Accessibility & Readability Check
Approximately 8% of men and 0.5% of women have some form of red-green color vision deficiency. A figure that distinguishes groups by red vs. green is uninterpretable for this audience. Licklider simulates deuteranopia and protanopia, checks contrast and readability, and suggests perceptually uniform alternatives —before the figure leaves the lab.
STEP 1 —The Pitfall: Figures That Exclude Part of the Audience
The default color palette in most scientific software —blue, orange, green, red, purple — was not designed with color vision deficiency in mind. The red-green pair (groups colored red and green) is the most common example: it is completely indistinguishable to individuals with deuteranopia (the most common form of red-green color blindness, affecting ~6% of men).
Beyond color blindness, readability is a general concern:
- Low contrast text. Gray axis labels on a white background may be readable on screen but disappear in print. WCAG 2.1 requires a contrast ratio of at least 4.5:1 for small text —a threshold many figure styles fail.
- Small fonts in print. A figure that looks readable at 100% zoom on a 27-inch monitor may have 5pt labels when printed at journal column width (89 mm). Most journals specify a 7pt minimum; readers typically need 8pt for comfortable reading.
- Thin lines invisible in grayscale. Many papers are printed in black and white by readers who print to save costs. A figure designed exclusively for color loses its information structure in grayscale if group distinctions rely only on color.
STEP 2 —Journal Requirement
Nature Methods states that "figures should be accessible to readers with color-vision deficiencies and should not rely solely on color to convey information." PLOS ONE requires that "figures should be legible to readers with color-vision deficiencies and to readers who will print the paper in black and white."
An increasing number of journals —including all Nature-family journals and many Cell Press journals —have adopted explicit accessibility requirements for submitted figures, in line with the WCAG 2.1 standards used in web publishing. The BMJ's editorial guidance states that "color should not be used as the only visual means of conveying information."
STEP 3 —Licklider's Solution
Input
- The figure being exported, including all color assignments for data series, annotations, and background elements.
- No additional setup required. The accessibility check runs automatically as part of the export pre-flight.
Output
- Color-blind simulation. The figure is rendered with deuteranopia and protanopia color filters applied, showing how the figure appears to affected readers. The simulation is displayed in the pre-flight check panel for review.
- Palette conflict detection. Color pairs that are indistinguishable under any of the simulated conditions are flagged: "Control (red, #CC0000) and Treatment (green, #00CC00) are indistinguishable under deuteranopia."
- Perceptually uniform palette suggestions. Licklider suggests replacements from the Viridis, Cividis, or ColorBrewer accessible palettes. The replacement can be applied with a single click.
- Contrast ratio check. The WCAG 2.1 contrast ratio between text elements (axis labels, tick marks, legend text) and their backgrounds is computed and reported. Elements below 4.5:1 are flagged.
- Grayscale simulation. The figure is rendered in grayscale to verify that group distinctions are preserved without color —through shape, pattern, or lightness differences.
Guard
Guard condition —Critical color conflict (advisory): When a color pair is flagged as completely indistinguishable under both deuteranopia and protanopia simulations, the figure is marked with an accessibility warning. Export proceeds, but the researcher must acknowledge the warning and either apply the suggested palette or provide a justification (e.g., the figures will be published alongside additional distinguishing markers such as symbols or patterns).
STEP 4 —Draft Output (Draft / Needs review)
Example: Accessibility check report
ACCESSIBILITY CHECK REPORT —Figure 2B
--------------------------------------------------
Check Status Details
Deuteranopia simulation [X] FAIL Control (#CC0000) and Treatment (#00CC00)
indistinguishable. Suggested: Viridis
#440154 (control) / #FDE725 (treatment)
Protanopia simulation [X] FAIL Same pair indistinguishable
Tritanopia simulation [OK]PASS Groups distinguishable
Grayscale legibility [X] WARN Groups not distinguishable without color.
Recommend adding marker shapes (circle vs. square)
Contrast ratio —axis labels [OK]PASS 8.2:1 (exceeds WCAG 4.5:1 minimum)
Font size —axis labels [OK]PASS 7pt (meets Nature minimum)
Line weight —axes [OK]PASS 0.5pt (meets Nature minimum)
--------------------------------------------------
Action required: Replace red/green pair with accessible palette.
[Apply Viridis palette] [Apply Cividis palette] [Acknowledge and keep]