Methodology

The survey begins with four colour-vision screening plates and one unrecorded practice question. It then presents 29 measured figure questions covering the design comparisons shown below.

Highlight all answer-area bounding boxes

Degradation crossing — Categorical Palette Robustness

Categorical Palette Robustness tested at both levels, under both normal and greyscale rendering. The normal and greyscale cells in the same row share identical underlying data -- only the colour transform differs -- so any difference between them is the degradation, not a data change.

	viewing condition: baseline	viewing condition: treatment
level: baseline	Scattered categorical point cloud: 14 samples split 8/4/2 across three categories. Categorical Palette Robustness = baseline (Arbitrary, CVD-confusable hues). tokens categorical_palette_robustness = baseline viewing normal prompt Which of the three categories has the MOST samples? options Category 1, Category 2, Category 3, Skip answer Category 1 pool degradation_crossing / scatter / normal / categorical_palette_robustness=baseline draw `degradation_crossing:scatter:normal:categorical_palette_robustness=baseline` · variant 1	Scattered categorical point cloud: 14 samples split 8/4/2 across three categories. Categorical Palette Robustness = baseline (Arbitrary, CVD-confusable hues). Rendered in greyscale to test whether the encoding survives colour loss. tokens categorical_palette_robustness = baseline viewing greyscale prompt Which of the three categories has the MOST samples? options Category 1, Category 2, Category 3, Skip answer Category 1 pool degradation_crossing / scatter / greyscale / categorical_palette_robustness=baseline draw `degradation_crossing:scatter:grayscale:categorical_palette_robustness=baseline` · variant 1
level: treatment	Scattered categorical point cloud: 14 samples split 8/4/2 across three categories. Categorical Palette Robustness = treatment (CVD-safe, luminance-spread palette). tokens categorical_palette_robustness = treatment viewing normal prompt Which of the three categories has the MOST samples? options Category 1, Category 2, Category 3, Skip answer Category 3 pool degradation_crossing / scatter / normal / categorical_palette_robustness=treatment draw `degradation_crossing:scatter:normal:categorical_palette_robustness=treatment` · variant 1	Scattered categorical point cloud: 14 samples split 8/4/2 across three categories. Categorical Palette Robustness = treatment (CVD-safe, luminance-spread palette). Rendered in greyscale to test whether the encoding survives colour loss. tokens categorical_palette_robustness = treatment viewing greyscale prompt Which of the three categories has the MOST samples? options Category 1, Category 2, Category 3, Skip answer Category 3 pool degradation_crossing / scatter / greyscale / categorical_palette_robustness=treatment draw `degradation_crossing:scatter:grayscale:categorical_palette_robustness=treatment` · variant 1