A new Coffee Brewing Control Chart relating sensory properties and consumer liking to brew strength, extraction yield, and brew ratio.

The classic Coffee Brewing Control Chart (BCC) was originally developed in the 1950s. It relates coffee quality to brew strength and extraction yield, and it is still widely used today by coffee industry professionals around the world to provide guidance on the brewing of coffee. Despite its popularity, recent experimental studies have revealed that sensory attributes and consumer preferences actually follow much more complicated trends than those indicated by the classic BCC. Here, we present a methodology to synthesize the results of these recent studies on drip-brewed coffee to generate new versions of the BCC: a new Sensory BCC that displays a broad array of statistically significant sensory attributes across typical total dissolved solids and percent extraction ranges, a new Consumer BCC that highlights the existence of two preference clusters with different likes and dislikes across those ranges, a new Sensory and Consumer BCC that combines both sensory descriptive and consumer preferences on the same chart, and a more streamlined BCC that omits consumer preferences and focuses on the overarching sensory descriptive trends. The new BCCs provide more accurate insight on how best to brew coffee to achieve desired sensory profiles. PRACTICAL APPLICATION: Through the manipulation of yield and extraction parameters, the new Sensory and Consumer Coffee Brewing Control Chart presented here can be used by brewers of drip coffee to design coffees with specific sensory profiles and match the preferences of different consumer types.

Acids in coffee: A review of sensory measurements and meta-analysis of chemical composition.

Coffee contains a variety of organic acids (OAs) and chlorogenic acids (CGAs) that contribute to overall sensory properties. Large variations in preparation and measurement methodology across the literature complicate interpretation of general trends. Here, we perform a systematic review and meta-analysis of the published literature to elucidate the concentrations of OAs and CGAs in both Coffea arabica (arabica) and Coffea canephora (robusta), for both green coffee and roasted coffee at multiple roast levels. A total of 129 publications were found to report acid concentration measurements, yielding 8,634 distinct data points. Analysis of the full data set reveals several trends. First, roasted robusta has considerably more acidic compounds than arabica with 2 to 5 times as much total OAs, and much larger amounts of formic and acetic acid. As for CGAs, in both arabica and robusta 5-CQA is the major component, and progressive roasting decreases the concentration of all CGAs. The total amount of CGA present was more dependent on roast level than the type of coffee (arabica vs. robusta). Overall, this meta-analysis suggests that the increases in certain OAs with roast level might play more of a role in the sensory profile of dark roast coffees than previously suspected.

The effect of post-brew holding time and carafe type on the sensory properties of drip brew coffee.

BACKGROUND: Coffee quality is believed to degrade quickly after brewing, and retail establishments discard unsold brewed coffee after a specified holding time period, sometimes as short as 30 min. We used trained sensory panels to evaluate the flavor profiles of light, medium, and dark roast coffees held in three different carafe types (glass on hot plate, thermal jacket, and vacuum insulated) for times ranging from 15 min to 3 h. Furthermore, a panel of 93 coffee-industry professionals performed a blind evaluation of fresh (30 min) versus held (180 min) coffee for overall liking and attribute level adequacy. RESULTS: Sourness increased over time, consistent with acidity increasing over time (i.e., higher titratable acidity, lower pH), but only for the light and medium roasts. Dark roasted coffee became significantly more acidic over 3 h post-brew but was not perceived as more sour over time by the sensory panel. Variations were observed between the thermal jacket and vacuum carafes for the light and dark roast, but few differences were observed with storage type in the medium roast. Surprisingly, the panel of coffee industry professionals showed no preference for fresh over held. CONCLUSIONS: More sensory attributes decreased than increased over time, suggesting that the primary concern with loss of quality during coffee holding may be the loss of volatile aroma compounds. Hedonic ratings suggest that even if the changes over time are noticeable, they may not negatively impact overall liking. © 2022 Society of Chemical Industry.

Impact of beverage temperature on consumer preferences for black coffee.

We recently performed a systematic investigation of consumer preferences for black coffee versus key brewing parameters, including total dissolved solids, extraction yield, and brewing temperature (Cotter et al. in J Food Sci 86(1):194-205, 2021. https://doi.org/10.1111/1750-3841.15561 ). An experimental goal in that work was for participants to taste the coffee at a beverage temperature of 65 °C, but the large sample size of more than 3000 individual tastings, combined with natural variations in the brewing and cooling processes, meant that coffees were assessed over a normally distributed range of temperatures between 56 and 71 °C. Here we use those data to provide a more detailed analysis of the impact of beverage temperature on consumer acceptance of the coffee, with a key objective of identifying beverage temperatures at which no consumers assess the coffee either as too hot or too cold. Using a 5-point just-about-right (JAR) scale, we find that a majority of consumers (> 50%) assessed the temperature as JAR at all temperatures tested up to 70 °C. A substantial fraction of consumers, approximately 6-12%, assessed the coffee as too cold over the range 56-68 °C. Only above 70 °C did a majority of consumers assess the coffee as too hot and none assessed it as too cold, albeit with 40% still assessing it as JAR. Complementary analyses indicate that beverage temperature over this range had little impact on assessments of the adequacy of flavor intensity, acidity, and mouthfeel, but did correlate slightly with overall liking and purchase intent. Overall, the results suggest that temperatures over the range of 58-66 °C maximize consumer acceptance, and that 68-70 °C is the minimum temperature range at which no consumers will assess black coffee as too cold.

Sensory Analysis of Full Immersion Coffee: Cold Brew Is More Floral, and Less Bitter, Sour, and Rubbery Than Hot Brew.

Cold brew coffee is often described as sweeter or less acidic than hot brew coffee. Such comparisons, however, are potentially confounded by two key effects: different brew temperatures necessarily change the extraction dynamics and potentially alter the resulting brew strength, and different consumption temperatures are well known to affect perceived flavor and taste. Here, we performed a systematic study of how extraction temperature affects the sensory qualities of full immersion coffee. The investigation used a 3 × 3 × 3 factorial design, with coffee from three different origins representing different post-harvest methods (washed, honey-processed, and wet-hulled), each roasted to three different levels (light, medium, and dark), and each brewed at three different temperatures (4 °C, 22 °C, and 92 °C). All coffees were brewed to equilibrium, then diluted to precisely 2% total dissolved solids (TDS) and served at the same cold temperature (4 °C). We find that four attributes exhibited statistically significant variations with brew temperature for all origins and roast levels tested, with bitter taste, sour taste, and rubber flavor all higher in hot brewed coffees, and floral flavor higher in cold brewed coffee. However, there were strong interactions with origin and roast, with several additional attributes significantly impacted by temperature for specific origins and roast levels. These results provide insight on how brew temperature can be used to modulate the flavor profile of full immersion coffee.

Roast level and brew temperature significantly affect the color of brewed coffee.

Beverage color significantly affects perceived sensory quality and consumer preference. Although the color of coffee beans is well known to vary strongly with roast level, little work has examined how roast level and brewing conditions affect the color of the final beverage. Here, we report that the color of full immersion brewed coffee is significantly affected by both roast level and brewing temperature. Coffees from three different origins were each roasted to three different levels (light, medium, and dark) and then brewed at three different temperatures (4, 22, and 92°C). Each sample was brewed toward full extraction and then diluted to precisely 2% total dissolved solids so that differences in concentration would not confound color measurements. Absorbance spectra (UV-vis) and color tristimulus values (L*a*b*) were then collected and analyzed. We find that roast level had the strongest impact on brew color, and that brew temperature had a significant impact on color for light and medium roasts, with less impact on dark roasts. Qualitatively, the cold brewed coffees tended to be redder, while the hot brewed coffees were blacker. The results suggest that there is an opportunity to manipulate and brand brewed coffee color through judicious choices of roast level and brewing temperature. PRACTICAL APPLICATION: Color serves as an indicator of coffee quality and potentially could affect perceived sensory characteristics. Our results suggest that appropriate control of roast level and brew temperature could yield desired colors for novel coffee products.

Food Choice and Waste in University Dining Commons-A Menus of Change University Research Collaborative Study.

The purpose of this multi-campus research was to investigate the relationships of food type and personal factors with food choice, consumption, and waste behaviors of college students at all-you-care-to-eat dining facilities. The amount of food taken and wasted was indirectly measured in units relative to the plate size using before and after photos taken by the diners themselves. Animal protein and mixed dishes (e.g., stir fry, sandwich) took up more of diners' plate space and these items were correlated to both greater hedonic appeal as well as a higher likelihood of the item being pre-plated. Greater confidence in liking an item before choosing it was correlated to a larger portion being taken. Finally, increased satisfaction with the meal and frequency of visiting the dining commons was correlated to less food waste. Understanding these potential food choice drivers can help dining facilities better target healthier meals to diners while reducing food waste.

Mothers' Perceptions and Attitudes towards Children's Vegetable Consumption-A Qualitative, Cross-cultural Study of Chilean, Chinese and American Mothers Living in Northern California.

This exploratory research focused on the cultural variables involved in children's vegetable consumption, through the analysis of mothers' perceptions, attitudes, and feeding practices regarding their children's intake, using qualitative consumer research methods. Twelve focus groups of mothers with children between 2-12 years old (Euro-Americans n = 20, Chinese n = 19, and Chilean n = 19) were conducted. All participants lived in Northern California, had higher education, and incomes that did not limit their vegetable purchase. Intercultural differences in vegetable preferences and consumption habits were found. Mothers across all groups agreed on the importance of children's vegetable consumption, the influence that mothers have over their children's vegetable intake, and how challenging it is to get children to eat a variety of vegetables. The ethnic groups differed regarding how they perceived the level of mothers' responsibility over children's vegetable intake, the way that mothers defined the amount of vegetables that children should eat, the constraints that mothers had on increasing their children's vegetable intake and mothers' recommendations to encourage vegetable consumption. Our study suggests that under similar socio-economic and parental education levels, culture-specific strategies should be considered to foster healthy dietary habits in children.

Consumer preferences for black coffee are spread over a wide range of brew strengths and extraction yields.

Brewing is the final and key step in the production of the coffee beverage. Extraction related metrics such as the total dissolved solids (TDS), percentage extraction yield (PE) of solutes, and brew temperature (BT) are widely believed to govern the flavor and corresponding consumer acceptance of the resulting brew, as summarized in the industry standard "Coffee Brewing Control Chart." In this study, we investigated how the three factors of TDS, PE, and BT affected consumer acceptance of a medium roast, single-origin coffee and whether consumer preference segmentation would be observed based on these variables. A cohort of 118 mostly college-age, self-reported consumers of black coffee tasted coffees that varied in BT, TDS, and PE. For each coffee, consumers rated overall acceptance on the 9-point hedonic scale; the adequacy of serving temperature, flavor intensity, acidity, and mouthfeel using 5-point just-about-right (JAR) scales; and described the flavor using a check-all-that-apply list of 17 attributes. Cluster analysis revealed two consumer segments whose preferences varied most strongly with TDS. Response surface methodology relating liking to TDS and PE produced dome- and saddle-shaped surfaces for the two segments, respectively. External preference mapping and penalty analysis indicated that overall flavor intensity as well as acidity heavily influenced the preferences of the two clusters. The Coffee Brewing Control Chart's "ideal" coffee should therefore be reconsidered to reflect consumer preference segmentation. PRACTICAL APPLICATION: This research informs the way coffee brewers manipulate brew strength and extraction of drip brew coffee for optimal consumer acceptance; and justifies a reform of the standard "Coffee Brewing Control Chart" in its representation of an "ideal" coffee as we uncovered two consumer preference segments with different positive and negative sensory drivers of liking.

Brew temperature, at fixed brew strength and extraction, has little impact on the sensory profile of drip brew coffee.

The brew temperature is widely considered a key parameter affecting the final quality of coffee, with a temperature near 93 °C often described as optimal. In particular, drip brewers that do not achieve a minimum brew temperature of 92 °C within a prescribed time period fail their certification. There is little empirical evidence in terms of rigorous sensory descriptive analysis or consumer preference testing, however, to support any particular range of brew temperatures. In this study, we drip-brewed coffee to specific brew strengths, as measured by total dissolved solids (TDS), and extraction yields, as measured by percent extraction (PE), spanning the range of the classic Coffee Brewing Control Chart. Three separate brew temperatures of 87 °C, 90 °C, or 93 °C were tested, adjusting the grind size and overall brew time as necessary to achieve the target TDS and PE. Although the TDS and PE both significantly affected the sensory profile of the coffee, surprisingly the brew temperature had no appreciable impact. We conclude that brew temperature should be considered as only one of several parameters that affect the extraction dynamics, and that ultimately the sensory profile is governed by differences in TDS and PE rather than the brew temperature, at least over the range of temperatures tested.

The Dessert Flip: Consumer preference for desserts with a high proportion of fruit and nuts.

American adults consume excess added sugar and saturated fat, and not enough fruit. A possible solution is the "Dessert Flip," a shift in which the proportions of more sustainable plant-based garnishes, such as fruit and nuts, are increased, while the proportion of conventional dessert is reduced. We hypothesized that the pleasing properties of fruits and nuts-color, variety, flavor, and texture-would allow the Dessert Flip to be liked as well or better than a conventional dessert by college-aged students in a dining hall. In a sensory test, participants (n = 118) significantly preferred plated flipped desserts-those containing 80% by weight fruit (strawberries, raspberries, blueberries, and mangoes) and nuts with 20% cake-over the conventional versions of those desserts with 80% cake and 20% fruit and nuts. No significant differences were found in overall liking. In a second experiment, after eating an entrée, participants (n = 150) chose pictures of multiple varieties of flipped cakes with 60% fruit and 40% cake significantly more than pictures of conventional 80% cake for dessert. Based on the U.S. average fruit prices, the Dessert Flip can be equal or lower in price per serving compared to an ungarnished typically sized cake. We conclude that the Dessert Flip offers a feasible option for increasing servings of fruits and nuts in desserts as well as reducing added sugar and saturated fat while maintaining sensory appeal. Future dietary guidelines could include this general strategy as a recommendation for helping adults include more fruits in their diets. PRACTICAL APPLICATION: The Dessert Flip offers an appetizing way for foodservice to promote healthier eating by increasing servings of fruits and nuts in their desserts while reducing added sugar and saturated fat. Consumers can try this strategy at home by cutting a smaller piece of cake and adding fruit on the side.

Sensory, physicochemical and volatile compound analysis of short and long shelf-life melon (Cucumis melo L.) genotypes at harvest and after postharvest storage.

Flavor is a key attribute defining melon fruit quality and driving consumer preferences. We characterized and compared fruit ripening patterns (ethylene, respiration), physicochemical properties (rind/flesh color, firmness, soluble solids, acidity), aroma volatiles, and flavor-related sensory attributes in seven melon genotypes differing in shelf life capacity. Fruits were evaluated at optimal maturity and after storage for six days at 5 °C plus one day at room temperature. Total volatile content increased after storage in all genotypes, with esters being dominant. Shorter shelf-life genotypes, displaying a sharper climacteric phase, correlated with fruity/floral/sweet flavor-related descriptors, and with esters, sulfur-containing compounds and a terpenoid. Longer shelf-life types were associated with firmness, green and grassy aroma/flavor and aldehydes. Multivariate regression identified key volatiles that predict flavor sensory perception, which could accelerate breeding of longer shelf-life melons with improved flavor characteristics.

The Salt Flip: Sensory mitigation of salt (and sodium) reduction with monosodium glutamate (MSG) in "Better-for-You" foods.

We tested the hypothesis that reduced-salt versions of four "better-for-you" dishes enhanced with monosodium glutamate (MSG) through a "Salt Flip" in an amount that still substantially reduced total sodium matched the consumer acceptance of normal-salt versions. Three versions each-standard recipe with normal salt, reduced salt, and reduced salt with MSG, of four dishes-roasted vegetables (RV), quinoa bowl (QB), savory yogurt dip (SD), and pork cauliflower fried rice (CR) were evaluated by 163 consumers for overall liking and liking of appearance, flavor, and texture/mouthfeel on the nine-point hedonic scale, preference, adequacy of flavor, saltiness, and aftertaste on just-about-right (JAR) scales, likeliness to order, and sensory characteristics by check-all-that-apply. For each dish, the MSG recipe was liked the same (or significantly more for SD, P < 0.05) than the standard recipe, and better than the reduced salt recipe for QB and CR. The same was true of likeliness to order. MSG recipes of QB and SD were significantly preferred to the standard recipes, with no difference for RV and CR. MSG recipes were consistently described as "delicious," "flavorful," and "balanced." Penalty-lift analysis showed that "delicious," "flavorful," "balanced," "fresh," and "savory"; and "bland," "rancid," and "bitter," were positive and negative drivers of liking, respectively. Two of three uncovered preference clusters, accounting for 68% of consumers, consistently liked MSG recipes, and the same or more so than standard recipes. We conclude that MSG can successfully be used to mitigate salt and sodium reduction without compromising consumer acceptance of better-for-you foods. PRACTICAL APPLICATION: The Salt Flip offers a promising dietary sodium reduction strategy through the addition of monosodium glutamate (MSG) to reduced-salt, savory, better-for-you foods that does not compromise consumer acceptance of their sensory profile.

Effects of brew strength, brew yield, and roast on the sensory quality of drip brewed coffee.

Drip brewed coffee is traditionally quantified in terms of its strength, also known as total dissolved solids (TDS), and its brewing yield, also known as percent extraction (PE). Early work in the 1950s yielded classifications of certain regimes of TDS and PE as "underdeveloped," "bitter," or "ideal," with the modifiers "weak" or "strong" simply correlated with TDS. Although this standard is still widely used today, it omits a rich variety of sensory attributes perceptible in coffee. In this work, we used response surface methodology to evaluate the influence of TDS and PE on the sensory profile of drip brewed coffee. A representative wet-washed Arabica coffee was roasted to three different levels (light, medium, or dark), with each roast then brewed to nine target brews that varied systematically by TDS and PE. Descriptive analysis found that 21 of the 30 evaluated attributes differed significantly across the brews for one or more experimental factors, yielding linear or second-order response surfaces versus TDS and PE. Seven attributes exhibited a significant response surface for all three roast levels tested: burnt wood/ash flavor, citrus flavor, sourness, bitterness, sweetness, thickness, and flavor persistence. An additional seven attributes also showed a significant response surface fit across some but not all roasts. Importantly, sweetness exhibited an inverse correlation with TDS irrespective of roast, while dark chocolate flavor and blueberry flavor decreased with TDS for medium roast. These results provide new insight on how to optimize brewing conditions to achieve desired sensory profiles in drip brewed coffee. PRACTICAL APPLICATION: This research provides guidance on how best to achieve specific flavor profiles in drip brewed coffee.

Sensory and monosaccharide analysis of drip brew coffee fractions versus brewing time.

BACKGROUND: The composition of drip brew coffee versus brewing time has been chemically characterized in previous studies, and it is known that the total dissolved solids (TDS) systematically decreases with each fraction during the brew. Little information exists regarding the corresponding sensory attributes versus time, however, and it is unclear how TDS correlates with flavor profile. RESULTS: Standard drip brews were fractionated into distinct samples by switching in an empty carafe every 30 s during the brew. Using a trained sensory descriptive panel, we found that most taste and flavor attributes decreased with brew time; for example, the earlier fractions were systematically more bitter and more sour than later fractions. Surprisingly, however, several flavor and taste attributes increased in time; for example, later fractions were systematically sweeter and more floral than earlier fractions. Since later fractions had lower TDS, these results indicate that perceived sweetness in drip brew coffee is negatively correlated with TDS. Mass spectrometry measurements of the monosaccharide content in the brews showed that none of the fractions had perceptible concentrations of any monosaccharide. CONCLUSION: The results of the sensory analysis and the monosaccharide analysis suggest that perceptible sweetness in coffee is a consequence of masking effects and/or the presence of sweet-associated aromas and flavors. The results further suggest that unique flavor profiles could be obtained from the same coffee grounds by judicious combinations of specific fractions. © 2020 Society of Chemical Industry.

Chemical and sensory analysis of commercial Navel oranges in California.

Seven lots of commercially available Navel oranges grown in California were evaluated with flavoromic, metabolomic, sensory descriptive analysis, and consumer testing techniques to identify sensory and chemical drivers of liking. Eight identified chemical clusters related to numerous sensory attributes and consumer preferences. Differences in adult and child preferences led to the discovery of six consumer clusters (four adult and two child). Sweetness, overall flavor, sourness, fruity flavor, and juiciness were identified as the main sensory drivers of liking for the consumers. Fructose, glucose, and proline were among the compounds that best explained perceived sweetness while sourness was correlated with citrate and ascorbate. Perceived fruity flavor increased with higher concentrations of ethanol. We conclude that consumers differ in their preferences for Navel oranges and desire fruit that is higher in both sweetness and sourness.

Evaluation of California-Grown Blood and Cara Cara Oranges Through Consumer Testing, Descriptive Analysis, and Targeted Chemical Profiling.

The acceptability of Moro, Tarocco, Cara Cara, Shahani, Bream Tarocco, Boukhobza, and Sanguinelli oranges from both commercial and research orchards was tested with adult (n = 152) and child (n = 72) consumers. Qualitative focus groups were also conducted to understand consumer familiarity and thoughts about the fruit. Sensory descriptive and chemical analyses were carried out to identify drivers of liking. Overall, consumers preferred the lighter colored varieties consisting of Tarocco, Cara Cara, and Boukhobza. One cluster of adults (n = 80) showed preferences towards sweet and fruity flavors and away from sourness and citric acid. The second adult cluster (n = 72) was tolerant of the sour fruit but did not like fruit high in bitterness and flavonoid content. The largest child cluster (n = 42) showed preferences for samples higher in orange and tropical flavors (Cara Cara, Tarocco, and Boukhobza varieties). The appearance of the Cara Cara was strongly liked by the consumer population in both quantitative and qualitative settings. Hunter scale a color values strongly correlated to the higher berry/dried fruit flavors, and concentrations of naringenin. Focus group participants noted that they were relatively unfamiliar with blood oranges. Growers and producers may want to invest in the lighter colored varieties, such as Cara Cara, Tarocco, Boukhobza and Shahani, as these were liked by a majority of consumers and were low in less desirable sensory characteristics, such as bitterness and sourness. PRACTICAL APPLICATION: Through consumer tests, sensory evaluation, and chemical analyses, this research uncovered which sensory properties may drive consumer acceptance of blood and Cara Cara oranges, and informed potential production and marketing strategies for increasing their consumption. This information should benefit the citrus industry as a whole and may enhance the use of specialty oranges by the food, beverage, and food service industries.

Sensory Profiling and Consumer Acceptance of Pasta, Bread, and Granola Bar Fortified with Dried Olive Pomace (Pâté): A Byproduct from Virgin Olive Oil Production.

An olive pomace (pâté) obtained from virgin olive oil production, was used for the fortification of pasta, bread, and granola bar. For each food, a control (without pâté) and a fortified sample (with pâté, 7% in pasta and 5% in bread and granola bar) were manufactured. Descriptive analysis showed that pâté strongly affected the appearance of pasta and bread and increased the bitterness of bread and granola bar but not pasta. Granola bar was less affected in general, likely because of its higher ingredient complexity. In a central location test with 175 Californian consumers, both the control and the fortified samples of all three foods were well accepted overall, with only the mean liking of the appearance of the fortified pasta falling below the "neither like nor dislike" mark. Approximately 30% of consumers preferred the fortified sample over the control for each food and 50% were willing to pay more for the fortified products. The percentage of phenols from pâté recovered in the prepared samples was such that 63 g of pasta, 18 g of bread, and 12 g of granola bar would be sufficient to meet the EFSA health claim for olive oil phenols. This study demonstrates that pâté can be used for fortification of foods for human consumption, thus adding potential economic value to the virgin olive oil production chain and allowing for a higher daily intake of phenols from Olea europaea L., whose beneficial health properties are well recognized. PRACTICAL APPLICATION: The dried olive oil pomace (pâté) that we developed and tested in this research can be used to fortify pasta, bread, and granola bars with health-beneficial phenols with only slight alterations of their sensory profiles and slight reduction in consumer acceptance. Virgin olive oil producers can use this byproduct and gain further economic value from olive oil production.

A cross-cultural analysis of children's vegetable preferences.

The successful promotion of vegetable consumption by children requires a deep understanding of children's vegetable preferences as well as the factors shaping them throughout childhood. This study analyzed children vegetable liking in four different age ranges (5-6, 7-8, 9-10 and 11-12 years old) in Chile, China and the United States. Three hundred and eighty-four children completed this study. All participants tasted and rated 14 different vegetables for liking and described the samples using Check-All-That-Apply (CATA). We found significant differences in degree of overall liking among children from the three countries (p < 0.001). Specifically, children in China gave higher overall liking scores than children in the US, and in the US higher than in Chile. Child age and gender did not influence children's vegetable overall liking across the three countries. Across all countries and age groups, liking of taste and texture were the best predictors of children overall liking. The penalty analysis of CATA selections by children showed that the mean impact of the attributes that children used to describe the samples on their liking varied among countries, with the descriptors having the least impact on liking for Chinese children.