Shape and texture differences survive greyscale output and colour-vision deficiency

What it is

Non-color redundant cues keep category identity from depending only on color differences that may shrink under grayscale conversion or a particular color-vision condition. Shape, texture, position, boundaries, and direct labels do not rely on the same hue difference and can provide another identification path. “Survive” does not mean these cues remain intact or legible to everyone; they still depend on size, contrast, crowding, vision, and the output pipeline.

Why it happens

Different types and degrees of color-vision deficiency alter discriminability for particular color pairs rather than disabling color wholesale; lightness perception should not be assumed invariant across people and conditions. Grayscale conversion depends on its mapping from color to luminance and may merge some colors. Contours and patterns provide structure different from color difference, but they too can become confusable when marks are small, contrast is weak, or print degrades.

Studying it

Test category identification, search time, and confusion in the original color, the target grayscale conversion, relevant output media, and relevant user groups. Simulators can flag candidate problems but are model-based diagnostics, not substitutes for real tasks and users. Toggle labels, position, patterns, and other redundant cues separately to estimate each path's contribution.

Where it stops holding

Some palettes remain discriminable under a particular color-vision condition, while shape or texture may fail first under low vision, dense scatterplots, or poor print. A grayscale check reveals dependence on one conversion and does not represent all color-vision variation. If color carries state, order, or brand meaning as well as identity, redundant cues must preserve the same semantics rather than merely look different.

Applying it

• Combine color with position, shape, pattern, boundary, or direct labels for consequential categories, ensuring the redundant cues express the same category relation.
• Check actual grayscale conversion, print, zoom, and minimum size, but treat these as output conditions rather than substitutes for color-vision testing.
• Use multiple simulations to find risks, then validate with target users and real tasks; record the color pairs, marks, and conditions that confuse.
• Provide sortable tables, category names, and structured data so nonvisual access does not depend on color, shape, or texture.

Related

• Same group: U1.07.1 Shape is categorical only, and reliably distinguishable shapes are few · U1.07.2 Shape judgment needs adequate mark size; tiny scatter dots defeat it · U1.07.3 Texture fills are unstable under scaling and printing · U1.07.4 Orientation encodes a few ordered values but rotational symmetry confuses it
• Nearby: U1.10.2 Redundant encoding is the main fallback for colour-vision deficiency and greyscale output · U4.04 Colour-vision-deficiency-safe palettes
• Search terms: redundant encoding · greyscale · colour-vision deficiency · pattern fill