Identifying Behavioral Strategies of Visually Impaired Users to Improve Access to Web Content
Darren Lunn, Simon Harper, Sean Bechhofer · 2011 · ACM Transactions on Accessible Computing · doi:10.1145/1952388.1952390
Summary
This paper presents a methodology for identifying how visually impaired screen reader users develop coping strategies when encountering inaccessible web content, then uses those strategies to inform the design of transcoding solutions. The researchers conducted observational studies with 11 visually impaired users at Henshaws Society for Blind People, using thematic analysis to identify a seven-factor framework for categorizing coping behaviors: the difficulty users experienced, what triggered the coping response, resources available, constraints limiting options, chosen actions, user goals, and outcomes. This framework was validated against data from 20 additional users in an independent NoVA study. From this analysis, six abstract coping patterns emerged: Candidate Chunk Discovery (scanning for important content sections), Masthead Avoidance (skipping repetitive navigation), Clustered Element Discovery (finding related content groups), Probing (testing links to verify destinations), Backtracking (returning to safe states when lost), and Withdrawal (giving up entirely when frustrated). The research then translated these behavioral patterns into transcoding functionality implemented via ARIA landmarks, allowing users to navigate using simple key presses: n/p for next/previous content chunks, m for menu access, and s for page summaries.
Key findings
The evaluation compared behavior-driven transcoding against both original web pages and structure-driven transcoding (which removes clutter and reorganizes content). With 8 participants across two groups performing tasks on 8 major websites (BBC News, CNN, Google, NY Times, etc.), behavior-driven transcoding outperformed both alternatives. All tasks were completed faster on behavior-driven transcoded pages than originals, and 5 of 8 tasks were faster than structure-driven transcoding. Users found the keyboard commands intuitive—one participant noted "I pressed 'm' and it gives me like the main menu isn't it?" Critically, users did not need prior familiarity with pages to navigate effectively; the transcoding surfaced important content chunks consistently. Structure-driven transcoding alone improved times by reorganizing content, but behavior-driven transcoding provided additional benefits by matching how users naturally want to interact with content rather than just adapting page structure.
Relevance
This research demonstrates that understanding actual user behavior—not just technical accessibility requirements—leads to better accessibility solutions. The coping strategy framework provides a reusable methodology for identifying how users adapt to inaccessible content, applicable beyond visual impairment to cognitive disabilities, motor impairments, or age-related access needs. The practical implementation using ARIA landmarks shows how semantic markup can enable adaptive interfaces without requiring changes to original content. For practitioners, this validates user research as essential for accessibility: automated testing catches structural issues, but observing real coping behaviors reveals interaction patterns that compliance alone cannot address. The Withdrawal Strategy findings particularly highlight the emotional toll of inaccessibility—users sometimes respond with frustration severe enough to abandon tasks or physically abuse equipment.
Tags: screen readers · web accessibility · transcoding · coping strategies · user behavior · visual impairment · ARIA
Standards referenced: WCAG 2.0 · WAI-ARIA