Ishihara Test Answers and Scoring

The answers to our 20 plates, what people with each type of colour blindness see, and how the real Ishihara test is scored.

Answer key for our 20 plates

Take the test first: reading the answers before you start makes the result meaningless. The table is generated from the same data the test uses, so it always matches the plates.

Answers for each plate; "nothing" means the number is expected to disappear.

PlateTypical visionRed-weak (protan)Green-weak (deutan)Blue-yellow (tritan)TestsFades at
Test plate 1112121212Control (everyone)—
Test plate 228nothing88Red-green screening84%
Test plate 332929nothing29Red-green screening83%
Test plate 445nothingnothing5Red-green screening83%
Test plate 5574nothing7474Red-green screening76%
Test plate 66266226Protan / deutan83%
Test plate 7733nothing3Red-green screening71%
Test plate 8845nothingnothing45Red-green screening66%
Test plate 996666Blue-yellow—
Test plate 101097nothingnothing97Red-green screening64%
Test plate 1111422442Protan / deutan71%
Test plate 121216nothingnothing16Red-green screening58%
Test plate 131373737373Control (everyone)—
Test plate 1414585858nothingBlue-yellow91%
Test plate 15152nothingnothing2Red-green screening53%
Test plate 161635nothingnothing35Protan / deutan58%
Test plate 171761nothingnothing61Red-green screening49%
Test plate 1818999nothingBlue-yellow75%
Test plate 191984nothingnothing84Red-green screening46%
Test plate 2020474747nothingBlue-yellow53%

How to read the answer table

  • Typical vision: the number most people read.
  • Red-weak and green-weak: what someone with a strong protan or deutan deficiency (dichromacy) is expected to read.
  • Fades at: the simulated severity at which the figure becomes too faint to see. A plate that fades at 50% is missed even by people with a moderate deficiency.
  • Two-digit classification plates show one digit to red-weak vision and the other to green-weak vision. That is how the test tells protan from deutan.

How the Ishihara test is scored

The printed Ishihara test uses a simple count of correctly read plates. These are the thresholds given in the instructions that come with the test.

EditionPlates countedNormalRed-green deficiency
38 plates1 to 2117 or more correct13 or fewer correct
24 plates1 to 1513 or more correct9 or fewer correct

What your score means

  • Normal: you read nearly all plates. One or two misses on faint plates is common.
  • Borderline: a score between the two thresholds (14 to 16 of 21 in the full edition) is rare and needs a retest or another test.
  • Red-green deficiency: you missed many plates. Classification plates then suggest protan or deutan; a clinic can confirm the type and severity.

How Ishihara results are written down

Clinicians usually record the edition and the score, for example "Ishihara 38: 21/21" or "Ishihara 24: 8/15, deutan". The plates read the colour-deficient way are often listed too, and whether both eyes or each eye was tested.

What to do after a failed screening

Book an eye examination. An optometrist can confirm the type, measure severity with tests such as the Farnsworth D-15 or an anomaloscope, and check for any eye condition if the loss is new or affects one eye.

Scoring questions

What is the passing score for the Ishihara test?

For the 38-plate edition, 17 or more of the first 21 plates read correctly. For the 24-plate edition, 13 or more of the first 15. Some employers set their own, stricter rule, such as no errors.

Which Ishihara score indicates a colour vision defect?

13 or fewer correct of the first 21 plates in the 38-plate edition, or 9 or fewer of 15 in the 24-plate edition.

Why can't I find the real Ishihara answers in a PDF here?

The Ishihara plates and manual are copyrighted. We publish answers only for our own original plates.

I saw a different number from the answer. Am I colour blind?

Not from one plate. Look at the pattern across all plates, check your screen settings, and retake the test. Several misses on red-green plates is what matters.

Haven't taken the test yet?

Take the colour blind test