Color Name Survey

Posted byRandall 369 Comments on Color Name Survey

I’d like your help for a color name survey!  The survey shows you colors, and you type a name (word or phrase) you might use for that color.  The names can be as broad or specific as you want.  The survey is here:

http://aram.xkcd.com/color/

My friend Finn (relsqui on #xkcd) wrote the frontend. I’m doing the analysis, though I won’t go too deeply into the details or purpose (to the extent that it has one) for now so as not to bias peoples’ answers.  Of course, RGB is a small and relative color space which varies depending on the device displaying it, so this survey has its limits, but it’s produced some cool data so far.

If any of  you want to help, you can fill out a few quick questions (don’t worry if you don’t know the answers—they’re highly optional and just help with calibration) and then come up with color names.  There’s no end to the survey; the more you answer, the better the data.  Thanks!

P.S. I’ve added a few much-requested prints and two new posters (Movie Narrative Charts and Gravity Wells) in the xkcd store.  The prints of Comic #150 (Grownups) are marked backordered right now while I replace a batch that was printed wrong, but they should be shipping again within a week.

369 replies on “Color Name Survey”

  1. There is a problem with the colour naming survey. I’m an undergraduate psychologist at Oxford University, so unless all my tutors and textbooks are lying to me, I think I’ve got a valid point.

    When you look at a particular pattern for a long time, your photoreceptors (the cones, in this case) become saturated. I won’t go into any more detail about it, but I’ll demonstrate the point with any afterimage illusion you might have come across. If you stare at an image for too long, and then look away, preferably onto white, you see the image’s complementary image in the white space. This also means that looking at one block of colour for a long time, and then immediately looking at a block of a different colour will alter your perception of the second colour.

    This is going to cause problems for the colour naming test – if the previous colour alters the perception of the next, then that’s obviously going to have implications for the naming of that colour. I assume that the results are going to be considered regardless of order, meaning that another variable will interfere with the results.

    Example: Staring at a red block for 30 second, and then looking at a light blue block will cause the light blue to appear to have a green tinge to it.

    The solution – simple. Insert a slight delay before the next block is presented. White light saturates all colour-photoreceptors equally, so should negate the effect.

    Like

  4. I tried the color experiment to take my mind off of boring reports and got to 130 when I realized that the colors were randomly repeating. I started out with simple descriptions ie:”Yellow Green” but towards the middle I was using “Light Chartreuse” and “Light Lilac Mauve”. Then towards the end, frustrated with the whole process, I was typing “IT’S FREAKING YELLOW GREEN!!!”, “GREEN!!!11ONE!!” and “ZOMG! VIOLET!!”. >.>

    I think it was a nasty experiment on human willingness to participate and their patience threshold. Probably one big bell curve.

    I see what you did there!!

    Like

  6. Is there a difference between salmon and peach? Well, of course: one is a fish, the other a fruit; as long as the colours go I can tell there is a difference, but I won’t be able to explain it nor name it.

    Like

  8. Oh, I agree strongly with Dakota. The raw data of this survey would be absolutely wonderful to see. *Pauses and realizes that I am a nerd, then shrugs* Also… I started the survey and I… I.. Cant stop!

    Like

  9. Man, I’d feel ridiculous if I didn’t know the difference between salmon and peach. Seriously, one’s a delicious flavor of ice cream and the other is a semi-edible juicy fruit.

    Like

  10. I wonder if people would give different answers for different shapes – for example, a t-shirt or car instead of just a square. Also, I have to agree that the background color makes a difference. I’ve seen quite a few black sqaures, but I haven’t seen anything yet that I would call white.

    Like

  11. It would be nice if there was a skip button. I’ve done a few colors, but I get a few colors which I just can’t classify properly. I got a shade of red and a shade of violet, but I wouldn’t have considered them as just “red” or “violet”, and I couldn’t think of any other name for them.

    Like

  14. I wrote a greasemonkey script to get the name of the closest color from the wikipedia list of colors based on the euclidean distance after transforming the RGB colors into L*A*B* space. Just in case you don’t want to answer one of the survey questions and instead see what wikipedia thinks : )

    http://userscripts.org/scripts/show/70375

    Like

  15. After the first 45 I got rather tired and started using names like “Bloose Change”, “Dark Green 2: Emerald Boogaloo” and “Medi-Ochre”.

    Like

  16. I did about 50. My color sense is decidedly skewed. My artist friends make fun of me all the time. I think I was most thrilled when I got something that was clearly “Gold” in color.

    Like

  17. I realized after answer a few that my laptop monitor’s color is *highly* dependent on the vertical viewing angle. Looking down at a sharp angle I would see a pale blue while the same color would look bright green when viewed looking up.

    Like

  18. Naming colours and doing so for 30 minutes absolutely is constructive to writing my skit for Italian class. It may very well be the most cohesive activity I’ve ran across yet, for sure. My professor will be thrilled if I were to divulge to her the creativity which blossomed during my naming spree. It’s obviously where the genius in the monologue came from.

    Like

  21. I got confused, because the perception of colours depends much on the context. I liked the instructions, as they allowed me not to worry that I called that light yellow just yellow, because I hadn’t seen this very bright one before.

    Like

  22. This was similar to an exercise in an undergrad anthropology class umpteen years ago: we were given a grid of colors, and we had to draw a line where “red” stopped and “orange” began, and so on. Even the males in the class were annoyed that “pink” wasn’t an option.

    Harder: drawing a line between where “white” stopped and “black” began, without an option of “gray.”

    Like

  23. I stopped trying at 105 and just started entering numbers after that and reached 300 before I quit…

    AAAAAAHHHHHHH!!!!!

    Like

  25. Just reread Kim Stanley Robinson’s Mars trilogy and this reminds me of the part where Maya and Sax play a game of naming the colours of the sky (when they’re old and retired and don’t have anything to do).

    Like

  26. You linked to relsqui’s website! Which links to The Forest.

    …where I’ve just spent the past few hours obsessively exploring and building. I’m up past my bedtime and it’s all your fault.

    Like

  30. I created colr.org for similar reasons (wanting to map words to colors). Let me know if you’re interested in using any of the data that colr.org has collected over the years.

    Like

  32. You should consider showing statistics every n colors about the colors they just named to keep people’s attention.

    Like

  34. I realized on about the sixth question that it may be a study in how long we would continue naming colors. If this was the purpose of the test, I suspect that the answer depends on the viewer’s feelings about the site. If there were repeats, this is a flaw in the test.

    I could be way off, but I doubt it has anything to do with naming colors. But then I’m a skeptic.

    Like

  36. As a color-weak person, let me just say that those colors are freaking hard to name. I didn’t think “I don’t know” would be too useful, but on the other hand, after a while, I just started to feel like, well, like part of my brain isn’t working like other people’s. Which, I gather, it’s not.

    Like

  37. You have the gall to ask people their monitor’s gamma and color temperature but you don’t bother to ask what type of colorblindness they may have?

    As a person (female, no less) with protanomaly, I can damn sure tell you that there is a difference between it and deuteranomaly, which is the more common form of color-blindness.

    I didn’t continue the survey after seeing that as I figured that all colorblind==yes answers will either be tossed out or be too wildly fluctuating to be useful.

    Like

  38. for gamma and temp maybe you should include sRGB which is what my monitor runs on

    Like

  39. I had a good laugh at this, as every single color presented was black. I figured the only data you were collecting was how many times people will type “black” before giving up (five, btw). Then I read the comments here, and realized that other people are seeing colors. I guess it’s just that my browser has been instructed to render pages in a way that’s easy on the eyes and not distracting to my autistic brain.

    Like

  40. I had to give up after about 12 for several reasons.

    1. I do light design for theatres, so my names for colors are way different than the norm. The very first one I saw (which, I assume they are random, so it will be different for others) I thought, Oh, that looks just like L181 – Congo Blue! which is actually much more of a purple than it is a blue. There was another that was predominantly red, but not true red which I would have pegged at probably R24 or R25. (L stands for Lee Filters. R==Rosco Gel).

    2. The white background is a killer. My rods and cones were going wild with all the white around the color sample. (Ironic, since the first thing I do with almost all sites that are black background and white/gray text is to press the zap colors bookmarklet.)

    3. Similar to The_Duck: Moving Horizontally, the colors were different from different angles.

    4. In the very few I did, I was getting multiple repeats, as if you were testing for optical illusion.

    4a. Which means, due to the white background and the color immediately before, it is hard to accurately name the colors, due to the previous input and the white background.

    It was fun, though!

    Like

  42. In the game of Color or Country, players take turns saying either a color or a country. If you say a country, you lose. Otherwise, you keep playing. Strategy is complicated by the existence of colors like peru*. I can’t help but think there’s some way these can be combined.

    * http://www.w3.org/TR/SVG/types.html#ColorKeywords

    Like

  43. My Theory: This is actually an experiment to find out how long it takes for the average person to go from being serious to typing in silly answers.

    I guess I’m sillier than average since I named the first color “Fred” and devolved from there.

    Like

  45. How DARE you have “the gall” to ask one question but not another? How dare you, sir?

    Like

  46. Did anyone else notice (or can they confirm) that there were no yellows or oranges??? What could this mean?

    Like

  50. I see a bit of a problem with this. After I stare at a color trying to think of a name for it, the color has been “burned into” my retina, which alters the perceived color of the next one.

    Like

Comments are closed.