Redundant encoding is the main fallback for colour-vision deficiency and greyscale output

What it is

A non-color fallback for color encoding uses shape, line style, texture, lightness, position, or text to preserve consequential distinctions when hue differences fail or output becomes grayscale. “Main fallback” means a common strategy, not that mechanically pairing every color with a symbol completes accessibility. Grayscale output and different types and degrees of color-vision deficiency are not one visual condition.

Why it happens

Color-vision differences alter discriminability for particular color pairs rather than uniformly removing color. Grayscale conversion maps colors to light and dark according to a conversion and print pipeline. A fallback must retain discriminable structure under the target failure mode and at delivered mark sizes. Shape and texture have their own size, clutter, and resolution limits; similar lightness values can still merge in grayscale. Direct labels and data access bypass color matching but cost space or interaction.

Studying it

Enumerate consequential color pairs and tasks, then test full color, the target grayscale output, and relevant color-vision conditions separately. Simulation can screen obvious collisions but cannot replace participants with lived visual experience. Measure category matching, trend tracing, anomaly detection, and value retrieval rather than preference alone. Record conversion algorithm, export or print chain, mark size, zoom, and environment, and measure confusion for the fallback channel itself.

Where it stops holding

Non-color redundancy protects a specified task, not an equivalent visual experience. Large shape or texture sets create new capacity limits, especially on thin lines and small points; adding lightness alone does not cover every low-vision need. Visual redundancy does not expose chart relationships to screen-reader users without structured data, summaries, or operable details. CVD simulation, grayscale preview, and automated contrast checks are only partial evidence.

Applying it

• Add a non-color cue to consequential categories and states where it survives target size, print, and zoom; decorative color need not be redundantly encoded item by item.
• Use line style or direct endpoint labels for lines, size-tested shapes for points, and boundaries, textures, or labels for regions. Do not rely on tooltips alone.
• Inspect actual exports and prints in grayscale, screen with several relevant simulations, then validate real tasks with target users.
• Provide focusable data points, keyboard navigation, a text summary, or a data table so information does not depend on a visual channel pair.

Related

• Same group: U1.10.1 Encoding one variable on two channels improves judgment robustness · U1.10.3 Two different variables sharing one visual dimension interfere with each other · U1.10.4 Redundant encoding spends channels that other variables could use · U1.10.5 Redundant channels that disagree semantically give contradictory cues
• Adjacent: U1.07.5 Shape and texture differences survive greyscale output and colour-vision deficiency · U1.05.3 Smaller colour patches need larger hue differences to stay discriminable
• Search terms: color vision deficiency · grayscale output · non-color cue