SENSORY PROFILING WITH PROBABILISTIC MULTIDIMENSIONAL SCALING

Variability is a fundamental characteristic of sensory profile data. Ignoring the variability may result in biased solutions that cannot be improved by the collection of additional data. Probabilistic multidimensional scaling (PMDS) models provide a means of accounting for the variability inherent in sensory data by using distributions, instead of points, to portray sensory objects. For profile data with high levels of variability, the probabilistic model recovers latent structure parameters very well — traditional deterministic MDS models and principal components analyses (PCA) do not. Advantages of the PMDS models include their parsimony, testability and extensibility. Two particularly attractive PMDS attributes are their ability to relate consumers' expressions of liking to product profiles and their ability to estimate a product's " perceptual share" from liking and profile data. Used as a criterion with what-if modeling, perceptual share estimates enable the evaluation of alternative product development strategies.     

SENSORY MEASUREMENT OF FOOD TEXTURE BY FREE-CHOICE PROFILING

Panelists were trained to measure texture of processed cheese analogues of known composition by free-choice profiling. Panel scores were analyzed by generalized Procrustes (a multivariate method). Plots of the Procrustes scores for the first 2 principal component axes from analysis of separate replicates reflected the experimental design. To compare replicates, scores from Procrustes analysis of the original data were subjected to Procrustes analysis. The results showed that panelists’ assessment of replicates was consistent. Regression of the scores from this second Procrustes analysis showed that they were well explained by the linear effects of either moisture in nonfat solids or fat content. Panelists who performed differently from the rest could be identified. Panelists were more sensitive to changes in moisture in nonfat solids content than fat. It was concluded that the sensory texture of the cheese analogues could be measured satisfactorily by free-choice profiling.     

A STUDY OF TEXTURE-FLAVOR INTERACTIONS USING FREE-CHOICE PROFILING

Interactions between perceptions of texture and volatile flavors were assessed by free-choice profiling using gelatin desserts as a model product. Gels were prepared with all combinations of 2 levels of gelatin, and 2 concentrations of 2 different flavors. A trained sensory panel profiled each sample for aroma, flavor, and oral and manual texture. Generalized Procrustes Analysis was carried out separately for each sensory modality. The results indicate that while gelatin concentration and flavor type both modified perception of flavor, flavor concentration did not. Gelatin content influenced instrumental, oral, and manual evaluations of texture; however, the type and level of flavoring had little influence on oral perception of texture.     

TEXTURE OF SWEET ORANGE GELS BY FREE-CHOICE PROFILING

Texture of orange gels prepared with 15% fruit pulp, sucrose up to 55° Brix and five different gelling agents — kappa-carrageenan, kappa-carrageenan plus locust bean gum, alginate, gellan gum, and gellan, xanthan and locust bean gums — was studied by Free-Choice Profile (FCP) analysis. Maximum rupture force and deformation at rupture were also determined by uniaxial compression in an Instron texturometer. Generalized Procrustes Analysis applied to FCP data permitted differentiation between samples and informed on the textural attributes responsible for the observed differences. Sensory differences were in general in accordance with mechanical differences. However, carrageenan and gellan gum gels were differentiated with the sensory method applied but not with mechanical tests.     

FREE-CHOICE PROFILING OF MILKS AND OTHER PRODUCTS PREPARED WITH MILKS OF DIFFERENT FAT CONTENTS

Free-choice profiling was used to develop a sensory profile of milks of different fat content and a range of food products prepared from them. The products were cornflakes, flavored milk, instant coffee, oat cereal, savory sauce, tea, and whipped dessert. The aim of the work was to investigate whether or not sensory differences brought about by using milk with a different fat content are noticeable in the context of use, and, if so how the differences can best be described. The milks of different fat content, the flavored milk and the cornflakes, savory sauce, and tea to which milk was added were significantly different from each other but the instant coffee, oat cereal, and whipped desserts were not. The differences found between the samples of different fat content were most often described with mouthfeel terms pertaining to viscosity, i.e., “thin/watery,”“buttery/fatty/ greasy/oily,”“coating/clinging,” and “creamy/rich.”     

MULTIDIMENSIONAL SCALING OF ODORANTS: EXAMINATION OF PSYCHOLOGICAL AND PHYSICOCHEMICAL DIMENSIONS

Nineteen odorants were arranged in a two-dimensional space byALSCAL, a nonmetric multidimensional scaling procedure, on thebasis of similarity of psychological odor quality. Adjectivescales concerning hedonic and ‘tactile’ qualitieswere related to the spatial arrangement Physiocochemical propertiessuch as Raman spectra, Laffort's parameters, and functionalgroup were related to the arrangement as well.     

ODOR QUALITY SIMILARITY SCALING AND ODOR-WORD PROFILE MATCHING

Multidimensional scalings of odor similarities in two conditionswere examined, with 7 odors, 15 subjects, and scaling in Euclideanspace with individual differences taken out as a third mode,by INDSCAL. It was shown that the presence of additional odorword profiles as dummy stimuli modified the relations betweenodors, thus attenuating the generality of verbally-based odorscaling methods.     

非度量多维测度及其在群落分类中的应用

排序与聚类分析法是近代最为常用的植被数量分析方法。原理上,排序较之于聚类法,一般具有较为严格的数学基础。在植被研究的应用方面,排序比聚类法能更好反映植被的连续性。同时,在反映植被分类结果上,排序图不仅象聚类树形图一样,使全部实体之间的划分关系得以反映,而且两两实体间的关系可以通过排序图上彼此间的距离比较而得以较好的反映。然而常规的排序方法主要适用于具有线性结构的数据分析,在2—3维排序图上常难以充分反映这些实体的生态学关系,造成大量生态学数据信息的损失。非度量多维测度(Non-metric multidimensional scaling)是近期发展起来的适用于非线性数据结构分析的一种复杂的迭代排序方法。它的基本思想是通过排序,n个实体在尽可能低维(t相似文献   

PERCEPTION OF FAT IN A MILK MODEL SYSTEM USING MULTIDIMENSIONAL SCALING

This study examined the contributions of stimulus fat content and flavor volatiles to the perception of fat in a milk model system. The model system was formulated by adding bland vegetable oil (0%, 5%, or 10% w/v) and natural cream flavor (0%, 0.5% or 1% w/v) to a skim milk base. Panelists judged pairs of samples for similarity on the basis of three attributes (fat content, mouthcoating and thickness) and the results were analyzed using a multidimensional scaling procedure. Two-dimensional solutions best represented the data. The stimulus spaces for fat content and mouthcoating were visually similar to each other and provided reasonable separation of the samples. Instrumental measures helped to define the underlying dimensions of the stimulus space for fat content. Dimension 1 related to texture and included the contributions of viscosity, and fat particle size and number distribution; dimension 2 related to flavor perception. The sample with a moderate fat content (5%) and the highest concentration of added cream flavor (1%) was perceived to be similar to the 10% fat samples with added flavor. Thus, the added flavor provided the sensation of higher fat content. These data suggest that flavor plays an important role in the preception of fat in dairy foods. A psychophysical model of fat perception in dairy foods is proposed which includes the contribution of viscosity, fat particle size and number distribution, and volatile flavor perceptions.     

EFFICIENCY OF A CYCLIC DESIGN AND A MULTIDIMENSIONAL SCALING SENSORY ANALYSIS TECHNIQUE IN THE STUDY OF SALT TASTE

Multidimensional scaling (MDS) procedures produce maps of stimulus points, the dimensions of which may be sensory and/or physicochemical attributes. Our objective was to test the efficiencies of a cyclic design and a MDS method in the study of saltiness perception. Two levels of NaCl (0.1% and 0.2%, w/v) were added to two nonionic and two ionic gum solutions prepared to two concentrations. Subjects scaled dissimilarities between pairs of solutions and rated each sample for flavor attributes. Solution pairs were selected using cyclic designs. Repeated measures ANOVA determined that added NaCl was the only significant factor affecting saltiness perception. In contrast, from KYST-2A MDS maps, we concluded that saltiness perception was influenced by gum property, gum concentration, presence of calcium and potassium, and related to binding of the sodium ion (Na⁺) as determined by ²³Na nuclear magnetic resonance (NMR) spectroscopy. The MDS cyclic design successfully reduced the number of samples and subjects when using experienced subjects.     

A REAPPRAISAL OF THE USE OF MULTIDIMENSIONAL SCALING TO INVESTIGATE THE SENSORY CHARACTERISTICS OF SWEETENERS

Previous investigations of the sensory characteristics of sweeteners using a multidimensional scaling (MDS) approach, have involved sweeteners which were not matched for sweetness. Under such circumstances, part of the estimated distance between sweeteners is attributable to sweetness differences. This detracts from the value of the consequent MDS space, when the main objective is usually to investigate sensory characteristics other than sweetness. In this study, the MDS approach was applied to sweetener solutions which were matched for sweetness with 5% sucrose. The direction of any residual sweetness differences was identified by including 1,3,5 and 7% sucrose solutions, all matched to equal viscosity, in the study. From the resulting three dimensional MDS sweetener space, it was evident that Dimension 1 was almost exclusively due to sweetness differences whereas Dimensions 2 and 3 were devoid of influence from sweetness and hence represent the sweeteners with respect to their other sensory characteristics.     

EVALUATION AND APPLICATIONS OF ODOR PROFILING

An odor profiling procedure was developed based on the ASTM odor profiling method. This modified procedure involved using approximately twenty panelists. Panel sessions and data collection were controlled by computer. The results obtained by this panel compared favorably to results obtained by the ASTM panel for which 150 panelists evaluated each compound, indicating that a small panel can be used to produce replicable results. Statistical methods of finding similarities and dissimilarities among compounds using profile data are discussed and compared to results from a multidimensional scaling (MDS) study in which degrees of differences among compounds were judged directly. These results indicate that profile data can be used to define and map the degree of similarity/dissimilarity among compounds, as well as to define the sensory dimensions on which these compounds differ. The use of factor analysis to study the underlying sensory dimensions of the odor space is also discussed. It is hoped that this type of research will lead to a better understanding of the underlying dimensions used to describe odorants.     

THE SCALING OF PLANT AND ANIMAL BODY MASS,LENGTH, AND DIAMETER

The interspecific scaling exponents of body mass M and diameter D with respect to length L were determined to evaluate the predictions of three scaling hypotheses (geometric, stress, and elastic similitude). The relation between M and L was determined for data from a total of 133 aquatic and terrestrial species (66 plant and 67 animal species); the relation between D and L was determined independently for a total of 753 aquatic and terrestrial species (667 plant and 86 animal species). Organisms were crudely classified as to their geometry (spheres, spheroids, cylinders) and shape (defined as the body slenderness factor, L/D) to examine whether geometry and shape evinced size-dependent variations. Regression indicated M = 1.29L^2.95 (r² = 0.91, N = 133; α_RMA = 3.09 ± 0.088). The stress and elastic similitude (which respectively predict α_RMA = 5 and α_RMA = 4) were rejected; geometric similitude was not (α_RMA = 3). For animals and plants, α_RMA = 2.81 ± 0.061 (r² = 0.98), and α_RMA = 2.95 ± 0.093 (r² = 0.94), respectively. For aquatics and terrestrial organisms, α_RMA = 2.82 ±0.134 (r² = 0.97, N = 36), and α_RMA = 3.08 ±0.111 (r² = 0.89, N = 97), respectively. These results were interpreted to support the hypothesis of geometric similitude. For the pooled plant and animals data, D = 0.05L^1.00 (r² = 0.95; α_RMA = 1.03 ± 0.009), which was compatible with the hypothesis of geometric similitude. For plants, D = 0.05L^1.06 (r² = 0.95; α_RMA = 1.09). For animals, D = 0.29L^0.98 (r² = 0.95; α_RMA = 1.01 ± 0.025). Also, for aquatics, α_RMA = 0.951 ± 0.151, whereas for terrestrial plants and animals, α_RMA = 1.03 ± 0.089. Although the scaling exponent for D differed among individual groupings of animals and plants, the results of regression analyses were interpreted to indicate that, on the average, body diameter scaled isometrically with respect to length as predicted by geometric similitude. For the pooled data set, organic shape varied over 3 orders of magnitude; L varied over 9 orders of magnitude reflecting 22 orders of magnitude of M. In terms of body geometry and the absolute numbers of species in the total data set: spherical shaped species (L = D) < unassigned species < prolate spheroidal species < cylindrical (squat < slender) species. The largest organisms in the data set were slender (L/D > 20) cylindrical plants; the smallest organisms were spherical plants and animals. Although not subject to statistical inference, these data were interpreted to indicate that organic shape and geometry evince size-dependent variations. These variations as well as size-dependent changes in bulk density are hypothesized to account for the scaling exponents of M and D determined for individual plant and animal clades and grades.     

SENSORY PROFILING AND POSITIONING OF COMMERCIAL SAMPLES OF MILK POWDER

Milk powders available commercially are of different qualities. The quality deviations may be due to inherent quality of the milk which in turn is dependent on various conditions like the fat content, the processing parameters, the storage time and the packaging materials used. The various brands of milk powders have been profiled sensorily for odor and flavor and also positioned using Canonical Discriminant Analysis (CDA). The study throws light on the relative qualities of the different brands, their positive and negative points and their relative position with respect to each other. The results indicate that the various types like whole milk powder, partially skimmed milk powder, skimmed milk powder and infant milk formula cluster together into the same group but the groups are separate from each other. The dominant quality attributes for the groups are different and are discussed in relation to the directional vectors.