Paired sensory difference tests

Analytical methods can be sub-divided into difference tests and descriptive analysis. Difference tests such as triangle and paired comparison tests are designed to identify differences between samples. In a triangle test, the assessor is given three samples, two of which are the same, and asked to identify which sample is different. In a paired comparison test, the assessor is asked to identify whether there is a difference in a particular sensory characteristic between a pair of samples. Descriptive analysis, in which the sensory characteristics of a sample are described and scored on a scale, is probably the most important analytical method. 

The brewer and his customer make a subjective assessment of beer flavour each time they taste but for a more objective appraisal it is usually desirable to submit the beer, with suitable controls, to a taste panel. Taste panels may be used to (/) select qualified judges, (//) correlate sensory with chemical and physical measurements, (Hi) study processing effects, maintain quality, evaluate raw material selection, establish storage stability, and reduce costs, (fv) evaluate quality, and (v) determine consumer reaction [Ij. The types of test used include (/) difference tests, (ii) rank order, (Hi) scoring tests, (iv) descriptive tests, (v) hedonic scaling, and (v/) acceptance and preference tests [1]. Difference tests are most commonly used in the brewing industry, the results of which are readily analysed by statistics. Several forms of difference test are used. The A-not-A form of test is perhaps the simplest. Assessors are first familiarized with a standard A and then presented, in a random manner, either with A again or with the comparative sample B. In the paired-comparison test two samples are presented simultaneously (AA, AB, BA, or BB) and assessors report either there is a difference or there is no difference . 

Paired comparison There are two analytical sensory forms of this test the difference paired comparison (also called as the simple difference test) and the... 

The first technique is used when the experimenter wants to determine whether a simple sensory difference exists between two products (two-sided test) without specifying the dimension(s) (attributes) of the potential difference. Two odd or matched pair coded samples (A, B) are simultaneously evaluated in a randomized order of presentation between the assessors with equal number of the four possible serving sequences. (AB, BA, AA, BB). About 25 to 50 presentations are required to determine differences. The probability (P) that an answer occurs by chance is 1 /2. 

In order to test whether the products are discriminated, it is possible to run a multivariate analysis on variance (MANOVA) or a discriminant analysis on the rotated data after GPA (i.e. the (product subject) x GPA axes table), using product as a dependent variable. The general discrimination level can then be estimated using multivariate F-ratio (Wilks A,). Note that pair-wise differences can also be tested in a multivariate way, which may be useful for decision making. Alternatively, confidence ellipses can be calculated and drawn on the sensory map (Husson et al, 2005), as in Fig. 6.3. 

In paired comparison tests two different samples are presented and one asks which of the two samples has most of the sensory property of interest, e.g. which of two products has the sweetest taste (Fig. 38.3). The pairs are presented in random order to each assessor and preferably tested twice, reversing the presentation order on the second tasting session. Fairly large numbers (>30) of test subjects are required. If there are more than two samples to be tested, one may compare all possible pairs ( round robin ). Since the number of possible pairs grows rapidly with the number of different products this is only practical for sets of three to six products. By combining the information of all paired comparisons for all panellists one may determine a rank order of the products and determine significant differences. For example, in a paired comparison one compares three food products (A) the usual freeze-dried form, (B) a new freeze-dried product, (C) the new product, not freeze-dried. Each of the three pairs are tested twice by 13 panellists in two different presentation orders, A-B, B-A, A-C, C-A, B-C, C-B. The results are given in Table 38.3. 

Sensory Analysis. A paired comparison test was run to determine if the difference in oil droplet size in the emulsion changed the perceived intensity of the orange flavor. The coarsest emulsion (3.87 pM) and the Microfluidized sample (0.90 pM) from the third set of spray dried samples were compared. The solutions were prepared using 200 ppm flavor in a 10% (w/v) sucrose solution with 0.30% of a 50% citric acid solution added. The amount of each powder required to attain 200 ppm orange oil was calculated on the basis of percent oil in each powder (determined by Clevenger analysis). A pair of samples at approximately 10 C was given to each of 24 untrained panelists. The samples were coded with random numbers. Half the panelists were asked to taste the coarsest sample first while while the other half tasted the Microfluidized sample first. This was done to determine whether or not adaptation was a factor. The panelists were asked to indicate which sample had the most intense orange flavor. 

Sensory Evaluation. Results on the sensory evaluation of the three encapsulated powders showed that all three powders developed oxidized flavor even at first sampling time (3 days). Since an expert trained panel was used, the recognition threshold of members for oxidized flavor was far below the expected value. In addition, since oven stored samples were evaluated against freezer stored samples in the pair comparison test, panelists could not characterize the degree of difference in oxidized flavor between various powders. It is therefore suggested that lower storage... 

Parametric statistics (t-test, ANOVA) are by far the most commonly used in studies of sensory-motor/psychomotor performance due, in large part, to their availability and ability to draw out interactions between dependent variables. However, there is also a strong case for the use of non-parametric statistics. For example, the Wilcoxon matched-pairs statistic maybe preferable for both between-group and within-subject comparisons due to its greater robustness over its parametric paired f-test equivalent, with only minimal loss of power. This is important due to many sensory-motor measures having very non-Gaussian skewed distributions as well as considerably different variances between normal and patient groups. 

---