Food color does more than guide us—it changes the experience of taste.

November 25, 2018 9:31 PM

The Colors We Eat: Tom Vanderbilt writes for Nautilus on the Official USDA Color Standards for a range of foods (including Tomato, Pumpkin/ Squash, Frozen French Fry, Frozen Cherries, Canned Tomato, Canned Ripe Olive (Munsell), Canned Apple Butter, Canned Lima Beans, Eggs, Canned Pimientos, Canned Clingstone Peaches (Visual Color Systems), and Veal (USDA text-only PDF)) and the scientific studies behind how we taste with our eyes. Beware: the eyes can deceive even the expert tongue. "Even the color of the plate might change our sense of taste: Subjects reported strawberry mousse tasted better on a white round plate than on a black square plate."
posted by filthy light thief (36 comments total)

This reminds me of the classic tests that have shown, again and again, that wine tasting is junk science. Expert judges regularly show bias in their tasting evaluations based on what the label looks like, what the cost is, and more directly relevant to this post, what color the wine is. (White wine dyed red with food coloring elicits the same evaluations as red wine.)

“Meanwhile the blind tasting contests go on. Robert Hodgson is determined to improve the quality of judging. He has developed a test that will determine whether a judge's assessment of a blind-tasted glass in a medal competition is better than chance. The research will be presented at a conference in Cape Town this year. But the early findings are not promising.

"So far I've yet to find someone who passes," he says.”
posted by darkstar at 9:46 PM on November 25, 2018 [6 favorites]

The eyes can deceive even the expert tongue. A study at Cornell University, for example, found that trained experts had trouble distinguishing the fat content of milk when they could not see it; the “whiteness” of milk was a far more important cue than whatever sensations came from actually having it in one’s mouth. “There is nowhere that is safe from expectation,” argues Spence.

Really?? I'm going to have to try this.
posted by polymodus at 10:47 PM on November 25, 2018 [1 favorite]

Some years ago I felt like I'd swear I could taste the difference between M&M colors—made a bit more obvious by the fact it was a Christmas mix that had only red, light green, and dark green colors. So I did a solo blind taste-test with just the red and dark green to make it even easier, and got 10/10.

So while I wouldn't doubt that color as appearance has a role in our experience of food, it can't be discounted that we might truly taste food coloring sometimes either.

Which is part of why I remain skeptical of the perpetual "wine tasting is bunk" studies, along with their tendency to play into the "take that so-called experts with your ivory towers and big words" mentality. I don't know wine myself, although I've had some I like and some I don't so clearly something about it is taste-able and judge-able. But I do know something about tea and if you dyed longjing blackish and called it puerh, I probably would take your word for it and not know to call your bluff, but I would also think that was an oddly young and grassy puerh.
posted by traveler_ at 10:55 PM on November 25, 2018 [10 favorites]

wine tasting is junk science.

I keep seeing this around and I keep feeling like these tests have to themselves be junk science when they take it so far as to claim red wine can't be distinguished from white. What are they dying it with? Tannin?
posted by RobotHero at 11:03 PM on November 25, 2018 [4 favorites]

Presumably, the researchers at the University of Bordeaux know something about wine, and how to run a research test. But I haven’t read the original paper, which is here, so I can’t confirm that it’s sound.

In the abstract, though, they suggest that the visual perception of the wine causes the subject to discount their olfactory perceptions, similar to the main thesis of the original post of this thread.
posted by darkstar at 11:40 PM on November 25, 2018 [5 favorites]

It’s a depressing idea that only one exact shade is correct for tomatoes/cherries/whatever, and anything else is inferior.
posted by Segundus at 12:10 AM on November 26, 2018 [1 favorite]

That Uni of Bordeaux paper does not show that people can't tell the difference between red wine and white wine dyed red. It shows that, on average, in a group of 54 undergraduates, if you ask them to describe the *scent* of a red wine and a white wine, and a week later present them with a dyed red white wine and a white wine, and their own list of descriptors to pick from to describe the *scent*, they are slightly more likely to use the 'real red wine' terms to describe the *scent* of the 'dyed red wine'. (You use their own list of descriptors because that is the major source of variation between individuals - there isn't a consistent 'right' answer).

The main reason this is significant is because, actually, it's thought to be quite difficult to tell red from white wine using only scent, so this test suggests that differences in scent word choice are shaped by visual as much as olfactory cues.

It does not suggest, at all, that people can't taste the differences between the wines. Nor, even, that they are imagining scents or flavours, rather that in, as a made up example, the choice between 'blackberry' and 'raspberry' to describe a bramble fruit, they are influenced by the wine colour.
posted by AFII at 12:16 AM on November 26, 2018 [20 favorites]

Which is part of why I remain skeptical of the perpetual "wine tasting is bunk" studies, along with their tendency to play into the "take that so-called experts with your ivory towers and big words" mentality.

This is often one sort of expert taking potshots at another sort, no?

I don't have much trouble believing that even experts are deceived by a fancy bottle. I'd be more skeptical of the claim that experts can't fairly consistently identify flavor elements (some of them, anyway) in fully blind tests.
posted by atoxyl at 12:40 AM on November 26, 2018 [1 favorite]

Just to follow up: there is no connection between Hodgson's work and the Bordeaux-dyed-wine paper. Hodgson's analysis is about how consistent - in terms of scoring for the purpose of wine competitions judges are over time. The answer seems to be 'not super consistent', e.g. a variation big enough to be the difference between bronze/silver and silver/gold is normal. This does not show that the wine tasters can't actually taste the wine - more whether they think this 'apricot fading into green tea' flavour they've identified is a 8/10 or a 9/10.
Some of his studies are also just about how well one wine does at different tastings, which isn't about consistency of judges as individuals at all, just about the fact that there isn't one single wine standard that human beings judge to.

Calling wine tasting 'junk science' as a consequence is basically the equivalent of calling book prizes 'junk science' because juries prioritise different things in different years.
posted by AFII at 12:47 AM on November 26, 2018 [8 favorites]

Every year my mother would bake me the same cake for my birthday: vegan chocolate cake with mint frosting. It was one of those eternal recipes that gets its teeth into a family kitchen and never lets go.

One year she'd run out of green food colouring, and asked me if it was okay if she left it white. I said "No way, let's go wild! Let's dye it purple!"

It was the same damn recipe as always. It was only at the very last minute, once the active ingredients of the frosting were already mixed, that I dropped in a few globs of red and blue food colouring and made the most unappetising cake I've ever eaten.

Subsequent years I enjoyed this cake as much as previous years, but we couldn't even face the thought of a second night of slices that year.
posted by rum-soaked space hobo at 1:22 AM on November 26, 2018 [5 favorites]

Conversely, who gives a shit if visual cues affect our appreciation of food? I realize in some sort of contest scenario it might matter, but at my personal dinner I am allowed to like what I like.

My older brother loves to buy super expensive and super hard to get beers. I call them quest beers. He will drive across multiple states in the middle of the night to try and get some limited edition brews. I used to really think this was a waste of time and money. Surely in a blind taste test he wouldn't be able to tell the difference between the quest beer and something cheaper and more available.

But then I thought a bit more about it and realized that the exclusitivity of it all is part of the enjoyment. He might not be able to tell the difference in a blind taste test, but when he is sitting down with his expensive hard to acquire beers he is actually happier.

So anyway, my point is that I think it is better to let people like what they like whatever the reason may be.
posted by Literaryhero at 2:38 AM on November 26, 2018 [4 favorites]

Expert judges regularly show bias in their tasting evaluations based on what the label looks like, what the cost is, and more directly relevant to this post, what color the wine is.

Sounds like a reason to do proper blind tests, with blindfolds!
posted by Dysk at 2:50 AM on November 26, 2018 [1 favorite]

I wonder why there's no blue food...
posted by fairmettle at 3:15 AM on November 26, 2018 [1 favorite]

Many tasting panels are done under red light so that visual differences between, e.g., pieces of meat are minimized. You would definitely want to conceal milk with differing levels of fat in taste tests because low-fat and skim milk, in my experience, tend to have a watery, blue-tinted appearance. I think that taking fat out alters the reflectance of the milk. I also believe that trained tasters could probably consistently tell whole from skim fat on mouth feel, but that most laypeople cannot. If you ever have the opportunity to take advanced dairy products judging, well, think carefully about that. It will forever alter your relationship with milk.

The Jersey breed of dairy cattle have intramuscular fat that tends to take on a yellow color which is detrimental to its sales. The taste isn’t notably different, but many consumers interpret yellow instead of white as indicative of spoilage or some other problem. My freezer has a bunch of lovely grass-fed Jersey beef in it, so let’s not tell anyone and send the price up. :-)
posted by wintermind at 4:05 AM on November 26, 2018 [2 favorites]

I also believe that trained tasters could probably consistently tell whole from skim fat on mouth feel, but that most laypeople cannot.

I'd believe you if you said whole versus 2%, but the difference in mouthfeel between skim and whole is way too different.
posted by explosion at 4:36 AM on November 26, 2018 [1 favorite]

Blonde stouts are a thing, too. It's a little weird when your mouth and eyes disagree.
posted by RobotVoodooPower at 5:06 AM on November 26, 2018 [2 favorites]

Taste is weird.

Flavor Science Explains How You Can Hear the Way Your Food Tastes
posted by Foosnark at 5:14 AM on November 26, 2018 [1 favorite]

Wait, what the heck is "inedible egg products denatured with caramel color?"

Okay, I looked it up. They're inedible in the sense that they're no longer for human consumption, and the caramel color is added to mark them as such. But they'll be used in animal feed.
posted by RobotHero at 6:17 AM on November 26, 2018 [1 favorite]

This reminds me of the classic tests that have shown, again and again, that wine tasting is junk science.

The conclusion to be drawn from TFA is not that expert tasters are being frauds when they can't distinguish red from white in a blind taste test, or when they say the same white wine tastes differently when dyed red, but that the actual sensation of taste can differ based on the colors of food that we perceive because sight plays such a large role in synthesizing taste.
posted by ultraviolet catastrophe at 6:34 AM on November 26, 2018 [3 favorites]

traveler_: Some years ago I felt like I'd swear I could taste the difference between M&M colors—made a bit more obvious by the fact it was a Christmas mix that had only red, light green, and dark green colors. So I did a solo blind taste-test with just the red and dark green to make it even easier, and got 10/10. So while I wouldn't doubt that color as appearance has a role in our experience of food, it can't be discounted that we might truly taste food coloring sometimes either.

Other people have talked about tasting red dye elsewhere on the 'net, but back on AskMe, the second answer about "bitter red dye" identified iodine. My wife and father-in-law dislike red- and purple-dyed frosting because it tastes bad to them, but I don't taste anything. I don't doubt what they taste, and it's interesting to find it backed up by AskMe :)


RobotHero: I keep seeing this around and I keep feeling like these tests have to themselves be junk science when they take it so far as to claim red wine can't be distinguished from white. What are they dying it with? Tannin?

darkstar: Presumably, the researchers at the University of Bordeaux know something about wine, and how to run a research test. But I haven’t read the original paper, which is here, so I can’t confirm that it’s sound.

In the abstract, though, they suggest that the visual perception of the wine causes the subject to discount their olfactory perceptions, similar to the main thesis of the original post of this thread.

RobotVoodooPower: Blonde stouts are a thing, too. It's a little weird when your mouth and eyes disagree.

Exactly! In addition to the strawberry mousse example I quoted in the OP, the article cites studies that found that "test subjects thought coffee in a white mug tasted less sweet than in a transparent or blue mug, and that 7up had a more lemon-limey taste when more yellow was added to the packaging." With those studies, and my own experience*, I fully accept that visual cues matter a lot when "tasting" food.

* In college, I had a delicious-looking brownie that was actually a fig bar. Because I bit into it expecting a brownie, it tasted awful, and I couldn't un-taste it as a bad brownie to try and enjoy it as a dark fig bar.

All that said, I have also enjoyed wine and beer tasting, taking the time to appreciate (or imagine :)) the flavors. Doing such tastings with people who know more about how wine and beer are made is fun, even if it's highly subjective. Like Literaryhero wrote: let people like what they like whatever the reason may be.
posted by filthy light thief at 7:13 AM on November 26, 2018 [2 favorites]

Would you eat green eggs and ham?

Because I bit into it expecting a brownie, it tasted awful

Though this also shows the limits of how much the appearance persuades you. You didn't convince yourself that it tasted like a delicious brownie because it looked like one. You judged its flavour according to brownie criteria and found it disappointing.

I'd heard the mutated rumour-mill version of the wine thing that they drank the wine, rather than just smelled it. If someone really drank a white wine believing it to be red, maybe they wouldn't realize it's "really" a white wine, but would they think it's really bad according to the criteria of red wine? (There's also temperature at play. To complete the illusion, you'd have to serve the wine at the "wrong" temperature for a white.)
posted by RobotHero at 8:43 AM on November 26, 2018 [1 favorite]

It was the same damn recipe as always. It was only at the very last minute, once the active ingredients of the frosting were already mixed, that I dropped in a few globs of red and blue food colouring and made the most unappetising cake I've ever eaten.

A few years ago I was at a benefit dinner with a Halloween theme, and they'd dyed a bunch of the food bright green. I can't quite remember what the decisive item was, although I think it was a pasta dish, but I couldn't eat it at all and became really nauseated. It was the weirdest thing, because of course I knew that the food was no worse than the usual activist benefit food and that the green food coloring was harmless (probably super harmless ultra pure co-op food-coloring, actually) but I still felt sick.

In a way it reminds me of how I felt when someone broke into our house while we were home one night (it all worked out) - I was so afraid for part of that experience that I could barely move. I had never known, to coin a cliche, what real fear was until that moment. It wasn't rational at all; I wanted to move and knew rationally (due to a quirk of circumstance, long story) that I was not actually in serious danger. It's so strange when you have that type of response - not just not a think response, but utterly unresponsive to thought, coming from a different part of yourself.
posted by Frowner at 9:05 AM on November 26, 2018 [1 favorite]

Know the neon bright colors of automobile coolant? I want to consume those colors, I want to drink them down.

There's no evolutionary psychology in the world that I think can account for how much I crave neon green foodstuffs.
posted by meese at 9:12 AM on November 26, 2018 [2 favorites]

meese: There's no evolutionary psychology in the world that I think can account for how much I crave neon green foodstuffs.

Nope, but I'd counter that it's a mark of success for marketing Mountain Dew ;)
posted by filthy light thief at 10:02 AM on November 26, 2018 [2 favorites]

The conclusion to be drawn from TFA is not that expert tasters are being frauds when they can't distinguish red from white in a blind taste test, or when they say the same white wine tastes differently when dyed red, but that the actual sensation of taste can differ based on the colors of food that we perceive because sight plays such a large role in synthesizing taste.

Very much this. The negative reactions to the article are perplexing - it is not telling you how to eat food, or what plate is correct, or making any other value judgement about food presentation. It simply provides more evidence to support what is an extremely well-known aphorism: perception is reality.
posted by grumpybear69 at 10:33 AM on November 26, 2018 [1 favorite]

If you compare us to dogs or pigs or other mammals, we have much more acute vision but they have much more refined senses of smell and taste, so it's not surprising that humans put more stock in our vision relative to our nose and taste buds.

I was involved in a research project* once that looked at this; I don't have the data to hand anymore, but we got a convenience sample of passers by in a public location to determine the flavour of clear fruit-flavoured sparkling water (back in the Clearly Canadian days). If the water was left clear, a fair number of people were able to determine the flavour. Once we coloured it, people's guesses were clearly determined primarily by the colour; orange water was thought to be peach even though it was actually cherry; purple water was thought to be grape or blackcurrant even though not only was it raspberry, but the same people had previously tasted it uncoloured and been able to determine it was raspberry then.

* sixth grade science fair
posted by Homeboy Trouble at 11:00 AM on November 26, 2018 [1 favorite]

Fact: toast tastes better when sliced on an angle.
posted by Robin Kestrel at 11:09 AM on November 26, 2018 [3 favorites]

I would like to eat a pink banana.
posted by BlueHorse at 12:12 PM on November 26, 2018 [2 favorites]

I recall being on skim milk for years, then switching to whole milk and thinking the taste and aroma was overpowering. So I am wary of any argument that we can't tell the difference between the two.
I get that the article is saying people generally can't judge the fat levels of milk. The insight seems rather to be that it's not the fat levels that causes people prefer the taste of whole over skim. So for kicks, a food scientist could do a Modernist thing of chemically binding aromas into skim milk to mimic whole milk. "Creaminess of milk is a myth" is a plausible claim.
And of course for cooks you can't just substitute one for the anyways.
posted by polymodus at 1:13 PM on November 26, 2018

There are pink bananas, but I can't figure out (but doubt) if their flesh is also pink. I too would eat a pink banana. There are pink lemons, if you want to drink pink lemonade. I saw them at the store last week, but they were so much more expensive that I didn't bother. Now I'm kind of regretting it.
posted by fiercecupcake at 1:16 PM on November 26, 2018 [1 favorite]

Also, yes, re milk: I kind of love milk, even though I "understand" that it's "gross" and all. Whole milk is a treat for me. 2% is daily. Skim is never. I can 100% tell the difference, because I enjoy it. I can imagine if you just served someone "milk" with a meal that many people would drink it and not be able to tell what % it is, but when I drink milk with something, it's a focus as much as the main entree.
posted by fiercecupcake at 1:18 PM on November 26, 2018

The colour of the plate also affects how appetizing food looks = I bought a set of rather stylish slate-grey dinner plates in a sale and found that none of the food I served on them looked particularly tasty, and when I served people food on them there was considerably more left on the plate than when I served the same food on white plates. The really odd thing is that I quite happily eat things like cheese off real slate slabs in restaurants. Is this because I know the cheese is on a real piece of slate rather than a slate-coloured plate?

I was once at a party where we were served ordinary bolognese sauce on spaghetti that had been dyed blue. With my eyes closed, it tasted absolutely normal, but I couldn't eat much, even with my eyes closed, because I knew it was blue. None of the other guests could eat it either Cake seems to be an exception to this, apart from when the colour is unexpected, as with rum-soaked space hobo's frosting. If we know a plain sponge cake is a pretty colour it doesn't much affect the taste, although I have noticed that US cake recipes use a lot more food colouring than British ones for some reason. The sort of cake that reveals rainbow striped layers when sliced open looks a bit off-putting to me, but I've never encountered one in the flesh, and I probably would eat it - it is cake after all.
posted by Fuchsoid at 2:09 PM on November 26, 2018 [1 favorite]

My brother figured this out at age 12 when he got into baking delicious treats. He would dye then bright green with food colouring because he found the rest of the family wouldn't eat them then and he could scoff the whole tray himself.
posted by lollusc at 2:36 PM on November 26, 2018 [3 favorites]

Sadly IME pink bananas are just off-yellow inside. They do actually taste pretty different.

I find any milk less fatty than whole milk to be unpleasantly sweet, but I grew up around fresh goat and cow milk, both of which are sweet in a different way. Fresh cow milk definitely looks a little weird if you're used to the homogenized stuff, and I can imagine that might affect how we taste it.

Sadly nowadays all of them make my digestive system complain.
posted by aspersioncast at 3:24 PM on November 26, 2018

So something about me that's really funny is that the way my Autism works all my food weirdness is texture based- but color means nothing to me. So in my experience the color of a food or drink means nothing based to the texture or flavor. It's one of those subtle ways in which I am wired differently I suppose- but it means I am loving these comments about how color affects neurotypical peoples perception of food! I suppose I'd be put off by blue tomato sauce... but It wouldn't taste any different to me if the texture was right. Human perception is great!
posted by Homo neanderthalensis at 4:16 PM on November 26, 2018 [1 favorite]

I'd counter that it's a mark of success for marketing Mountain Dew

I literally screamed and cried and flung my cup away in horror the first time I realised what colour Mountain Dew was -- I'd only had it in cans, up til then, and enjoyed it, but the first time I had it dispensed from a fountain.... NOPE

admittedly i think i was about five or six years old but i stand by my reaction and believe it was sound
posted by halation at 5:39 PM on November 26, 2018 [1 favorite]