCI accessibility design paradigms including ability-based design, social accessibility, interdependence, and crip technoscience. The analysis used a combination of inductive and deductive coding by four researchers, with affinity diagramming for deeper thematic analysis. From this empirical foundation, the authors develop a process for "modeling accessibility" that captures how individuals assess and navigate access barriers in their daily lives, emphasizing that accessibility is fundamentally situated—shaped by identity, context, available tools, and personal priorities.
Key findings
The paper identifies four distinct types of access barriers: (1) failure point barriers, where a task is simply impossible without assistive technology; (2) usability barriers, where a task can be completed but not with the desired quality or efficiency; (3) bodymind barriers, where performing a task leads to an undesirable physical or mental state (pain, fatigue, distress); and (4) future impact barriers, where completing a task now will cause negative consequences later. The authors introduce the concept of technology repertoires—the collections of tools people use in coordination to achieve access. These take two forms: combination repertoires (multiple tools working together for one task) and option repertoires (multiple tools providing alternative approaches for the same task in different contexts). The modeling process involves two phases: first, assessing the situation by identifying the barrier type, available repertoire, and contextual factors; then performing "consequence calculus"—weighing all options against personal priorities to choose the best path forward. Critically, contextual factors including race, gender, class, queerness, language, religion, and community membership profoundly shape what tools are available, how they are experienced, and what constitutes an acceptable access solution. Participants described tools being unusable due to not matching their skin tone, cultural stigma around disability preventing technology adoption, and financial barriers limiting repertoire options.
Relevance
This is a landmark theoretical contribution to accessibility research, providing the field with shared vocabulary and a structured process for understanding how accessibility actually works in practice. For practitioners, the four barrier types offer a practical framework for evaluating whether a technology truly addresses access needs or merely resolves one barrier type while creating others. The concept of technology repertoires challenges the assumption that a single tool can solve an access need—designers should consider how their tools fit within broader ecosystems of use. The emphasis on contextual factors, particularly non-disability identities, is a crucial corrective: accessibility solutions that ignore race, gender, income, and cultural context risk serving only the most privileged disabled users. The paper's design implications—including designing for interoperability within repertoires, considering multiple sites of change (including the bodymind itself), and identifying under-served identity intersections—provide actionable guidance for inclusive design practice. Limitations include the US-centric perspective and focus on technology-mediated access.
Tags: accessibility theory · models of disability · intersectionality · assistive technology · disability justice · qualitative research · access barriers · technology repertoires · contextual factors