Published Method

THE SCIENCE
BEHIND THE
TEST

Most online vision tests reuse the same scanned plates and never say how they work. OpticQuiz does the opposite: the method behind our color vision test is published open-access, with a DOI, and the source code is open — including an honest account of exactly what a screen can and cannot measure.

Open-access preprint
A Procedural Method for Generating Pseudoisochromatic Plates in the Browser, and the Calibration Limits of Screen-Based Colour-Vision Screening
Gonzalez, V. (2026). https://doi.org/10.5281/zenodo.21310578 · CC BY 4.0 · ORCID 0009-0005-3640-014X
Companion source code
OpticQuiz PPPG — Procedural Pseudoisochromatic Plate Generator
Software deposit: https://doi.org/10.5281/zenodo.21311369 · repository: github.com/zengineco/opticquiz.com

What the method actually is

Our color plates are not stored images. Each one is generated procedurally in your browser the moment you take the test: dots are packed live, the hidden figure is separated from the background only along a red-green or blue-yellow color-confusion axis, and every dot's brightness is randomized so that no lightness cue leaks the answer. No two runs are identical, so a plate can't be memorized or traced.

The paper specifies the full algorithm — figure masking, dot packing, palette construction, and scoring — reproducibly, and then does the thing most tools avoid: it states the limits. On an uncalibrated, three-primary screen, and under observer metamerism, a color test is an honest screen, not a measurement instrument. It can suggest a likely red-green deficiency; it cannot classify the type or grade severity. That boundary is the point.

Procedural generationConfusion-line colorsRGB / calibration limitsObserver metamerismScreening ≠ diagnosis

Why this matters for an online vision test

Publishing the method is a claim anyone can check. It is the difference between "trust our score" and "here is exactly how the score is produced, here is where it breaks, and here is the code." For a free online vision test, that verifiability — not a bigger marketing claim — is what makes a result worth trusting.

Explore the concepts in the paper

Each idea the method builds on has a hands-on test or a plain-English guide here:

The color vision test the method powers Take the procedurally generated screen, per eye. Color test for kids (shapes, not numbers) The same method with shape figures for pre-readers. How Ishihara plates work The pseudoisochromatic principle the method implements. How accurate are online color blindness tests? The calibration problem, in plain English. Which color vision test is most accurate? Ishihara vs D-15 vs anomaloscope vs online. The anomaloscope test The clinical gold standard the paper cites for red-green. The Farnsworth D-15 test The arrangement test referenced for severity and axis. Color blindness, explained Types, confusion lines, and prevalence.
Take the color vision test → ← All free vision tests