*3.4. Comparison of Facial Expression Recognition (FER) Method*

Facial expressions of consumers in response to the different yogurt products were measured using the FaceReaderTM software. Tables 4 and 5 show the mean values for the eight facial emotion terms, together with the valence and arousal estimates for Western and Asian consumers, respectively. No significant differences were observed among the FER attributes for Western consumers, whereas the emotions 'surprised' and 'disgusted' were significantly different (*p* < 0.05) for Asian consumers, where 'surprised' was rated highest for soy and lowest for berry, whereas 'disgusted' was rated highest for berry yogurt.

'smiling face ' and 'savoring delicious food ' (cookies); 'fearful ' (berry). The reference, drinkable, soy and cookies yogurts were related to positive emojis, whereas berry was related to negative emojis, by both the cultural groups. The emojis 'unamused ' and 'disappointed ' were similarly rated by both cultures for all the yogurt

products. The berry yogurt was chosen to be more 'pensive face ' by Western con‐ sumers as compared to the Asians, and the cookies yogurt was related to 'relieved face ' more often by Western consumers as compared to Asians, as seen by the Fisher's

The correspondence analysis biplot for Western consumers (Figure 2a) indicates that 86.93% of the variability observed could be attributed to the first two principal components (F1: 76.42%, F2: 10.51%). Meanwhile, a biplot for Asian consumers (Figure 2b) found that 89.76% variability (F1: 68.94%, F2: 20.83%) could be attributed to these components. Both yogurt products, coconut and berry, were related to more negative emojis by Asian as well as Western consumers, and the cookies was placed in the posi‐ tive region by both cultural groups. In the case of Western participants, the reference, drinkable and soy yogurts were placed in between the positive and negative emotional areas, whereas Asian participants related the drinkable with positive emojis, and the ref‐


**Table 4.** Mean values for the biometric emotions for Western consumers.

NS indicates no significant differences in that column (*p* > 0.05) by the Fisher's least square difference test. The data are presented as mean ± standard deviation.

**Table 5.** Mean values for the biometric emotions for Asian consumers.


a,b Means with different superscripts in each column indicate significant differences (*p* < 0.05) by the Fisher's least square difference test. NS indicates no significant differences in that column (*p* > 0.05). Highest value indicated with 'a' in superscript. The data are presented as mean ± standard deviation.

The PCA biplot (Figure 3) shows that 86.35% of the variability in Western consumers' responses could be attributed to the two principal components (F1: 50.25%, F2: 36.11%) and 73.28% of the variability could similarly be attributed to Asian consumers (F1: 47.52%, F2: 25.76%), respectively. 'Happy' was positively related to 'disgusted' and was negatively related to 'surprised' and 'neutral' in the case of Western consumers. In contrast, 'happy' was positively related to 'neutral' and 'valence' in the case of Asian participants and was negatively related to 'angry'. The coconut and berry incited 'arousal' in the case of Western consumers and were linked to 'angry' emotion. The coconut was related to 'contempt' and berry was placed in between 'angry' and 'disgusted' for Asian consumers. The drinkable and cookies incited 'arousal' in the case of Asians and were linked to 'surprised' and 'scared' emotions. *Foods* **2021**, *10*, 1237 11 of 18

**Figure 3.** Biplots for facial expression recognition analysis comparing for the two cultural groups: (**a**) Western consum‐ ers, (**b**) Asian consumers. **Figure 3.** Biplots for facial expression recognition analysis comparing for the two cultural groups: (**a**) Western consumers, (**b**) Asian consumers. ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to

#### *3.5. Comparison and Relationship between Methods 3.5. Comparison and Relationship between Methods Foods* **2021**, *10*, 1237 9 of 18 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 compare terms across cultures. compare terms across cultures.

The three methods—CATA emotions, CATA emojis and facial expression recogni‐ tion—were examined using a linear mixed model, with overall liking as the response variable and the individual attributes as fixed effects. The percentage variance explained by each model was calculated by fitting the same explanatory variables in a linear re‐ gression model. The difference in means and standard error for each of the term in the models is shown in Table 6. The three methods—CATA emotions, CATA emojis and facial expression recognition were examined using a linear mixed model, with overall liking as the response variable and the individual attributes as fixed effects. The percentage variance explained by each model was calculated by fitting the same explanatory variables in a linear regression model. The difference in means and standard error for each of the term in the models is shown in Table 6. **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35a 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001

CATA emotions explained 66.0% of the variation, with 'cheerful', 'neutral', 'trust‐ ed', 'indifferent' and 'uplifting' being positively correlated with overall liking and show‐ ing an increase in the mean rating value of 2.29, 0.84, 1.16, 0.93 and 0.81, respectively. The terms 'artificial' and 'nasty' decreased the mean rating value by 0.87 and 2.31 units, CATA emotions explained 66.0% of the variation, with 'cheerful', 'neutral', 'trusted', 'indifferent' and 'uplifting' being positively correlated with overall liking and showing an increase in the mean rating value of 2.29, 0.84, 1.16, 0.93 and 0.81, respectively. The terms 'artificial' and 'nasty' decreased the mean rating value by 0.87 and 2.31 units, respectively, and were negatively correlated with overall liking. ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 b0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 a0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 b0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 b0.07 b0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 b0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 **Table 3.** Cochran's Q test for emoji terms shownfor the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers**  \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 **Attributes Reference Coconut Drinkable Soy Cookies Berry***p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.07 <sup>b</sup> 0.14 ab0.03 <sup>b</sup> 0.07b0.03 <sup>b</sup>**0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup>**0.12x** 0.255ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03a 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03xy 0.18 x0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09x 0.03x 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001

respectively, and were negatively correlated with overall liking. The CATA emojis accounted for 67.8% of the overall variance with the icons, with 'heart‐shaped eyes ', 'relieved face ', 'smiling face ' and 'smiling face with mouth open ' positively related to overall liking, showing an increase in the mean rating value of 2.08, 1.28, 1.90 and 1.35, respectively. The other emojis, including 'angry face ', 'fearful ', 'pensive face ', 'persevering face ', 'confounded face ' and 'screaming in fear ', were negatively linked to overall liking, with a reduction in the mean rating value by 1.88, 1.69, 1.32, 1.60, 1.05 and 0.13 units, respectively. Only The CATA emojis accounted for 67.8% of the overall variance with the icons, with 'heart-shaped eyes \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 ', 'relieved face \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. ', 'smiling face ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00b 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ' and 'smiling face with mouth open ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18x 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 ' positively related to overall liking, showing an increase in the mean rating value of 2.08, 1.28, 1.90 and 1.35, respectively. The other emojis, including 'angry face **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry***p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 ', 'fearful ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \*0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 ', 'pensive face **Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12xy 0.24x 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ', 'persevering face ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 x0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 x0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 y0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 ', 'confounded face \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 a0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 x0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 a0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 b0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 a0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \*0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 x0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 ' and 'screaming in fear ns0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 b0.14 ab0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 y0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10ab 0.17 ab0.24 ab 0.21 ab0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 x0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 y0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 ', were negatively linked to overall liking, with a reduction in the mean rating value by 1.88, 1.69, 1.32, 1.60, 1.05 and 0.13 units, respectively. Only model 2

compared to Asian consumers. In other models (1 and 3), culture was not a significant (*p*

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

The facial emotion recognition explained only 8.8% of the variation, with the term 'surprised' linked positively to overall liking, with an increase of 5.73 units in the mean value, and the term 'disgusted' was negatively linked to overall liking, decreasing the

paired comparisons may all be not significant even if the overall test is significant.

\* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

cant differencesbetween pairs of samples according to Fisher's exact test are indicated by bold terms.Fisher's exact test is generally more conservative than Cochran's Q test, so the

ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06y 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy0.42 x0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicatessignificant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

\*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test areindicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 a0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

\* 0.00 0.000.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 y0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

**Western consumers Asian consumers** 

\*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\*Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

The multifactorial analysis (MFA) plot combined all three data sets, estimating how closely each factor was linked to other factors and the yogurt products. The MFA plot for all participants (Figure 4) explained 90.43% of the total variability (F1: 80.66%, F2:

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

> 0.05) effect.

mean rating value by 4.73 units.

compare terms across cultures.

showed a significant effect (*p* < 0.05) of culture (Asian versus Western consumers). Western consumers had a decrease in the overall liking score of 0.36 in the model compared to Asian consumers. In other models (1 and 3), culture was not a significant (*p* > 0.05) effect. **Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 *Foods* **2021**, *10*, 1237 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers** 

*Foods* **2021**, *10*, 1237 9 of 18

**Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to

**Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values**

**Western consumers Asian consumers** 

\*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06y 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\*0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

**Western consumers Asian consumers** 

\* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

**Western consumers Asian consumers** 

\*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

paired comparisons may all be not significant even if the overall test is significant.

cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact testis generally more conservativethan Cochran's Q test, so the

\*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

\*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014

cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the

paired comparisons may all be not significant even if the overall test is significant.

**Table 6.** A linear mixed model for each of the methods, taking 'overall liking' as the response variable and representing percentage variance explained by each method. *Foods* **2021**, *10*, 1237 9 of 18 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 *Foods* 9 of 18 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 *Foods* **2021**,*10*,1237 9 of 18


cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. cant between of samples according test are indicated terms. is more Q test, \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. ns 0.28 <sup>a</sup> 0.14a0.240.35 <sup>a</sup> 0.07a0.10a 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.090.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03x0.146\*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \* 0.00 0.00 0.03ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>**0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ \* 0.10ab 0.17ab ab 0.21ab 0.35 <sup>a</sup> 0.00b 0.005\*\* 0.06 y0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09xy 0.42x0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 b0.00 <sup>b</sup> 0.00 b0.00 <sup>b</sup> 0.31a<0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \*0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 y0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 cant differences betweenpairs of samples according to Fisher's exact are indicated by bold terms. Fisher'sexacttest isgenerally more conservativethan Cochran's Q test,so the paired comparisons may all be not significant even if the overall test is significant. The facial emotion recognition explained only 8.8% of the variation, with the term 'surprised' linked positively to overall liking, with an increase of 5.73 units in the mean value, and the term 'disgusted' was negatively linked to overall liking, decreasing the mean rating value by 4.73 units. **Table 3.** Cochran's Q testfor emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 a0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy CookiesBerry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 **Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy CookiesBerry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 a0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 *Foods* **2021**, *10*, 1237 9 of 18 *Foods* **2021**, *10*, 1237 9 of 18 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 z0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>**0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers)comparing terms within the culture, and using Fisher's test to compare terms across cultures.

\*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. The multifactorial analysis (MFA) plot combined all three data sets, estimating how closely each factor was linked to other factors and the yogurt products. The MFA plot for all participants (Figure 4) explained 90.43% of the total variability (F1: 80.66%, F2: 9.76%). Overall liking was positively linked to 'cheerful', 'uplifting' and 'trusted' in case of the emotion terms, 'smiling face **Western consumers Asian consumers Attributes Reference Coconut DrinkableSoy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ', 'relieved face \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences betweensamples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. ', 'smiling face with mouth open \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 ' and 'heart-shaped eyes ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 ' in case of the emojis, and 'surprised' in case of the biometrics. The overall liking was negatively linked to 'artificial', 'nasty', 'indifferent' and 'neutral' emotion terms, 'fearful ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>**0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06y 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 ', 'persevering face ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 a0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 x0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 x0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>**0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 y0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 ', 'angry face **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21a 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 a0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07b 0.14 ab 0.03 <sup>b</sup> 0.07b0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \*0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 ', 'pensive face **Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to compare terms across cultures. **Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 a0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35a**0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00a0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31a0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ', 'screaming in fear ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55a** 0.03 <sup>b</sup> <0.001\* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 ' and 'confounded face **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes ReferenceCoconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001 \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>**0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ' emojis, and 'disgusted' for the biometrics measurements. The soy, drinkable, reference and cookies were all positively linked to the overall liking, and the berry and coconut were linked to negative terms.

ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairsof samples according to Fisher's exacttest are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 \*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03b0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 a0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ns0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* <0.001. \* Indicates significantdifferencesbetween samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generallymore conservative than Cochran's Q test, so the paired comparisons may all be notsignificant even if the overall test is significant. ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 The MFA plots comparing the responses of Western and Asian participants (Supplementary Figures S1 and S2) showed that the most liked yogurt (cookies) and the least liked yogurts (berry and coconut) were similarly placed across both the cultures. Apart from this, soy and drinkable showed minor differences (although positive) across the two groups.

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

paired comparisons may all be not even if the overall test is significant.

paired comparisons may all be not significant even if the overall test is significant.

*Foods* **2021**, *10*, 1237 13 of 18

However, the reference yogurt was related to more positive terms by Western participants and associated with slight dislike by Asians.

**Figure 4.** Multifactor analysis for a combination of check‐all‐that‐apply (CATA) emotions, check‐all‐that‐apply (CATA) **Figure 4.** Multifactor analysis for a combination of check-all-that-apply (CATA) emotions, check-all-that-apply (CATA) emojis and facial expression recognition (FER) with overall liking ((**a**)-variables, (**b**)-yogurt products).

#### emojis and facial expression recognition (FER) with overall liking ((**a**)‐variables, (**b**)‐yogurt products). *3.6. Price Perception and Purchase Intent of Yogurts*

*3.6. Price Perception and Purchase Intent of Yogurts* Consumers perceived the cookies yogurt to be "somewhat" to "very likely" to be purchased from a supermarket, with an average purchase intent score of 3.6 (on a scale of 5), although they considered its price to be the same as the regular yogurts that they generally consume (rated by 59.7% participants). Only the berry yogurt received the lowest score of 1.1, indicating that it was 'not at all likely' to be purchased. The refer‐ ence, drinkable and soy yogurts were rated to be in between 'not so likely' and 'some‐ what likely', although their perceived price was rated to be the 'same' by 41.9, 38.7 and 35.5% of participants, as compared to their regular yogurts. The data are shown in Sup‐ plementary Table S4 and S5. The interaction for culture and yogurt products was signifi‐ Consumers perceived the cookies yogurt to be "somewhat" to "very likely" to be purchased from a supermarket, with an average purchase intent score of 3.6 (on a scale of 5), although they considered its price to be the same as the regular yogurts that they generally consume (rated by 59.7% participants). Only the berry yogurt received the lowest score of 1.1, indicating that it was 'not at all likely' to be purchased. The reference, drinkable and soy yogurts were rated to be in between 'not so likely' and 'somewhat likely', although their perceived price was rated to be the 'same' by 41.9, 38.7 and 35.5% of participants, as compared to their regular yogurts. The data are shown in Supplementary Tables S4 and S5. The interaction for culture and yogurt products was significant for purchase intent (*p* < 0.05). Asian consumers perceived the price of cookies and drinkable yogurts to be higher. Western consumers perceived the price of the reference yogurt to be higher.

## cant for purchase intent (*p* < 0.05). Asian consumers perceived the price of cookies and **4. Discussion**

#### drinkable yogurts to be higher. Western consumers perceived the price of the reference yogurt to be higher. *4.1. Traditional Sensory Method*

**4. Discussion** *4.1. Traditional Sensory Method* The advantage of using a 9‐point hedonic scale is that it is simple to use by con‐ sumers and is effective in predicting the acceptance of a food product [2]. In the present study, it was successfully used to predict the liking of the six different yogurt formula‐ tions, suggesting that the cookies was the most liked and the berry was the least liked The advantage of using a 9-point hedonic scale is that it is simple to use by consumers and is effective in predicting the acceptance of a food product [2]. In the present study, it was successfully used to predict the liking of the six different yogurt formulations, suggesting that the cookies was the most liked and the berry was the least liked yogurt product. However, these scales could not predict any differences in the emotional responses of participants, or differences in liking across cultures, which is in contrast to some of our previous works, in which cultural differences influenced the affective responses towards different product types [29].

yogurt product. However, these scales could not predict any differences in the emotional responses of participants, or differences in liking across cultures, which is in contrast to some of our previous works, in which cultural differences influenced the affective re‐ sponses towards different product types [29]. The hedonic scales have some drawbacks, including less freedom for the partici‐ pants to express a full range of emotions [2]. Furthermore, these liking scores are limited The hedonic scales have some drawbacks, including less freedom for the participants to express a full range of emotions [2]. Furthermore, these liking scores are limited in predicting market success. Hence, more elaborate methods, that measure consumer emotions (conscious or unconscious), need to be included for a more complete understanding of consumer food choices [30].

in predicting market success. Hence, more elaborate methods, that measure consumer emotions (conscious or unconscious), need to be included for a more complete under‐

standing of consumer food choices [30].

#### *4.2. Self-Reported Responses (CATA Methods)* compare terms across cultures.

Apart from hedonic scale, self-reported methods of emotions or emojis has gained popularity for sensory segmentation of emotions in sensory experiments. This study shows that this approach allows differences in liking towards the same yogurt products to be better understood and is particularly useful where Western and Asian consumers use different terms to express liking towards the same yogurt type. For example, in this study, Asians selected 'common' to describe the reference yogurt, and the berry yogurt was associated **Western consumers Asian consumers Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values Attributes Reference Coconut Drinkable Soy Cookies Berry** *p***‐Values** ns 0.14 <sup>a</sup> 0.21 <sup>a</sup> 0.21 <sup>a</sup> 0.10a **0.14a** 0.14 <sup>a</sup> 0.817 ns 0.12 xy 0.24 <sup>x</sup> 0.12 xy 0.15 xy **0.00 <sup>y</sup>** 0.09 xy 0.087 \* 0.28 ab 0.21 ab 0.24 ab 0.31 ab 0.41 <sup>a</sup> 0.00 <sup>b</sup> 0.007 \*\* 0.15 yz 0.09 <sup>z</sup> 0.42 xy 0.30 xyz 0.52 <sup>x</sup> 0.00 <sup>z</sup> <0.001 ns 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.03 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.10 <sup>a</sup> 0.267 \* 0.00 <sup>y</sup> 0.09 xy 0.00 <sup>y</sup> 0.03 xy 0.00 <sup>y</sup> 0.18 <sup>x</sup> 0.001

*Foods* **2021**, *10*, 1237 9 of 18

**Table 3.** Cochran's Q test for emoji terms shown for the two cultural groups (Western consumers, Asian consumers) comparing terms within the culture, and using Fisher's test to

more with the 'pensive face \* 0.07 <sup>b</sup> 0.14 ab 0.03 <sup>b</sup> 0.07 <sup>b</sup> 0.03 <sup>b</sup> **0.35 <sup>a</sup>** 0.001 ns 0.06 <sup>x</sup> 0.15 <sup>x</sup> 0.06 <sup>x</sup> 0.09 <sup>x</sup> 0.00 <sup>x</sup> **0.12 <sup>x</sup>** 0.255 ns 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.00 <sup>a</sup> 0.03 <sup>a</sup> 0.10 <sup>a</sup> 0.00 <sup>a</sup> 0.098 \* 0.03 xy 0.00 <sup>y</sup> 0.03 xy 0.03 xy 0.18 <sup>x</sup> 0.00 <sup>y</sup> 0.004 \*\* 0.10 ab 0.10 ab 0.00 <sup>b</sup> 0.03 <sup>b</sup> 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \* 0.21 <sup>x</sup> 0.09 <sup>x</sup> 0.03 <sup>x</sup> 0.12 <sup>x</sup> 0.00 <sup>x</sup> 0.21 <sup>x</sup> 0.026 ns 0.03 <sup>a</sup> **0.21 <sup>a</sup>** 0.21 <sup>a</sup> 0.17 <sup>a</sup> 0.17 <sup>a</sup> 0.00 <sup>a</sup> 0.051 \* 0.09 xy **0.00 <sup>y</sup>** 0.06 <sup>y</sup> 0.09 xy 0.27 <sup>x</sup> 0.00 <sup>y</sup> 0.001 ' by Western consumers. A similar effect was found for the selection of the emotional terms in a study of chocolate tasting, as Asian and Western consumers expressed emotions relating to the consumption of chocolate differently [28]. Hence, CATA terms, when used in combination with a hedonic scale, can better explain consumer preferences and segmentation, without creating any bias [31].

ns 0.28 <sup>a</sup> 0.14 <sup>a</sup> 0.24 <sup>a</sup> 0.35 <sup>a</sup> 0.07 <sup>a</sup> 0.10 <sup>a</sup> 0.050 ns 0.15 <sup>x</sup> 0.18 <sup>x</sup> 0.09 <sup>x</sup> 0.18 <sup>x</sup> 0.03 <sup>x</sup> 0.03 <sup>x</sup> 0.146 \*\* 0.00 <sup>b</sup> 0.17 ab 0.03 <sup>b</sup> 0.14 ab 0.00 <sup>b</sup> 0.31 <sup>a</sup> 0.000 \*\* 0.06 <sup>y</sup> 0.15 xy 0.00 <sup>y</sup> 0.09 <sup>y</sup> 0.00 <sup>y</sup> 0.33 <sup>x</sup> <0.001 \* 0.10 ab 0.17 ab 0.24 ab 0.21 ab 0.35 <sup>a</sup> 0.00 <sup>b</sup> 0.005 \*\* 0.06 <sup>y</sup> 0.03 <sup>y</sup> 0.15 <sup>y</sup> 0.09 xy 0.42 <sup>x</sup> 0.00 <sup>y</sup> <0.001 \*\* 0.07 <sup>b</sup> 0.07 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.00 <sup>b</sup> 0.31a <0.001 \*\* 0.06 <sup>y</sup> 0.09 <sup>y</sup> 0.06 <sup>y</sup> 0.06 <sup>y</sup> 0.00 <sup>y</sup> 0.49 <sup>x</sup> <0.001 \* 0.00 0.00 0.03 ab 0.03 ab 0.00 0.17 <sup>b</sup> 0.004 \* 0.06 xy 0.03 <sup>y</sup> 0.03 <sup>y</sup> 0.00 <sup>y</sup> 0.00 <sup>y</sup> 0.21 <sup>x</sup> 0.001 Further, it has been found that emojis and emoticons can be an easy and effective way to understand food-related emotions, as consumers can relate to visual cues better than words [12]. Jaeger, Xia, Lee, Hunter, Beresford and Ares [7] found emojis (or emoticons) to be suitable for understanding the liking of a wide variety of food products by consumers. In another study by Jaeger et al. [32], emojis were successfully used to assess emotional liking towards the consumed food products (where a variety of products were served according to culture), and emojis had the potential to be used as an approach for direct measurement.

\*\* 0.28 ab 0.07 <sup>b</sup> 0.45 <sup>a</sup> 0.35 ab **0.55 <sup>a</sup>** 0.03 <sup>b</sup> <0.001 \* 0.18 xy 0.18 xy 0.30 <sup>x</sup> 0.12 xy **0.15 xy** 0.00 <sup>y</sup> 0.014 \*\* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.001. \* Indicates significant differences between samples according to Cochran's Q test at *p* < 0.05. ns Indicates no significant difference between samples according to Cochran's Q test at *p* > 0.05, with 'a' the largest in case of Western and 'x' in case of Asian consumers. Signifi‐ cant differences between pairs of samples according to Fisher's exact test are indicated by bold terms. Fisher's exact test is generally more conservative than Cochran's Q test, so the paired comparisons may all be not significant even if the overall test is significant. There are limitations of the self-reported methodology of emojis, as subtle differences between the samples cannot be easily detected, although this method does offer a simple and easy approach for understanding the sensory characteristics of food products [13]. Furthermore, it is important to define an emotion, so that it is easier to understand consumers' expectations from the tasted product and also to keep a balance of positive and negative emotion terms, in order to eliminate bias [33].
