Abstract
Sensometrics assesses sensory perspectives in consumer research using statistics and various methodologies. This study evaluated consumer responses to hot and cold germinated-wheat beverages in check-all-that-apply (CATA) and rate-all-that-apply (RATA) assessments using sensometric statistical approaches, including Cochran’s Q test, penalty-lift analysis, and multiple factor analysis. Hot beverages (HB) were prepared by infusion using different amounts of germinated wheat (HB_1: 0.8 g/100 mL, HB_2: 2 g/100 mL, and HB_3: 4 g/100 mL), while cold beverages (CB) were made using cooled boiled germinated wheat with varying concentrations (CB_1: 25 g/L, CB_2: 50 g/L, and CB_3: 75 g/L). Results of the CATA study suggested that consumers preferred HB_1 and CB_1 because they expressed the sensory characteristics associated with liking, including “barley tea flavor”, “neat taste”, and “nutty taste”, while “bitterish taste”, “stuffy taste”, and “astringent taste” were undesirable attributes. “Browning index”, “barley tea odor”, and “nutty taste” showed significant differences (p < 0.05) in both favorable and unfavorable rating scores. Overall, CB_1 elicited a clear taste and odor with fewer negative emotions. These findings demonstrate the usefulness of the sensometric approach combined with CATA and RATA analyses to obtain more easily interpretable results on the sensory perception of consumers to new food products.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13197-023-05884-z.
Keywords: Multiple factor analysis, Regressor vector coefficient, Penalty-lift analysis, Germinated-wheat beverage, Hot and cold beverage
Introduction
Globally, trends in beverage preference are shifting from sugary to functional health and wellness formulations that improves nutritional intake, gastrointestinal tract health, stress relief, and brain health (Tireki 2021). This trend has inspired increasing interest in the development of “functional beverages” containing various bioactive components, including phenols, flavonoids, minerals, vitamins, prebiotics, and probiotics (Maleš et al. 2022). Wheat (Triticum aestivum L.) has been successfully incorporated in cereal-based beverages through various procedures, including steaming, germination, and roasting, and can impart certain beneficial properties, including the enhancement of bioactive constituent levels, such as phenolic and flavonoid compounds, gamma aminobutyric acid, and branched-chain amino acids (Aung et al. 2022a).
Apart from beneficial health properties, the sensory perception of newly developed beverages is key to their success. Generally, the sensory quality of beverages differs depending on the processing conditions, such as temperature and pH, which determine the fate of volatile and non-volatile compounds, ultimately affecting the chemical composition and sensory perception of the final product (Turgut et al. 2022). We previously assessed the dynamic changes in aromatic metabolites associated with different preparation methods for cold and hot germinated-wheat beverages using electronic sensor technology coupled with chemometric analysis (Aung et al. 2022b). In previous study, hot and cold germinated-wheat beverages presented different flavor profiles associated with varying levels of amino acids and bioactive components (Aung et al. 2022b). Although we previously conducted consumer acceptability tests to discriminate consumer preferences for these beverages (Aung et al. 2022a), a detailed sensory profile and mental representation, considering both desirable and undesirable sensory attributes, remains lacking. To fill these gaps and improve the consumer perception of new beverages requires consumer-based sensory methodologies that can differentiate consumer liking and sensory characteristics.
The consumer description of product sensory characteristics can influence the formulation and processing techniques to achieve the desired outcomes in the development of new products (Hunaefi and Marusiva 2021; Cho and Moazzem 2022). Among consumer-based sensory analysis methods, the check-all-that-apply (CATA) and rate-all-that-apply (RATA) methods are simple and rapid sensory profiling techniques that require little cognitive effort (Xia et al. 2020; Pineau et al. 2022). The CATA method comprises versatile and simple questions that are appropriate for naive consumers because they are provided with a predefined list of attributes (Vidal et al. 2019; Hunaefi and Marusiva 2021). In addition, CATA allows for product rating using a liking scale that encompass the consumer’s ideal product based on the list of attributes (Vigneau et al. 2022). In addition to being simple and reliable, CATA is limited in the direct measurement of sensory attribute intensity, in which case RATA is considered an effective alternative approach (Baião et al. 2022). Emotional response to food samples can also be determined through a rating scale to discriminate among samples and improve the prediction of food choice beyond overall acceptance (Rini et al. 2022).
Sensometrics—the statistical assessment of sensory profile data—has attracted attention in the fields of consumer product development in that it provides greater insight into consumer choice and behavior (Qannari 2017). In the statistical analysis of sensory data from rapid profiling procedures, various statistical approaches, such as Cochran’s Q test, Pearson’s χ2 tests, penalty-lift analysis (PA), multiple factor analysis (MFA), and correspondence analysis (CA), can be utilized to accurately obtain the most relevant attributes associated with consumer likes and/or dislikes and optimize the preference profile of a product (Meyners et al. 2013; Qannari 2017; Vigneau et al. 2022). Therefore, the combined use of sensometrics with CATA and RATA sensory profiling is a promising tool in the development of germinated-wheat beverages.
According to our previous studies, we have already explored the compositional changes during beverage preparation including steaming, germinating, roasting, and the subsequent serving methods-both hot and cold (Aung et al. 2022a, b, 2023a). Furthermore, we have undertaken a neuroimaging approach to differentiate the variances in brain waves following the consumption of hot and cold germinated wheat beverages (Aung et al. 2023b). Another neuroimaging study using electroencephalogram (EEG) reported that higher GABA levels can enhance cognitive function (Murthy et al. 2022). Regarding the outcomes of these studies, it has become evident that beverages made from germinated wheat with increased GABA levels due to germination elicit distinct responses in human cognitive function. This disparity can likely be attributed to the divergent compositions inherent to hot and cold servings. Consequently, the sensory perception on these beverages, whether served hot or cold, holds the potential to significantly influence consumers’ sensory and emotional responses. To the best of our knowledge, the measurement of sensory responses using CATA, RATA, and sensometrics have only been compared in the development of germinated-wheat beverages but never combined. Thus, the explicit and implicit measurement of consumer responses to germinated-wheat beverages using a combination of the CATA, RATA, and sensometric techniques represents an innovative approach. In this study, we evaluated the sensory attributes and liking scores of different concentrations of germinated-wheat beverages prepared under hot and cold conditions using the CATA method. Second, the intensity of the sensory attributes was rated using the RATA method. Finally, Cochran’s Q test, PA, MFA, and one-way analysis of variance (ANOVA) were used in the sensometric analysis of the CATA and RATA results. These outcomes would contribute the product quality optimization through a sensometric analytical approach in new germinated-wheat beverage development.
Materials and methods
Sensometric approach
Beverage sample preparation and tasting
Wheat (Triticum aestivum L. from Jinju, Republic of Korea) was germinated at 17.6 °C for 46.18 h, dried, steamed, and roasted at 180 °C for 44.56 min (Aung et al. 2022a, b). Hot and cold beverages were prepared as described earlier (Aung et al. 2022b). Hot beverages were prepared by infusing tea bags filled with different amounts (HB_1: 0.8, HB_2: 2, and HB_3: 4 g) of roasted, steamed, and germinated wheat in 100 mL boiling water for 25 min. Cold beverages were obtained by boiling different amounts (CB_1: 25, CB_2: 50, and CB_3: 75 g) of roasted germinated wheat in 1 L water, followed by cooling to 4 °C. Each of the hot and cold beverages (100 mL) were randomly presented to panelists at 5 min intervals; the panelists were required to rinse their mouths with bottled water during each interval.
Participant recruitment
Voluntary recruitment was carried out among students and staff of the Changwon National University and sensory evaluation was conducted at the sensory lab of the Food and Nutrition Department on April 2021. Ethical approval for the involvement of human subjects in this study was granted by the Institutional Review Board, Reference number 7001066-202002-HR-004. Before the sensory evaluation, panelists were surveyed based on demographic data and consumption patterns, including frequency of intake and ranking of sensory preferences related to cereal-based tea (Table S1). A total of 200 participants (44 males and 156 females) participated in the survey. All participants were over 20 years of age, and the majority of respondents were between 20 and 29 years old. Approximately 34.5% of the participants consumed cereal tea at least once to twice per month, 17.5% reported a rate of once every 6 months, 17% reported a rate of once a week, and 10% reported a rate of once every 3 d. Less than 10% of the participants reported the most and least frequent consumption rates. Only 6% of participants did not drink cereal tea. Preference ranking revealed that taste and odor were important characteristics determining the consumption of cereal teas.
CATA survey
We employed the rapid sensorimotor procedures CATA and RATA to generate an in-depth sensory profile of germinated-wheat beverages. CATA lexicons were selected based on the characteristics of roasted brown rice and Korean traditional tea (Kim et al. 2009; Lee et al. 2010; Kim 2020). CATA ballots with 25 attributes (11 odor and 14 taste terms) were used to assess all terms that apply to each sample. Analysis of CATA datasets was performed using the CATA analysis tool in XLSTAT version 2021.2 (Addinsoft, Paris, France). Cochran’s Q statistic, a non-parametric test, was used for the comparative analysis of binary data (Cochran 1950; Meyners 2016). Correspondence analysis (CA) was applied to determine the differences between each beverage in terms of their sensory profiles based on chi-square distance with significance set at p < 0.05. Principal coordinates analysis (PCoA) was used to assess the distribution of liking scores associated with specific attributes. Penalty-lift analysis (PA) was applied to evaluate the liking scores expressed for different attributes with significance set at p < 0.0001(Ares et al. 2014).
RATA survey
The RATA assessment was conducted to complement the CATA assessment in determining the intensity of CATA attributes at three rating scales, i.e., low, medium, and high (Meyners et al. 2016; Jaeger et al. 2018). The RATA questionnaire was comprised of two sections: favorable and unfavorable sensory characteristics with emotional responses. Each RATA sensory ballot was divided into appearance, odor, and taste. RATA ballots (19 attributes) were generated from previous CATA data (25 attributes) by implementing sensory dimensions to reduce cognitive effort (Baião et al. 2022). Favorable and unfavorable emotions for each beverage were assessed based on 12 emotional responses. The data were analyzed using parametric methods, including ANOVA, F-test, and t-test, because these tests are considered to be reliable in discriminating between products (Meyners et al. 2016). Multiple factor analysis (MFA) was used to assesses the distribution of data associated with each attribute (Meyners et al. 2013). The relationship between RATA attributes was calculated using a regressor vector (Rv) coefficient ranging from 0 to 1. All RATA statistics were calculated using XLSTAT version 2021.2 (Addinsoft).
Results and discussion
CATA
The counts and proportions (%) of CATA terms for odor attributes, including nutty, grass, milk, flower, barley tea, fishy, earthy, grain, burnt, flour, and wood, and taste attributes, including stuffy, gritty, bitter, sweet, neat, sour, tart, bitterish, metallic, artificial, astringent, chemical, refresh, and nutty, are presented in Table 1. Of the 25 attributes, 13 terms showed significant differences between beverages (Cochran’s Q test, p < 0.05) and all beverage samples differed significantly (p < 0.0001). Among the odor attributes, five showed significant differences between beverages, including “grass odor”, “barley tea odor”, “fishy odor”, “burnt odor”, and “flour odor” (Table 1). Interestingly, there was no response for the “milk odor” attribute in any beverage sample. These odor attributes may relate to the processing steps used for beverage preparation. The “grass odor” may be due to the presence of hexanal, which is produced by germination of wheat as reported in our previous study (Aung et al. 2022b). There is a possibility that the roasting of wheat may produce an aroma that is reminiscent of barley tea, leading to the identification of a prominent “barley tea odor” in wheat beverage samples. This similarity in aroma is likely due to the utilization of a similar roasting process in the production of barley tea and wheat beverages (Tatsu et al. 2020). The “burnt odor” may result from roasting, which causes the Maillard reaction to occur, leading to the formation of volatile compounds such as furan (Liu et al. 2022). In addition, “flour odor” and “fishy odor” may be due to the presence of aldehydes, ketones, and esters, which are formed during the roasting process (Aung et al. 2023a). Regarding taste attributes, “stuffy taste”, “bitter taste”, “neat taste”, “tart taste”, “bitterish taste”, “astringent taste”, “refresh taste”, and “nutty taste” varied significantly between beverages. These taste attributes can be attributed to various compounds, including polyphenols, and alkaloids, presence in wheat beverages. The “tart taste”, “astringent taste”, “bitter taste”, and “bitterish taste”, may result from the presence of bitter compounds in wheat due to the roasting of wheat. The “neat taste” and “nutty taste” are often attributed to the presence of amino acids and peptides in wheat beverages, which has already been represented in our articles (Aung et al. 2022b, 2023a).
Table 1.
Counts of check -all- that- apply (CATA) for sensory profile of roasted germinated wheat beverages n (%)
| Hot-brewed beverages | Cold-brewed beverages | ||||||
|---|---|---|---|---|---|---|---|
| HB_1 | HB_2 | HB_3 | CB_1 | CB_2 | CB_3 | ||
| Odor | Nutty odor | 64 (14.10) | 66 (14.13) | 60 (12.93) | 61(13.77) | 68 (14.14) | 64 (11.79) |
| Grass odor* | 8 (1.76) | 7 (1.50) | 14 (3.02) | 9 (2.03) | 6 (1.25) | 14 (2.58) | |
| Milk odor | 0 (0.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| Flower odor | 4 (0.88) | 3 (0.64) | 6 (1.29) | 3 (0.68) | 2 (0.42) | 1 (0.18) | |
| Barley tea odor* | 49 (10.79) | 55 (11.78) | 52 (11.21) | 76 (17.16) | 69 (14.35) | 67 (12.34) | |
| Fishy odor* | 11 (2.42) | 14 (3.00) | 16 (3.45) | 0 (0.00) | 2 (0.42) | 1 (0.18) | |
| Earthy odor | 12 (2.64) | 12 (2.57) | 12 (2.59) | 6 (1.35) | 6 (1.25) | 8 (1.47) | |
| Grain odor | 55 (12.11) | 54 (11.56) | 54 (11.64) | 48 (10.84) | 63 (13.10) | 54 (9.94) | |
| Burnt odor* | 3 (0.66) | 7 (1.50) | 6 (1.29) | 3 (0.68) | 8 (1.66) | 21 (3.87) | |
| Flour odor* | 3 (0.66) | 7 (1.50) | 4 (0.86) | 1 (0.23) | 3 (0.62) | 0 (0.00) | |
| Woody odor | 13 (2.86) | 11 (2.36) | 13 (2.80) | 6 (1.35) | 12 (2.49) | 10 (1.84) | |
| Taste | Stuffy taste* | 17 (3.74) | 21 (4.50) | 27 (5.82) | 17 (3.84) | 21 (4.37) | 34 (6.26) |
| Gritty taste | 9 (1.98) | 8 (1.71) | 12 (2.59) | 2 (0.45) | 5 (1.04) | 9 (1.66) | |
| Bitter taste* | 11 (2.42) | 14 (3.00) | 9 (1.94) | 20 (4.51) | 18 (3.74) | 42 (7.73) | |
| Sweet taste | 22 (4.85) | 21 (4.50) | 21 (4.53) | 20 (4.51) | 17 (3.53) | 18 (3.31) | |
| Neat taste* | 53 (11.67) | 30 (6.42) | 32 (6.90) | 51 (11.51) | 42 (8.73) | 34 (6.26) | |
| Sour taste | 3 (0.66) | 4 (0.86) | 6 (1.29) | 4 (0.90) | 6 (1.25) | 8 (1.47) | |
| Tart taste* | 4 (0.88) | 2 (0.43) | 7 (1.51) | 2 (0.45) | 4 (0.83) | 14 (2.58) | |
| Bitterish taste* | 19 (4.19) | 20 (4.28) | 15 (3.23) | 20 (4.51) | 21 (4.37) | 38 (7.00) | |
| Metallic taste | 1 (0.22) | 2 (0.43) | 2 (0.43) | 2 (0.45) | 3 (0.62) | 1 (0.18) | |
| Artificial taste | 8 (1.76) | 9 (1.93) | 6 (1.29) | 2 (0.45) | 4 (0.83) | 7 (1.29) | |
| Astringent taste* | 15 (3.30) | 19 (4.07) | 23 (4.96) | 12 (2.71) | 25 (5.20) | 37 (6.81) | |
| Chemical taste | 2 (0.44) | 2 (0.43) | 4 (0.86) | 1 (0.23) | 0 (0.00) | 1 (0.18) | |
| Refresh taste* | 21 (4.63) | 17 (3.64) | 11 (2.37) | 21 (4.74) | 13 (2.70) | 9 (1.66) | |
| Nutty taste* | 47 (10.35) | 62 (13.28) | 52 (11.21) | 56 (12.64) | 63 (13.10) | 51 (9.39) | |
HB_1, hot beverage 0.8 g/100 mL; HB_2, hot beverage 2 g/100 mL; HB_3, hot beverage 4 g/100 mL; CB_1, cold beverage 25 g/L; CB_2, cold beverage 50 g/L; CB_3, cold beverage 75 g/L. * indicates significant differences between attributes according to Cochran’s Q test (p < 0.05)
We conducted a CA to assess the distribution of attribute and beverage concentration data (a symmetric plot of CATA attributes is illustrated in Fig. 1a). A higher quality of examination was obtained, as 87.16% of the total inertia was explained by F1 and F2. A bi-plot was classified into four groups: HB_1 and CB_1, HB_2 and HB_3, and CB_3 and CB_2. Panelists reported that HB_1 and CB_1 were associated with a “nutty taste”, “neat taste”, “refresh taste”, and “barley tea odor”, HB_2 and HB_3 were associated with a “fishy odor”, “flour odor”, and “gritty taste”, CB_3 was associated with a “tart taste”, “burnt odor”, “astringent taste”, “stuffy taste”, and “bitterish taste”, and CB_2 was associated with a “bitter taste”. The association of these attributes with liking was determined by PCoA using the correlation coefficients (Fig. 1b). Because “barley tea odor”, “neat taste”, and “refresh taste” were closely related to liking, beverages associated with these attributes would be preferred by consumers. Astringency and bitterness were the least preferred by panelists. Therefore, HB_1 and CB_1, which were characterized by these liking terms, were considered the most preferred beverages based on the CATA assessment. Hot and cold beverages were commonly separated in terms of their sensory attributes, except for HB_1 and CB_1, which is similar to previous findings of sensory intensity and flavor metabolites being determined by preparation temperatures (Aung et al. 2022b).
Fig. 1.
Evaluation of CATA attributes and beverage concentrations according to correspondence analysis, principal coordinate analysis (b), penalty-lift mean impact analysis (c), and mean drop analysis (d). HB_1, hot beverage 0.8 g/100 mL; HB_2, hot beverage 2 g/100 mL; HB_3, hot beverage 4 g/100 mL; CB_1, cold beverage 25 g/L; CB_2, cold beverage 50 g/L; CB_3, cold beverage 75 g/L
PA, which indicates the association of liking and a deviation of each attribute, was performed as a supportive CATA data analysis to determine the mean impact and mean drop in Fig. 1c, d. To identify the main attributes associated with liking, penalty-lift attributes were arranged from highest to lowest in the mean impact penalty chart (Meyners 2016). The significant attributes (p < 0.05) were arranged according to positive or negative impacts. We found that “barley tea odor”, “nutty taste”, and “neat taste” had a positive penalty and “bitterish taste”, “stuffy taste”, and “astringent taste” had a negative penalty (Fig. 1c). “Barley tea odor” was the main positive attribute associated with the liking of the beverage and “astringent taste” was the main negative attribute. We assessed the “must have” attributes by mean drops using an X–Y response graph. Because ideal product identification was not determined in this study, an analysis of the presence and absence of attributes was conducted in mean drops of PA (Fig. 1d). Not surprisingly, “barley tea odor”, “nutty taste”, and “neat taste” were positive drivers of liking, while “astringent taste”, “stuffy taste”, and “bitterish taste” were the main negative drivers. These findings suggest that the panelists preferred HB_1 and CB_1, specifically because of their neat and refresh taste and barley tea odor.
RATA
In the RATA survey, we prepared ballots from CATA responses to be rated according to intensity, and favorable and unfavorable sensory characteristics were displayed along with the respective emotional responses. Ratings of between 1 and 3 were included on the list of sensory attributes, including appearance (1), odor (8), and taste (12) (Table 2), and emotional responses, including favorable (12) and unfavorable emotions (12) (Table 3), for hot and cold beverages. Both parametric and non-parametric approaches were used to interpret the data. We performed a parametric analysis of sensory intensities using a one-way ANOVA through Fisher’s F-test and multiple comparisons using Tukey’s HSD test. The least squares means of favorable and unfavorable sensory intensities of hot and cold beverages are compared in Table 2. For all pairwise comparisons, “browning index”, “barley tea odor”, and “nutty taste” varied significantly (p < 0.05) in both favorable and unfavorable rating scores while “nutty odor” was only significant in favorable attributes and “fishy odor”, “burnt odor”, and “tart taste” were only significant in unfavorable attributes.
Table 2.
Favorable and unfavorable sensory intensity scores of hot and cold beverages rated by participants in RATA analysis
| Sensory attributes | HB_1 | HB_2 | HB_3 | CB_1 | CB_2 | CB_3 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Favor | Disfavor | Favor | Disfavor | Favor | Disfavor | Favor | Disfavor | Favor | Disfavor | Favor | Disfavor | ||
| Appearance | Browning index | 0.684 c* | 0.286 bcd* | 1.031 b* | 0.602 a* | 1.306 b* | 0.459 ab* | 1.204 b* | 0.092 d* | 1.622 a* | 0.153 cd* | 1.796 a* | 0.327 bc* |
| Odor | Nutty odor | 1.204 ab* | 0.214 a | 1.408 a* | 0.112 a | 1.429 a* | 0.143 a | 0.969 b* | 0.184 a | 1.367 a* | 0.102 a | 1.398 a* | 0.173 a |
| Grass odor | 0.133 a | 0.092 a | 0.143 a | 0.143 a | 0.153 a | 0.173 a | 0.184 a | 0.082 a | 0.204 a | 0.102 a | 0.173 a | 0.224 a | |
| Fishy odor | 0.061 a | 0.429 a* | 0.122 a | 0.357 a* | 0.102 a | 0.388 a* | 0.153 a | 0.133 b* | 0.133 a | 0.143 b* | 0.133 a | 0.153 b* | |
| Barley tea odor | 0.847 bc* | 0.102 a | 0.847 bc* | 0.112 a* | 0.755 c | 0.143 a | 1.133 ab* | 0.112 a | 1.429 a* | 0.061 a | 1.296 a* | 0.102 a | |
| Earthy odor | 0.204 a | 0.173 a | 0.224 a | 0.153 a | 0.163 a | 0.143 a | 0.224 a | 0.082 a | 0.184 a | 0.153 a | 0.184 a | 0.184 a | |
| Burnt odor | 0.122 a | 0.122 b* | 0.143 a | 0.143 b* | 0.194 a | 0.153 b* | 0.143 a | 0.092 b* | 0.184 a | 0.112 b* | 0.163 a | 0.306 a* | |
| Grain odor | 0.857 b | 0.112 a | 1.173 ab | 0.071 a | 1.235 a | 0.143 a | 0.837 b | 0.133 a | 1.051 ab | 0.082 a | 0.898 ab | 0.122 a | |
| Woody odor | 0.184 a | 0.194 a | 0.194 a | 0.102 a | 0.143 a | 0.112 a | 0.214 a | 0.082 a | 0.235 a | 0.082 a | 0.224 a | 0.153 a | |
| Taste | Sweet taste | 0.408 a | 0.204 a | 0.582 a | 0.173 a | 0.551 a | 0.204 a | 0.439 a | 0.173 a | 0.612 a | 0.082 a | 0.378 a | 0.112 a |
| Bitter taste | 0.153 a | 0.143 a | 0.204 a | 0.112 a | 0.184 a | 0.092 a | 0.194 a | 0.122 a | 0.194 a | 0.163 a | 0.184 a | 0.163 a | |
| Astringent taste | 0.122 a | 0.276 a | 0.153 a | 0.408 a | 0.204 a | 0.296 a | 0.143 a | 0.398 a | 0.204 a | 0.367 a | 0.163 a | 0.510 a | |
| Nutty taste | 0.908 c* | 0.194 a | 1.286 ab* | 0.102 ab | 1.173 bc* | 0.102 ab | 1.092 bc* | 0.122 ab | 1.531 a* | 0.041 b | 1.071 bc* | 0.133 ab | |
| Clean taste | 1.112 ab | 0.031 a | 0.929 b | 0.071 a | 0.898 b | 0.051 a | 1.388 a | 0.051 a | 1.061 ab | 0.051 a | 1.051 ab | 0.020 a | |
| Stuffy taste | 0.102 a | 0.500 a | 0.112 a | 0.582 a | 0.143 a | 0.571 a | 0.122 a | 0.408 a | 0.184 a | 0.459 a | 0.204 a | 0.684 a | |
| Refresh taste | 0.316 a | 0.102 a | 0.306 a | 0.041 a | 0.378 a | 0.071 a | 0.378 a | 0.031 a | 0.520 a | 0.031 a | 0.276 a | 0.092 a | |
| Tart taste | 0.112 a | 0.143 b* | 0.112 a | 0.235 b* | 0.082 a | 0.163 b* | 0.122 a | 0.194 b* | 0.133 a | 0.327 ab* | 0.122 a | 0.439 a* | |
| Bitterish taste | 0.092 b | 0.184 b | 0.163 ab | 0.153 b | 0.163 ab | 0.163 b | 0.194 ab | 0.184 b | 0.184 ab | 0.316 ab | 0.286 a | 0.378 a | |
| Gritty taste | 0.061 a | 0.245 ab | 0.071 a | 0.224 ab | 0.071 a | 0.173 ab | 0.102 a | 0.082 b | 0.071 a | 0.235 ab | 0.102 a | 0.276 a | |
HB_1, hot beverage 0.8 g/100 mL; HB_2, hot beverage 2 g/100 mL; HB_3, hot beverage 4 g/100 mL; CB_1, cold beverage 25 g/L; CB_2, cold beverage 50 g/L; CB_3, cold beverage 75 g/L. Values are LS means by ANOVA, n = 98. Different small letters indicate significant differences between beverages for each favorable and unfavorable sensory attribute according to all pairwise comparison (p < 0.05)
Table 3.
Favorable and unfavorable emotional responses for hot and cold beverages rated by participants in RATA analysis
| Emotional responses | HB_1 | HB_2 | HB_3 | CB_1 | CB_2 | CB_3 | |
|---|---|---|---|---|---|---|---|
| Just don't like | 6.67 | 6.00 | 5.66 | 4.88 | 4.76 | 4.48 | |
| Strong taste | 26.67 | 8.00 | 15.09 | 36.59 | 26.19 | 16.42 | |
| Strange | 8.89 | 18.00 | 11.32 | 2.44 | 7.14 | 10.45 | |
| Provocative | 6.67 | 6.00 | 0.00 | 0.00 | 0.00 | 7.46 | |
| Favorable | Artificial | 2.22 | 6.00 | 3.77 | 7.32 | 7.14 | 7.46 |
| Detrimental to health | 2.22 | 0.00 | 1.89 | 0.00 | 0.00 | 1.49 | |
| Feel bad | 2.22 | 8.00 | 7.55 | 7.32 | 0.00 | 5.97 | |
| Don't want to drink again | 15.56 | 12.00 | 11.32 | 17.07 | 11.90 | 11.94 | |
| Burdensome | 4.44 | 6.00 | 3.77 | 0.00 | 7.14 | 10.45 | |
| Dissatisfied | 13.33 | 14.00 | 13.21 | 7.32 | 16.67 | 8.96 | |
| Awkward | 2.22 | 4.00 | 11.32 | 7.32 | 9.52 | 5.97 | |
| Not harmonious | 8.89 | 12.00 | 15.09 | 9.76 | 9.52 | 8.96 | |
| Just good | 4.76 | 6.92 | 5.06 | 7.57 | 4.69 | 5.73 | |
| Familiar | 22.45 | 21.38 | 22.47 | 29.19 | 25.00 | 29.30 | |
| New | 2.04 | 1.26 | 2.81 | 1.08 | 1.56 | 1.27 | |
| Incentive | 0.00 | 0.00 | 0.56 | 0.00 | 0.00 | 0.00 | |
| Unfavorable | Luxurious | 1.36 | 1.89 | 2.25 | 0.54 | 0.00 | 0.00 |
| Be good for health | 12.24 | 13.84 | 9.55 | 8.65 | 11.98 | 10.83 | |
| Feel better | 5.44 | 6.92 | 6.18 | 2.16 | 3.13 | 3.18 | |
| Drink more | 3.40 | 5.03 | 6.18 | 7.57 | 7.29 | 7.01 | |
| Not burdensome | 18.37 | 13.21 | 11.80 | 19.46 | 19.79 | 17.83 | |
| Satisfied | 6.12 | 6.29 | 10.11 | 6.49 | 6.25 | 8.92 | |
| Plain | 19.05 | 18.87 | 16.85 | 14.59 | 16.67 | 14.01 | |
| Harmonious | 4.76 | 4.40 | 6.18 | 2.70 | 3.65 | 1.91 |
HB_1, hot beverage 0.8 g/100 mL; HB_2, hot beverage 2 g/100 mL; HB_3, hot beverage 4 g/100 mL; CB_1, cold beverage 25 g/L; CB_2, cold beverage 50 g/L; CB_3, cold beverage 75 g/L
MFA was performed as a non-parametric approach to compare the sensory characteristics and emotional responses with respect to their ratings. In MFA, the MFA Rv coefficient is a measure used to determine the correlation between two sets of variables, ranging from 0 to 1, with a higher value indicating a stronger correlation (Josse et al. 2008; Kennedy 2010). Table 4 shows the Rv coefficients of the sensory attributes and emotion scores obtained through the MFA analysis. The MFA Rv scores of odor, taste, and emotion were very high and comparable for both favorable (0.943, 0.901, and 0.838, respectively) and unfavorable aspects (0.915, 0.840, and 0.948, respectively). When comparing individual attributes and emotions, the association between appearance and taste attributes and emotion showed low Rv values for both aspects. In contrast, higher Rv scores were observed between odor and emotion for both favorable and unfavorable aspects. This indicates that participant emotions were affected more by beverage odor than taste and appearance.
Table 4.
Rv coefficient of favorable and unfavorable RATA scores by MFA
| Appearance | Odor | Taste | Emotion | MFA | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Favorable | Unfavorable | Favorable | Unfavorable | Favorable | Unfavorable | Favorable | Unfavorable | Favorable | Unfavorable | |
| Appearance | 1.000 | 1.000 | 0.577 | 0.346 | 0.579 | 0.225 | 0.398 | 0.617 | 0.724 | 0.619 |
| Odor | 0.577 | 0.346 | 1.000 | 1.000 | 0.809 | 0.824 | 0.804 | 0.818 | 0.943 | 0.915 |
| Taste | 0.579 | 0.225 | 0.809 | 0.824 | 1.000 | 1.000 | 0.644 | 0.704 | 0.901 | 0.840 |
| Emotion | 0.398 | 0.617 | 0.804 | 0.818 | 0.644 | 0.704 | 1.000 | 1.000 | 0.838 | 0.948 |
| MFA | 0.724 | 0.619 | 0.943 | 0.915 | 0.901 | 0.840 | 0.838 | 0.948 | 1.000 | 1.000 |
In Fig. 2, the MFA plots present 71.11% and 58.01% of the variability of the first two dimensions for favorable and unfavorable RATA ratings, respectively. The favorite sensory attributes and emotions were clearly discriminated between beverages (Fig. 2a), and the observed variables with the corresponding projected points are displayed in Fig. 2b. HB_2 and HB_3 were positively loaded in F2 and associated with “sweet taste”, “grain odor”, and “nutty odor”. In addition, positive emotions, such as “incentive, new, harmonious, feel better, luxurious, plain, be good for health” corresponded with HB_1 and HB_2. Participants rated the “earthy odor” in HB_1 as their favorite odor. However, according to the projected points, the emotion for HB_1 corresponded with “be good for health”. The taste attributes of CB_2 corresponded with “astringent taste”, “nutty taste”, refresh taste”, “bitter taste”, “stuffy taste”, and “bitterish taste”; the odor attributes corresponded with “burnt odor” and “grass odor” as well as higher “browning index”; and the emotions corresponded with “satisfied” and “drink more”. For CB_1 and CB_3, the odor attributes included “fishy odor”, “barley tea odor”, and “woody odor”; taste attributes included “gritty taste”, “tart taste”, and “clean taste”; and emotions included “just good”, “not burdensome”, and “familiar”. The appearance of CB_3 and HB_3 was projected towards the “browning index” location and was negatively correlated to the appearance of CB_1 (Fig. 2b). Based on the ratings for favorable responses, our findings indicate that the highest germinated-wheat concentration showed the highest browning index, and hot beverages exerted more positive emotions on panelists than cold beverages did.
Fig. 2.
Multiple factor analysis (MFA) of the correlations between beverage concentrations and appearance, odor, taste, and emotional response determined by the RATA survey. Variable loading plot and projected points based on observations of favorable (a, b) and unfavorable attributes (c, d). HB_1, hot beverage 0.8 g/100 mL; HB_2, hot beverage 2 g/100 mL; HB_3, hot beverage 4 g/100 mL; CB_1, cold beverage 25 g/L; CB_2, cold beverage 50 g/L; CB_3, cold beverage 75 g/L
The corresponding loading of variables and projected points of observations for unfavorable RATA ratings are displayed in Fig. 2c, d (F1 and F2: 58.01%). Unfavorable responses for all hot beverages were comparable and included “detrimental to health”, “feel bad”, “not harmonious”, and “just don’t like” emotions. Cold beverages were distributed separately in each loading plot of CB_1 on the negative side of F1, and CB_2 and CB_3 were segregated on each side of F2. Negative emotions for CB_1 included “strong taste” and “do not want to drink again” and were rated as “nutty odor”, “grain odor”, and “clean taste”. Unfavorable emotions (“artificial” and “awkward”) elicited by CB_2 were influenced by “bitterish taste”, “bitter taste”, and “astringent taste”, which were negative attributes in the CATA analysis. Panelists segregated CB_3 according to “tart taste”, “gritty taste”, “stuffy taste”, “refresh taste”, “burnt odor”, “grass odor”, “earthy odor”, and “woody odor” attributes for the emotional responses including “burdensome”, “provocative”, “dissatisfied”, and “strange”. These results indicate that CB_1 was associated with a clear taste and odor, which in turn elicited fewer negative emotions.
Conclusion
This study evaluated the sensory characteristics and emotional perception of hot and cold germinated-wheat beverages using the rapid sensory analysis methods CATA and RATA. The interpretation of the complex dataset was performed using a sensometric approach to elucidate the consumer perception of the attribute profile and emotional representation of each product. The CATA approach clearly discriminated the liked and disliked attributes of each product, identifying “barley tea odor”, “nutty taste”, and “neat taste” as positive drivers for liking and “astringent taste”, “stuffy taste”, and “bitterish taste” as negative drivers. The participants preferred HB_1 and CB_1 because of the contributions of these positive liking drivers. The RATA approach described the favorable and unfavorable attributes coupled with emotional responses for each beverage. CB_1 received the best rating for taste and odor and elicited fewer unfavorable emotions from consumers. Collectively, our study findings provide valuable insights into the sensory attributes and consumer perception of germinated-wheat beverages based on a combined approach of CATA and RATA surveys interpreted using sensometric analytical techniques.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
Not applicable.
Authors’ contributions
TA: Methodology, Validation, Formal analysis, Investigation, Writing—original draft, review & editing, Visualization. B-RK Methodology, Validation, Formal analysis, Investigation, Visualization. M-JK: Conceptualization, Writing—original draft, review & editing, Supervision, Project administration.
Funding
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2018R1D1A1B07049760). This research was funded by the Financial Program for Self-Directed Research Capacity in 2022.
Data availability
Not applicable.
Code availability
XLSTAT software version 2021.2 (Addinsoft, Paris, France) was used for data analysis.
Declarations
Conflicts of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Ethical approval
Ethical approval for the involvement of human subjects in this study was granted by the Institutional Review Board, Reference number 7001066-202002-HR-004.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- Ares G, Dauber C, Fernández E, et al. Penalty analysis based on CATA questions to identify drivers of liking and directions for product reformulation. Food Qual Prefer. 2014;32:65–76. doi: 10.1016/j.foodqual.2013.05.014. [DOI] [Google Scholar]
- Aung T, Kim BR, Kim MJ. Optimized roasting conditions of germinated wheat for a novel cereal beverage and its sensory properties. Foods. 2022;11:481. doi: 10.3390/foods11030481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aung T, Kim BR, Kim MJ. Comparative flavor profile of roasted germinated wheat (Triticum aestivum L.) beverages served hot and cold using electronic sensors combined with chemometric statistical analysis. Foods. 2022;11:3099. doi: 10.3390/foods11193099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aung T, Kim BR, Kim MJ. Comparative volatiles, amino acids, and phenolic compounds and characteristics of roasted germinated wheat (Triticum aestivum L.) during beverage preparation. LWT. 2023;173:114412. doi: 10.1016/j.foodres.2022.110976. [DOI] [Google Scholar]
- Aung T, Kim BR, Kwak HS, Kim MJ. Neuroimaging approach: effects of hot and cold germinated wheat beverages on electroencephalographic (EEG) activity of the human brain. Foods. 2023;12(18):3493. doi: 10.3390/foods12183493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Baião LF, Rocha C, Lima RC, et al. Development of a Rate-All-That-Apply (RATA) ballot for sensory profiling of sea urchin (Paracentrotus lividus) gonads. Food Res Int. 2022;153:110976. doi: 10.1016/j.foodres.2022.110976. [DOI] [PubMed] [Google Scholar]
- Cho S, Moazzem MS. Recent applications of potentiometric electronic tongue and electronic nose in sensory evaluation. Prev Nutri Food Sci. 2022;27(4):354–364. doi: 10.3746/pnf.2022.27.4.354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cochran WG. The comparison of percentages in matched samples. Biometrika. 1950;37:256. doi: 10.2307/2332378. [DOI] [PubMed] [Google Scholar]
- Hunaefi D, Marusiva W. Sensory profile of 3 in 1 instant coffee using emotional-sensory mapping, flash profile, and CATA (Check-All-That-Apply) methods. J Teknol Dan Ind Pangan. 2021;32:169–180. doi: 10.6066/jtip.2021.32.2.169. [DOI] [Google Scholar]
- Jaeger SR, Lee SM, Kim KO, et al. CATA and RATA questions for product-focused emotion research: five case studies using emoji questionnaires. Food Qual Prefer. 2018;68:342–348. doi: 10.1016/j.foodqual.2018.04.001. [DOI] [Google Scholar]
- Josse J, Pagès J, Husson F. Testing the significance of the RV coefficient. Comput Stat Data Anal. 2008;53:82–91. doi: 10.1016/j.csda.2008.06.012. [DOI] [Google Scholar]
- Kennedy J. Evaluation of replicated projective mapping of granola bars. J Sens Stud. 2010;25:672–684. doi: 10.1111/j.1745-459X.2010.00302.x. [DOI] [Google Scholar]
- Kim H, Cho M, Lee K. Studies on the sensory characteristics of korean wine and imported wine. J Korean Soc Food Cult. 2009;19:593–602. [Google Scholar]
- Kim MK. Sensory profile of rice-based snack (Nuroongji) prepared from rice with different levels of milling degree. Foods. 2020;9:685. doi: 10.3390/foods9060685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lee B-Y, Kim M-H, et al. Influences of roasted or non-roasted brown rice addition on the nutritional and sensory properties and oxidative stability of sunsik, Korean heated cereal powder. Korean J Food Cook Sci. 2010;26:872–886. [Google Scholar]
- Liu S, Sun H, Ma G, et al. Insights into flavor and key influencing factors of Maillard reaction products: a recent update. Front Nutr. 2022;9:2014. doi: 10.3389/fnut.2022.973677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Maleš I, Pedisić S, Zorić Z, et al. The medicinal and aromatic plants as ingredients in functional beverage production. J Funct Foods. 2022;96:105210. doi: 10.1016/j.jff.2022.105210. [DOI] [Google Scholar]
- Meyners M. Testing for differences between impact of attributes in penalty-lift analysis. Food Qual Prefer. 2016;47:29–33. doi: 10.1016/j.foodqual.2014.11.001. [DOI] [Google Scholar]
- Meyners M, Castura JC, Carr BT. Existing and new approaches for the analysis of CATA data. Food Qual Prefer. 2013;30:309–319. doi: 10.1016/j.foodqual.2013.06.010. [DOI] [Google Scholar]
- Meyners M, Jaeger SR, Ares G. On the analysis of Rate-All-That-Apply (RATA) data. Food Qual Prefer. 2016;49:1–10. doi: 10.1016/j.foodqual.2015.11.003. [DOI] [Google Scholar]
- Murthy SV, Fathima SN, Mote R. Hydroalcoholic extract of ashwagandha improves sleep by modulating GABA/histamine receptors and EEG slow-wave pattern in in vitro-in vivo experimental models. Prev Nutri Food Sci. 2022;27(1):108–120. doi: 10.3746/pnf.2022.27.1.108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pineau N, Girardi A, Lacoste Gregorutti C, et al. Comparison of RATA, CATA, sorting and Napping® as rapid alternatives to sensory profiling in a food industry environment. Food Res Int. 2022;158:111467. doi: 10.1016/j.foodres.2022.111467. [DOI] [PubMed] [Google Scholar]
- Qannari EM. Sensometrics approaches in sensory and consumer research. Curr Opin Food Sci. 2017;15:8–13. doi: 10.1016/j.cofs.2017.04.001. [DOI] [Google Scholar]
- Rini L, Lagast S, Schouteten JJ, et al. Impact of emotional state on consumers’ emotional conceptualizations of dark chocolate using an emoji-based questionnaire. Food Qual Prefer. 2022;99:104547. doi: 10.1016/j.foodqual.2022.104547. [DOI] [Google Scholar]
- Tatsu S, Matsuo Y, Nakahara K, et al. Key odorants in Japanese roasted barley tea (Mugi-Cha)—differences between roasted barley tea prepared from naked barley and roasted barley tea prepared from hulled barley. J Agric Food Chem. 2020;68:2728–2737. doi: 10.1021/acs.jafc.9b08063. [DOI] [PubMed] [Google Scholar]
- Tireki S. A review on packed non-alcoholic beverages: Ingredients, production, trends and future opportunities for functional product development. Trends Food Sci Technol. 2021;112:442–454. doi: 10.1016/j.tifs.2021.03.058. [DOI] [Google Scholar]
- Turgut SS, Entrenas JA, Taşkın E, et al. Estimation of the sensory properties of black tea samples using non-destructive near-infrared spectroscopy sensors. Food Control. 2022;142:109260. doi: 10.1016/j.foodcont.2022.109260. [DOI] [Google Scholar]
- Vidal L, Antúnez L, Ares G, et al. Sensory product characterisations based on check-all-that-apply questions: further insights on how the static (CATA) and dynamic (TCATA) approaches perform. Food Res Int. 2019;125:108510. doi: 10.1016/j.foodres.2019.108510. [DOI] [PubMed] [Google Scholar]
- Vigneau E, Cariou V, Giacalone D, et al. Combining hedonic information and CATA description for consumer segmentation. Food Qual Prefer. 2022;95:104358. doi: 10.1016/j.foodqual.2021.104358. [DOI] [Google Scholar]
- Xia Y, Song J, Lee PY, et al. Impact of consumption frequency on generations of sensory product profiles using CATA questions: case studies with two drink categories. Food Res Int. 2020;137:109378. doi: 10.1016/j.foodres.2020.109378. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Not applicable.
XLSTAT software version 2021.2 (Addinsoft, Paris, France) was used for data analysis.