Test Blindness Color

June 28, 2006 8:14 PM

Reverse Color Blindness Test "normal vision humans have a lower degree of color contrast detection in the red spectrum. A colorblind person shouldn't be burdened by that lowered contrast sensitivity and should be able to see the object immediately by picking out the change in contrast at the objects edges" A small oddity that takes but a few seconds of time. (via The Presurfer)
posted by caddis (52 comments total)

This is interesting, but I'm not sure I buy it entirely. I saw both images without much of a problem, but I have no reason to believe I have any degree of colorblindness.
posted by feathermeat at 8:27 PM on June 28, 2006

Now, my girlfriend and I could both clearly see the embedded object in the first image, so I wonder how accurate a test that actually is. I couldn't see the embed in the second until I knew what it was, however.
posted by Dipsomaniac at 8:27 PM on June 28, 2006

Saw them. Not colorblind.
posted by _sirmissalot_ at 8:27 PM on June 28, 2006

I could see the first one, but I'm not colorblind. Probably just some sort of superhero.
posted by amro at 8:27 PM on June 28, 2006

uhhhhhh.
You people could actually see them?
Are your monitors set to weird settings? (Is mine? Hrm.)
I couldn't see them at all.
posted by blacklite at 8:31 PM on June 28, 2006

I could see part of the first one, but the second one pretty much eluded me. I do have a slight colorblindness, although it thankfully hasn't led to me wanting to wear bright orange and purple together, yet (although strange combinations of orange and black are somehow an annual addiction come June).
posted by caddis at 8:32 PM on June 28, 2006

Aha, when I had brightness up to 100 (out of 100) I could see the first one, barely.
posted by blacklite at 8:33 PM on June 28, 2006

I think this probably has a lot to do with your monitor.
posted by borkingchikapa at 8:33 PM on June 28, 2006

, yet?

I need a personal copyeditor.
posted by caddis at 8:34 PM on June 28, 2006

I thought I could just barely detect a round shape in the first one. Turned out to be slightly different than what I thought. Saw nothing in the second.
posted by ijoshua at 8:34 PM on June 28, 2006

I could see neither at all. How clearly could you see them?
posted by null terminated at 8:35 PM on June 28, 2006

I could see both no problem... and I have no degree of color-blindness...

flawed test
posted by WhipSmart at 8:35 PM on June 28, 2006

I turned my brightness settings up and could see both images.
posted by null terminated at 8:42 PM on June 28, 2006

I saw the first image with no difficulties. For the life of me I could not see the second image even after referring to the answer several times. I am colorblind although not excessively so. I agree that this test is flawed
posted by ob at 8:43 PM on June 28, 2006

I could see the 'hidden' components of both of 'em, because I'm more sensitive to color variation than average. The test seems fun, but not meaningful.

This is a good read about the most common types of colorblindness and their causes -- including examples of standard tests and how to read them.
posted by whitelight at 8:47 PM on June 28, 2006

I could see them more clearly by looking at my LCD monitor from an angle. Printed out, I probably wouldn't be able to see them.
posted by alidarbac at 8:55 PM on June 28, 2006

Now, of those who saw them and those who didn't, who's using LCD monitors?
posted by Dipsomaniac at 8:56 PM on June 28, 2006

I saw both with no problem. I know I'm a little colorblind. When I was younger (seven or eight) and first told I was colorblind my eye doctor told me a story that the military used colorblind people to detect camouflage.
posted by telseth at 8:56 PM on June 28, 2006

I could see the bottom of the first image. The second image was completely invisible to me. I am not colorblind, although it does run in the family.
posted by Malor at 8:57 PM on June 28, 2006

Your monitor definitely makes a difference, I have dual monitors and I could see the images much more clearly on one than the other. Surprisingly they were much easier to pick out on my 5+ year old CRT than on my brand new LCD.
posted by chicken nuglet at 8:59 PM on June 28, 2006

If you can easily see the images and you aren't colorblind, your monitor is likely not calibrated properly.
posted by Justinian at 9:06 PM on June 28, 2006

What is in the second one? I barely see a red circle in the bottom half of the first one, but the second one is just clouds (iMac 20" LCD, calibrated, not color blind)
posted by mathowie at 9:06 PM on June 28, 2006

Try changing the angle at which you view the image (up/down) if you're using an LCD.
posted by Steven C. Den Beste at 9:12 PM on June 28, 2006

I could see it in photoshop, the first one is a 6. The second one is a circle. Just up the contrast on the second, in the first you'll need to tweak the curves.
posted by Paris Hilton at 9:12 PM on June 28, 2006

If you can easily see the images and you aren't colorblind, your monitor is likely not calibrated properly.

That's probably what it is. If I ajust the curves in photoshop to 'undo' the calibration of my monitor, I see the shapes.
posted by Paris Hilton at 9:15 PM on June 28, 2006

I can see them both, and I am deaf.
posted by brain_drain at 9:48 PM on June 28, 2006

You don't have to photoshop it all, if you scroll down to the bottom of the page there's a link to the 'answers' (adjusted images).
posted by jacalata at 9:53 PM on June 28, 2006

Saw them once I switched my monitor to "graphics/video mode" which ups the gamma I reckon.
posted by Matt Oneiros at 10:02 PM on June 28, 2006

On my PowerBook G4 I tried that trick I learned on AskMe the other day to switch to grayscale (ctrl-option-'apple'-8) the first image showed up perfectly, the second was not visible to me at all.
posted by jamjam at 10:06 PM on June 28, 2006

Not color blind, couldn't see a thing, even after I peeked at the answers, hideously expensive LCD, which I hope is properly calibrated.
posted by unrepentanthippie at 10:10 PM on June 28, 2006

Intriguingly I could see the second one perfectly well but couldn't see the first at all.
posted by Veritron at 10:26 PM on June 28, 2006

Uh... I am extremely colorblind and couldn't see a thing.
posted by ORthey at 10:27 PM on June 28, 2006

Oddly enough, I was just theorizing the possibility of a reverse-colorblind test to my wife a few months ago. I always wanted to send one of my color-blind friends a birthday card that only he could read. Maybe someone will make a generator someday.

I could barely make out two of the outer curves of the first image (but thought it was a circle), and nothing of the second even after I knew what to look for.

After upping the brightness on my monitor I could barely make out the first image. I have color calibrated my monitor in the past.
posted by BrotherCaine at 10:37 PM on June 28, 2006

Too bad that site didn't include a chart to adjust your monitor, or a link to one.

no suspicion of colour blindness - could see the 1st easily, the 2nd barely - NEC LCD adjusted (for my eyes) against a few different tests - although I prefer my brights bright and darks dark so that's probably the reason
posted by porpoise at 10:37 PM on June 28, 2006

I cheated and tilted the LCD to find the first one (indpendently discovering Steven CDB's trick.) Even with the tilting, I could only find the second one after seeing what it was supposed to look like. So I guess I'm perfectly normal.

Man, that hurts.
posted by Opposite George at 11:06 PM on June 28, 2006

hmm I could sort of make out the first one but I wasn't sure if it was a 6 or B.

The second one I couldn't really see at all. Strangeness!
posted by Deep Dish at 11:08 PM on June 28, 2006

BrotherCaine , that birthday card is a gret idea. I do hope that someone makes a generator. :)
posted by dabitch at 11:19 PM on June 28, 2006

Another data point: not colour-blind as far as I know, could see both from some angles on my LCD monitor.
posted by TheophileEscargot at 11:27 PM on June 28, 2006

I saw an image, but then got scared when i read the first sentence here:
": If it's painfully obvious that there's another image there, then you're
: probably colorblind to some degree in the red part of the spectrum. Can't
: see it? Try looking at the white space at either side of the image, you might
: be able to see the object by using your contrast-sensitive rods (which are
: concentrated more heavily in your peripheral vision). Don't give up if you
: can't see it, that's the whole point - this is an *anti* colorblind test."

I've decided it's an editing error on their part... i hope
posted by wumpus at 11:51 PM on June 28, 2006

Could see both images with difficulty- no colourblindness, LCD monitor, image adjusted against several test screens a short while back.
posted by YAMWAK at 11:57 PM on June 28, 2006

From the answers page:

Humans, however, have the luxury of being able to learn to intellectually compensate for their vision, so someone who is colorblind may be able to guess from the context if something non-colorblind people see as red, is actually red.

I think this hypothesis is underplayed. Instead of saying that the colorblind "intellectually" compensate for their vision, and "guess" that the object is red, I suspect it is more correct to say that the colorblind subconsciously compensate for their vision and perceive the object as red.

The standard story about the conscious experience of the colorblind goes like this: we have three different types of photopigment in the cones in our retinas, and their outputs allows us to compute the three axes of our phenomenal color space (the axes are blue-yellow, green-red, and black-white). Dichromats, lacking one type of cone, can't process the difference between red and green. So they have only a two-dimensional color space.

However, dichromats can make perceptual distinctions between red things and green things -- as this experiment tries to show -- by subconsciously using luminance information. Red things tend to be darker than green things. This is why black and white photographs "work". We evolved trichromacy extremely recently, and it probably evolved directly out of our abilities to tell differences in luminance. Our primate ancestors needed to see fruit amidst leaves, and could do so using luminance cues. It behooved evolution to find a more efficient way to make those same discriminations. (Alternately, some think that trichromacy evolved to let us distinguish young nutricious red leaves from older green ones.) Thus, mutations in the genome led to a new type of photoreceptor in the eye.

What's the reason for thinking that the trichromatic apparatus in the retina (the third cone) isn't simply a more efficient way of making certain perceptual judgments and having certain conscious experiences? It would be weird if such a low-level entry-point into the brain were responsible for a totally different plenum of color experience.

It's interesting to hear colorblind people talk about their conscious experience. Contradicting the standard story, many say that they do see red and green as qualitatively different colors -- they just often confuse them for one another. This is what would be expected if all they lack is an efficient and reliable mechanism for doing the exact same job. Roger Shepard has some neat experiments on this score. Also, many colorblind people don't realize that they're colorblind until a very late age. Wouldn't it be weird if they saw red and green as the same but never realized that they did?

If anyone's interested, Kathleen Akins and Kimberly Jameson are two fascinating researchers in this field. I found this paper of Jameson's, in particular, to be extremely thought-provoking.
posted by painquale at 12:09 AM on June 29, 2006

And another data point. Not colour-blind. Could dimly see the top image on both my home and work LCD monitors, strongly from some angle. The bottom image though was completely invisible from any angle on my work monitor, though visible at home.

I think this test depends a lot on what monitor you have.
posted by TheophileEscargot at 1:17 AM on June 29, 2006

Not color blind, dimm LCD monitor, could see the first one very clearly. The second one was much harder to see.
posted by vertriebskonzept at 2:36 AM on June 29, 2006

I barely saw the second, saw nothing on the first. But after I saved them to my desktop, the icon previews showed the images pretty clearly.
posted by Mcable at 4:21 AM on June 29, 2006

A better test. Rather than trying to determine what kind of colorblindness you might have (which is difficult to impossible with non calibrated monitors,) this one merely tries to show if you have a color deficiency.

The test, here, is simple. Watch the movie, follow the colored square. If you see the square through the whole movie, you have close to normal color vision. If it disappears, you have some sort of colorblindness.
posted by eriko at 4:50 AM on June 29, 2006

I can no longer watch videos like this on the internet without expecting a horrible, screaming face to appear.
posted by Justinian at 5:11 AM on June 29, 2006

The industry-standard Ishihara charts for testing colour blindness also include 'anti' colour blindness tests, i.e. patterns that you can only see if you have anomalous colour vision. Ishihara developed them to screen military recruits. For this kind of screening, it's important to know whether the subject really has a visual abnormality, or is simply answering 'I can't see it' to every question. By including patterns that only the 'color-blind' can see, Ishihara was able to distinguish between people with anomalous vision and liars.

The image at the bottom of this page on the right is an example. It is seen as an 8 by people with normal colour vision, and as a 3 by people with anomalous red-green colour vision. [The section 'The different types of plates' describes 'transformation' and 'hidden digit' plates that are both designed to force a response other than 'Can't see it'.]

The Ishihara charts are fairly expensive to buy, because the printing is carefully calibrated. Given variability between monitors, there's no way the linked page provides a meaningful diagnosis of colour blindness. [I score very normally on the Ishihara test, but can see the 6 in the first image.] The City University test is much more convincing, because it tests many different parts of colour space, and controls for luminance variations that result from adjusting the 'brightness' and 'contrast' settings on your monitor.
posted by beniamino at 5:43 AM on June 29, 2006

Cool. I saw them, I'm colorblind. {And amro is a superhero!}
posted by OmieWise at 6:59 AM on June 29, 2006

I'm colorblind. I could see part of the image in the top example and all of the circle in the bottom. A coworker, who is not colorblind, was able to only see the figure in the top image. The bottom image contained no figure, and I was hard pressed to convince her that there was one.
posted by NationalKato at 7:06 AM on June 29, 2006

I saw both images, but I am not colourblind. Actually, I usually think of myself as somewhat sensitive to colour - I use it to see (since even with glasses things are slightly blurry more than a few feet away).

Tipping my LCD monitor made the first image clearer, but did not change the second at all. The second image I saw faintly, and thought "All I see is a faint thin circle, perhaps I'm missing something". But I wasn't.
posted by jb at 7:39 AM on June 29, 2006

painquale writes "It's interesting to hear colorblind people talk about their conscious experience. Contradicting the standard story, many say that they do see red and green as qualitatively different colors -- they just often confuse them for one another. This is what would be expected if all they lack is an efficient and reliable mechanism for doing the exact same job. Roger Shepard has some neat experiments on this score. Also, many colorblind people don't realize that they're colorblind until a very late age. Wouldn't it be weird if they saw red and green as the same but never realized that they did?"

This is me, I see colours different that most people with many distinct shades in some areas and no differentiation in others. I no longer have discussions with people about how good a certian colour looks or about most art for that matter because it is obvious that almost everybody sees colours different than I do. Though it took me a good ten years to figure this out. I used to constantly have discussions like:

GF: What should I wear?
Me: Put on your green dress.
GF: I don't have a green dress.
Me: Yes you do.
GF: No, I don't.
Me Yes you do. Remember, the one that buttons all the way up the front.
GF: That dress is brown.
Me: (Just now remembering this conversation from the last time) Whatever, put on your _brown_ dress.
GF: The colour is ugly.
Me: I think it looks good on you.
GF: You would.

Rinse, lather repeat on almost all conversations about colours.
posted by Mitheral at 8:24 AM on June 29, 2006

As a pretty severely colorblind person, I would have to say that the "guess" description is pretty accurate. Without someone else confirming a color, an ambiguously colored object can flip back and forth between red and green (or maroon and brown, or green and pink) until I am told what it is. At that point it pretty much settles down and stays one color. I also simply do not notice especially loud colors until pointed out, at which point I can recognize them as such.

FWIW, I saw all the images.

And I thought it was the conventional wisdom that CRT's tend to be better at displaying color range than LCD's.
posted by mzurer at 9:26 AM on June 29, 2006