Alternatives to data averaging of consumer preference data

The relationship between a consumer preference data set and a corresponding sensory profile on eight cooked wheat noodles with different formulas was examined using several multivariate techniques. Individual consumer hedonic responses (100 noodle/pasta consumers) and eight appearance and texture sensory attributes were collected. The consumer preference data were treated in two different ways: mean values averaged across all consumers or principal components extracted from individual responses. The mean preference scores were submitted to both principal component stepwise regression and partial least squares regression (PLS1), whereas the summarized major preference components were subjected to canonical correlation analysis, as well as partial least squares regression (PLS2). The results suggested that in case of complex consumer data, using mean value can only capture the most manifest trends in consumer preference patterns, while studying individual responses and by further categorizing major preference patterns provide an opportunity to discover the hidden information that are masked by data averaging.     

Comparison of differences between lexicons for descriptive analysis of Cheddar cheese flavour in Ireland,New Zealand,and the United States of America

Flavour lexicons for cheese provide a way to document cheese flavour for both research and marketing. The objective of this study was to compare differences and similarities in three independently developed sensory languages for Cheddar cheese flavour at three different locations (Ireland, New Zealand, United States of America) using an international selection of Cheddar cheeses. Twelve Cheddar cheeses (four from each country) were evaluated by the three panels using the respective sensory languages. Sensory space patterns obtained by principal component analysis were consistent between the three sites indicating that the overall differentiation of the cheeses by each panel was similar. The key flavour characteristics among the cheeses were described by different attributes. Cheeses were grouped by each site by country of origin suggesting international differences in Cheddar cheese flavour. Cross-cultural differences can exist in sensory language and perception, but highly trained panels using standardized, representative languages can provide comparable results.     

南瓜的感官品质、质构及生化分析

在对127份南瓜材料进行感官鉴定的基础上,选取10份口感明显不同的南瓜材料为研究对象,分别测定其质构指标及生化指标,并分析感官属性和质构指标以及感官属性和生化指标的相关性。结果表明：南瓜的感官属性可以分为3个主成分,第1主成分为粉质、干湿情况和甜度,第2主成分为面度和纤维度,第3主成分为硬度和脆性;南瓜的质构指标也分为3个主成分,第1主成分为硬度、回复性和剪切力,第2主成分为内聚性,第3主成分为黏附性和弹性。感官属性与质构指标及理化指标不同项目之间具有不同的相关性。南瓜感官评价的关键指标为粉质,干湿情况和甜度。     

Behavioural extensions to preference mapping: the role of synthesis

It is suggested to tackle unsolved methodological problems in preference mapping by taking account of the behavioural processes underlying preference formation. To this end an information processing model of preference formation is proposed. Within this model we particularly consider the process of synthesis, which is the way sensory information about product similarities and differences is analysed and processed. Although current applications of external preference mapping assume that consumers and trained judges synthesise sensory stimuli similarly, accumulating evidence in the literature suggests that this is not so. We, therefore, propose that analysing preference data in terms of a product space that accurately reflects how consumers see products will improve preference mapping methodology. Differences in synthesis are accounted for by applying sets of synthesis weights reflecting differences in the relative weighting given to each sensory attribute. An empirical analysis of sensory and preference data pertaining to eating apples supports the hypothesis that consumers use only a few key sensory attributes rather than synthesise a large number of attributes during preference formation.     

Sensory profiles and preference analysis in ornamental horticulture: The case of the rosebush

The context of ornamental horticulture is considered in order to extend the techniques of sensory and preference evaluation by taking the rosebush as a plant model. In a preliminary study (Boumaza, Demotes-Mainard, Huché-Thélier, & Guérin, 2009), a sensory evaluation was conducted in order to set up a list of attributes. Subsequently, this list was adapted to assess 10 rosebushes. After the control of the panel performance using a multivariate strategy of analysis, the average scores were used in product mapping. The evaluation of the preferences with regard to these rosebushes was undertaken: 253 subjects were asked to rank the products by decreasing order of liking. Thereafter, the preference data were subjected to an internal preference mapping and a cluster analysis. Six homogeneous segments of consumers were eventually retained. By way of performing an external preference mapping, the average ranks were regressed upon the sensory attributes using principal component regression: the preferences of 67% of the consumers were satisfactorily explained by the attributes.     

An integrated method for the emotional conceptualization and sensory characterization of food products: The EmoSensory® Wheel

Although acceptability is commonly used to examine liking for food products, more studies now emphasize the importance of measuring consumers' conceptualizations, such as emotions for food products. It is also important to identify how consumers perceive the sensory attributes of food products, as illustrated by the increasing involvement of consumers in product characterization. The objective of this paper is to examine the use of a wheel-format questionnaire to obtain both an emotional and sensory profiles for food products using a hands-on consumer tool. Terms selected were product-specific and the rate-all-that-apply (RATA) approach was used as a scaling technique. Three different experiments demonstrated that the EmoSensory® Wheel could discriminate within and between food product categories. The added value of the RATA approach was illustrated in the sample discrimination for some food products when using the weighted attribute scores for analysis. The tool was used in both blind and informed conditions to illustrate its applicability across different experimental designs. In general, the respondents did not find the task tedious when using the wheel-questionnaire format, demonstrating the potential for collecting information in a more facile way. Although further studies with other food products are needed, this paper shows the potential for using this wheel format to obtain information about consumers' emotional and sensory profiling of food products.     

Combining extrinsic and intrinsic information in consumer acceptance studies

This paper proposes a methodology for combining extrinsic and intrinsic attributes in consumer testing of food products. The paper attempts to focalize on the main sensory drivers of liking or choice probability and their interaction with additional information, and to investigate effects related to the population as well as how consumers differ in their assessments. Two different data analysis approaches are considered and compared on choice probability data from a consumer study of orange juice. One of the methods is based on mixed model ANOVA of individual differences, the other approach is based on fuzzy clustering related to regression residuals. The main results show that extrinsic consumer attributes are easily and efficiently related to the sensory properties of products, allowing for interactions. The methodology estimates population or segment means and gives an overview of individual differences in choice intent or liking.     

Consideration of Thurstonian Scaling of Ratings Data

In the analysis of sensory data, the use of parametric statistics, specifically analysis of variance (ANOVA), is standard. However, such parametric analyses often make assumptions that are not valid. Recent advances have made available an alternative analysis that does not make these assumptions, specifically Thurstonian Scaling of Ratings Data (SoR). This study compared these two methods of analyses on a single dataset. To further clarify the differences between analyses, assessments of a subset of the stimuli were also made in a triangle task. Findings indicate that the differences between ANOVA and SoR are minimal, with SoR tending to be slightly more conservative. Regardless of analyses used, triangle tests were found to be superior at differentiating the stimuli. The d′ values determined using Thurstonian SoR did not agree with those determined with the triangle tests. Possible reasons for the discrepancy include boundary variance, actual strategy used during triangle tests, as well as other possible sources of variance.