Data from online product reviews offers a novel avenue for the sensory characterisation of food. But, little research has investigated the quality of sensory information in the online reviews. The aim of this research was to investigate consumer online reviews as sources of sensory attributes of food products (starting with a minimum of 1000 online reviews per product), and to assess how the resulting sensory product profiles compare to profiles obtained in a central location test with 105 consumers using rate-all-that-apply (RATA) questions. A case study was conducted with five unflavoured coffee samples. A semi-automated approach, combining natural language processing and sensory science expertise was used to clean online review data, develop a sensory lexicon, and analyse the frequency of attributes used by consumers. It was possible to develop online review-based sensory product profiles and discriminate the five samples on this basis. Consumers used a small set of broad, mostly intensity related sensory terms (e.g. 'Bold/rich', 'Strong/intense', 'Smooth', 'Weak/bland') more frequently than descriptive terms. Canonical analysis showed high agreement between new method and RATA for product discrimination and between two group of descriptors. The first group (including 'Coffee Flavor,' 'Rich in Flavor', and 'Smell of Coffee') is associated with the intensity of flavour of coffee, while the second group describes characteristic flavour of coffee (including 'Bold', 'Dark' 'Body'). Furthermore, care should be taken when implementing these findings in food categories with lower levels of consumer engagement, where consumer comments relating to sensory properties in online reviews may be less frequent.

Replacing animal-based protein food products with plant-based protein products is essential for human health and the environment. For this reason, a number of plant-based (PB) yoghurts have been developed to replace dairy yoghurt. However, many of these products suffer from poor sensory quality. In order to improve the quality of these products, data on the sensory drivers of liking, emotional, and conceptual responses to PB yoghurts, PB alternatives (PB blends, PB/dairy blends) and dairy yoghurts are required. In addition, in order to optimise the design of these products for sustainability-oriented consumers, research is needed on the drivers of consumers' conceptualisations of sustainability. To provide actionable product development guidance on these issues, multiattribute research was undertaken in a central location test with New Zealand consumers, using a variety of commercial PB and dairy yoghurts (blind sample presentation) that were expanded in their range of sensory attributes through ingredient modifications, with the goal of quasi-replicating and extending prior research on the sensory drivers of liking, emotions, and conceptualisations of PB yoghurts. Results of the research confirmed previous literature findings on the sensory drivers of liking and non-sensory variables in the PB yoghurt space. Among the salient findings related to sensory drivers are that white and lighter coloured yoghurts with smooth and creamy appearance and mouthfeel, sweet taste, and vanilla or coconut flavours maximise liking and positive emotional responses to yoghurt products. In addition, a number of these and other sensory characteristics elicited specific positive conceptualisations of the yoghurts. Although sustainability conceptualisations were evoked less frequently, the major positive drivers of these conceptualisations were "white colour," "sweet," "even/smooth" appearance and texture, "oat-like flavour" and "coconut flavour." "Dairy flavour" evoked conceptualisations related to "traditional" and "authentic," but also elicited a number of negative sustainability concepts, such as "resource intensive," "commercially farmed raw materials" and reduced conceptualisation of "animal friendly." Other important sensory drivers of liking, emotions and conceptualisations are discussed, providing consumer insights and direction to product developers in their efforts to improve the sensory characteristics, acceptance and consumption of PB and other alternative yoghurt products.

Consumer-sensory methods based on check-all-that-apply (CATA) for texture and flavour terms and conceptual descriptors offer alternatives for characterising eating quality. In this study, 'Red Delicious', 'Gala' and 'Scilate' apples (Malus domestica Borkh.), which were of comparable instrumental firmness, and similar in crispness and apple flavour, evoked profoundly different consumer responses. Mean liking scores from 122 consumers were 7.1 (similar to like moderately), 5.8 (similar to like slightly), and 5.1 (similar to neither like nor dislike) for 'Scilate', 'Gala', and 'Red Delicious' respectively, and cluster analysis indicated three consumer-segments: one that preferred 'Scilate' over the other cultivars (62 % of consumers), another that liked 'Scilate' and 'Gala' equally over 'Red Delicious' (26 %), and a small third segment that liked 'Red Delicious' and 'Gala' equally over 'Scilate' (12 %). In the associated experimental auction, consumers were provided with $US3, and the mean bids for 'Scilate', 'Gala', and 'Red Delicious' reflected the pattern of consumer liking at $US1.72, $US0.79, and $US0.52, respectively. Correspondence analysis plots of CATA data indicated that 'Scilate' was associated with sensory descriptors sweet, fruity, tropical and pear flavours and positive conceptual descriptors including exciting, complex, unique, premium, yummy, intense, and powerful. 'Gala' and 'Red Delicious' were associated with sensory descriptors including old, mealy, bland, green-grassy and sour-acidic and negative conceptual descriptors dissatisfied, inferior, boring, disappointing, and uninspiring. A best-worst scaling methodology was used to elicit consumers' priorities regarding the aspects of apple quality and production sustainability. Despite the profound differences in eating quality uncovered in the hedonic and CATA tasks, the statement 'tastes like the best apple eaten today' was assessed as being only 23 % as important as the statement 'grown in a way that is sustainable and does not harm the environment' in consumer choices. This notable outcome was contradicted in a matching online study changing the eating quality statement 'tastes like the best apple eaten today' to 'Crisp, juicy and sweet' (the attributes most frequently identified in the sensory CATA). This suggests that trade-offs between eating quality and sustainability are subject to context effects associated with whether taste is an experienced or imagined factor. There is an urgent need to better understand how consumers' needs for sustainability in the food production system might alter what is acceptable in terms of eating quality.

The present research contributed to a better understanding of how check-all-that-apply (CATA) questions work by examining the relationship between likelihood of selecting a term and perceived attribute intensity. Seven consumer studies were conducted (147-157 people per study) using within-subjects experimental designs where participants twice evaluated the same set of stimuli on the same set of terms (or attributes), respectively with CATA questions and intensity scaling (7-point category scale; 1 = "not at all," 7 = "extremely"). As a function of perceived intensity, the average CATA citation frequency tended to follow a sigmoidal-like relationship where likelihood of selecting a CATA term increased more slowly at the extreme ends of the intensity scale (1-2 and 6-7) and linearly otherwise. This illuminates why for a given term, CATA questions are less suited for discriminating between samples that are of similar "low" or "high" intensity. Practical ApplicationsCATA questions are popular for sensory product characterization tasks with consumers. Despite their simplicity, they accurately discriminate among samples, and term citation frequency is a proxy for perceived intensity, albeit not a direct measure hereof. Versatility and applicability of CATA questions to characterize diverse stimuli using diverse types of terms/attributes was demonstrated. By showing that likelihood of CATA term selection typically increases with perceived intensity according to a sigmoidal-like shape, the present research shows that CATA terms best discriminate between samples when these vary in intensity rather than being of similar "low" or "high" perceived intensity.

To assess the role of factual personal health and environmental benefit information on consumer responses to plant-based (PB) yoghurts, a central location test (CLT) was conducted with 338 New Zealand consumers. Participants tasted a series of commercially available PB yoghurts and provided sample evaluations according to a multi-variable measurement approach where liking, emotional response, sensory characterisation (appearance and taste), holistic and conceptual perceptions, and attitudinal evaluations were obtained for each sample. A between-subjects experimental design was used, with participants allocated to one of three experimental con-ditions (ECs) in which the PB yoghurt samples were presented blind (B), with factual information about personal health (PH) benefits of PB foods and yoghurts, or with factual information about environmental (ENV) benefits of PB foods and yoghurts. The two key results were low sample acceptability of the commercially available yoghurts and a minimal impact of factual health and environmental information on judged variables. The results were interpreted as showing the limitations of providing health and environmental benefit information for changing consumers' behaviour toward PB yoghurt when sensory product quality is poor. Against this background, research to improve PB yoghurt quality is paramount. Possible reasons for the limited effect of factual PH and ENV information are discussed, and suggestions are made for future research on possible uses of extrinsic var-iables to modulate choice and consumption of PB yoghurt and other PB foods.

Plant-based (PB) yoghurts is a rapidly expanding food category with the potential to reduce many of the negative environmental impacts associated with the farming of dairy cattle and related greenhouse gas emissions. Within the context of growing consumer demand for PB products, this research aimed to identify the main drivers of consumer acceptance of PB yoghurts through data drawn from a consumer study of 113 adult New Zealand participants who tasted a representative range of commercially available vanilla-flavoured PB yoghurts, and provided sample evaluations using a multi-variable measurement approach comprised of liking ratings, emotional responses, sensory characterisations, and holistic / conceptual perceptions. The primary finding of this study was that the overall acceptability of currently available PB yoghurts is low, leading to poor attitudes and low willingness to consume PB yoghurts. The examined products were characterised by the presence of several sensory attributes that negatively impacted consumer liking, either because they were unexpected or because their intensity level was too high. Among the main sensory issues identified were perceived sourness (acidity), lumpy appearance and non-white colour. In contrast, vanilla flavour and a creamy/smooth texture were identified as the main drivers of liking. The multi-response approach used in this study enabled a comprehensive characterisation of the products. One of the key insights gained was that negative emotional activation, caused by the presence of unexpected sensory attributes in the samples, was the main driver of product rejection. This was likely due to the disparity between consumers' sensory expectations, formed primarily based on their experience with dairy yoghurts, and the actual experience upon tasting the PB yoghurts. Overall, the findings indicate that the sensory properties of currently available PB yoghurts are a significant barrier to their adoption by most consumers and that addressing these sensory deficiencies is essential to achieving mainstream market acceptance of these products.

Text highlighting is a new method for consumer research, especially in attitude measurement. Presently, the method is evolved into a mixed method approach through a combination with open-ended questions. While participants read a text they use highlighter functions to select content that they like/feel positive about, and content which they dislike/feel negative about. Immediately hereafter they answer open-ended questions about the text they selected and why they did so. The mixed method approach was implemented in two consumer studies (1,554 participants) and the results showed that the combination of text highlighting and open-ended questions achieved greater "completeness," that is, more comprehensive insights on the topic of the research. The supplementary qualitative component of the mixed method design (i.e., the open-ended questions) helped to elaborate the insights from text highlighting (i.e., the quantitative core component). Practical ApplicationsThis research showcases how the recent method of text highlighting can be combined with open-ended questions, and that doing so is simple and beneficial. This is illustrated in two case studies on the topics of, respectively, biodynamic agriculture and vertical farming. The open-ended questions were added to identify reasons underlying participants' "like" and "dislike" text highlights and helped to better interpret the highlighting responses. The combination of text highlighting with open-ended questions is a general purpose method that can be used in many areas of consumer research-including novel foods/food technologies and product research.

Product-focused emotion research remains popular and continues to advance, both with respect to product insights and methods for obtaining these. The present research primarily contributes to the latter by showcasing consumer segmentation based on product-elicited emotional associations obtained using the circumplex-inspired valence x arousal single-response emotion word questionnaire (CEQ). This gives rise to three-way data, where responses from each consumer can be represented by a matrix of p products x 12 emotion word pairs. A multistudy (n = 6) strategy was pursued in the present research, with 197 to 1014 people per study. Four of the six studies used tasted products and two studies used written stimuli (i.e., names of foods and beverages). For each study two or three emotion-based clusters were retained, and the differences between these could be classified into three patterns: i) differences in the emotional associations with specific sensory characteristics (tasted products); ii) differences in the emotional associations of the category (tasted products); and iii) differences mainly driven by liking (written stimuli). Fitting expectations, differences between emotion-based clusters within a study extended beyond the insights gained from liking scores. Taken together, the results from the six case studies suggest that consumer segmentation based on emotional responses deserves further exploration.

Product-focused emotion research has been in ascendency for more than a decade, but methodological research into questionnaires, the most popular tool for measuring emotional associations is still needed. The present research was situated in this context and pertained to the valence x arousal circumplex-inspired emotion questionnaire (CEQ) which includes 12 pairs of emotion words. Variants of the CEQ that pertained to response options (single vs. multiple) and layout (circular vs. list) were compared in four consumer studies with a total of 1857 participants. In Studies 1 to 3, the CEQ variant that featured a circular layout and single response format (C&S) was compared to a list variant with multiple response format (L&M) (i.e., a check-all-that-apply question). Selection of CEQ emotion word pairs and degree of sample discrimination was contingent on CEQ variant, meaning that conclusions about sample similarities and differences depended on the CEQ variant that was used. This was also the result in Study 4 where two circular CEQ variants were compared, with single and multiple response formats, respectively. A further important finding was that stimuli format - tasted samples in Studies 1-2 and written stimuli in Studies 3-4 - as well as degree of sample differences influenced the extent to which the results from the CEQ variants differed. Our recommendations at this stage are to be aware that conclusions about sample profiles and discrimination is CEQ variant dependent and avoid directly comparing results from studies that use different CEQ variants. Implementing the CEQ as a multi-response variant is supported since it improves sample discrimination relative to single response data, particularly for pairs of emotion words that primarily relate to the dimension of arousal.

The negative impact of food neophobia (FN) on food and beverage (F&B) liking/disliking is well documented. Against this backdrop, the present research sought to raise awareness of data analysis strategy in FN research. Extreme groups analysis currently dominates as evidenced from the many studies that compare groups of neophobic and neophilic consumers. However, when a variable, like FN, is naturally continuous, full-data analysis has been recommended. We present both approaches using data from online surveys in three Western countries where consumers rated degree of liking/disliking for F&B names (USA, UK and Australia, n = 3454). The first objective was to explore how results from extreme groups analysis were influenced by group extremity (more/ less neophobic vs. more/less neophilic). In the case of seafood and ethnic food items, the average difference in liking score between neophobic and neophilic groups ranged from 2 to 4.5 points on the 9-pt hedonic scale depending on FN group definition (less extreme vs. more extreme). To acknowledge this influence contextualisation of degree of FN group extremity is encouraged. For example, FN groups defined by M +/- 0.5 x SD and M +/- 1.5 x SD should be differently named to convey greater FN extremity in the latter. In the second objective, full-data analysis was performed, treating FN as a continuous variable. Additional knowledge about the impact of FN on F&B liking/disliking beyond that from extreme groups analysis was gained. Linearity in the relationship between FN and degree of F&B liking/disliking was generally observed. Moderating factors were included in additional analyses and illustrated the possibility of accounting for participants' country of living or differences between consumer populations in overall level of FN (central location vs. online). We suggest that it may be time to rely less on extreme groups analysis in FN research, and if used that adequate justification be given and care taken to describe the FN groups to facilitate more robust interpretations and between-study comparisons. The insights and recommendations have relevance for other continuous psychological traits relating to food liking food-related consumer behaviour in general.

Emoji have been argued to have considerable potential for emotion research but are struggling with uptake in part because knowledge about their meaning is lacking. The present research included 24 emoji (14 facial, 10 non-facial) which were characterized using the PAD model (Pleasure-Arousal-Dominance) of human affect by 165 consumers in New Zealand and 861 consumers in the UK. The results from the two countries were remarkably similar and contributed further evidence that emoji are suitable for cross-cultural research. While significant differences between the emoji were established for each of the PAD dimensions, the mean scores differed most on the Pleasure dimension (positive to negative), then on the Arousal dimension (activated to deactivated), and lastly on the Dominance dimension (dominance to submissive). The research also directly measured the perceived appropriateness of the 24 emoji for use with foods and beverages. The emoji face savoring food, clapping hands and party popper were in the top-5 for the highest appropriateness in food and beverage context for both studies, as was a strong negative expression linked to rejection (Study 1: face vomiting; Study 2: nauseated face). On the other hand, zzz and oncoming fist were considered as the least appropriate to be used in a food and beverage context in both studies. Again, the results from the UK and NZ were in good agreement and identified similar groups of emoji as most and least suitable for food-related consumer research.

Perceived situational appropriateness crucially informs decisions about what to eat and drink, and this contextual-cognitive factor of food choice helps to understand why in certain situations, despite high acceptability, a non-consumption decision is made. Being shaped primarily by accumulated experience and familiarity, appropriateness is influenced by culture, but inter-individual variability within populations also exists. Consumer segmentation based on appropriateness ratings was recently reported in work with Australian consumers, with segment differences relating to how adaptive or conforming consumers' perceptions were of appropriateness for a range of foods and beverages (F&B) for use in daily main eating occasions [Jaeger et al.; Fd. Qual. Pref., 2019, e103701]. Believing in the strength and value of cumulative evidence, the present research was informed by the broad paradigm of replication and extension. This directed that empirical execution, data analysis and reporting strategy directly and extensively defined this work. In two studies with consumers from Denmark (n = 780) and New Zealand (NZ, n = 448), partial replication was achieved for situationally Adaptive and Conforming consumer segments in regard to breakfast, lunch and dinner (morning, midday and evening meals, respectively). Extending to three different eating occasions in NZ, partial replication of Adaptive and Conforming consumer segments was obtained. It appeared that these segments exist for some but not all eating occasions and that cultural differences are also at play. The present research also replicated the relationship between perceived appropriateness and F& B liking/disliking, whereby liked products can be both appropriate and inappropriate, depending on the eating occasions. Consumer segments based on appropriateness ratings had similar profiles, which also replicated Jaeger et al. (2019).