Plate 1 of 14
Screening tool, not a clinical diagnosis
What number do you see?
Why might I see a deficit even with normal vision?
- Display calibration. Phone screens vary 5-15 ΔE between manufacturers. Older or budget panels may compress the color gamut enough to mask plate boundaries.
- Night Shift / True Tone / blue-light filters. These warm the white point and shift hues, especially affecting blue-yellow plates. Turn them off and retake.
- Ambient lighting. Bright sunlight or warm/yellow indoor lighting can wash out the chromatic contrast the test relies on.
- Screen brightness. Below ~50% brightness the dark dots blend together, especially on the subtle red-green plates.
- Fatigue or fast clicking. Color discrimination is harder when you're tired or rushing. Slow down on retake.
- Mild anomalous trichromacy you didn't know you had. About 5-8% of men have a mild form, often undiagnosed.
Your results plate by plate
This is a screening, not a medical diagnosis. See an eye care professional for a clinical evaluation.
What Color Is This?
A color identification app built for how you actually see.
This is free and there's no marketing list. Signing up is actually a favor to me — Google Play requires 12 testers. Privacy
You're in! I'll email you when the beta is ready.
About this screening
This is a 14-plate pseudoisochromatic screening that tests for red-green and blue-yellow color vision deficiency, and attempts to classify red-green deficiency as protan-type or deutan-type. It combines two established clinical methodologies with confusion-line geometry from Brettel et al. (1997):
Ishihara-style plates present numbers hidden in fields of colored dots. The foreground and background colors are placed so that hue is the only available cue (matched luminance). People with color vision deficiency cannot distinguish the figure from the background.
HRR-style plates use geometric shapes instead of numbers, which avoids digit-recognition bias and works for participants who cannot read numerals. The Hardy-Rand-Rittler test is the standard clinical complement to Ishihara for identifying both type and severity.
Protan-only and deutan-only classifier plates use color pairs that vanish under one form of dichromacy but remain visible under the other. Their colors were generated by simulating each candidate pair under Brettel et al.'s 1997 dichromacy model and selecting pairs with very low CIEDE2000 distance for the target type and high distance for the opposite type. These plates let the screening separate protan-type from deutan-type rather than reporting both as undifferentiated red-green CVD.
Tritan plates use pairs that map to the same color under tritan simulation, which is a much stronger filter than the b*-axis separation a naive design would use. A b*-axis pair leaks ~30 CIEDE2000 units of contrast in tritan-simulated space; the confusion-line pairs here leak under 1 unit.
Plate 1 is a luminance-contrast control: it should be visible regardless of color vision type. If it is missed, the result is flagged as inconclusive (likely a screen brightness or lighting issue).
Limitations: This is a screen-based approximation. Clinical Ishihara and HRR tests use precisely calibrated printed pigments under standardized illumination (CIE Illuminant C). Display color rendering varies by device, panel type, brightness, and ambient lighting. This screening can indicate the likely presence and type of color vision deficiency, but it cannot replace a clinical evaluation.
References