Color consistency check

How to use it

1. Choose what to check: letters and numbers, only letters, only numbers, days, months, or a shorter check of 12 letters and numbers picked at random. The page tells you how many trials that is and gives a rough time. A shorter check has fewer items, so its numbers are less reliable.
2. Press Start the check. You see one item at a time on a plain card. Choose its color from the named colors, mix your own with the red, green and blue sliders, or type a hex color. Then press Next.
3. If no color comes to mind, press Skip this one. Nothing is recorded for that trial. If you press Next or Skip by mistake, press Undo my last answer.
4. Each item comes up 3 times, mixed in with the others, and no item comes up twice in a row. You will not see your earlier colors while you go. There is no timer.
5. Press Pause and finish later at any time. Your colors so far are saved in this browser. Come back to this page and press Resume. Start over deletes the check in progress, after a second tap.
6. At the end you see each item with its three colors, their names and hex codes, an item number, and an overall number. Sort the table, copy the results as text or download them as a CSV file. Past checks are listed under Past results.

Good to know

• This is not a diagnosis and it is not a test result. It does not show whether someone has synesthesia, and it gives no pass or fail. It shows your own numbers and says plainly how they are worked out.
• How the numbers are worked out: each color is a point described by red, green and blue, each from 0 to 255. The distance between two colors is the straight-line distance between those points. An item number is the average of the 3 distances between an item's 3 colors. The overall number is the average of the item numbers. 0 means the same color every time. The largest possible distance, between black and white, is 441.7. Numbers are shown with 1 decimal.
• Researchers use computer tests of this kind. A 2023 article in Behavior Research Methods says that in the test battery of Eagleman et al. (2007) each grapheme is repeated three times in random order, as each item is here. It says their score is based on "taxicab distances" in RGB color space, and that they proposed a cut-off score but stress that it is "merely an optimal dividing line between two populations whose scores vary along a distribution". This page is not that test. It does not use that battery's software, data or scoring, its distance is a different one, and it has no cut-off. Its numbers cannot be compared with scores from that battery or with anyone else's numbers.
• Skipped trials are left out. An item is counted only if all 3 of its colors were chosen, so every item number is worked out the same way. The page shows how many trials were skipped and how many items were left out.
• The Synesthesia guide on this site gives the sources for these points: young children's colors are still settling (Simner and Bain, 2013), a research test called the Test of Genuineness-Revised proved unsuitable for synesthetes with autism (Baron-Cohen et al., 2013), and Simner (2012) presented evidence that might argue against the claim that synesthesia gives rise to consistent associations over time.
• Color names are this tool's own plain names. Each one is the nearest color in a list of about 150 names, so a name is approximate. The hex code next to it is exact. Every color in the table is shown with its name and its hex code as text, so nothing depends on seeing the color.
• Everything is saved in this browser on this device and nothing is sent anywhere. Clearing your browser data erases it. Up to 30 finished checks are kept, and when there are more, the oldest is removed. A check you have not finished is kept until you finish it or start over.

Where this comes from

• A 2023 article in Behavior Research Methods describes the test battery of Eagleman et al. (2007) and says "Each grapheme is randomly repeated three times throughout the test.", says that Eagleman et al. suggest a score "based on response colors' taxicab distances", and says that they proposed a cut-off but stress that it is "merely an optimal dividing line between two populations whose scores vary along a distribution.": Wilsson, van Leeuwen and Neufeld (2023), synr: An R package for handling synesthesia consistency test data, Behavior Research Methods
• The Synesthesia guide
• My synesthesia map

More tools

General support only. These tools are not medical, legal or diagnostic advice, and they do not replace professional help. If you are in crisis, go to Crisis help.