Issue link: http://winesandvines.uberflip.com/i/438260
26 p r a c t i c a l w i n e r y & v i n e ya r d J a n U a r y 2 0 1 5 W I N E M A K I N G that the RWMW was used as a "start- ing point" in this study. However some significant differences are apparent and discussed below. The clear division of the wheel into dis- crete and integrated sensations partially addresses concerns with the RWMW that some terms are compound in nature and/or contain a hedonic component, and perhaps confound a common under- standing of some attributes. However, the lexicon is not free from this potential criticism, and further work is required to establish the reliability and descriptive value of the integrated descriptors. Fresh meringue, solid meringue, whipped cream and dessert mousse, for example, may seem unusual terms to describe wine. The panel felt strongly that these products, when assessed orally, elic- ited very similar sensations to many spar- kling wines with respect to the impression of fullness and dynamic feeling of expan- sion. Their actual utility in assisting with describing and discriminating sparkling wine remains to be determined. The RWMW contains descriptors that normally carry flavor connotations (retro- nasal odor — such as pepper, chili or resin- ous), even though they are defined without reference to olfaction, and a "slice" of the wheel is dedicated solely to flavor attributes (lift, activity and concentration). We attempted to avoid descriptors related to flavor in order to simplify the task of the assessor(s) and develop a lexicon as focused and unambiguous as possible. The ecological validity of using nose clips should be further investigated. For instance, how well do terms generated in the absence of olfactory cues describe the oral sensations experienced under normal wine evaluation conditions, and particularly the absence of nose clips? Numerous interactions between aroma, taste and tactile sensations have been doc- umented in the literature. For instance, taste may enhance aroma intensity or suppress it. 16,32 Conversely, aroma may enhance or suppress taste intensity. 8,28 Mediating factors in the direction and extent of these interactions include prior association with the particular flavor-taste combination 28 and whether a "gestalt" or analytical paradigm is used to assess stimulant combinations. 22 Irritants have also been shown to interact with both taste and smell perception, for instance suppressing sweetness 23 and some odors. 2 Studies have confirmed that tactile stimuli can also modify taste and aroma perception. 3 While the role of texture in controlling flavor release and thus aroma and taste intensities is significant and well-documented, 33 there is evidence that a perceptual integration of these modali- ties may also be involved. 34 Much of the research investigating tex- tural/tactile interactions with other sen- sory modes has been confined to solid and semi-solid foods/model foods or relatively simple food systems. However, given these previous findings, it is pos- sible that aroma cues will influence per- ception of some mouthfeel attributes elicited by white wine, although recent work from our lab found no effect on the perceived astringency and bitterness of catechin from the addition of odorants 35 Descriptors are ordered clockwise on the wheel, starting at "12 noon," accord- ing to the approximate order of percep- tion or maximum intensity experienced. This should assist judges in finding and rating terms more easily, particularly important given the large number of descriptors presented here and when assessing complex wines eliciting mul- tiple sensations. In contrast with the use of the RWMW by R. Gawel et al., 11,12 terms on the outer edge of the white wine wheel generally represent points on the scale for inner-tier sensations. The panel found this orga- nization useful when conceptualizing mouthfeel attributes and in subsequent scoring of wine, as they were already familiar with the line-scale anchor terms and qualities to which they related. Structuring of the wheel into two time components (early and finish) assists in more fully characterizing mouthfeel sen- sations, particularly given the known changes in intensity during evaluation of wine astringents. 10 The decision to reduce six original time-segments down to two was made primarily for prag- matic reasons, to increase the ease to use the wheel. However, more research is encouraged to fully define how oral sen- sations elicited by wine change in both quality and intensity over time. Scales and data analysis During preliminary trials using this wheel, the judges rated discrete sensa- tions using traditional, 15 cm visual ana- log scales anchored with adjectives and, where appropriate, reference standards (Figure II). Data collected in this way can be assessed using powerful parametric statistical techniques such as analysis of variance. Additionally, multivariate methods such as principal components analysis can be applied, which allow for thorough exploration of the data, includ- ing defining of relationships between the descriptors. Integrated terms are best rated as sim- ply present or absent. The data can then be analyzed using a number of non-para- metric options. Simply tabulating the fre- quency of use of each integrated term across the panel can provide valuable information of how wine samples differ for those attributes. Results for both dis- crete and integrated terms can be graphed using cobweb plots/radar graphs, which can provide an intuitive and clear presen- tation and description of each wine. Visual analog scales used for discrete sensations are typically scored by having each judge place a vertical mark on the scale that corresponds to their perception of the intensity or quality of the attribute under consideration. The distance from the origin (left hand anchor mark) to their mark is then measured, and that distance (intensity score) is recorded. The assessment is made relative to the reference standards and anchor terms, and the location of both on the scale can be altered depending on specific wines under examination. For example, if a panel is evaluating heavily extracted or acidic wines that might elicit coarser sensations, the chalk standard can replace talc to anchor the left-hand end of the particulate scale (see Figure II), thus increasing the range of scale available that is appropriate to wines under consideration. This can increase the likelihood of establishing differences between samples. In developing the lexicon, the panel rated some descriptors multiple times across different time-points. Particular sensations can be measured more than once during an evaluation to fully describe a sample, if needed. For exam- Figure II: Representative scales used for measuring discrete oral sensations elicited by white wine Sweetness Reference std. (standard) Reference std. 3 Reference std. 2 Reference std. Low Medium High Expansion (mousse evolution) Slow Medium Fast i V scosity Thin Thick Smoothness — in mouth Satin Silk Chamois Low Moderate High Mouthcoat Baby Oil Sunflower Oil Olive Oil Absent Light Moderate Heavy Length Short Medium Long Particulate — after expectoration Talc Chalk Plaster Grainy Fine Coarse