"Beyond Perceivability": Critical Requirements for Universal Design of Information
Takashi Kato, Masahiro Hori · 2006 · Proceedings of the 8th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '06) · doi:10.1145/1168987.1169058
Summary
This paper argues that accessibility discussions focus too narrowly on perceivability — making information visually, audibly, or tactilely available — while neglecting the equally critical requirement of cognitive understandability. The authors contend that information should not be considered "accessed" merely because it is perceivable; it must also be cognitively internalised or understood by the user. They propose an extended HCI model that explicitly distinguishes between perceiving and understanding at each stage of human-computer interaction. Based on Norman's Seven Stages of Action model, the extended model incorporates two key distinctions: "object vs. action" and "perceiving vs. understanding." Where the standard cognitive walkthrough (CW3, the third version) poses 4 questions at each step of a correct action sequence, the authors' Extended Cognitive Walkthrough (ECW) generates 9 questions by requiring analysts to explicitly address whether the user can both perceive and understand relevant interaction targets (objects such as icons and links), actions (such as clicking and dragging), and system states at each step.
Key findings
A comparative study with 48 third- and fourth-year undergraduate students evaluated the ECW against the standard CW3 on two fictitious websites (an online jeans shop and a university site). The ECW was significantly more effective at identifying accessibility and usability problems, with a higher mean proportion of correct identifications (.58) compared to CW3 (.42), a statistically significant difference (F(1,46)=10.39, p<.005). Neither website type nor the interaction between factors was significant. When two questions (Q2-c and Q2-d, about perceiving and understanding "mouse click" actions) were excluded as trivially obvious, the ECW remained significantly more effective (.57 vs CW3). Critically, the ECW achieved this improvement without taking significantly longer: task completion times were comparable (ECW 1748 sec vs CW3 1709 sec, F<1 when the trivial questions were excluded). The results demonstrate that explicitly prompting evaluators to consider understanding separately from perceiving leads to the discovery of more accessibility and usability problems at no additional time cost.
Relevance
This paper makes a fundamental conceptual contribution to how we think about accessibility: perceivability is necessary but insufficient. A screen reader can make text perceivable to a blind user, but if the content is poorly structured, uses jargon, or relies on implied visual relationships, it may not be understandable. This distinction directly anticipates WCAG 2.0's separation of Perceivable and Understandable as distinct principles (published two years after this paper). For practitioners, the ECW method offers a practical tool for catching cognitive accessibility problems that standard evaluation methods miss. The finding that simply asking "can the user understand this?" in addition to "can the user perceive this?" significantly improves problem detection suggests that many accessibility evaluations are systematically overlooking cognitive barriers — a gap that remains highly relevant as web content grows more complex.
Tags: cognitive accessibility · cognitive usability · universal design · cognitive walkthrough · evaluation methods · web accessibility · perceivability · understandability