Cut the Calories, Keep the Flavor

Designing a reduced-calorie version of a popular food is a little like gambling with chaos theory—especially that bit where the butterfly flaps its wings in Grand Forks and sets off a tornado somewhere in Malaysia. Of course, this is common knowledge among product developers, who’ve long accepted that the tiniest tweaks in a food’s formula can unleash undreamed of changes down the pike. But when the components in line for tweaking serve as the foundation for a product’s function and formulation—as is often the case in reduced-calorie foods—the changes that ensue can plunge a product’s entire existential balance into ... well, chaos.

Naturally, none of this matters a whit to Joe Consumer, who has as little patience for the exigencies of low-calorie product design as he does for the nonlinear dynamics of chaos theory. All he knows is that when presented with a low-calorie revamp of his favorite food, one part of his brain appreciates the merits of the reduced-calorie declaration while the other just expects the product to taste the same as all the soft drinks, salad dressings, yogurts, snack cakes or ice creams he’s eaten a million times before. It’s in getting these two brains to think alike that product developers of low-calorie foods face their steepest challenge.

Where calories come from

Understanding what calories are and where they come from is the first step toward meeting that challenge. And strictly speaking, the “calories” noted on product packages aren’t calories per se, but rather kilocalories (kcal), the unit of measure by which we denote the energy it takes to raise the temperature of 1 kg of water 1°C.

To our bodies, though, a calorie by any other name is just energy, and the more of it that a food has, the more we extract for our own use. Four major food sources provide us with caloric energy: digestible carbohydrates (4 kcal per gram), lipids (9 kcal per gram), proteins (4 kcal per gram) and alcohol (7 kcal per gram). Organic acids and sugar alcohols contribute some calories as well (usually fewer than 3 kcal per gram), but in light of their slight caloric value, and their generally minor role in most foods, we usually set them aside when discussing food energy.

Given that four major nutrients account for the lion’s share of a food’s calories, calorie reduction, by definition, means reducing a food’s quantity of those nutrients (while also customarily replacing them with something less calorically dense). The composition of the American food supply being what it is, we’ve tended to focus our calorie-reduction efforts on the two most-prominent contributors: carbohydrates and fats.

But removing these substances from a formula removes more than just calories. Such is the case with sugar, whose functional duties span from freezing-point depression, water-binding and the addition of bulk solids, to antimicrobial activity, staling postponement and participation in critical browning reactions. As for fats and oils, they, too, prevent staling, and also transfer heat, lubricate, emulsify, aerate foams, build structure, lend moisture and make possible a repertoire of food textures, from creamy to crisp. “You have to look at calorie reduction in a complete context—what you’re doing and what you’re replacing,” notes Markus Eckert, vice president of technical for flavors, Mastertaste, Teterboro, NJ.

The physiology of flavor

Nowhere is context more important than in considering carbohydrate and lipid contribution to flavor— which we all know is the real reason we eat in the first place. (Our superegos may eat to live, but those devilish ids, to which most of us have given over menu selection, live to eat.)

We’re by now familiar with the five basic tastes: sweet, sour, salty, bitter and, the latest addition to the family, umami, whose character is variously described as “brothy,” “savory,” “mouth-filling” or just plain “delicious” (the word’s approximate translation from the Japanese). The sensation of taste results from the direct interaction of dissolved chemical stimuli and chemoreceptor cells in the mouth. Our minds register sweetness, for example, when simple sugars, high-intensity sweeteners and select amino acids bind taste receptors on the tongue, setting off a cascade of reactions that splits compounds called G-proteins into subunits, activates enzymes that convert intracellular chemical precursors into “second messengers,” closes the cell’s potassium ion channels and, ultimately, triggers the transmission of neural messages to the brain to be decoded as “sweet.”

While taste works through this direct detection of chemical composition, the physiology of flavor maps out a more-circuitous path. Only when you couple the chemical senses of gustation and olfaction (the response of nasal cavity receptors to volatile compounds) with the effects of physical stimuli, such as food texture and temperature, can you approach an understanding of total flavor perception.

Or, as Otis Curtis, business development manager, DSM Food Specialties, Savory Ingredients, Eagleville, PA, puts it: “Total flavor perception depends not only on the relative contribution of taste stimuli, but also on their interaction. This, together with the recognition of five basic taste components and the interaction between taste, smell and the trigeminal senses—which detect chemical irritants in the mouth and throat—creates a complex stimulus-response reaction in the brain that helps us recognize and distinguish different foods.”

Sweetness is just the start

In other words, lose the substances responsible for a food’s taste and you risk losing not only that taste but myriad associated flavors and, quite possibly, the food’s very identity —a not-so-trivial risk in low-calorie products, which often fail for their inability to replicate their full-calorie counterparts.

Such has been the case with low-calorie foods whose calorie reduction has come by way of less sugar. When we reduce calories by trimming a food’s sugar content, we need to restore more than just its sweet taste—which, if it were the only element missing, would make low-sugar formulation a much simpler nut to crack. While it’s easy to think of sugar solely in terms of sweetness, the nutritive sweetener category encompasses specimens as diverse as table sugar and tupelo honey, and each carries its own trademark flavor. The effect of altering or removing those trademarks is akin to sketching an easily recognized portrait and then erasing the signal features.

Sweeteners owe these trademarks to a number of factors, including how they’re processed; their quantity and concentration in a formula; their interactions with organic acids, flavors and other ingredients; and any amino acids or minerals associated with their chemical composition (key contributors to molasses and honey flavor). Each plays “a critical role in how you perceive not only the sweetness, but the flavor characteristics of the sweetener system,” says Kevin Riley, technical director of the sweet flavor category, Mastertaste. Thus, while sucrose may give you “a very distinct sweetness, as well as some additional sugar-related flavor character,” he says, “if you use, say, a high-fructose corn syrup or some other type of carbohydrate- based sweetening system, the overall flavor perception can change dramatically.”

Consider, for instance, how changes in carbohydrate composition influence flavor release. “Carbohydrates help to force flavor compounds out of an aqueous solution and into the air and nose,” according to Shane T. McDonald, Ph.D., dairy flavor chemist, Edlong Dairy Flavors, Elk Grove Village, IL. Strip these carbohydrates from a formula, then, and the resultant flavor will seem flat by comparison. On the other hand, larger polysaccharides are known to bind flavors, he says. “So, systems that have had the starches removed may seem over-flavored or unbalanced.” The upshot for the formulator is the perpetual need to fine-tune flavor levels to compensate for even minor shifts in sweetener quantity or variety.

Also often missing from low-sugar foods are Maillard browning products generated when amino acids react with a reducing sugar’s carbonyl group under high-temperature conditions. Everything from toast to condensed milk to roast beef owes some of its flavor to these compounds, but if a sugar-free formula lacks reducing sugars to participate in the reaction, or if the substitute sweetener has no reactive carbonyl group of its own, the proper profile goes wanting. Some low-calorie alternative sweeteners— tagatose, short-chain inulin, polydextrose —will function as Maillard reactants. But more often, restoring brown notes to low-sugar foods is “where the flavor companies are in demand,” Eckert says. “We can simulate flavor formation externally just by reacting sugar and amino acids and making that part of our flavor formulation.” Cocoa, coffee, roasted nuts, caramel, dulce de leche, créme caramel: “All these specific profiles can be simulated by Maillard-reaction technology.”

Taste modifiers are also useful to formulators of reduced-sugar foods. These are “flavor combinations that have little or no taste or smell of their own, but complement, enhance or otherwise modify the flavor of a food product,” says Mariano Gascon, flavor lab director, Wixon, Inc., St. Francis, WI. “They may improve overall flavor, modify the character of off flavors, enhance desirable ones or inhibit the impact of undesirable ones.”

Because their sophistication rivals the complexity of flavor perception itself, “many times these systems need to be designed specifically for different applications, because of the physics of how these molecules interact on the receptor site,” Riley says. The target profile—and the formulation strategy that will achieve it—is “going to be very different in a beverage than in a baked good or a soup, or a full-fat ice cream vs. a low-fat ice cream,” he says.

The perils of proxies

A product developer who had hoped to effect a stealthy caloric reduction simply by switching out a nutritive sweetener for a high-intensity or low-calorie sugar substitute should think twice, especially since sugar substitutes can beget as many flavor problems as they solve in terms of taste. Notes Eckert: “If you go with a pure high-intensity sweetener such as aspartame or sucralose, and if you use it as the only sweetness ingredient, it will signal to the brain that the product is as sweet as a full-sugar product. But the sweetness profile—the flavor, the aroma—will be very different than if you sweetened with honey vs. maple syrup vs. molasses vs. even just cane or beet sugar.”

Saccharin, the granddaddy of high-intensity sweeteners, is a case in point, with its bitter, metallic after-taste illustrating the limits of the first generation of sugar substitutes. (Worth noting, though, are the legions of long-term devotees who’ve grown so used to the sweetener’s profile that they prefer it to sugar.) Aspartame has suffered similar criticism from some on account of the allegedly tinny, astringent tone to its profile. While the cooling properties of xylitol and some of its polyol siblings may lend a refreshing boost to chewing gum and breath mints—a result of their negative heat of solution, which literally saps warmth from the palate as they dissolve—this effect doesn’t sit as well in, say, an oatmeal cookie or hot-fudge sauce.

High-intensity sweeteners can also upset the flavor balance that formulators carefully crafted in the full-sugar original—an effect that often manifests itself in swings of acid perception. Sucrose’s intense sweetness and later onset, for example, can compete with and even drown malic acid’s tartness, while fructose, whose sweetness profile hits the palate earlier, often enhances it. Thus, observes Curtis, “You totally have to readjust the formula when you do a diet soda vs. a full-calorie version.”

Those differences in temporal development —when a sweetener’s taste first hits, where it peaks and how long it lingers—suggest yet another consideration that complicates reduced-sugar formulation. “Sweeteners like aspartame, and especially sucralose, exhibit a long sweetness build and a pronounced and long-lasting sweet aftertaste,” Gascon says. “On the contrary, sugar is characterized by a balanced profile with a fast sweetness onset and without a pronounced sweet aftertaste.” (See this month’s Food Product Design Elements article, “Customizing Sweetness Profiles,” for additional information.)

Sweetener solutions

As Doris Dougherty, senior food scientist, Tate & Lyle Americas, Decatur, IL, counsels: “You always need to think about the sweetness profile, whether your sweeteners are nutritive or high-intensity, in order to balance those flavors.”

In the case of reduced-sugar products, balancing flavors often translates to balancing sweetener options themselves, as one of the best ways to counter the drawbacks of one is to pair it with the strengths of another. Thus, sugar alcohols have frequently accompanied high-intensity sweeteners in low-sugar formulations, not only taming the latter’s flavor flaws but, because we use them in similar quantities to nutritive sweeteners, restoring needed bulk. Mannitol’s negative heat of solution, for example, actually masks high-intensity sweeteners’ bitter notes (although it’s limited to applications where its cooling effect doesn’t seem out of place). Lactitol also often appears as a counter-part to high-intensity sweeteners, particularly in fruit preps, where its sugar-like sweetness and ability to promote flavor release show off fruity notes to best effect. And isomalt, which, like lactitol, also encourages flavor transfer, dissolves slowly on the palate, extending sweetness and flavor perception in reduced-sugar formulas.

Because of its clean, sucrose-like flavor, tagatose—which is an isomer of galactose whose incomplete digestion yields only 1.5 kcal per gram— mitigates the bitterness of aspartame, acesulfame-K and sucralose while also hastening sweetness attack and enhancing citrus and mint notes in confectionaries.

Formulators can improve sweetness profiles while also padding roughage levels by supplementing high-intensity options with multifunctional fibers that replace both the bulk and the taste of sugar. Litesse, from Danisco Sweeteners, Redhill, England, is a 1 kcal per gram, colorless, mildly sweet liquid polydextrose recognized as a dietary fiber in the United States. Its taste-masking properties minimize off notes attributable to high-intensity sweeteners. In addition, it bolsters the profile of fructose, and its low residual acidity amplifies delicate sweet notes and cocoa flavors, according to the company.

Inulin, another dietary fiber, is a naturally occurring storage carbohydrate whose linear-chain fructan polymers vary in sweetness according to length—the shorter the chains, the sweeter the inulin. Thus, says Connie Lin, Ph.D., applications manager, Sensus America LLC, Monmouth Junction, NJ, chains whose degree of polymerization averages around 4 to 5 can exhibit as much as 50% sucrose’s sweetness, allowing them to be part of a system that can “effectively replace sugars in confections, fruit preparations, sweet baked goods, cereal bars and dairy beverages to reduce sugar and calories and enrich fiber.” Because the fiber “doesn’t interfere with the natural flavor profile of products,” she adds, “it helps enhance pleasant flavors —for example, fruit flavors in fruit preparations.” Because of what she describes as the “good synergy” between inulin and high-intensity sweeteners, formulators can mask the latter’s off notes while enhancing sweetener efficiency, allowing a cut in sweetener use overall.

Even high-intensity sweeteners themselves, when strategically matched with their brethren, can make up for some of their own flavor drawbacks. According to Shuji Matsui, associate director, marketing and business development, Ajinomoto Food Ingredients LLC, Chicago, “The aspartame molecule’s three-dimensional structure fits almost perfectly into the sweet taste receptor and therefore has the most sugar-like taste.” That near-perfect fit, he says, “can help to mask the taste limitations of other sweeteners, including the bitter aftertaste of saccharin and acesulfame- K”—making these sweeteners logical partners in sugar-free soft drink formulations.

Sucralose, because it’s made from sugar, “has a profile more like that of sugar,” Dougherty points out. “So, if you balance that with another high-intensity sweetener like acesulfame-K, which has more of an upfront sweetness, you get the benefits of both.” Formulators who want to reduce nutritive sweetener levels rather than eliminate them outright can also achieve well-rounded profiles by pairing sucralose with fructose, whose “more-intense, upfront sweetness” complements the stronger tail-end sweetness of sucralose, she says.

Rebalance LF3, a sweetener blend from Tate & Lyle, puts this strategy into action, combining fructose, sucralose and appropriate acids in such proportion that, according to the company, it can reduce calories by up to 66% vis-à-vis full-sugar formulations while also enhancing fruit and cola notes and contributing mouthfeel to carbonated and still beverages.

Another sweetener blend pairs maltodextrin and sucralose in a coprocessed powder that increases sucrose’s sweetness 60-fold. At this year’s IFT Annual Meeting + Food Expo® in Orlando, FL, the company featured this in a lemonade that, at 5 kcal per 8-oz. serving, had “all the taste and mouthfeel of a full-sugar product,” per a company press release.

The mouthfeel-receptor link

That a high-intensity sweetener can return mouthfeel to a reduced-sugar product might come as a surprise to some. In fact, that a reduction in sugar might create a mouthfeel deficit in the first place isn’t always intuitive to those who associate this sensory attribute primarily with palate-coating fats and oils. But, as Curtis says, “There are two elements that people always seem to overlook beyond just the sweetness of sugar.” One “is the impact of these macrocomponents on the interplay of flavors and their release”—the subject just covered. The other concerns the effect we get from the sweetener solids themselves, “that tactile sense of structure and bulk that you might compare to the difference between water and a 10% sucrose solution,” he continues.

Quite simply, reduction in sugar solids attenuates mouthfeel. As Eckert explains, “Because of the intensity of high-intensity sweeteners vs. sugar, you’re replacing much more sugar with less sweetener at the same intensity. That means you’re missing solids in your formulation already, so you’re definitely missing texture and mouthfeel.”

However, as we solve more of flavor’s physiological puzzle—and mouthfeel definitely ranks highly as part of total flavor perception—we’re learning that sugar’s contribution isn’t a matter of sheer bulk alone. The effect of sugar on mouthfeel also “has to do with perception and the receptors on your tongue,” Eckert says. The tongue, he explains, has two different types of sweetness receptors. “One is more for very small, polar molecules, such as sucrose and some of the artificial sweeteners,” he says. “The other one is a receptor that most of the larger molecules—like sweet proteins or larger saccharides—attach to. And it’s usually a combination of the two that gives us the overall sweetness signal, the overall intensity and the overall profile.”

But, Eckert continues, umami-producing compounds also bind this second receptor. What’s more, this sweetness/umami receptor “is also known as a receptor that provides mouthfeel”—indicating that mouthfeel comprises receptor-mediated chemical sensations as well as physical ones. “It’s mouthfeel on a molecular basis,” he says of the mechanism. “You’re fooling your mouth that it gives you more mouthfeel and more impressions just by adding some molecules that bind to the various different taste receptors. You perceive it on the tongue, vs. texture, which is more like touch on the skin.”

So, in addition to restoring mouthfeel through the addition of non- or low-calorie bulk solids and perhaps gums and hydrocolloids for their textural effects, “You might reintroduce another molecule that binds to the second sweetness/umami receptor to give you the impression of having a bit more mouthfeel,” Eckert says, suggesting polysaccharides, polypeptides and sweet proteins, such as thaumatin, as suitable candidates.

Umami and beyond

The revelation of umami’s relationship to mouthfeel should encourage formulators to explore this sensation’s potential as more than just a savory generator—the role into which we’ve largely typecast it. Being stuck in that role, it’s gotten most of its play in straightforward, savory applications, such as gravies, sauces and broths. But while “most people think of umami as being soupy or bouillon,” Curtis says, “that’s not all umami is.”

It is also, he contends, a valuable addition to reduced-calorie foods, where its ability to expand existing flavors makes up for some of what gets lost when we take out calories in general and fat in particular. “There’s something very fundamental about removing fat,” Curtis says. “As soon as the formulator takes some of these gold standards and starts pulling out the butter or pulling out the fat, they lose complexity. The profile becomes very transparent, or simple.”

The best example of this is a reduced- fat salad dressing. “If you took a regular-formula, full-fat dressing and pulled out the fat portion,” Curtis explains, “all of a sudden, all of the flavor perception is changed,” including the mouthfeel qualities. “You have a textural challenge, so you have to use starches or gums to build back the bulk. But what you learn at the bench top is that there’s still an aspect of complexity and overall mouth-filling that goes away or is hard to build back just with something like starch or gum. And then you have the acidity issue—that whenever you take out the fat, all of a sudden the pH you’ve targeted for stability becomes a sensory problem because it screams of acid.” That’s where high-umami ingredients, such as yeast extracts, come to the rescue.

Being rich in the nucleotides and amino acids that are “known to interact with other umami-contributing factors to create a synergy,” Curtis says, yeast extracts “drive umami taste.” This compensates for the thinness and lack of depth that plagues reduced- fat foods. In a low-fat mayonnaise featured at IFT, the company’s yeast extract ingredient Maxarome Select “gives you the mouth-coating character. It tempers some of the perception of the acidity. And it also changes some of the perception of the flavor notes,” he says.

Furthermore, it does all of this without introducing an overtly meaty taste of its own, so you needn’t worry about low-fat mayo tasting like chicken broth. “You wouldn’t want a brothy or soupy character in a mayo,” Curtis allows, “and yet you might want some of the desirable effect of enhancing umami: We want the mouthfeel. We want some of the mouth-coating and fullness. And we would want to balance the flavors, as opposed to having some of the spikes, like with the acid that you sometimes have when you pull out the fat. So, with yeast extracts, you definitely get the balancing of the overall flavors, as well as the mouthfeel, without the soupiness.”

While enhancing umami provides one avenue for improving the flavor of low-fat foods, formulators can tap into yet another sensory concept with the potential to rebalance total flavor perception. Matsui calls this concept kokumi, a Japanese term that’s even trickier to translate than umami. Kokumi, he says, “goes beyond the five basic tastes. Kokumi could be translated as ‘roundness’ or the ‘fullbodied taste,’ which is used to evaluate wine, though it is still vague.” What’s important, he says, is that kokumi distinguishes itself by taking into account “factors of time, expansion and harmony, whereas the five basic tastes are simply evaluated by their intensity.” Consider the way a Cabernet blooms on the palate moments after a sip, the unity and continuity among flavors in a three-cheese sauce blend, or a premium ice cream’s capacity for filling the mouth with richness... Each gives us a little snapshot of kokumi in action.

Where, precisely, does kokumi— which isn’t a tidy, receptor-mediated sensation—come from? “We know some substances that contribute to kokumi,” Matsui says, “but the process is not dependent on a single substance.” Certainly, a balance of the amino acids and peptides so active in other aspects of taste, flavor and sensory perception comes into play. But, he continues, oil and fat also contribute to kokumi, which is why “lowcalorie products generally lack kokumi, or full-bodied taste. Their taste is, in general, very thin.”

Thus, his company targets such products—especially reduced-fat applications —for kokumi enhancement using their taste enhancer, Koji-Aji. “The key component of Koji-Aji is a partially hydrolyzed wheat protein,” Matsui says, and that, combined with high-grade yeast extract, “creates varieties of peptides, varieties of amino acids and varieties of oligosaccharides, all of which contribute to kokumi.” With a free nucleotide content of 2.9% and free glutamic acid at 18.1%, the product’s unique composition apparently gives it an ideal edge in delivering this mouth-filling sensation. “Typical dosages for soups, marinades and dressings are around 0.03% to 0.10%,” he says. “For gravies, sauces and snack seasonings, they’re 0.05% to 0.2%.”With its nondescript flavor, neither these levels nor higher ones threaten to edge a product’s natural flavor profile into the bouillon-cube realm.

Please release me

While kokumi enhancers and yeast extracts deserve credit for improving the flavor of reduced-fat foods, they can’t replace the fat itself—nor did anyone intend them to. For that, we still turn to fat mimetics, barriers, extenders and substitutes. The latter, says Gascon, “are synthetic chemicals with similar properties to fats.” (Think olestra and salatrim.) Fat barriers, on the other hand, are usually based on starch or cellulose and form films across a product’s surface that lock out oil that would otherwise seep in during frying. Fat extenders take advantage of emulsifiers to spread the textural effects of a reduced fat level throughout a lower-fat food.

But, it’s the fat mimetics— starches, proteins, soluble fibers, gums and hydrocolloids, in particular—that have become the common currency of fat replacement. Fat mimetics simulate fat by binding water into gels whose viscosity isn’t too far off from fat’s. Long-chain inulin, for example, “forms nice particle gels due to its excellent gel-forming and water-binding capacities,” Lin says. “The inulin gels are three-dimensional networks of inulin particles, which have a remarkable resemblance to the rheology of fat and are thermo-reversible.” Thus, she suggests using long-chain inulin (that with a degree of polymerization greater than 20) as a stand-in for fat in spreads, margarines, desserts, dairy items and baked goods.

Finding the right fat replacement is only the first step to perfecting a reduced-fat formula. As was the case when subbing out sugar with its various substitutes, once you’ve taken fat out of a system and woven its proxy in, then comes the challenge of readjusting the flavor profile to make up for all the havoc that the substitution has likely wrought.

That havoc starts with the flavor of the missing fat itself. “Depending on the system,” Eckert says, “some of the fatty acids in it will be present in free form. And they are the ones that make a specific contribution to the taste, the flavor and the aroma of the system. So there are very specific fatty acids that you’d find in a chicken fat vs. a beef fat vs. a dairy fat.” In other words, a fat-free yogurt will need a dose of the esters, lactones and identifying dairy compounds responsible for characterizing the full-fat original in the first place. he suggests reintroducing them “as part of the flavor portion of your formulation, where you would bring back some of the creamy notes into the system to create the sense that there’s more fat to it than there is.”

Don’t think the flavor adjustments stop there. Just like chaos theory and the flapping butterfly wings, fat, by its very presence or absence, sends out shockwaves that shake up every other flavor in a product. Like chaos theory, the reasons why have to do with bare-bones science.

“In a food system with both fats and carbohydrates,” McDonald explains, “flavor compounds will tend to equilibrate between the fat phase and the water-loving phase that contains the carbohydrates. The relatively hydrophobic compounds will partition into the fat phase, and hydrophilic compounds will partition into the aqueous phase.”

Because taking the fat out of a system exposes flavor volatiles to a more-hydrophilic environment, he continues: “The hydrophobic compounds will want to ‘jump’ out of the hydrophilic environment and quickly move into the gaseous phase to be perceived by the nose. A flavor that is balanced for a fat-containing system will seem too strong, spiky, unbalanced and possessing of little ‘linger’ in a fat-free one.”

We can get a clearer handle on all these effects by referencing a chemical property called a partition coefficient, the measure of a compound’s differential solubility in two solvents. According to Riley, an imperative aspect of product formulation is “how one uses the knowledge of that property to design a flavor system for a particular fat content in a food, because that’s really going to define how that flavor will be released and perceived on the palate.”

Basically, Eckert says, “the partitioning of every single chemical in the flavor is important.” With a typical flavor comprising 15 to 50 individual components—each of which has a unique coefficient—that makes for some pretty complicated calculations. That doesn’t even take into account the total formulation of the product itself. So, Eckert continues, “you’ve got to understand your food matrix overall, too—the carbohydrates, protein and fat—because the flavor release and perception will depend on how that flavor is partitioned between those ingredients.”

So, when comparing three different yogurts at 0%, 3.5% and 10.0% fat each, “The partitioning will look completely different for every single chemical and in every single yogurt,” Eckert says. “And that’s why your overall flavor perception will be very different if you use the same exact flavor, and at the same rate, in those three different systems. You would perceive a totally different flavor initially, as well as over time.”

Anethole, a flavor with a slightly sweet anise note featured widely in vanilla, honey and several other profiles, provides an example of how this works at the level of the individual flavor compound. Its intensity in a 0%-fat yogurt, Eckert says, would be four times higher than in a 3.5%-fat formula. “However, over time, the flavor will fade very quickly vs. in a 3.5% yogurt,” he continues,” because there, the chemical will be perceived much lower but for a longer period of time overall.”

Because each chemical in each flavor in a formula similarly does its own thing, so to speak, “if a developer wanted the 0%-fat product to have exactly the same flavor profile as the 3.5%-fat product,” Eckert says, “the total flavor system has to be reformulated to reflect those influences.”

All this necessary number crunching isn’t for the faint of heart (or the slow of brain). But you’ve got to trudge through it to churn out an accurate prediction of flavor partition. What’s more, it yields invaluable benefits for other aspects of low-fat product design.

The partition coefficient also provides insight into how processing and ingredient behavior factor into a product’s flavor. “When you change the fat level in an ice cream,” Riley offers, “it alters the size and growth of ice crystals, which will also have a significant effect on flavor release, as well as on the overall texture of the ice cream in your mouth.” While he says that formulators tend to take “an endrun approach” around these changes by hedging against ice-crystal growth through the use of hydrocolloid stabilizers, “If not selected correctly or formulated at the right usage rate, they can have a tendency to depress flavor release.”

Indeed, gums can act like flavor blankets, smothering flavors under their weight. The effect, Eckert says, actually involves not only their relationship to partition coefficients, but to the mouth’s flavor receptor sites, as well. Hydrocolloids, he explains, “really coat your mouth and block some of your receptors, making it fairly difficult for you to detect the flavor in the first place.”

That’s why, Eckert says, “there’s not a perfect system out there. In a lot of cases, it’s better to work with a combination of different fat replacers, because gum-based replacers are usually water-dispersible, whereas fat is insoluble in water. So you would go for maybe a combination of a protein-based fat replacer and a carbohydrate-based fat replacer,” because that way, “at least you’re offering two or three different systems to replace one, and that way you minimize your problems a bit,” he says.

Another way to minimize those problems is to take a reasonable approach to reducing fat, sugar and calories in the first place. Maximizing the amount removed may not be as clear-cut as it first seems. “I spent a lot of time trying to formulate a nonfat muffin, and it’s very difficult because of the texture,” says Dougherty. “But a low-fat muffin, while I wouldn’t say it’s easy, is very feasible—and you can get a very good product. You don’t get that rubbery texture. You get much more of the expected mouthfeel. And so it’s what the consumers want. It’s almost to the point where consumers are shying away from the fat-free product.”

Riley agrees: “The market changes. A lot of it is what the consumer wants—what is the fad of the day, whether it’s Atkins or fat elimination. You have to work within the framework of the consumer and design products that taste good.” And in the end, that shouldn’t be all that hard after all. As Eckert is quick to add, “We’re consumers, too.”

Kimberly J. Decker, a California-based technical writer, has a B.S. in Consumer Food Science with a minor in English from the University of California, Davis. She lives in the San Francisco Bay area, where she enjoys eating and writing about food. Contact her at [email protected].