WHAT...is your favorite color?

May 17 2010 Published by under Behavioral Neuro, Evolution

ResearchBlogging.org Palmer and Schloss. "An ecological valence theory of human color preference" PNAS, 2010.
Sci will admit that she didn't really know all that much about color preference theory until she read this paper. And that until she read this paper...she thought a lot of it was silly.
Also, she doesn't have a favorite color. That might have something to do with it. Can someone have a favorite color palette instead?
Anyway, let's talk color preference theory.

One thing Sci really liked about this paper was the way the authors laid out the background in the discussion. Concise, well worded. No space wasted. Hot.
Anyway, the two big theories an color preference right now are the biological adaptation theory, and the emotion theory. And the big question they address is not "WHAT" is your favorite color, but WHY.
Basically, when you ask men and women for their color preferences, you get certain similarities and differences. Here's what we're going to be talking about:

This is a common color block used in some types of color preference testing. Here the S quadrant is saturated, the L quadrant is light, the D quadrant is dark, and the M quadrant is muted. You will notice that the same colors are represented in each quadrant in different ways. As for the colors themselves, the four corners are red, yellow, green, and blue, and the four sides are each of their color bisectors (the color in between the two corners) which are orange, chartreuse, cyan, and purple.
So, when you ask people which of these colors they prefer, both men and women prefer colors closer to blue and violet than those close to yellow and green. However, women tend to prefer colors closer to reds than men, who prefer colors closer to blues. This difference has occupied the minds of many scientists for a while, and several theories have come out.
1) The signal theory: this is the idea that certain colors portray certain signals (like yellow meaning a wasp, while red means a ripe berry). Colors that are attractive to us are those which signal "approach" while ones we don't like signal "avoidance". This theory has been losing ground, because it doesn't tend to match too much with data on color preference.
2) The biological adaptation theory: this is the idea that women needed to like red more because they were doing all the gathering early in our society and had to be able to pick out red berries very readily, and therefore they prefer red. Sci will go into this a little more later.
3) The emotional theory: this is the idea that certain colors evoke certain emotions, like activity, passivity, heaviness, or warmth. However, no one really has a theory on how certain emotions would arise from certain colors.
And 4) The theory proposed in this paper, the ecological theory. The authors here propose that humans prefer colors like blues and greens because those colors and ecologically healthy (blue skies, clean water, healthy vegetation), and do not prefer colors like brown because it's associated with stuff that is ecologically unhealthy (like crap and things that are rotting). This is somewhat like the biological theory in that they are both adaptive. Red berries are more healthy than green ones (probably), and clean water is more healthy than not, thus relating color preference to survival. It relates slightly to the emotional theory in that certain colors "look good" while others "look bad".
The authors decided to test their theory, and got a bunch of people together to look at colors. They ended up with the same kinds of colors preference that the previous studies had shown. They ALSO had the participants look at the names of objects (each of which was associated with a particular color) and rate them positively or negatively. Not only that, they showed the participants a color, and then had them write down all the things that they could associate with that color (like "apples", "feces", or "grapes"). They then tallied this all together, and looked at the color preferences, as well as what people associated the colors they looked at with.
They got a good correlation between favored colors and favored things. So for instance, red with cherries or apples, brown with feces. And their theory appears (they say) to fit the data better than other theories to date.
Why is this important, you may ask. Well it's very important to people who do marketing. Observe the stark color differences between magazines targeted toward men:

(The first thing Sci noted about this picture is that if they'd put the Carolina guy next to the Duke guy, it would have looked like the Carolina guy was smacking the Duke guy in the face. Sci can't help but think that this would have (a) generated a ton of press and (b) been really hilarious)
And women:

(Taylor Swift, honey, let's talk about your eyebrows. Those things look like they could boomerang a kangaroo at 200 paces. You're too young for such things, and I bet your natural eyebrows are just fine. You should try them sometime. Love, Sci.)
Note the blues in the Sports Illustrated, vs the pinks in the Self magazine (the uniforms on the players are probably not coincidental either). Knowing color preferences on average for women and men can go a good way to selling magazines, and coloring, and heck, maybe even food (packaging on food marketed to women tends to be a lot redder and pinker than packaging on food marketed to men).
Sci isn't really sure what to think of this article. Their interpretation does appear to fit the data, but as the authors themselves point out, there are a LOT of caveats. All the people in the study were American, so this isn't humans preferences, it's American preferences. Sci really wonders if people across cultures and around the world show similar color preference. In addition (and particularly important to Sci) they don't know what role socialization plays in color preference. Sci thinks it probably plays a very big role indeed, and seriously complicates the issue. After all, many women are raised from the time they are very small on reds, pinks, and purples, while boys are raised on blues. This is going to have a pretty strong effect on your color preferences.
Not only that, there's the association between colors and things. Do you like the color because it evokes the thing you like (like cherries) or do you like the thing because it evokes the color? If you just like cherries, are you more likely to like red? If colors always indicated the value of a thing, things like chocolate and coffee, which are similar to things like feces in color, wouldn't come out well at all.
But still, it's not like the other theories are much better. After all, the idea of some colors signaling approach and others signaling avoidance doesn't really work very well. Many colors which signal things that are good to eat (bananas) also signal nastiness (wasps), and vice versa. The biological adaptation theory may explain a little, but early in evolution men and women both did plenty of gathering. In addition, women's preference for red may be very strongly socialized. As for the emotional theory, Sci personally thinks this theory sounds like hooey, but she'll also admit she doesn't know a lot about it. But really, the emotional aspect of a color would, in my mind, stem from something else which signals its salience. I feel like the emotional aspect doesn't scratch the surface. So of all these, the ecological theory seems like it might be a better contender, with colors we like indicating good things in the environment. But still, it doesn't cover everything. Blue can mean good weather and good water, and green can mean lush vegetation, but green can also mean rot. Brown can mean decay, but it can also mean fertile soil, or well done meat. Red can mean ripe berries or health, but it can also (on humans) mean illness.
So the reality, Sci thinks, is probably a lot more complicated than any of these theories suggest.
Palmer, S., & Schloss, K. (2010). An ecological valence theory of human color preference Proceedings of the National Academy of Sciences, 107 (19), 8877-8882 DOI: 10.1073/pnas.0906172107

28 responses so far

  • Mr Z says:

    I don't think that the human mind sees color devoid of context. That is to say that green vegetation is in a context of lush, healthy plant life. Green on a piece of meat is in a context of wrong, unexpected, badness. In the example with the magazines, both have context in which the colors should be interpreted. The mindset of sports fans wanting information and young women wanting information about women is fairly different contexts.
    I think it will make much more sense when given color perception is taken in context.

  • Fargo says:

    I have to say that none of the colors presented in those blocks are very appealing to me. S is like the best of a bad situation, with L being best runner up. Seriously now, what kind of ass shade of green is that?
    My favorite color has always been deep, dark greens. Like the way emeralds should look but never do. Blues are nice and all, but they run a very distant second for me. Really blacks and maroons typically attract me more than blue.
    I think it'd be interesting to put the colors in more of a pattern context, and also test various age groups. We already know the younger peoples see more blue, since the sun hasn't yellowed the cornea yet, so would that impact color choice?
    While it seems very likely that a wide swath of color perception has been beneficial to us, since it's a very well preserved trait, I think it's also important to consider that any color preference is going to be solely due to our free-associating hunk of grey matter and the actual mechanical bits are something like better depth perception or detail discernment.
    I'm going to stop rambling now.

  • Mary H says:

    Green is it for me. A spring baby, I find the re-leafing of the trees, the grass growing, the beauty of all in bloom is the most lovely color of all. I also like yellow and orange, and green-blues. The colors I like least? I HATE pink. I despise pink. I think it's the ugliest color in the whole spectrum. I suppose I can attribute my dislike of pink to the very thing it stands for most: girly. I grew up without much interest in gender roles, believing I could do anything and be anything I wanted without resort to my gender.
    BTW, a couple of years ago, when Cory Aquino passed away, I was reminded that she often dressed in yellow suits--a color which is supposed to stand for power. Perhaps it's true in the sense that the color makes one stand out--you can never be considered a shy person if your thinking is bold enough to accept it.
    The other colors I dislike are brown and baby blue. With respect to the brown, I don't necessarily think of feces, but simple mud. It signifies to me that a person who likes brown has "muddled" thinking, and none of the decisiveness that clearer thinking can evoke. The baby blue is because it's been done to death. It's also such a mediocre color in general.

  • The "pink for girls" thing has only applied in the last 50-80 years or so, and in western culture. Before that in the west, it was blue for girls, like Alice in Wonderland. So it's pretty unlikely to be a human universal.
    Also I agree with Fargo that the colours are crappy, except that I like D best of the bad bunch. And my favourite colour is violet.

  • snarkyxanf says:

    Doesn't it seem like the distribution of preferences is important, not just the averages?
    Maybe women on average are a bit red-shifted because there is a handful of pink fanatics. Maybe the male aversion to reds lines up closely with male-pattern red-green color blindness/impairment. Is color preference stable over time, either by cohort or individual. If not, is there any reason to believe that observed gender differences have biological bases?
    Has this been reproduced in other cultures? What do people from countries with strong color associations think? Do the Irish like green more (and the Northern Irish like orange)? What about people from cultures that have different color-word schemes (e.g. Chinese uses a term meaning "bluegreen", while Hungarian has two different words for what we would call "red")?
    In short, is there any reason to even assume there is a biological phenomenon to explain?

  • Black. Blue. I don't know...

  • John says:

    In the UK we have weets called Smarties which are chocolate covered in different coloured candy. Only the orange candy is flavoured, the rest all taste exactly the same. But, from memory, most children prefer the black or purple smarties. That clearly seems to indicate that colour is driving a perceived difference in taste.

  • John says:

    Sorry, make that "sweets", not "weets".

  • ted says:

    Very nice article, thanks.
    Very annoying that you speak at the third person, though.

  • CRM-114 says:

    My favorite color is clear. As for spectral colors, no preferences. I find saturation to be irritating.

  • Shadow Of A Doubt says:

    No, I love the way she uses the third person!
    And I too wondered if red-green colour blindness had been accounted for.
    However, if you look at countries' flags you see a definite preference for red and blue.

  • Stephanie Z says:

    Shadow Of A Doubt, my understanding is that red and blue are the colors of power because they were the colors of the expensive dyes that commoners couldn't afford.

  • hectocotyli says:

    Example of speaking *in* the third person: hectocotyli is very annoyed that you leave drive-by grammatical style criticism on Scicurious' hilarious blog.
    Example of speaking *at* the third person: hectocotyli invites the commenter named ted to bite me.

  • Scicurious says:

    Sci would assume that any type of color blindness would have been controlled for. Would have been very silly to do it any other way. 🙂
    Also, none of the colors pictured are really SUPPOSED to be your favorite colors, they are examples of common, easily labeled colors with saturation and light dark differences. You then rank them in order of preference. You'll notice that crap brown isn't on there, neither is black or white. It's a measure of relative color preference rather than absolute.
    Ted: Sci thinks your concern trolling is annoying. One of the best things about the internet is that no one is tying you to a chair, duct-taping your eyes open, and forcing you to read Neurotopia. You'll get over it in a little.

  • Stephanie Z, there's a long fascinating history of the economics of colour and how they affected paintings.
    One of my favourite anecdotes, is that for much of the Renaissance, true Ultramarine pigment was made from crushed lapsi lazuli. A patron would give an artist money to purchase a quantity of ultramarine from the apothecary, usually for a Biblical scene.
    The expensive colour would be used on the most important figure in the scene. But typically, Christ was portrayed as either a baby or semi-nude on the cross. So the blue would go to the Virgin Mary. A lot of the symbology of the colour can be explained that way.
    Purples were hard to produce too, especially as dyes. A mollusc from one part of the Mediterranean could produce purple dye. It's expense was a reason it was reserved for royalty: the fact it's a complimentary colour to gold is a bonus.

  • Dr Becca says:

    I agree with Dr Z that context is important. I wonder if anyone's tried priming experiments to test this? For example, show some people "good" blue things like a clear sky, the ocean, blueberries, etc, and some people "bad" blue things like mold, bruises, etc--would you then see a shift in color preference in a palette test like the one here?

  • I recently saw my cousin's 12yo daughter at a barbecue lunch. She talked to me about her favourite colours, one of which was maroon. I told her that I remember the day she first learned what maroon was: years ago, playing a travelling game where one player nominates a colour and everyone else tries to spot a car of that colour. (I picked maroon because I have that sort of devious mind, and because I knew there was a chance it would be educational.) She'd forgotten this entirely, and asked me how old she was at the time.
    As for me, anyone who's visited my blog and seen the domestic photographs I upload from time to time knows that I like all sorts of colours, but rather than choosing between them I prefer to use them all...

  • AmoebaMike says:

    I wonder what role sex plays in this? It seems like women (at least white women) tend to wear red lipstick and powder on their cheeks. Didn't that have something to do with imitating being "flush" with hormones and being a good mate? But why then do men prefer blue to red? Is it different with women of color? What about baboons of multi-colored butt fame? Sci curious, indeed!

  • Stephanie Z says:

    Heh. Glendon, I now want to know how the relative expense of dyes versus pigments played out as important people were clothed, then painted, then clothed, etc. Seeing as pigment dyes are really technologically recent, I'm guessing dyes had priority, but it's entirely a guess.

  • McJohnny says:

    Liturgically, red is the color of Christ (think of the blood-red glass candle globes you see in church) and blue is the color for Mary (Marian Blue is a popular color in vestments and she is usually seen in blue garb).

  • Waltz says:

    Such theory would explain a global picture of color preference in humans but fail miserably in describing intersubjects' preferences.

  • aidel says:

    Interesting discussion. I think the nature of color preference *has* to be a social phenomenon, based largely on the group(s) with which you identify.

  • I think that colors have an emotional impact and it is very important the contrast and the context in which they are presented... here I agree with Mr Z

  • Liisa says:

    If I may add my two cents, I fail to see a control for cultural perception. This 'pink is for girls, blue is for boys' thing.
    I would need to look up the exact source, most likely it was some fashion mag or some Handbook of Good Housewife thingy, but until sometime like 1920's, the idea was the exact opposite and pale blue was perceived as good colour for girls because it was sweet and unmanly, and vice versa for pink.
    The others' ideas about expensive dyes are pretty interesting, too, although I strongly doubt that the Medieval usage would be relevant to modern man (unless they're historians or some such).
    One thing has caught my attention: yellow seems to be hardly worn, at least in the parts of Europe where I live. Unhealthy colour of dandelions, I wonder?

  • Andreas Johansson says:

    Has their been any studies on how colour preferences change over an individual's life?
    Looking at myself, as a little kid I had a very strong preference for red (to the point of rejecting blue toys), but over time I've my preferences have moderated and migrated, somewhat depressingly, to stereotypically manly blues (little red in my wardrobe these days - lots of blue and black).

  • cyberroth says:

    A 2007 study found similar gender differences in color preference. The researcher speculated that "Evolution may also drive females to prefer redder faces (male faces are ruddier than female faces across all races). A reddish face often [means] good health; this may be a good cue to help the females to select mates."
    That study also found a cultural effect on color preferences: its Chinese men liked red more than its British men, probably because red stands for luck and happiness in that culture.
    Study reported here: http://www.scientificamerican.com/article.cfm?id=women-prefer-pink

  • Stanton says:

    Hahaha that's utter crap! I prefer yellow and green colors and all related to them. I don't really like red (except for some cars) and I am indifferent to most shades of blue.

  • Stanton says:

    Oh, and talking about men and women separately is called sexism.