Color affects the judgmentof food, the perception of freshness, the intensity of flavor, andthe overall acceptability of food. Dubose, Cardello, and Maller(1980) have shown that one's attitude towards and one's choice offood is greatly influenced by its color. A person's first judgment offood is based on color, which then leads to the decision of whetheror not to consume that particular food. In a study conducted by TheGood Housekeeping Institute, 29% of their readership ranked goodappearance or color as important in making a decision about aproduct, second only to freshness (Good Housekeeping, 1984).
Knowing that food coloringis important to consumers, psychologists such as Dubose et al. (1980)have conducted research and found that colors can affect a change inthe flavor of a product. However, as Stillman (1993) points out, "Theinfluence of color on flavor is interesting, because light reflectedfrom a tastant does not directly influence any of the nerve fibersactivated by the chemical and textural properties of food or drink"(Stillman, 1993, p 810). Nevertheless, there have been numerousstudies that show that the occurrence of light waves bouncing offedible objects can in fact have a profound effect on how one tastesfood stimuli. Alley and Alley (1998) show that specific coloring,color intensity and color-related expectations can modify the impactof color on taste. Clydesdale (1993) does specify that the effect ofcolor on flavor most likely results from learned associations, ratherthan innate knowledge. Furthermore, color may be more prominentperceptually than flavor because color generates a stronger visualneural response than flavor generates a gustatory neural response(Oram, Laing, Hutchinson, Owen, Rose, Freeman, et al.,1995).
Solids andLiquids
When looking at how colorinfluences taste, it is important to study if solids are perceiveddifferently from liquids, or if cooler products are perceiveddifferently from warmer products. A study conducted by DuBose et al.(1980) had participants rate the overall acceptability, coloracceptability, flavor acceptability, and flavor intensity for 44samples of liquid and solid stimuli. The experimenters tested 16orange flavored beverages and 16 cherry flavored beverages. The foodstimuli were 12 cake samples that were a combination of four yellowcolor levels, and three lemon flavor levels. The results showed thatthe perceived intensity in beverages increases as color increases forboth the orange and the cherry beverages. This held true in thecolored but flavorless samples where participants rated a moreintense color as a more intense flavor as well, despite the fact thatthere was no flavor. This seems to also apply in solid samples wherethe perceived lemon taste increased when the yellow additive wasincreased (Dubose et al., 1980). This indicates that when tastingfood products where color and flavor are both present, color may bethe dominant influencer on the perception of taste. This experimentshows that the judgment of flavor intensity is directly affected bycolor level in both solid and liquid products. Clydesdale (1993)found the same results in fruit flavored beverages and cake as well.
These studies did notmeasure, however, if the same levels of sweetness were perceiveddifferently between liquids and solids. Alley and Alley (1998) arguethat liquids may produce more intense sensations than solids becauseliquids can cover more of the taste receptors faster and morethoroughly than solids can. The commercial world has taken this intoconsideration and added more sweeteners to solid sweets than toliquid sweets (Alley & Alley, 1998).
Temperature
Another factor that affectsthe color influence on flavor intensity is the temperature of thestimuli. Research has shown that cold temperature foods tend todecrease one's sensitivity to sucrose (Hyman, 1983). Additionally,taste sensitivity was prominently increased in low concentrations ofsucrose as opposed to higher levels because "the various functionsthat relate perceived sweetness to concentration for varioustemperatures converged" (Bartoshuk, Rennert, Rodin, & Stevens,1982). Since jellybeans are used in the current study, the sucroselevel is high but they are tasted at room temperature; therefore,sensitivity should be relatively high.
FlavorIdentification
Identification of flavorsbased on color has been tested to determine how well participants canidentify flavors in a variety of color-flavor combinations. Specificcolor-flavor combinations have also been studied to determine howsweet the different colors are perceived to be. Various studies haveshown that matched color flavor combinations are important inidentifying flavors and are perceived to be more intense in flavor.Clydesdale's study (1993) showed that when colored food had matchingcolor and flavor combinations they were perceived as having astronger intensity than mismatched color flavor combinations(Clydesdale, 1993). Further, Oram et al. (1995) conducted researchwhich showed that when beverages are color-masked or mismatchedadults are worse at identifying the beverage's flavor. Additionally,the participants' identification is biased in the direction of theflavor that matches the color. In essence, if a participant is givena red-colored drink that contains some amount of grape flavoring, theparticipant is more likely to perceive that drink as some sort of redflavor (i.e. cherry) rather than a purple flavor. This bias is moreevident when the flavor is less distinctive (Oram et al., 1995).Philipsen, Clydesdale, Griffin, & Stern (1995) also found thatthe absence of color could greatly reduce or eliminate one's abilityto identify flavor. Furthermore, familiar food items may have agreater impact on the sensory judgment of color, resulting in greaterintensities than non-familiar food items (Alley & Alley, 1998).In general, red colored food is perceived as sweet cherry orstrawberry, yellow and green are perceived as sour/citrus tasting,and blue coloring is characteristic of sweet foods (Alley &Alley, 1998).
Taster Type
One area of taste perceptionthat has yet to be considered as an influence of color on tasteintensity is that of the different types of tasters. Research hasshown that a person is a taster or a non-taster to certain bittercompounds, and that this status is inherited (Tepper, Christensen,& Cao, 2001). The bitterness of PROP probably follows anincomplete dominant pattern; non-tasters have 2 recessive alleles andperceive the least, super tasters probably have 2 dominant allelesand taste the most, while tasters probably have 1 dominant and 1recessive allele and taste an intermediate degree (Bartoshuk, 2000).PROP (6-n-propylthiouracil) is a chemical relative tophenylthiocarbamide (PTC), which produces a bitter taste, and is usedto determine taster type (Bartoshuk, Duffy, & Inglis, 1994). PROPis normally tested with NaCl as the control substance since NaClintensity has been shown to remain stable across taster types(Bartoshuk et al., 1994; Tepper et al., 2001).
A person is defined as ataster or a non-taster by using the taste detection threshold forPROP (Ly & Drewnowksi, 2001). Research has shown thatapproximately seventy per cent of the US adult Caucasian populationcan taste PROP which categorizes them as tasters, while about thirtyper cent cannot taste PROP, thus categorizing them as non-tasters(Tepper et al., 2001). Within the taster type there are alsosupertasters, who have a lower threshold (high sensitivity) for PROP,and regular tasters who have a higher threshold (lower sensitivity)for PROP (Drewnowski, Henderson, Shore, & Barratt-Fornell, 1997).
Tasters are also generallymore sensitive than non-tasters to various compounds such as benzylalcohol, caffeine, potassium chloride, sucrose, and saccharin (Tepperet al., 2001). However, the research is inconclusive as to whether ornot the different taster types actually perceive sweet tastesdifferently (Drewnowski, Henderson, Shore, & Barratt-Fornell,1997; Bartoshuk, 1979; Drewnowski, Henderson, & Barratt-Fornell,1997; Ly & Drewnowski, 2001; Tepper et.al, 2001). The researchwhich indicates that the taster types respond differentially to sweetstimuli do support that this is due to super-tasters' highersensitivity to sucrose and saccharin water solutions than non-tasters(Drewnowski, et al., 1998)
Hypotheses
In the current study sixteenflavors and colors of jellybeans, half exotic and half typicalflavors, will have different color intensities, to determine theeffect color intensity has on flavor intensity between the tastertypes. Also, half of the participants will receive labeled jellybeansand half will receive unlabeled jellybeans. There are severalhypotheses adopted in the present study. The first hypothesis statesthat the non-tasters will rate the jellybeans as lower in flavorintensity. The second hypothesis states that non-tasters will be moreinfluenced by color-enhanced foods than super tasters, and thus willrate lighter and white colored jellybeans as less intense than theyrate the medium or darker shades. Non-tasters will also be lesslikely to be able to determine the non-labeled jellybeans, while thesuper tasters will have a higher likelihood of correctly assigning alabel. One general hypothesis is that all participants who receivelabeled jellybeans will rate higher intensity than those that aregiven masked jellybeans.
Participants
Participants will berecruited from the SFASU student psychology pool using sign upsheets. There will be 96 participants with an age range of 18 yearsto 24 years. There will be a representation of ethnicity and an equalnumber of males and females in each taster type. Upon completion ofthe experiment each participant will be given a sample bag of JellyBelly Jellybeans for their participation.
Materials
In accordance with Tepper's2001 study, the taste stimulus used to determine taster status willbe a solution of PROP with a concentration of .32 mmol/l and asolution of NaCl with a concentration of .1 mol/l. The solutions willbe prepared in advance by dissolving the powder in spring water onmildly heated hot plate, and served at room temperature (Tepper etal., 2001).
A questionnaire and a pencilwill be supplied with each sample. The questionnaire will ask theparticipants' name, whether the current sample is Solution 1 (PROP)or Solution 2 (NaCl) and the rating of the sample on the Green scale.The Green scale, developed by Green, Schaffer, & Gilmore (1993)will be used to determine the perceived intensity of the solutions.The Green scale is an adjective-labeled vertical line that has itsupper limit as "strongest imaginable" taste and the lower limit is"barely detectable". The participant is to rate the intensity of thecurrent taste compared with all other tastes found in daily lifealong this vertical line. Unlike other magnitude scales, the Greenscale controls for ceiling effects by using the "strongestimaginable" adjective as the upper limit, thus not limiting one to afixed number range (Bartoshuk, 2000).
A separate questionnairewill be used for tasting jellybeans. Questionnaires and a pencil willbe provided on each desk for each individual participant. Everyquestionnaire will state the participants number (as explained in theprocedure) and a sample number which corresponds to a number on theDixie cup of each sample, as well as the Green scale for theparticipant to rate flavor intensity (as described on the previouspage). The only difference in the questionnaires will be whether ornot the flavor is stated on the questionnaire. The questionnaire forthe non-labeled group will ask the participant to identify the flavorwhile the questionnaire for the labeled group will state the flavorof each sample.
White Dixie cups will beused to supply each participant with samples of the PROP and NaClsolutions, the jellybeans and spring water to drink between eachsample.
Eight different colored andflavored Jelly Belly Jellybeans will be used as the taste stimuli.Table 1 shows the flavor/color combinations that will be used.
Table 1
Color Flavor Combinations
Color Typical Exotic Red Cherry Cinnamon Orange Orange Cantaloupe Yellow Lemon Banana Green Apple Watermelon Blue Blueberry Cotton Candy Purple Grape Wild Blackberry Pink Strawberry Bubble Gum Black Licorice Chocolate Pudding
First Day: DeterminingTaster Type
The experiment will takeplace over two days. The first day of the experiment will be used todetermine taster type. Taster type will be determined using a methoddetermined by Tepper et al. (2001) that uses one level of PROP andone level of NaCl. They have shown that this method is as effectivein determining taster types as normally employed methods which usethree levels of PROP (Tepper et al., 2001). The current study willuse the same instructions as Tepper et al. (2001) shownbelow.
"You will rate the intensityof each solution by placing a mark on the labeled scale that bestdescribes what you are experiencing. You can use any part of the linescale that seems appropriate for judging intensity. In making yourjudgments of intensity, you should rate the solution relative to thestrength of all sensations you have experienced in your mouth ineveryday life. Thus, 'strongest imaginable' refers to the mostintense sensation you have experienced putting food and non-fooditems in your mouth. This includes such varied taste and mouth feelsensations that come from hot and cold foods an beverages, spices,and spicy foods, toothpaste, mouthwash, medicines etc (Tepper et al.,2001, p573)."
Spring water will beprovided for each participant to rinse their mouth before the PROPand NaCl solutions. The participants will be told to place the entiresample in their mouths, expectorate the sample into an empty Dixiecup, and then rate its intensity on the Green scale. The participantswill not swallow the sample as research has shown that salivaproduced by food may have an effect on the taste and intensityperception of stimuli (Guinard, Zoumas-Morse & Walchak, C.,1997). Subjects will be given one sample at a time (to allow for timebetween samples), as well as water before each sample. Uponcompletion of tasting and rating both samples the participants willbe asked to return the next day at the same time. Taster type will bedetermined by visually comparing the data (Tepper et al.,2001).
Second Day: Using ColorEffect on Flavor Using Jellybeans
On the second day of theexperiment each participant will be given a number based on his orher taster type (example: 1n1A: first non-taster in group 1 receivingorder A). The participant will then be escorted to a partitioned deskthat corresponds to their number. The participants will be dividedinto eight groups with four non-tasters, four tasters, and foursupertasters in each group (see Table 2). Four groups will receiveunlabeled jellybeans and four groups will receive labeled jellybeans.Each of these groups will then be divided so that one labeled and oneunlabeled group will receive jellybeans that are either white, lightcolored, medium colored, or dark colored. Each of these 8 groups willreceive eight differently colored jellybeans of which there will befour typical flavors and four exotic flavors that are red, orange,yellow, green, blue, purple, pink, or black. To ensure that not allparticipants receive jellybeans in the same flavor or color order theflavors and colors will be rotated within each group. In each group,two of each taster type will receive the flavors rotated betweentypical and exotic. However, one will begin with red typicaljellybeans and progress through the colors as stated above towardsblack exotic jellybeans (Order A1) and the other will begin with ablack typical jellybean and end with a red exotic jellybean (OrderA2). The other two of each taster type will receive the flavorsrotated between exotic and typical. One will begin with black exoticjellybeans and progress towards red typical jellybeans (Order B1),while the other will begin with a red exotic jellybean and end with ablack exotic jellybean (Order B2) (See Table 3).
Table 2
Color Flavor Order and GroupAssignments
One of each taster typereceives one of the following within each group.
Order A1 Order A2 Order B1 Order B2 Cherry Licorice Chocolate Pudding Cinnamon Cantaloupe Bubble Gum Strawberry Orange Lemon Grape Wild Blackberry Banana Watermelon Cotton Candy Blueberry Apple Blueberry Apple Watermelon Cotton Candy Wild Blackberry Banana Lemon Grape Strawberry Orange Cantaloupe Bubble Gum Chocolate Pudding Cinnamon Cherry Licorice
Table 3
The Groups will be assignedas follows (4 of each taster type per group):
Group 1 Group 2 Group 3 Group 4 Labeled Labeled Labeled Labeled White Light Colored Regular Colored Dark Colored Group 5 Group 6 Group 7 Group 8 Unlabeled Unlabeled Unlabeled Unlabeled White Light Colored Regular Colored Dark Colored
When all participants havebeen seated they will be given a consent form to read and sign, aswell as a demographic form to complete. All participants will be toldthe same directions as a group. The directions will be as follows:
"You will taste differentflavors of jellybeans. Simply chew the jellybean, but do not swallowit. Spit it back into the Dixie cup when you are finished, and thentake a drink of water from the other Dixie cup. You will be given 1jellybean at a time and a cup of water with each sample. Be sure todrink the water between each sample. You will also be given a formthat asks relevant questions pertaining to the sample. I will collectboth cups and the form after each sample. Please proceed at your ownpace. Any questions?"
Again, to reduce the effectsaliva has on taste and intensity perception, the participants willnot swallow the sample (Guinard, Zoumas-Morse & Walchak, C.,1997).
The participants will begiven one cup of water to drink before the first sample is given.Each participant will be given samples of jellybeans one at a time,as determined by their group, a cup of water to drink after thesample is rated, and the questionnaire. The groups given labeledjellybeans will rate the flavor accuracy and the intensity level,while the groups given unlabeled jellybeans will determine the flavorand rate the intensity of flavor. After each sample is completed thenext sample will be distributed individually by the experimenter.Upon completion of the eight samples the experiment will be finished,and the participant will receive a sample bag of Jelly BellyJellybeans for their participation.
Results will be analyzedusing a 4(color intensity) X 3 (taster type) X 2 (flavor type) X 2(label) between subjects ANOVA. The color intensities are white,light colored, medium colored, and dark colored. The taster types arenon-taster, taster, and super-taster. The flavor types are defined astypical or exotic flavors. These two flavor types were used to ensurethe results were similar across the two types. No hypotheses werestated and thus no results are concluded for this variable, as it wasused to diminish confounding effects. The last independent variableis whether the jellybean is labeled or not labeled with the jellybean flavor. The two dependent variables are the individual ratingsof intensity measured in the number of centimeters from the x-axiswhere the maximum is 100 cm, and the number of correct labels ofunidentified stimuli.
The hypotheses stated forthe current study will be supported. Non-tasters will rate theintensity of the jellybeans (M = 32.5) much lower than the tasters (M= 55.75) or the supertasters (M = 63.5). In general, this may be dueto the idea that non-tasters have less sensitivity to certain taste,thus this may influence how the non-tasters perceive the sweet tasteof the jellybeans. If a certain compound is not tasted, then theentire stimuli can be distorted. Supertasters have the ability totaste a variety of compounds and thus the flavor of the stimuli willbe complete, giving rise to a higher intensity. Complete taste versusincomplete taste may influence the way in which these two tastertypes perceive stimuli.
The hypothesis thatnon-tasters rate lighter shades of jellybeans as less intense thandarker shades more so than do tasters or super-tasters will also besupported. Again, this may be due to the idea that non-tasters do nottaste as many compounds as tasters do. Thus, non-tasters may usetheir visual system to aid in determining the intensity of thestimuli. Therefore, a non-taster will perceive a lighter coloredjellybean as less intense than the tasters will because their visualsystem perceives less intensity.
Further, once intensityreaches a certain level very little can actually have a significantimpact on intensity. In this case, super-tasters already have ahigher sensitivity to certain foods so they will be less influencedby the colors of the jellybeans. However, since non-tasters have amuch lower intensity rating they are more vulnerable to outsideinfluences on taste, such as color.
The hypothesis thatnon-tasters will have less ability to determine the non-labeledjellybeans, while the super tasters will have a higher likelihood ofassigning a correct label will also be supported. Since non-tastersdo not perceive all the compounds tasters perceive, their ability tocorrectly label unidentified stimuli is decreased. The results willshow that non-tasters labeled the jellybeans incorrect more oftenthan tasters or super-tasters did. When one does not receive all theinformation possible for an unidentified stimulus, the ability tolabel that stimulus specifically may be hindered. Therefore, theresult that non-tasters will not able to correctly identify as manystimuli as tasters is reasonable.
Across all subjects thelabeled jellybeans received higher intensity ratings than non-labeledjellybeans. This is in accordance with past research which indicatesthat unlabeled stimuli have lower intensity ratings because humanshave a difficult time identifying taste when no color is present(Clydesdale, 1993; Oram et al., 1993; Philipsen, et al., 1995). Whenboth color and taste are present, visual and taste receptors are bothactivated suggesting that more sense receptors are activated andtherefore provide more information to the brain. When the brain hasmore information it can utilize top-down processing to assess thecurrent situation.
In conclusion, past researchon the influence of color on taste perception has been extensive.However, the current study is able to assess whether different tastertypes will be differentially influenced by various shades of colorsfor the same flavor. In essence, non-tasters should more influencedby color than super tasters, and are less able to identify the flavorwhen the jellybean was not labeled. Additionally, the non-labeledjellybeans will be rated as less intense than the labeled jellybeans.Outside of the lab this in turn suggests that food manufacturers needto increase colorants so that the foods appear more intense to ensurethat taste intensity levels are fairly even across taster types.Additionally, specific labeling of flavors will increase theintensity of the flavor. This research produces a greaterunderstanding of the impact of physical state and color on thejudgment of foods and beverages. Thus, it is beneficial in improvingfood products, and allows psychologists a better understanding of howsense receptors influence perception.
Alley, R.L., & Alley,T.R. (1998). The influence of physical state and color on perceivedsweetness. The Journal of Psychology, 132(5), 561-568.
Bartoshuk, L.M. (2000).Psychophysical advances aid the study of genetic variation in taste.Appetite 34(1), 105.
Bartoshuk, L. M. (1979).Bitter taste of saccharin: Related to the genetic ability to tastethe bitter substance 6-n-Propylthiouracil (PROP). Science 205,934-935.
Bartoshuk, L. M., Duffy, V.B. & Miller, I. J. (1994). PTC/PROP tasting: Anatomy,psychophysics, and sex effects. Physiology & Behavior 56(6),1165-1171.
Bartoshuk, L. M., Rennert,K., Rodin, K. & Stevens, J. C. (1982). Effects of temperature onthe perceived sweetness of sucrose. Physiology & Behavior 28(5),905-910.
Clydesdale, F. M. (1993).Color as a factor in food choice. Critical Reviews in FoodScience andNutrition, 33(1), 83-101.
Drewnowski, A., Henderson,S. A., Barratt-Fornell, A. (1998). Genetic sensitivity to6-n-Propylthiouracil and sensory responses to sugar and fat mixes.Physiology & Behavior 63(5), 771-777.
Drewnowski, A, Henderson,S.A., Shore, A. B., & Barratt-Fornell, A. (1997).
Nontasters, tastes, andsupertasters of 6-n-propylthiouracil (PROP) and hedonic response tosweet. Physiology & Behavior 62(3), 649-655.
DuBose, C.N., Cardello,A.V., & Maller, O. (1980). Effects of colorants and flavorants
on identification, perceivedflavor intensity, and hedonic quality of fruit-flavored beverages andcake. Journal of Food Science, 45, 1393-1399,1415.
Good Housekeeping Institute.(May, 1984). Consumer food and nutrition study: A good housekeepinginstitute report. New York: Consumer Research Department.
Green, B. G., Shaffer, G. S.& Gilmore, M. M. (1993). A semantically-labeled magnitude scaleof oral sensation with apparent ratio properties. Chemical Senses 18,683-702.
Guinard, J., Zoumas-Morse,C. & Walchak, C. (1997). Relations between parotid
saliva flow and compositionand the perception of gustatory and trigeminal stimuli in foods.Physiology & Behavior 63(1), 109-118.
Hyman, A. (1983). Theinfluence of color on the taste perception of carbonated water
preparations. Bulletin ofthe Psychonomic Society, 21(2), 145-148.
Ly, A., & Drewnowksi, A.(2001). PROP (6-n-Propylthiouracil) tasting and sensory responses tocaffeine, sucrose, newhesperidin dihydrochalcone and chocolate.Chemical Senses 26, 41-47.
Oram, N., Laing, D. G.,Hutchinson, I., Owen, J., Rose, G., Freeman, M., & Newell, G.(1995). The influence of flavor and color on drink identification bychildren and adults. Developmental Psychobiology, 28(4),239-246.
Philipsen, D. H.,Clydesdale, F. M., Griffin, R.W., & Stern, P. (1995). Consumerage affects response to sensory characteristics of a cherry flavoredbeverage. Journal of Food Science, 60(2), 364-368.
Stillman, J. A. (1993).Color influences flavor identification in fruit-flavored beverages.Journal of Food Science, 58(4), 810-812.
Tepper, B.J., Christensen,C.M., & Cao, J. (2001). Development of brief methods to classifyindividuals by PROP taster status. Physiology & Behavior 73 (4),571-577.