*2.3. Sensory Evaluation and Data Collection*

Participants were asked to taste each of the six yogurt samples (~15 g), which were served in plastic containers at an internal temperature of 10 ± 2 ◦C, labeled with 3-digit codes and presented in a random order. Participants were seated in individual sensory booths (under white natural LED light); the temperature of the booths was maintained at 22 ± 2 ◦C. Each booth was equipped with a Samsung 18-inch tablet (Samsung Group, Seoul, South Korea) for recording videos to obtain consumer responses using the Bio-sensory app [26], which did not require any calibration. The armed tablets were placed near to face of the participants, with an in-built camera, and could be adjusted according to the participant's height. When a participant was seated in the sensory booth, it was ensured that their faces were recorded properly in the set-up, at an approximate distance of 45– 50 cm from the participant. The investigators who had expertise in sensory ensured at the beginning of the tasting that the participants' faces were recorded properly in the set-up.

There was a mandatory break of at least 30 s in between the tasting of any two samples for cleansing of the palette with crackers and water. For each product, participants were asked to evaluate overall liking using a continuous 9-point hedonic scale (1—dislike extremely, 5—neither dislike nor like and 9—like extremely). The self-reported emotional responses of participants were recorded using the CATA methodology, where they were asked to choose the terms from a list of emotions provided (Table S1). These emotion terms were divided into positive, negative and neutral categories and were selected using the perceptual mapping technique [27,28]. A similar procedure was implemented for choosing emojis from another list (Table S2), which were representations of popular face scales from other tasting studies [7]. Preliminary research group discussions also helped in reaching the final list of terms of emotions and emojis, which were fitting for the cross-cultural component of the study.

Facial expressions were recorded during the tasting of each yogurt (Table S3). The videos were cut from the point when participants placed the spoonful of yogurt in their mouth until the time it was swallowed (or consumed), showing their first immediate reaction, which was covered in 5–7 s or even less time in most cases. Further, the data were processed using FaceReaderTM 8.0 software (Noldus Information Technology, Wageningen, Netherlands) for recording facial emotions. No calibration was required for the FaceReaderTM software as it has pre-built models to detect faces and analyze the expressions. The video analysis was carried out using the "general" settings for all the participants, except for the participants from South East Asia, for whom specific settings were used for analysis. Participants were also asked to assess purchase intent and willingness to pay for each product.
