Preferred Flavors

Exploring the links between flavor perception, nutrition and health in today’sfood and beverage market.

By Stephen Goff
Senior Syngenta Fellow
Syngenta Biotechnology Inc.
Research Triangle Park, NC

People have innate and learned preferences for flavors of certain foods, and aversions to others. These flavor preferences and aversions are intimately linked, and learned from the nutritional value of the foods we eat.

Millions of years of evolution have provided humans with the ability to sense the chemical composition of foods and develop appropriate flavor preferences or aversions. Perception of food flavors by both taste and smell is achieved by a very ancient sensory system. Even simple animals like insects with a few hundred sensory cells are able to recognize and avoid specific detrimental foods and forage selectively to acquire dietary or health-enhancing meals. For example, both spiders and beetles have been shown to select meals rich in fat or protein to make up for dietary deficiencies. Likewise, caterpillars have enhanced taste preferences for specific plants that help eliminate parasites when they’re infected. Like insects and lower animals, humans regulate their feeding behavior to achieve dietary needs, but this highly evolved system can be fooled by current food production approaches. Only recently are researchers beginning to understand flavor perception and how current food preferences may be at odds with food products that combine low nutrient value with flavors that naturally occur in less processed foods.

Flavor Perception: Beyond Taste and Smell

The perception of flavor in humans is often described as a mixture of taste and smell. This definition, however, is an oversimplification since visual input, texture or mouthfeel, temperature and pain sensations, hearing, and past experience all play a role in flavor perception. These different sensory inputs are integrated in the brain to generate an overall assessment of a food. It has become increasingly clear that nutritional sensory input from the gastrointestinal tract is also integrated in the brain with these various sensations to shape the overall preference for specific foods.

Taste provides humans with information on the basic macro-nutritional content of a meal. The basic macro-nutrients—proteins, starches, fats—are sensed by a taste sensory system composed of receptors activated by amino acids, sugars, and lipids, the breakdown products of proteins, complex carbohydrates and fats respectively. Studies of perception de-monstrate that taste is an innate or “hardwired” sensory system, whereas smell is primarily learned. Newborn or even premature infants have clear responses to sweetness indicating they are born with a preference for sweets. They can even discriminate between different types of sugars in their foods. On the other hand, repeated experience with the smells of specific food items as we consume them shapes longer-term preference or aversion for a specific food. The smell of food helps provide a definitive identification of each specific food item.

The 2004 Nobel Prize in Physiology and Medicine was awarded to Dr. Richard Axel and Dr. Linda Buck for their discovery of olfactory (smell) receptors approximately 15 years ago. Since this discovery the understanding of the molecular mechanisms of taste and smell has advanced significantly. The Human Genome Project led to the identification of approximately three dozen taste receptors and a few hundred smell receptors. The activity of the taste sensory system signals to the brain the presence of salt (sodium chloride), small sugars, organic acids, proteins and nucleic acids in a food. Receptors similar to taste and smell receptors have also been found throughout the gastrointestinal tract. The olfactory system consists of approximately 350 functional receptors, although humans can detect thousands of different odorants. As is suggested by the relatively large number of olfactory receptors, the range of flavors perceived by humans is strongly influenced by the odor of foods.

Flavor Compounds are Linked to Nutritional Content

Animals and humans learn to discriminate between safe and unsafe foods by correlating flavor to the nutritional or health consequences associated with eating a particular food. For example, rodents fed a specifically flavored food and then made artificially ill through irradiation immediately develop a strong aversion to the flavor associated with that food. Likewise, humans and other animals have extremely efficient learning mechanisms to identify and avoid foods that cause illness.

Food preferences, however, are learned more slowly, as anyone with small children can attest. Early learning of what is safe to eat has obvious advantages later in life. In fact, studies of newborns’ breastfeeding behavior demonstrate that the developing fetus acquires flavor preferences to the mother’s diet In Utero beginning the establishment of these associations early in development. These learned preferences function to identify specific safe foods and allow for the association between a food and the nutritional content or health value. Although humans have developed an intricate system to identify specific individual foods through their flavor profiles, it’s very common today to experience flavors dissociated from their original context and associated with foods of very different and varied nutritional content.

More than 7000 flavor volatiles have been identified and catalogued from various foods and beverages. Many of these are plant volatiles produced during fruit or vegetable ripening or maturation. Only a small number of the volatile compounds produced by a fruit or vegetable contribute to those sensed as food flavors. For example, in tomato, which produces more than 400 volatiles, only a few of these volatiles impact the flavor profile detected by humans. Studies on the biochemical synthesis of important fruit and vegetable flavors have revealed that the majority are derived from essential nutrients, vitamins, or other health-promoting compounds, collectively referred to as micronutrients. For instance, one class of very important tomato flavors is derived from the two essential fatty acids linoleic and linolenic acid. These flavors provide information on the content of fatty acids essential to the human diet and are also found in apples, sweet cherries, bay leaves and teas. Likewise, important flavors in pear and banana are derived from the essential fat linoleic acid. Essential fatty acids also give rise to flavors in peaches, apricots, coconuts, cucumber and other fruits and vegetables. These nutritionally important fatty acids are also reported to be sensed as fat taste, while non-essential saturated and trans-fatty acids do not have the same taste-stimulating qualities. These findings support the interpretation that mammalian sensory systems are tuned to provide information on essential nutrients in the diet.

A second class of volatiles that contribute to tomato flavor is derived from the essential amino acids leucine, isoleucine and phenylalanine. These volatiles are indicative of the protein content of a food product, and are important flavor constituents of many fruits, as well as processed foods such as breads, cheeses, wines and beer. Non-essential amino acids rarely serve as precursors for important flavors. One notable exception is the garlic flavor allicin, which is derived from the non-essential amino acid cysteine. Allicin is of particular interest since it is a flavor component with antibacterial and antifungal activity, and garlic has been in medicinal use for several centuries. Garlic and similar spices with antimicrobial activity are believed to be important for food preservation in many parts of the world. As with insects shown to preferentially feed on plants with anti-parasitic properties, preference for the flavors of various plant-derived spices may therefore be attributed to their health-promoting properties.

Another class of tomato flavors is de-rived from carotenoids, including compounds such as vitamin A, which serve important antioxidant roles in the human diet and are implicated in many aspects of human health. Humans are very sensitive to the flavors derived from carotenoids. For example, they can detect b-ionone, a carotenoid derivative, when present in concentrations as low as a few parts per billion. Several other plant flavor volatiles are derived from carotenoids and have very low odor thresholds. For example, b-damascenone, which is found in tomatoes, berries, apples and grapes, can be detected at concentrations as low as two parts per trillion. Safranal, found in saffron and green tea, is derived from carotenoids and also has a very low odor threshold.

It is worth noting that tomato taste tests have demonstrated that the visual cue of “redness” has a major impact on the perceived flavor of the fruit. The red color is due to the presence of lycopene, a caro-tenoid with demonstrated health benefits. These findings are consistent with the conclusion that flavor perception is achieved via the integration of sensory inputs from a variety of sources.

Flavors Today: Lots of Flavor, Little Nutritional Value

The precise physiological mechanism responsible for detecting the nutritional or health value of a specific food is not completely understood. Disruption of flavor perception by disease or by medical treatments, such as chemotherapy often results in eating disorders and malnutrition. Likewise, the decrease in flavor perception common with age is frequently associated with weight loss and deficiency of specific nutrients. The recently identified sensory system located throughout the gastrointestinal tract may be responsible, in part, for assessing the quality of a meal, but much more research is needed to describe this newly identified system.

Until recently, humans did not have the easy access to food products with out-of-context flavors, high energy density, and large amounts of sugar and/or fat as they do today. In ancient times, limited food access created a preference for rare energy-dense foods and a natural response to eat when food could be found. Eating whenever food is available, and eating very energy-dense food is no longer necessary, nor is it healthy. In studies conducted over 60 years ago it was shown that when mainly healthy food choices were available, people subconsciously settled on a balanced healthy diet without training. Energy intake and energy expenditure in this situation is very closely balanced, resulting in minimal weight loss or gain over long periods of time. These early studies showed that humans have the natural ability to select a balanced diet when minimally processed foods are the only options.

Today it’s a different story since many less healthy choices are available. Surprisingly, calorie intake today is actually less than it was 20 or 30 years ago, even though the incidence of obesity has risen dramatically in that same time period. The probable explanation is that physical activity has decreased even more significantly than caloric intake has.

Many food products produced today are made with familiar flavors from natural sources, but have been associated with lower and variable nutritional value relative to the original flavor source. Based on current understanding of the biochemical origin of flavors and the development of flavor preferences through learned re-sponses, consumption of modern highly processed food products could lead to confusion on the nutritional value of these foods. It could even decrease the learned preference for that flavor in the original natural food. The expected result would be less consumption of natural, healthy, minimally processed foods, and a higher reliance on taste for sweet and fat in the evaluation of foods.

Lack of sufficient micronutrient content of highly processed food products could diminish consumer loyalty for such products over time. Does this force the food production business into a pattern of rapid introduction of new food products? Are consumers fickle about highly processed food products due to their subconscious assessment of the health value of such foods?


There is a definite correlation between health and the flavor components that contribute to the perception of foods. Perception of food is subconsciously integrated with vision, texture, pain, hearing, and post-ingestive feedback from the gastrointestinal tract through mechanisms researchers are only just beginning to explore.

Many important flavor compounds are derived biochemically from essential nutrients, and therefore flavors provide information on the nutrient content of foods. Obesity and diet-related health issues are becoming an increasingly problematic, and the highly processed foods that are so prevalent today often combine preferred flavors with low nutrient content. Dissociation of flavors from their natural nutritional context may lead to undesirable health consequences, such as the over consumption of highly pro-cessed starch or saturated fats, and in the longer term, a diminished preference for less processed foods with natural flavors and good nutritional content.

Of course, many more years of research are needed before a complete understanding of the flavor preferences that drive food consumption can be elucidated. In the meantime, efforts should be made to enhance the nutrient content, health value and flavor composition of food products available today. The design of food products with low nutritional value using flavors out of their natural context should be carefully reconsidered.

About the author: Stephen Goff is a senior fellow at Syngenta Biotechnology Inc., Research Triangle Park, NC. He can be reached at [email protected].

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.