Appearances Count

  "Depending on the product type and the project objectives, the appropriate appearance target is one that meets your consumers' expectations," says Barbara Booth, sensory research manager, NSC Technologies, a division of The NutraSweet Company, Mount Prospect, IL.

  "Nearly all cheddar and processed cheeses contain added orange coloring and consumers have come to expect the resulting appearance," says Carol Locey, color product manager, Kalsec Inc., Kalamazoo, MI. "Another example is chicken soup. Without the added yellow coloring -- provided by annatto and turmeric -- the product would not have the expected appearance."

  According to Booth, consumers have traditional expectations as well as situational expectations. Certain products should look like they came from the kitchen, not the oven. What would Mom make? Chocolate chip cookies and apple pies should look homemade; Twinkies and Oreos should not. For "fun" foods, anything goes.

  "It's important to look at products in the category or pick one or two that are key and analyze them for unique characteristics," says Leora Hatchwell, flavor technology manager at NSC Technologies. "If possible, determine the attributes that relate directly to consumer acceptability."

  Sometimes the arena is so wide, like with barbecue sauce, designers have to make a decision to target a specific attribute, a type of flavor, a degree of cling. On the other hand, a product like vanilla ice cream essentially has two categories: either artificial vanillin-type or gourmet-type. But even that is difficult. It used to be that high quality meant french vanilla, made with egg yolks. Today some consumers consider white with flecks of vanilla bean as high end."

  For long-term appeal, however, Rule No. 2 should take precedence. Consider the example of clear products. Consumers seem to have reconciled clear with New Age-type beverages, but does clear cola really fit the concept?

  "For a while there was a fad for brightly colored and over-colored products," notes David Frick, laboratory supervisor, color service laboratory, Warner Jenkinson Co. Inc., St. Louis. "The over-colored, mouth-staining colors fad has come and gone. Bright neon colors seem to be a trend rather than a fad. Many companies are trying to achieve that look, especially for children's products, but it's hard to get there with the colors that are approved."

  Though these are difficult challenges, they must be addressed at the onset of a project. Marketing is supposed to take on the task, but R&D should share the responsibility of translating concepts into concrete attributes. After all, it's the product designer who ultimately decides how to achieve the desired product.

  When marketing says "homemade," for example, what does that really mean? "Normally homemade means things that are irregular," Hatchwell offers. "Things that are uneven, that do not come out of a die. Browned if that is appropriate."

  Food colors fall into two main categories: certified and noncertified. These are often referred to in the food industry as artificial and natural, respectively. However, from a regulatory standpoint, non-certified colors cannot legally be termed natural colors on a food label unless they are used to color the same type of product. For example, beet juice is really only a natural color if it is used to color beets. If it colors cherry juice, the FDA considers the product to be artificially colored and the beet juice is deemed a color additive.

  The full regulations concerning food colors can be found in 21 CFR parts 70-80. This lists the approved colors, purity specifications and requirements, and their uses and restrictions.

  "Processors are allowed to add paprika as a colorant to processed meats, such as bologna and pepperoni," says Locey. "They may not add color to fresh meat such as ground beef. One notable exception is chorizo sausage. Oleoresin paprika is specified in the regulations as an acceptable coloring agent."

  Certified or artificial colors consist of water soluble synthetic dyes or the aluminum salts of these dyes, called "Lakes." The FDA has approved seven synthetic dyes and their salts for use in food. Blending the seven produces a wide spectrum of color, including purple, black, brown, and variations of the primary colors.

  Several other soluble dyes are approved for use in countries other than the United States. Colors permitted by the EEC (European Economic Community) and WHO (World Health Organization), for example, include Ponceau 4R (red), Quinoline Yellow, Patent Blue V, Green S, Brown HT, Brilliant Black BN, and Carmoisine (red). Amaranth (red) is permitted in Canada, as well as EEC and WHO countries. The EEC does not permit the use of Green #3 and Red #40 in foods. Even annatto is restricted in some EEC products, such as confections. Specific international restrictions should be investigated for a product designed for worldwide sale.

  The FD&C dyes display colors when dissolved in the aqueous phase of a food product. However, their solubility varies with temperature and often with the solute. Although at typical usage levels this will not ordinarily affect the finished product, it could make a difference when color solutions are prepared. If the temperature of the solute changes, all the dye may not be in solution, and this could affect the color of the finished product.

  The Lakes are insoluble in most solvents and instead color by dispersion. They are used in low moisture, often high-fat applications, but are not fat soluble. They may bleed color slightly in water, but in most applications they minimize color bleed into adjoining areas. The shades produced depend on the method of production and to some extent on particle size. Lakes are more resistant than dyes to fading when exposed to high heat and light.

  "Dyes go into solution and function on a molecular level," Frick says. "Lakes are different because the particles are dispersed, and you are dealing with a physical property rather than a chemical property. Particle size, particle; shape and how well the particles are dispersed in the finished product all influence the outcome. In some cases, simply increasing the shear during dispersion may intensify the color."

The industry refers to a number of the non-certified colors as "natural" because they are derived from natural sources rather than chemical synthesis. None of the non-certified colors are natural. Canthaxanthan, for example, is a synthetic carotenoid. Most of the colors fall in the yellow-to-red range. No non-certified blue or green colorants have been approved for food use in the United States.

  "Natural colors are finding a pretty good following," says Frick. "They are not as technically demanding as once thought, but they do require a little more specialization for individual applications. Normally you wouldn't use one natural yellow across your entire product line."

  Looking to differentiate your beer? Try using a malt that imparts a reddish hue. The roast on the malt can produce caramelized pigments that contribute to the product color.

  "The ingredients greatly influence a product's appearance," states Hatchwell. "Look at a barbecue sauce. The amount of tomato paste, and the type and amount of spice very clearly affect the color. They can affect other appearance attributes: the brightness, the sheen, even the viscosity."

  Even some of the noncertified colors have a dual purpose, notably beta carotene (fortification) and the spice oleoresins (flavor). This can be an asset or a liability, depending on the application.

  "Many of the salad dressings, like french and thousand island, have ground paprika or oleoresin paprika to produce a characteristic color," Locey points out. "In cases like this, both the flavor and color of the paprika are desirable characteristics."

  Dark-colored fruits, like blueberries, can color a food product, but this is a double edged sword. Tinting yogurt by adding fruit can make an appealing product. However, tinting the batter for a blueberry muffin can be a disaster. In cases like this, unwanted color migration can be kept to a minimum by modifying the process. Using frozen berries, removing moisture from the berries, and minimizing mixing after the addition of fruit are often used for products like this. Still, unless there is a moisture barrier, which can exist to some extent in dehydrated products like raisins and infused fruit, moisture migration will cause soluble colors to bleed into the surrounding area.

  Washes also directly affect product appearance. These color a product, provide a flavor, and can even improve the product's texture when exposed to heat. Simple washes take advantage of the Maillard browning reaction. Amino acids and reducing sugars in the presence of heat result in brown pigments and characteristic flavors. Since these are desirable results for many foods, designers may choose to apply a surface coating of reactive materials to enhance the reaction and, subsequently, the product appearance. This approach is often taken to provide a consistent appearance for baked or roasted products, especially if the heat process used creates insufficient color. This can happen in high-volume operations that make high moisture products like bread or pies, parcooked products (brown-and-serve products), and microwave products.

  While the results are generally successful on conventionally browned products, microwave products present more than a visual challenge. The color and even the flavor of conventional products can be simulated, but usually not the texture.

  In most cases, the goal is maximum visibility. Contrasting colors often provide the desired effect. Red peppers, for instance, are the cake sprinkles of the vegetable world. Parsley is universally popular because it contributes a great deal of visibility at a relatively low cost, with minimal flavor impact. ("Madge, look at all these herbs and spices in this dip!")

  "Blanched almonds are popular in many applications," says Richard Meridith, executive vice president, Paramount Farms, Bakersfield, CA.

  "The process of blanching doesn't add much to the flavor, but it creates a different visual effect. The color is unique and that's popular now."

  Although these visual cues are appealing to the consumer, more often than not they are technically challenging, especially when designing a product that must be transferred successfully to manufacturing. How do you get chunky salsa instead of puree? Minimize the shear and make certain the ingredients used are resistant to the shear that is left. How do you get consistent, evenly distributed topical application? Because it depends on the product, ingredient and process, this may be one of the great unanswered questions of our time.

  In general, manufacturing abhors a particulate. Often it is a matter of shear, which takes its toll on the particulate in the mixing and pumping stages. Or, the particulates create problems in the cutting or filling stage. If a particulate is too large, in many cases it's difficult to split. This makes overfills/underfills or doubles, depending on the process, common. On the other hand, small almonds mean one will fit in each candy bit. Small raisins or bits of fruit minimize doubling or misshaped cookies.

  The shape of a particulate also can be important, especially in topical applications. Adherence is usually a problem, so having maximum surface area for adhesion -- a characteristic seen in flake salt versus granular -- improves the chances that the particulates will make it through the process without falling off. Weight can factor, too.

  "For snack foods, you've got to use the right size granule of salt," points out Wilbur Gould, technical consultant to the Snack Food Assoc. "If it is too fine, it blows all over the plant during application. If it's too large, it doesn't adhere well. Usually you need something in the 200 mesh size range."

  The shape also can help maximize visual impact, according to Meridith, especially if cost becomes a factor. "A teardrop-shaped almond when sliced can result in whole slices, as well as pieces," he explains. "A lot of products can use these smaller pieces to increase the number of pieces without increasing the amount of almonds. You can slice as thin as 0.023 inches, but you have to remember this increases their fragility."

  Consumers want visual reassurance that what they are buying, or what they think they are buying, is there. That can involve extra formulation work to reconcile the appearance with finished-product and process goals. A high-fiber bread, for example, must contain a certain level of fiber in the finished product, but it also may require the visual impact of brown, grainy particulates. A white, highly refined, high-fiber ingredient like cellulose may work best from a technical standpoint, but it would not provide the desired visual effect. Using a certain percentage of an ingredient with a lower fiber content but with highly visible particles, like bran, can increase the visual appeal with minimal impact on the process and eating quality.

  Although physical comparison to another sample is one method, relying solely on this can lead to inaccuracies. Product appearance, especially color, changes with time. Replacing the product with freshly made samples can cause the target to drift. Quantitative measurements can help provide product consistency.

  Two types of instruments can measure color objectively. Colorimeters use a tristimulus method similar to the human eye, while spectrophotometers measure the specific wavelengths of light.

  Color has three attributes. Hue indicates which primary colors or mixtures of primary colors we see. Red, orange, bluish purple, and so on, are all hues. Lightness indicates how bright or dark a color appears. The extremes are white and black. Saturation indicates whether a color is dull or vivid. This ranges from the most intense color of a particular hue to gray. Because the human eye has receptors for three colors (red, blue and green), the theory of color vision follows that all colors are seen as mixtures of these three.

  The attributes listed can be ranged in a three-dimensional space on which each axis is assigned numerical values. The two most widely used methods for measuring these values were established by an organization known as the CIE (Commission Internationale de l'Eclairage): the Yxy color space, which was first based on the tristimulus values (XYZ) but is independent of lightness; and the L*a*b* color space (or CIELAB), which was developed to provide more uniform color differences in relation to visual differences. Several variations of these methods are used, including the L*C*h and HunterLab color spaces. These methods all measure the same thing, a space on a three-axis spectrum, but actual values are different.

  "The human eye is very selective for shade differences, but it's not very selective for strength," observes Frick. "For instance, the human eye cannot discern a 10% strength difference in a color. If you use a yellow and blue combination with the yellow constant, but the blue varying by 10%, you can change the shade sufficiently so that the human eye can pick it up."

  Colorimeters measure these values. The numbers generated indicate the differences in the direction and the degree of the difference between two samples. Colorimeters can pick up differences that may be difficult to discern with the human eye.

  "Historically, the instrument of choice in the food industry has been a colorimeter," says Maria Repaci, marketing manager, instrument systems division, Minolta Corp., Ramsey, NJ. "Colorimeters tend to be simpler to use. They see colors the way the human eye sees them. Unless you are trying to match colors a spectrophotometer is not necessary, although they have become easier to use. Spectrophotometers have been used to evaluate maturity in apples. The color is one indication of how that process progresses."

  Rather than measuring differences, spectrophotometers measure absolute colors based on the light reflected from a product at each wavelength. The instrument picks up wavelengths outside of what the human eye can see. The results can be plotted on a graph (percent reflectance vs. wavelength) to provide an accurate measurement of the color. Spectrophotometers generate tristimulus values that correspond to those obtained by colorimeters. They also can measure reflectance under a wide variety of light sources, which may change how a product looks. They can pick up metamerism, the phenomenon where colors can look the same under one light source but different under another.

  While these methods provide valuable information on many products, they may not be appropriate for all foods. Currently these systems work best on uniformly colored products. They can tell you if your peas are within specification, but they won't tell you if a pie has the right degree of browned highlights. Those cake sprinkles would pose quite a problem, too. It is possible to measure a very small area, though, so that in many cases the confounding areas could be avoided -- measuring between the chocolate chips in a cookie, for example. Technique and translation of the results play an important role.

  "Spectrophotometric values do not always directly correlate to the effect of a color in a food product," says Locey. "Annatto, for example, is available as either an oil solution or as a suspension of pigments in an oil. If they contain the same amount of pigment, they will produce the same number on a spectrophotometer. But when incorporated into a food, the oil solution produces a more yellow hue and the suspension will be more orange."

  Another visual characteristic that can be measured is gloss, which is a function of the amount of reflected light rather than any particular wavelength. All spectral components except gloss are filtered out.

  "If you measured a black flat mirror and a white flat mirror with a spectrophotometer, the white would measure two to three times higher than the black," says Tim Allen, vice president, business development, Tricor Systems Inc., Elgin, IL. "Such an instrument is not fooled by color, but the human eye is. Different people may pick different samples as having more gloss, but a machine can consistently differentiate between them."

  In many food products, gloss can provide visual appeal and indicate certain conditions, including ripeness, oil content or the proper tempering of chocolate.

  "When given a choice, a consumer will often pick a product that is glossy," points out Allen. "Manufacturers are finding that gloss is indicative of other characteristics, as well. With cooked rice, the gloss relates to the amount of starch or stickiness of the product."

  While it is possible to quantify many attributes of product appearance, quantifying appeal in terms of acceptance takes research. That's when sensory evaluation becomes invaluable.

  First, the correct procedures for the evaluation are needed. Choosing the right descriptors, understanding what they mean, and putting them on the correct scales all help produce meaningful results. The product designer and sensory analysts must decide which attributes are important and if the questions will give data that can help improve the product. Sometimes conducting sensory research at the onset of a project and setting up acceptance optimization studies provides the necessary information, especially if the project involves a new concept. Screening the various factors can save time.

  Getting specific information on attributes such as hue or opacity may be too difficult for a consumer panel. For this a descriptive panel or instrumental analysis can help focus the development effort. However, these methods still will not indicate consumer preference.

  "You can ask consumers if a product has a homemade appearance or if they like the color," says Booth. "Some of our best research is done when consumers tell us what they like. But if you need more detailed information, use a different kind of panel assessment to get at the real attributes. A descriptive panel can give a word-picture of the product, which can then be correlated with preference."

  The manner in which the sample is evaluated affects the outcome. Some techniques, such as random sampling, are always appropriate; some are not.

  "There is a time and a place for conducting studies under masked lighting and canceling out color differences," says Booth. "But that's not the real world, and it's important that you acknowledge and measure the interactions. If there is something wrong with the appearance, all of the other attributes are affected. We call that a halo affect."

  Although appearance is only one of the characteristics that promote appeal, it should not be overlooked. It can make or break a product. "If the appearance doesn't at least meet minimal expectations, you won't be able to get a good fair evaluation of the flavor or texture," warns Booth.

TABLE:

© 1995 by Weeks Publishing Company

3400 Dundee Rd. Suite #100
Northbrook, IL 60062
Phone: 847-559-0385
Fax: 847-559-0389
E-mail: info@foodproductdesign.com
Website: www.foodproductdesign.com