Food Color: More than Meets the Eye

 Food Color:
More than Meets the Eye
October 1996 -- QA/QC

By: Ray Marsili
Contributing Editor

  Flavor gets a lot of attention from food product designers, but color often doesn't get the attention it deserves. For fresh fruits and vegetables, for example, color is an indication of taste and flavor quality (e.g., freshness, over-ripeness or under-ripeness), and it is usually the primary attribute consumers consider in making purchasing decisions. Color measurements are useful for grading commodities.  The association of certain colors with the acceptance of specific types of foods begins early in childhood development and is maintained throughout life. The blue-green mold on bread and the off-color of meats, fruits and vegetables are warning signals that the food may be microbiologically contaminated, or at least contain off-flavors.  Fruit and vegetable processors often base decisions to harvest on color measurements. Furthermore, measuring the color of various parts of the plant may provide information that allows farmers to optimize applications of fertilizers and herbicides, according to James DeGroff, president of Clinton, NJ-based ColorTec.  Meat processors also use color measurements to assess the quality of their products. Measuring the extent of red and brown is a good indicator of meat freshness, as well as consumer appeal and acceptance.  Measuring the color of processed foods can be a critical quality control tool. For example, the color of extruded cereal and snack products is a good indication of over- and under-processing conditions. Also, color measurements can be a useful indicator of the quality of incoming raw materials and ingredients used to prepare processed foods. Color influences  "All animals eat, but humans feed," says Fergus M. Clydesdale, Ph.D., professor and department head, Department of Food Science, University of Massachusetts, Amherst. "Eating is a festive occasion, and color adds to the enjoyment of food."  Color can dramatically influence the flavor perception of food. Food technologists often use various colored lights in their tasting sessions to mask the color of the foods being taste-paneled so tasters will make their acceptability judgments based on flavor rather than color. One study dramatically illustrates the influence of color on food acceptability (J. Wheately, "Putting Color into Marketing," Marketing, October 26, 1973). In this study, panelists were offered a dinner of steak, peas and french fries abnormally colored but served under color-masked conditions. Panelists enjoyed the meals until normal lights were switched on. The appearance of blue steak, red peas and green french fries was so overwhelmingly objectionable to some of the panelists that they became ill.  According to Clydesdale, studies show that when red color of fruit-flavored products is enhanced, the perceived sweetness level increases. Food product designers could potentially reduce sugar and calorie levels in fruit-flavored products without reducing perceived sweetness by increasing the level of red color in these products. This phenomenon has been observed in strawberry- and cherry-flavored beverages containing various levels of red color. Other colors also have been observed to affect sweetness perception, depending on the appropriate combination of color, flavor and sucrose.  In addition, Clydesdale (in "Color as a Factor in Food Choice," Critical Reviews in Food Science and Nutrition, 33(1): 83-101, 1993) reports that color affects taste thresholds. Studies indicate that the threshold concentrations at which the basic tastes (salt, sour, bitter and sweet) are perceived are color dependent. In the study, the effects of red, green and yellow on the threshold concentrations of sweet, sour, salty and bitter tastes were measured. The yellow-colored sweet solution was detected at a significantly higher concentration than the colorless control. Therefore, tasters did not associate yellow color with a sweet taste. However, the green-colored sweet solution was detectable at a concentration significantly below that of the control.  With sour solutions, the red-colored sour solution was detected at a slightly higher concentration than the colorless control sour solution. However, the sour flavor in yellow and green solutions was detected at significantly higher concentrations than in the control.  With salt solutions, color had little influence on the perception of saltiness. However, with the bitter solutions, yellow and green solutions were detected at significantly higher concentration than the colorless bitter control, but the red bitter solutions required the highest concentrations before they were detected.  One potentially useful application of color manipulation is with foods for the elderly. After age 40 the senses of smell and taste tend to deteriorate. Clydesdale says that paying more attention to color can help food processors compensate for the loss of taste perception and significantly improve the appeal of foods to the elderly. Food processors' interest in this area of color application is likely to escalate since the average age of consumers is rapidly increasing as the massive baby boom generation ages. Clydesdale has recently initiated a study to determine if color affects mouthfeel perception of foods. Limitations of the eye  The human eye has its limitations as a color-differentiation device. Eye fatigue, color blindness and viewing conditions are examples of limitations of the human eye. Furthermore, the eye does a poor job in differentiating three important characteristics that account for how we perceive and judge color acceptability: (1) hue -- red, yellow, green, blue, etc.; (2) chroma or saturation; and (3) lightness/darkness.  On average, the human eye perceives hue differences first, chroma or saturation differences second, and lightness/darkness last. To determine if a color is acceptable, colorimeter readings are converted into acceptability values. These numerical values can be used as guidelines or tolerance limits to aid in controlling color in a processed food product, to ensure color consistency within a production run, to minimize lot-to-lot variation, to judge how closely the color of your new prototype product matches competitor products, etc. (See sidebar for explanation of the meaning of colorimeter numerical values.) More to measuring  To say, "We eat with our eyes," is trite but true. Foods have to look "right" before consumers will buy them. The challenge facing food designers is to determine what "right" is and then to ensure that the desired color is not only achieved at the time a food product is manufactured, but also that it is maintained during the product's shelf life. To do this, the food designer has to choose some type of tool that measures the product's color accurately, precisely, conveniently and quantitatively. There are two basic types of color-measuring instruments (spectrophotometers and tristimulus colorimeters) and two basic modes of operation (transmittance and reflectance).  "If measuring transparent solutions, measuring the color transmitted through -- or absorbed by -- the solution with a spectrophotometer works best," says Dave Frick, operations manager, Color Service Laboratory, Warner-Jenkinson, St. Louis.  A spectrophotometer has a narrow band width and measures transmitted or reflected light one wavelength at a time over the entire visible wavelength range, approximately 300 wavelengths or data points. A plot of wavelength versus intensity can be generated.  "The problem is this information doesn't measure how humans actually perceive color," says Frick. "Small changes may represent something important to human color perception, and large changes may be insignificant."  "A 16- or 31-data point (wavelength) spectrophotometer or a tristimulus colorimeter are more realistic tools for accurately measuring the color of foods," says Brian Tennis, product manager, X-Rite Inc., Grandville, MI.  A tristimulus colorimeter measures wider spectral band widths than a spectrophotometer and concentrates on three major colors: red, green and blue. When connected to a personal computer equipped with the appropriate software, the instrument can provide three-dimensional color plots and other information that is more relevant to the way humans perceive the reflected color of a food product.  But spectrophotometers and colorimeters work better for paints and dyes than they do for food products. The key problem that prevents accurate, reproducible color measurement of foods is that most foods have non-uniform surfaces. This has a profound effect on how light and color are reflected and perceived, according to Frick. For example, the color measurement of a cherry-flavored jelly bean depends on where on the curvature of the jelly bean surface the measurement is made. If the angle of measurement is different from reading to reading, the quantitative color reading will be different. Measuring the color of porous foods, like a piece of cake, is also challenging because when the sample is compressed during the color-measurement process, it turns darker.  "To get reproducible and meaningful color measurements, you need to present the food sample exactly the same way to the instrument," says Frick.  Different types of reflectance spectrophotometers are available, according to Teunis. In one type of spectrophotometer, the 0/45 spectrophotometer, the detector that measures the reflected light is placed at a 45° angle to the incident light source. These instruments work well when the surface of the sample being tested has uniform texture and gloss -- for example, paints and some types of foods. If significant texture or granulation is present on the surface of the sample, however, some reflected light may be scattered at different angles and escape detection by the 0/45-type spectrophotometer.  To compensate for this problem, other types of colorimeters use a spherical geometry which diffusely illuminates samples, eliminating the directionality of the light. In these instruments, the detector is mounted on top of the sphere at 8° from center. One application for spherical geometry-based instruments is measuring the color of granulated powders. Colorimeters that use spherical geometries tend to be more expensive. Deciding which type of instrument to use requires experience.  Frick offers this advice: "Colorimeters are a wonderful tool for measuring the color of foods, but just don't turn them over to a technician and walk away." There is more to making accurate and meaningful color measurements than meets the eye. Trends in instrumentation  Few significant changes or new developments have occurred in recent years with the optical systems of color-measuring instruments. However, DeGroff has identified the following trends and recent developments in color-measurement technology: New software. New software upgrades by color instrument manufacturers allow staff to run programs in the user-friendly Windows operating system. New software versions are more easily networked, and exchanging data with various types of quality control programs is much simpler than it used to be. For example, color-measurement data can be easily exported to QC programs that include other types of quality control data such as pH, fat, moisture, etc. Portability. Instrument manufacturers are now making powerful hand-held colorimeters that can be taken conveniently to the product to make color measurements (e.g., in an apple orchard or on the processing line). They can then be brought back to the lab where the stored data can be extensively analyzed with a computer. Improvements in user-friendliness and accuracy. New-generation instruments are easier to use than older versions. They can be programmed to provide easily understood outputs -- e.g., pass/fail, good/bad, over-ripe/under-ripe, etc. No longer do technicians have to worry about the meaning of esoteric number values. Some instrument manufacturers have taken into account the potential for error when sample presentation is inconsistent. They have compensated for this problem by making measurements at various angles simultaneously and then averaging results. One instrument maker, ColorTec, uses an unusual light source in its hand-held colorimeter: a series of 12 light emitting diodes (LEDs) that cover the entire visible wavelength range. The advantages of LEDs include their reliability and long life, according to the manufacturer. Lower price. Prices of quality color-measurement instruments are dropping dramatically. It is now possible to purchase a reliable, accurate instrument for under $3,000. Increasing demand. One trend noted by DeGroff, Frick and Teunis is that interest by food companies in measuring color has skyrocketed in recent months. With instrument improvements and enhancements increasing and price decreasing, more and more food companies are discovering just how useful and practical color-measuring instruments can be as a quality control tool.   The first internationally accepted method for measuring color was developed by the Commission International de I'Eclairage (CIE) and was known as L*a*b* color tolerancing (also called CIELAB). A later modification to the L*a*b* scale was made by CIE, resulting in the L*C*H* color-tolerancing scale. The scale system currently in vogue, the CMC color-tolerancing scale, was developed by the Colour Measurement Committee of the Society of Dyes & Colourists in Great Britain.  L*a*b* values used in the system are calculated from the tristimulus values (X, Y, Z) which are the backbone of all mathematical color models. The location of a color in the CIE color space is defined by a three-dimensional Cartesian (rectangular) coordinate system. The lightness value (L*) indicated how light or dark the color is. The a* value indicated the position on the red-green axis, and b* is the position on the yellow-blue axis. Once the L*a*b* position of a standard color is determined, a rectangular tolerance "box" can be drawn around the standard.  However, since human visual perceptability is more accurately defined by an ellipse, there are some places in the L*a*b* color space where setting a rectangular box around a color standard can lead to inaccurate color-tolerance results. A rectangular tolerance around the ellipse can give good numbers for unacceptable color. If the tolerance rectangle is made small enough to fit within visual acceptability, it would be possible to get bad numbers for acceptable color.  L*C*H* color-difference calculations are derived from the L*a*b* values. L*a*b* values are converted from the rectangular coordinate system to a cylindrical coordinate system. The L* value is the same as the L* value in the L*a*b* color space and represents the lightness plane on which the color resides. The C* value is the calculated vector distance from the center of the color space to the measured color. Larger C* values indicate higher chroma. H* measures hue. Using the L*C*H* polar coordinate system allows instrumental readings to match more closely the color perception of human observers.  CMC is not a new color space, but rather a tolerancing system. CMC tolerancing -- a modification of CIE -- provides better agreement between visual assessment and instrumentally measured color differences. Briefly, the CMC calculation mathematically defines an ellipsoid around the standard color with the same semi-axis corresponding to hue, chroma and lightness. The ellipsoid represents the volume of acceptance and automatically varies in size depending on the position of the color in the color space.  "CMC is the best system to evaluate human color perception," says Brian Teunis, product manager, X-Rite Inc. According to him, human color perception is more tolorant in the green region but less tolerant in the dark blue region.   *(Source: "A Guide to Understanding Color Tolerancing," X-Rite, Inc., 3100 44th St. SW, Grandville, MI 49418.)Back to top<