Scientists are beginning to unravel why we love some types of food and hate others. It's a vastly more complex topic than they once thought.
COPENHAGEN—When Per Møller, a Danish food scientist, was in the United States a few years back, he tried a well-known U.S. chocolate bar. "It tasted awful to me, like vomit," he says. His American colleagues disagreed. "They told me this is what chocolate is supposed to taste like," he says. To Møller, who works at the University of Copenhagen in a yellow building known as "the cheese," the episode illustrates how little we understand why some things taste great and others awful—or why people can disagree so strongly on the issue.
Møller is one of a growing group of researchers trying to answer these questions. Using a variety of approaches—including genetics, physiology, and psychology—they are beginning to piece together a picture of how genes, receptors, nerve cells, and experiences interact to create a flavor experience in the brain.
They find questions everywhere. It's well known that humans are born liking the tastes of sweet and fat, and disliking bitterness—but it's a mystery why steak and Béarnaise sauce go well together, or gin and tonic, and why some people learn to like Campari, olives, and cilantro, and others don't. Even more important, says Katrin Ohla, who heads a research group at the German Institute of Human Nutrition in Potsdam, little is known about the path a stimulus takes from the moment food touches the tongue to the diner's decision that it's disgusting, delicious, or somewhere in between.
Compared to the other senses, science has long neglected taste, Ohla says. The receptors for umami, which lets humans taste glutamate and other amino acids, were discovered only in the 2000s, for instance, and scientists are still looking for fat receptors. Only in February did they discover why too much salt in your soup tastes bad: In high concentrations, it activates sour and bitter receptors on the tongue, presumably to prevent intake of dangerous amounts of salt.
In recent years, scientists have found a number of genetic variants that seem to affect how people perceive certain foods. A mutation that influences whether cilantro tastes soapy, for instance, or one that determines whether someone can taste a bitter chemical called phenylthiocarbamide.
But DNA tells only a very small part of the story, says geneticist Danielle Reed of the Monell Chemical Senses Center in Philadelphia. She compares taste to vision. There are individual differences in color perception, and an impressionist painting may be less appealing to the colorblind. "But overall, your vision does not affect what kind of art you like," Reed says. "It is the same for food." In fact, the researchers who discovered the genetic variant affecting cilantro taste calculated that it accounted for only half a percent of the variance in its perceived soapiness.
What makes flavor so difficult to study is that it consists of much more than molecules sensed by taste receptors. For one, when you swallow a piece of food, some air is forced up at the back of the mouth, so that receptors in your nasal cavity can detect its odor, a process called retronasal olfaction. "We used to think that was the same as smelling, but in fact, the brain records which way the odor came in and sends it to a different place," says Linda Bartoshuk, a psychologist at the University of Florida's Center for Smell and Taste in Gainesville. "Flavor is created in the brain through integrating taste and retronasal olfaction," she says.
Just how important volatile compounds are became clear from an experiment in which Bartoshuk studied the flavor of tomatoes. She found that subjects perceived a tomato variety called matina as twice as sweet as one called yellow jelly bean, even though it contained less sugar. The effect appears to be due to the fact that matina has higher levels of six volatile compounds that enhance the perception of sweetness.
Even things as difficult to measure as expectations seem to influence flavor. In a study published last year, Ohla showed participants photos of high-calorie or low-calorie food; afterward, a short electric current was passed through their tongue to create a standardized, slightly metallic sensation. Participants who had seen pictures of high-calorie food described that taste as more pleasant than those who had not. Food color and temperature also have a strong effect, Ohla says: "It's as if we're not really trusting our sense of taste alone."
Past exposure appears to play an important role as well. In one experiment, Møller gave pureed artichokes to 2- and 3-year-old kids. Over ten consecutive sessions, they then received either the normal puree, a sweetened version, or one fortified with sunflower oil. The idea was that the children would learn to like the puree through association with sweetness or the energy-density of oil, both known to affect food choices. But at the end of the experiment, those exposed to the unmodified puree the whole time liked it the most. "We come to like what we eat," Møller says—which may be why Americans liked the chocolate bar he found repulsive. This type of learning starts in the womb: In several studies, babies from mothers who ate anise or garlic during pregnancy showed an increased preference for those foods.
That taste is conservative isn't surprising from an evolutionary viewpoint. "You need a system that will protect you from eating foods that will poison you and kill you," Møller says. That's the reason a preference for sweet and an aversion to bitter are hardwired in our brain, says Bartoshuk. But it appears that retronasal olfaction allows learning and refining food choices during life. "What makes lasagna loved is that the odors have been paired to a source of calories," she says.
As the field of flavor perception develops, it's seeking collaborations with the world of gastronomy. Møller started a new journal last year, Flavour, that seeks papers not just from scientists, "but also from the growing number of chefs and other food professionals who are introducing science into their kitchens." Flavor scientists are also joining hands with culinary labs associated with top restaurants, such as the Nordic Food Lab, not far from Møller's office (see sidebar). Psychologists at the University of Sussex worked together with famed chef Heston Blumenthal of The Fat Duck, a topnotch eatery in Bray, U.K., to examine how people reacted to ice cream with a smoked salmon flavor. (Not surprisingly, they liked it better if they were told what to expect).
In the long run, understanding why things taste well will allow chefs to come up with better food, Møller believes. "Romans built bridges but they didn't know anything about Newtonian mechanics. We do, and, therefore, we can build enormous bridges that do not collapse," he says. Similarly, humanity has developed paella and pizza, without knowing why they are so good. If chefs understood the principles, they too, could build bolder and bigger, Møller argues. And maybe create a chocolate bar that everyone likes.