A Study of Blind Drawing Practice: Creating Graphical Information Without the Visual Channel
Hesham M. Kamel, James A. Landay · 2000 · Proceedings of the Fourth International ACM Conference on Assistive Technologies (Assets '00) · doi:10.1145/354324.354334
Summary
This paper from UC Berkeley investigates how blind users create drawings and proposes a grid-based model for non-visual drawing tools. The researchers began by observing that existing drawing tools for blind users fail because they lack adequate contextual feedback — blind users cannot perceive the state of their drawing as they work. To understand what feedback is actually needed, the team conducted a tactile drawing study at the Orientation Center for the Blind in Albany, California, with five participants (three with low vision, two totally blind, ages 18-42). Participants used a Sewell raised line drawing kit — a clipboard with a rubber surface where pressing an inkless stylus on a plastic sheet creates a tactile raised impression — to complete four drawing tasks: drawing an uppercase "D," drawing greater-than signs, drawing two squares, and a free drawing exercise. The study revealed that existing tactile freehand drawing provides no feedback for relocating important points, determining line lengths, assessing curvature, or maintaining consistent angles. Participants independently invented compensatory strategies like counting aloud, measuring with knuckles, and using thumb-to-pinky span as a ruler — all attempts to impose a frame of reference that the drawing medium itself did not provide.
Key findings
The study identified four essential properties that any successful drawing program for blind users must support: (1) targeting an abstract point in relation to the drawing — the ability to relocate a previously selected point, which participants failed at when trying to close figures like the letter "D"; (2) evaluating the length of a line, since participants could not make equal-length sides without visual feedback; (3) assessing the curvature of a drawing, as participants drew curves that did not match their mental model; and (4) determining angles, evidenced by participants' inability to maintain consistent 90-degree corners. Based on these findings, the authors proposed a grid-based model using the familiar 3x3 telephone keypad layout as a spatial reference frame. The grid can be applied recursively — selecting a cell and then subdividing it into another 3x3 grid — for increasingly fine-grained point selection. This model was implemented in IC2D (Integrated Communication 2 Draw), a prototype drawing program where a blind user (age 23, blind since birth) successfully drew a cube by selecting grid points and connecting them with lines, demonstrating the viability of the approach.
Relevance
This research challenged the assumption that graphical information creation is inherently visual and therefore inaccessible to blind users. The key insight — that the problem is not blind users' inability to conceptualise spatial relationships, but rather the lack of appropriate feedback mechanisms in drawing tools — reframes graphical accessibility as a design challenge rather than a limitation of blindness. The grid-based model using the telephone keypad metaphor was elegant in its use of a spatial layout already familiar to blind users from daily life, demonstrating how accessible tools can build on existing mental models rather than requiring users to learn entirely new paradigms. The findings have implications beyond drawing: the same grid-based spatial reference approach could be applied to any graphical interaction including data visualisation, map navigation, diagram comprehension, and GUI interaction. The work also highlighted the value of observing blind users' existing compensatory strategies to inform tool design — a user-centred methodology that remains a best practice in assistive technology development.
Tags: tactile graphics · blindness and low vision · non-visual interaction · drawing tools · accessible graphics · user research · graphical accessibility