Is seeing the same as eating? How what you see affects what you taste

Heinz's green ketchup nothwithstanding, we generally like our foods to be predictable colors: milk, white; bananas, yellow; oranges—well, you get the idea. But when foods are the "right" color, do they actually taste any different? We all know that food coloring is tasteless, so what happens when we dye foods different colors?

The results so far have been difficult to pin down. A study in 1960 found that green pear nectar tasted less sweet than colorless pear nectar. A study in 1962 found no such result. A 1982 study revealed red dye made strawberry juice sweeter; a 1989 study found it did not. Studies have found that brown liquids are refreshing, but brown beers are not. Perhaps the impact of color depends on the specific flavor being tested. Perhaps color matters for some people but not others, depending on their likes and dislikes.

It was in this foggy terrain that Debra Zellner of Montclair State and Paula Durlach of Army Research Institute devised a new experiment to see if they could find some order in the chaos of taste/color research ("Effect of Color on Expected and Experienced Refreshment, Intensity, and Liking of Beverages," American Journal of Psychology, 2003). Their experiment presented people with glasses of liquid in eight different colors (clear, red, blue, green, yellow, orange, purple, and brown). Each set of participants tried only one flavor of liquid: lemon, vanilla, or mint. They were asked to rate each drink they sampled on a scale of refreshing—not refreshing, intensity of flavor, and like—dislike. A separate group was told the flavor of the liquids but was not allowed to taste them, and then rated each glass of liquid based on how they thought it would taste.

The large number of different colors and choices helped clear up the picture: the group that didn't taste the drinks had the strongest opinions, but both groups indicated that they did not like the brown mint and lemon drinks, and that they didn't think they would be refreshing. The brown vanilla drink, by contrast, was rated as good tasting. The same people who said the brown mint drink did not taste good believed the clear mint drink was fine—even though each drink contained the same amount of flavoring. This result makes sense, since vanilla extract and vanilla beans are brown, but nothing minty or lemony is. Oddly, green mint drink and yellow lemon drink weren't rated as any better-tasting than other colors.

So, sometimes, especially in the case of brown, color makes a difference. If color does make a difference in how things taste, we tend to prefer colors that are appropriate for the thing we're tasting. I'm still not sure what all this means for the sales of green ketchup, but I'd definitely suggest Heinz not try selling brown ketchup anytime soon!