Saturation supports fewer discriminable levels than lightness

What it is

Conditional discriminability of chroma and saturation describes a common result for sequential palettes and ordering tasks: varying only vividness versus gray often supports fewer robust steps than a monotonic lightness path. “Saturation” is not the same coordinate in HSV/HSL, CIELAB chroma, and color-appearance models. The title is therefore a relative tendency under common design conditions, not a universal limit of four or five levels.

Why it happens

Available chroma depends on hue, lightness, reference white, background, and device gamut; even at fixed L*, hues have different distances to the gamut boundary. HSV S is an algorithmic coordinate, so equal numeric steps are not equal appearance steps. Lightness and chroma also interact in appearance, and viewers may read chroma change as salience rather than precise order.

Studying it

Declare the chroma or saturation definition, then test pair discrimination, multilevel ordering, and lookup while holding or systematically varying hue and lightness. Manipulate step distance, mark size and shape, background, density, display gamut, and visual ability. Define “discriminable” with a preregistered accuracy criterion and interval. Include a lightness-ramp comparison; otherwise the study cannot support the claim “fewer.”

Where it stops holding

Large swatches, adjacent legends, wide chroma gaps, and stable displays can support several distinctions; no count transfers across hues and devices. Binary emphasis is a different task from precise multilevel order. Even when coordinate steps are controlled, color-vision differences, gamut clipping, and surround induction can change achieved discriminability.

Applying it

• Do not generate sequential ramps from equal HSV/HSL saturation steps. Record color space, reference white, gamut mapping, and target background.
• Use a lightness path as the main carrier for many ordered levels, with chroma only as a tested supporting variation. Specify binary emphasis separately from quantitative ramps.
• Set the number of levels from the smallest mark, weakest device, and target surround. If steps fail, merge data bins rather than only stretching coordinates.
• Add redundant text, shape, or position cues and provide bin name, source value, unit, and order to screen readers.

Related

• Same group: U1.06.1 Lightness carries order — the first-choice colour dimension for ordinal data · U1.06.2 The same lightness reads as different values on light and dark backgrounds · U1.06.4 Lightness and saturation changes are read as one strength dimension · U1.06.5 Saturation differences are nearly invisible on small marks
• Adjacent: U1.09.1 Every visual channel supports a limited number of reliably discriminable levels · U4.02.3 Luminance carries the sense of order
• Search terms: chroma discrimination · saturation scale · ordinal color levels