Search

People differ in their awareness for odours surrounding them. Body odours are a special category because they are a vehicle for social communication. Body odours evoke approach and avoidance behaviours such as withdrawing from social interaction, and personal hygiene behaviours like washing or using fragranced products. So far it has remained unclear what the role of conscious awareness of body odours is in guiding social behaviour. Here, we present a new psychological scale on odour awareness, focusing specifically on body odours: the Body Odour Awareness Scale (BOAS). The scale was validated measuring body odour awareness in two dimensions (valence and source) over four facets: awareness for one’s own body odours, both favourable and unfavourable, and awareness for other persons’ body odours, both favourable and unfavourable. An explorative follow-up study suggested regional differences exist in body odour awareness, but these are not the same for every dimension of body odour awareness. Taken together, these results suggest the new BOAS is a useful tool to assess differences in awareness for body odours, and uncover an application potential for this new and validated scale.

Many studies address the effect of the COVID-19 restrictions on food consumption and health, focusing on one or two measurements. Whether or not any effects are permanent or change over a longer period of restrictions has not been assessed in such studies. This study presented a survey containing questions on food consumption, exercise and self-assessed physical and mental health, repeatedly for six times over a 20-week period (July to November 2020) to a representative sample of 258 Dutch consumers. The majority of consumers reported no change in food consumption compared to before the COVID-19 restrictions, two smaller groups report a change to a more, or a less, healthy choice. This trend appears stable over the course of the measurements. The ‘healthy changers’ seem to couple a healthy and more diverse diet to healthy exercise habits, in contrast to the ‘unhealthy changers’. No change was observed in self-assessed physical health over the measurements. Overall a decline in time spent exercising showed, as well as a decline in self-assessed mental health. Clearly a lockdown situation affects food choice, exercise habits and (self-assessed) mental health. The fact that habits are able to suddenly change, for better or for worse, and that a decrease in mental health was reported, suggests that such insights need to be further explored to help individual consumers retain a healthy diet and lifestyle, and governments devise effective public health recommendations.

How do situations influence food desire? Although eating typically occurs in rich background situations, research on food desire often focuses on the properties of foods and consumers, rather than on the situations in which eating takes place. Here, we take a grounded cognition perspective and suggest that a situation that is congruent with consuming a food increases simulations of eating it, which, in turn, affect desire, and the expected and actual liking of the food. We tested this idea in four pre-registered experiments (N = 524). Participants processed an image of a food presented in a congruent situation, an incongruent situation, or no background situation. Compared to the incongruent situation, the congruent situation increased expected liking of the food and desire, and this was partially or fully mediated by eating simulations. The congruent situation also increased salivation, a physiological indicator of preparing to eat. However, there was only weak and indirect evidence for congruence effects on actual liking of the food when tasted. These findings show that situational cues can affect desire for food through eating simulations. Thus, background situations play an important but understudied role in human food desires. We address implications for research using food images, and for applications to promote healthy and sustainable eating behaviour.

The primary goal of this study is to be able to discern specific types of consumers in terms of their psychosocial characteristics who may need different ways of receiving dietary advice. Knowing these types will enable a better fit of advice to consumers’ psychosocial characteristics, hereby stimulating healthy eating as the probability of compliance to the advice can potentially increase. The study draws upon several psychological theories to distinguish unique underlying factors that can subsequently be used to personalize nutrition information for consumers. A number of general psychological scales (self-regulation, action and coping self-efficacy, social comparison, intrinsic motivation, health info processing, need for cognition and for affect, and regulatory focus) are filled out by 988 respondents, including their preferences for receiving personalized forms of nutrition advice. The set of joint items from various psychological constructs is analyzed using a Principal Component Analysis to find underlying psychological characteristics. The PCA produces four components (explaining 51% of variation), that could be interpreted as ‘intrinsic interest and capabilities for healthy eating,’ ‘perceived difficulty to eat healthily,’ ‘self-worth insecurity,’ and ‘seeking positive challenges,’ respectively. By means of a Logistic Regression these components are able to predict preferences for different forms of receiving nutrition advice. This first component shows that a mind set for maintaining a healthy diet goes together with an interest in receiving an advice on what do to and on how that will affect one’s health. The second component predicts a preference for a fixed moment to receive information/advice. This may be a strategy of those that perceive difficulties to eat healthily, to help them control their healthy food intake. The insecurity that the third component models seems to lead to a wish for receiving specific advice about their health situation at fixed moments in time. The fourth component is a small component, therefore its prediction of a wish for an advice focussing on prevention of negative consequences is probably not a strong result. The study does point out that there appear different psychosocial types of consumers, that may benefit by being addressed according to their preference for receiving nutrition advice on specific moments, of a specific level of detail or pointing at the type of consequences the advice has. A better fit of the advice to the psychosocial characteristics of the recipient, captured in the identified components in the current study, may lead to an increase in compliance, although that will have to be further investigated in subsequent work.

Prior research using immersive technology has shown that consumption contexts can be successfully recreated to elicit effects on hedonic food evaluations, but the food-specificity of these effects is as of yet unclear. The current study investigates how repeated exposure to foods in congruent and incongruent immersive contexts affects hedonic perception over time. Two groups of participants (N = 23, N = 18) were exposed to three food products (sushi, popsicle and iced tea) in either an immersive beach context or an immersive sushi restaurant context for seven days. On the eighth day, participants were exposed to the same foods once more, but switched to the other context. Hedonic evaluations were collected prior to and during tasting at initial exposure (day 1), after repeated exposure to the same context (day 7) and in the switched context (day 8); consumption behavior was monitored on exposure days (days 2–6). Results showed that prior to tasting, expected liking and desire to eat were higher for congruent food-context combinations (popsicle at the beach, sushi in the sushi restaurant) than for incongruent combinations (popsicle in the sushi restaurant, sushi at the beach). Upon tasting, no differences were found in average food liking, but individual liking ratings for congruent (vs. incongruent) food-context combinations were more consistent over time. Immersive contexts also elicited general effects, such that a typical consumption environment (sushi restaurant) increased food consumption and yielded more consistent product liking ratings over time. Findings underline the importance of taking a holistic view on consumer testing, and showcase the potential of immersive technology.

Initial research indicates that the use of immersive technologies may improve the predictive validity and reliability of liking scores in consumer testing. However, how immersive technologies impact Just-About-Right ratings is hardly known. Forty-five participants took part in three tasting sessions, each in a different context: 1) laboratory, 2) immersive context simulating a café using audiovisual cues, and 3) real café. Each session, participants tasted four tomato soups varying in salt content preceded by a warm-up sample. Liking, optimal levels of sensory attributes (JAR) and engagement were measured. Results showed that there were no differences in liking or JAR ratings on sensory attributes of the soups across the three contexts. Nevertheless, participants felt more engaged in the real café and simulated café than in the laboratory. These results contribute to a better understanding of how sensory differences as assessed in a laboratory or immersive context relate to sensory differences that consumers would notice when they use the products in real-life.

International audience; Background: Real-life human eating behaviour does not take place in a vacuum, rather it happens in context. The context in which consumers eat their foods influences the acceptance of the consumed foods. Consequently, consumers' hedonic and sensory ratings elicited in a natural consumption context will differ from those elicited under controlled sensory laboratory conditions. Moreover, foods are rarely consumed on one single occasion but are typically consumed repeatedly and ratings may change over repeated consumptions as well. Often, consumer acceptance is tested explicitly, for example with liking ratings, especially when the testing is done outside the laboratory. Implicit tests such as facial expressions and physiological measurements of the autonomic nervous system can provide additional information on consumer acceptance. As a result of technological advantages, such tests are no longer limited to the laboratory but can also be used in natural consumption contexts. Method: Eighteen healthy Dutch consumers (18–65 years of age) tested four test foods plus a warm-up sample ten times on consecutive weekdays and on similar hours using their own laptop and webcam. Test locations alternated between the sensory laboratory and the participant's own home. Explicit measures included liking scores and scores on ten sensory taste/flavour/texture attributes, and implicit measures included facial expressions, heart rate and consumption duration using Face ReaderTM. This study was the first to validate the Face ReaderTM for usage at home. Results: The liking scores and sensory profiles varied between test foods (p < 0.05), but not between test locations and only some specific sensory attributes showed systematic variation over repeated consumption. In contrast, implicit measures showed systematic effects of test foods, test locations, and repeated consumptions (p < 0.05). Compared to consumption in the laboratory, consumption at home was faster, triggered higher heart rates, and triggered more intense facial expressions of happiness, contempt, disgust and boredom. Conclusions: Implicit tests were more sensitive to effects of test location and repeated consumption than explicit tests. Additional research is required to investigate the relevance of these measures to long term consumer acceptance of food products.

Flavor is multisensory; several interacting sensory systems—taste, smell, and mouthfeel—together comprise “flavor,” making it a cognitively constructed percept rather than a bottom-up sensory one. In this chapter, some of the complications this entails for flavor priming are introduced, along with a taxonomy of different priming situations. In food-related applications of flavor, both bottom-up (sensory) as well as top-down (expectations) processes are at play. Most of the complex interactions that this leads to take place outside the awareness of the perceiving subject. A model is presented where many, past and current, aspects (sensory, surroundings, social, somatic, sentimental) of a (flavor) perception, together result in the perception of a flavor, its liking. or its choice. This model borrows on ideas from priming, situated/embodied cognition, and (food-related) perception.

Expectations about a food can impact on its taste, but this may represent a perceptual change or a bias in response at the decision-making stage. We hypothesised that expectation of taste intensity should be underpinned by modulation of activity in primary taste cortex. Using functional magnetic resonance imaging, we found that expecting a very sweet drink, but receiving a less sweet drink, enhanced the reported sweetness and bolstered activity in taste cortex, relative to a less sweet drink without this expectation. The activation overlapped with primary taste cortex activation found in 11 recent taste studies. Our findings provide evidence that taste expectation modulates activity in an area consistently reported as primary taste cortex, implying that expectation effects do indeed impact on taste perception.

The present study investigated the factors that affect the intention¿behaviour consistency of healthy snack choices. Intended snack choice was assessed by asking participants (N = 538) to choose a snack on paper, out of 8 snacks (4 healthy, e.g. melon and gingerbread, and 4 unhealthy, e.g. crisps and chocolate). The next day participants chose one out of 8 different snacks for actual consumption. Participants completed a questionnaire about attitudes towards taste and health, habitual snack use, self-control, anticipated regret, and pleasantness of the snacks. Results showed that 24% of the participants with a healthy snack choice intention chose an unhealthy snack instead. Female gender, a high education level, a strong habitual healthy snack use, and a strong self-control increased the healthy intention¿behaviour consistency. To facilitate healthy choices, interventions should target males and lower educated people, and focus on increasing their healthy snacking habit and self-control.

Objective To investigate dietary constructs that affect the discrepancy between intentioned and actual snack choice. Design Participants indicated their intentioned snack choice from a set of 4 snacks (2 healthful, 2 unhealthful). One week later, they actually chose a snack from the same set. Within 1 week after the actual choice, they completed a questionnaire that evaluated several dietary constructs. Setting Worksite cafeterias. Participants Office employees in the Netherlands (N = 585, 65% male, mean age 39.6 years [standard deviation = 9.2], 83% highly educated). Main Outcome Measures Snack choice intentions and actual snack choices (healthful vs unhealthful). Demographic and dietary constructs. Analysis Student t tests, chi-square tests, and logistic regression (P

The creation of a sensory descriptive panel for dry crusted, ‘crispy’ and ‘crunchy’ food products is presented. A sensory vocabulary comprising appearance, odour, taste, texture and sound is developed and the panel is trained to use these attributes. Model deep-fried battered snack and baked bread food products are used in the experiments with this panel. Special attention is given to ageing of the model products, and the effects this has on the sensory perception. The separation of the sensory assessment in a ‘first bite’ and a ‘during chewing’ phase turns out to be useful for this type of products, as does the inclusion of sound attributes. The vocabulary allowed both the effect of ageing and the differences between model products to be investigated in considerable detail.

Generalised Procrustes analysis (GPA) is a method for producing a group average from rotated versions of a set of individual data matrices followed by bi-linear approximation of this group average for graphical inspection. Partial Least Squares Regression (PLSR) is a method for relating one data matrix to another data matrix, via bi-linear low-rank regression modelling. The merger of these methods proposed aims to produce an average (e.g. a sensory group panel average), which balances an “intersubjective”, internal consensus between the individual assessors’ data against an “objective” external correspondence between the sensory data and other types of data on the same samples (e.g. design information, chemical or physical measurements or consumer data). Several ways of merging GPA with PLSR are possible, of which one is selected and applied. The proposed “GP–PLSR” method is compared to a conventional GPA followed by an independent PLSR, using a data set about milk samples assessed by a group of sensory judges with respect to a set of sensory descriptor terms, and also characterised by experimental design information about the samples. The GP–PLSR gave a more design-relevant group average than traditional GPA. The proposed algorithm was tested under artificially increased noise levels.

Texture and mouthfeel sensations of three groups of semisolid foodstuffs (mayonnaises and dressings, custard desserts and warm sauces) were characterised, measured, evaluated and compared based on quantitative descriptive sensory analysis of commercial products. Six groups of texture attributes were generated and used by the sensory panel. These were related to: viscosity, surface feel, bulk homogeneity, adhesion/cohesion, wetness-dryness and fat. Two groups of nontextural mouthfeel attributes were found to be relevant, which were related to perceived temperature and oral irritation. The importance of the attributes for each product category is discussed.

Temporal effects of dairy and vegetable fats (0 to 18%) on perception of strawberry flavor release and melting of ice cream were studied using the time intensity sensory method. Also, aroma and flavor attributes of the ice cream samples were evaluated. Only slight effects of fat on the rate of flavor release and flavor intensity were perceived. A slightly faster flavor release from the vegetable fat compared with dairy fat was noticed. Polydextrose and maltodextrin as bodying agents in the fat-free ice cream significantly increased flavor release and melting rate of the ice cream. Increasing fat content slightly retarded melting of ice cream in the mouth. No significant effect of the fat quality on perceived melting was noticed. Significant differences in aroma and flavor attributes of the fat-free and other samples were perceived. Intensity and sharpness of the strawberry aroma and flavor were greater in fat-free samples and they were perceived as nontypical. Fattiness and creaminess were highly correlated. Maltodextrin and polydextrose increased perceived fattiness and creaminess of fat-free ice cream.

Evaluation of the pleasantness of odours is an important function of the olfactory system. Electrodermal recordings monitor activity in the autonomic nervous system and it has been hypothesized that such measurements could be especially valuable for investigations of affective processes in the olfactory system. In this study we measured changes in skin conductance on both hands while subjects were exposed birhinally to two pleasant and two unpleasant odours. All odours were iso-intense and non-trigeminal. We wanted to test the hypothesis that skin conductance is contra laterally governed. Furthermore, we tested the hypothesis that unpleasant odours produce larger skin conductance responses than pleasant odours. Our results indicate that there are differential skin conductance responses to different odours, but we did not find that unpleasant odours in general produce larger skin conduction responses than pleasant odours. Neither did we find support for the hypothesis that skin conduction is contralaterally governed. Rather, our data are consistent with the existence of an ipsilateral system for the control of emotional responses.

The objective of this study was to investigate the effect beverages have on subsequently perceived basic tastes. Two types of milk, Cola and water were used to rinse the mouth prior to conducting a Time Intensity study with basic taste solutions. From this some effects of the prior beverage onto the TI parameters was found. Especially Cola showed an effect on the intensity of the subsequently tasted sweet taste, and whole milk showed a stronger effect than skim milk. The effects showed both in size and timing of the basic taste intensities.

A statistical method to reduce a large set of variables to a smaller subset of variables was published by Krzanowski [Krzanowski, W. J. (1987). Selection of variables to preserve multivariate data structure, using principal components. Applied Statistics , 36 (1), 22–33]. An application in the field of sensory science is presented in this paper. The method selects the subset from all possible subsets by matching the multidimensional configuration of objects of the subset to the full set of variables. To this end a Procrustes rotation is used and the subset which produces the lowest Procrustes loss in this matching is selected as the optimal subset. For two data sets the loss values of all possible subsets of all possible sizes are studied. It is concluded that considering the subsets corresponding to a range of lowest loss values should be considered instead of only the subset producing the lowest loss value. The method can easily be extended to include fitting methods other than Procrustes rotations and other optimality criteria than the Procrustes loss employed here.

Reducing salt levels in food products is an important motivation for research, as the general intake of salt by consumers is too high. Finding strategies for salt reduction, while maintaining salty taste of products remains a big challenge. In this paper we show the effect that a perceptual expectation of a taste can have on subsequent taste perception, in the same product. A perceptual expectation is based on previous experience, memory, or other information from the product. Our hypothesis is that, if a product looks the same, smells the same and has the same texture as the product one is used to eat, small variations in taste will go unnoticed. In a consumer study, we investigated if the expectation, or implicit assumption, that saltiness remains constant across mouthfuls, can indeed reduce the perception of variation in salt concentration. We were able to demonstrate the existence of such an effect. We found that perceptual expectation, based on the first bite, can influence saltiness perception.

A sensory vocabulary of 20 terms each with a corresponding reference material was developed over 7 sessions using pork patties derived from the meat of carriers and non-carriers of the RN − gene. Patties were oven-cooked at 150 and 170°C and chill-stored for up to 5 days to facilitate warmed-over flavour development. Generalised Procrustes Analysis (GPA) was used to investigate sensory terms and their individual use by panellists over the sessions. GPA explained variance indicated that the final vocabulary displayed a similar amount of information to that of the initial vocabulary of 42 terms. Individual panellists scale use was found to converge over the sessions. Panel agreement on many odour and flavour terms appeared to be enhanced as term synonyms were removed in vocabulary development. Sample discriminability decreased from sessions 1–4, where term concepts were verbally communicated to the panel. Term reference introduction in session 5 caused a levelling in sample discriminability and a reduction in agreement, most likely related to perceptual confusion. Subsequently, references enhanced both discriminability and agreement. Thus, it may be more useful to introduce reference materials earlier, if not in the first session, of the vocabulary development process.

The focus in this study was to study time intensity (TI) methodology and procedures of getting the sensory panel acquainted with this technique. By means of a descriptive profiling exercise effects of altering the structure of a β-lactoglobulin gel, laced with banana aroma on the perception of banana flavor were obtained and a suitable attribute for TI was selected. Samples made up by protein, without any fat are not ideal for a TI study of flavor release because of the fast release of aroma. However, although the concentration of banana aroma was identical in the four groups of gels the intensity was perceived as different. This was reflected in the descriptive profiling as well as in the dynamic study. The TI study showed that it is important to minimize distractive, noncrucial information during the exercise. The assessors had different abilities to connect with the dynamic data acquisition procedure and training made the assessments more congruent. The results also indicate a relationship between the appearance of the TI-curve and the score of the total banana aftertaste.

Human perception systems are generally geared towards detecting change. Constant stimulation is of low survival value, hence uninteresting. Most sensory methods focus on static judgements, but there is a class of methods especially adapted to measuring perceived change in stimulation from food. Most processes involved in eating, e.g. mastication and salivation, are dynamic processes, so methods acknowledging dynamic properties of eating are likely to produce results more valid than static methods. Food components as texturing agents, flavour systems, etc., have an impact on the dynamics of food breakdown and flavour release. Both from a fundamental—food perception and appreciation— and, from an applied—product development—viewpoint, dynamic sensory methods are worthwhile studying and employing.

An international study has been conducted in order to determine the respective contributions of androstenone and skatole to boar taint and their possible variations according to production systems and consumer populations. The presentation of the study and the main results concerning skatole and androstenone levels and data from sensory evaluation or consumer surveys are reported in companion papers. The present paper summarises the main conclusions of the study and gives tentative recommendations. A simulation study was conducted, based on the skatole and androstenone levels currently observed in European populations of entire male pigs and on the results of the consumer surveys. The first part of the simulation study demonstrated that, overall, 6.5% (odour) and 3.0% (flavour) more consumers were dissatisfied with entire male than with gilt pork. The differences were, however, very variable according to countries. Consumer dissatisfaction for the odour of entire male pork was mostly associated with high skatole levels, while androstenone had little influence on it. On the other hand, androstenone and skatole had similar contributions to the level of dissatisfaction for flavour. From the present study it is not possible to determine clear cut-off levels for androstenone/skatole. The regression equations presented in [Matthews, K. R., Homer, D. B., Punter, P., Béague, M. P., Gispert, M., Siret, F., Leask, H., Fonti i Furnols, M., & Bonneau, M. (2000). An international study on the importance of androstenone, skatole for boar taint: III. Consumer survey in seven European countries. Meat Science, 54, 271–283] provide a basis for decision making. However, due to methodological limitations, the results may underestimate consumer reaction to entire male pork. The second part of the simulation study demonstrated that sorting carcasses on the basis of androstenone/skatole would reduce, but not eliminate, differences in consumer dissatisfaction between entire male and gilt pork. For odour, taking androstenone into account did not improve the efficiency obtained from sorting using skatole only. For flavour, sorting using both compounds was more efficient than sorting using skatole only. Sorting out 15% of the entire males, on the basis of skatole only, would result in a difference in the proportion of dissatisfied consumers of 4.2% (odour) or 2.0% (flavour) between entire male and gilt pork. The results of the last part of the simulation study demonstrated that decreasing skatole in entire male pig populations, to levels as low as 0.10 ppm, would still result in a difference in the proportion of dissatisfied consumers of 3.2% (odour) or 1.6% (flavour). To reduce this difference further, the levels of both compounds would have to be reduced still further. The lowest difference that can be achieved is 2.3% (odour) or 0.4% (flavour). The conclusions of the present study may differ according to whether immediate commercial applications or long-term goals are considered. On the basis of the skatole and androstenone levels currently observed in entire male pig populations, sorting out procedures based on skatole is the easiest way to rapidly achieve a significant decrease in consumer dissatisfaction with entire male pork. In most countries, however, this will not be sufficient to obtain the same level of acceptability as with gilts. In the long term, a sharp reduction in both skatole and androstenone would have to be achieved in entire male pig populations to obtain low differences in acceptability between entire male and gilt pork.

The influence of a perceptual expectation, based on the first bite of a model product, on the perception of the taste of subsequent bites is investigated for bitter taste. A model product (chocolate ice cream) consisting of three regions of differing concentrations of a bitter compound (theobromine) was evaluated by a sensory panel using a multiple-bite Time–Intensity (TI) method. TI bitterness parameters were compared to those obtained with the model product containing the same amount of bitter compound but distributed homogeneously through the product. The relevant parameters of the resulting TI bitterness curves, i.e. the area under the curve, the time to return to zero, the overall maximum intensity, show that when the first bite is of a low bitter intensity, the perception of subsequent regions in the product assimilate towards a lower perceived bitterness intensity.

OVERALS is a technique for canonical correlation analysis with two or more sets of variables. Any three way table can be used as input for the OVERALS program. In OVERALS terminology the ways are called objects, variables and sets. Three measurement levels of the data can be handled: numerical, ordinal and nominal. They can be defined for each variable separately. Also the conditionality of the data is defined variable-wise. The OVERALS technique searches for what is common between sets of variables measured on the same objects. The mathematical model and the algorithm are discussed. In addition an illustration of the technique is given in the form of an application.

Perceived bitterness in drinks is known to fade slowly over time. Time‐Intensity curves are a means of studying this fading process. K subjects record the perceived bitterness intensity by moving a slider on a monitor using a mouse. Usually average TI‐curves are calculated to give a representation of the TI‐curve for one particular object. The problem is that there often are large individual differences, so the average TI‐curve is not always a good representation. An alternative is to perform a Principal Component Analysis on the matrices of objects by individual TI‐curves for each object. The resulting, so‐called, Principal Time Intensity Curves (PTIC's) are better representations than the average curves. Often the PTIC's for the different products are hard to distinguish. In this case a noncentered PCA of the matrix with curves gives results which show more differences between the products.

Here, we investigate whether the expectation that a drink tastes consistent across mouthfuls influences reported taste. A paradigm was developed to prompt the expectation that drink stimuli tasted identical to each other. Participants sipped two drinks and indicated whether they tasted the same. In some trials, the drinks appeared to be poured from the same jug, prompting the homogeneous taste expectation. These 'same-jug' drink pairs were reported more similar than 'different-jug' drink pairs, where the drinks were seen to be poured from different jugs, even when sweetness differed. This effect did not occur when the difference in sweetness between the drinks was too great. This suggests that expectancy effects act to smooth out variation in taste within a window of taste differences, until differences become perceptually noticeable. The expectation effect also dwindled over the course of the experiment, suggesting that expectations of similarity were updated with increasing experience of incongruous same-jug drinks. We conclude that flavor perception is best considered in terms of windows of flavor variation within which sensation changes to be more in line with expectations derived from past and current consumption experiences.

In their article Scriven and Mak (1991) claim to have used Generalized Procrustes Analysis (GPA) to analyze their data. However, the Consensus program they used does not perform a GPA. Since 1989 there is a very fast GPA program that runs on microprocessors.

Generalised Procrustes Analysis (GPA) is a method for matching several, possibly large, data sets by fitting them to each other using transformations, typically rotations. The linear version of GPA has been applied in a wide range of contexts. A non-linear extension of GPA is developed which uses Optimal Scaling (OS). The approach is suited to match data sets that contain nominal variables. A database of a Dutch power supplier that contains many categorical variables unfit for the usual linear GPA methodology is used to illustrate the approach. (c) 2009 Elsevier B.V. All rights reserved.

We discuss various issues surrounding the use and interpretation of Generalized Procrustes Analysis and related methods. Included are considerations that have to be made before starting an analysis, how to handle different dimensionalities of data, when to consider fitting scaling factors and when not to, and the distinction between the number of dimensions that are needed to give an adequate fit and the number of dimensions needed for graphical representation. The distinction between signal and noise plays an important part in explaining how different methods are suitable for exploring different aspects of the data, rather than being viewed as competing methods with the same general objectives. Explanations are largely set in a geometrical context, thus keeping technical mathematics to a minimum; a common Analysis of Variance framework allows all the methods to be considered in a unified way and suggests some new ways in which these kinds of data may be analysed. The whole is illustrated by example analyses.

Two different methods for generalized Procrustes analysis are introduced. One of the methods is used to illustrate the interpretation of the results of a conventional and of a free choice profiling experiment. The different sources of variance are partitioned over the products and judges. A relative measure corresponding to the percentage ‘variance accounted for’ is proposed as an alternative to the variance measures usually reported with GPA analyses. This relative measure is easy to interpret, graphical display of the measures helps in interpreting the ‘analysis of variance’ tables.

An analytical sensory test and an authenticity test were used to evaluate subtle flavour differences in milk. Consumers who habitually drink milk were selected for the tests. These consumers, we assume, possess implicitly learned knowledge about the taste of milk. In the authenticity test the consumers were able to differentiate milk samples which they were less able to discriminate in the analytical test. These results indicate that the consumers' ability to perceive subtle flavour differences is sharpened in this test as compared to the analytical test. The two tests are compared using d'-values, for which a novel calculation is suggested for the authenticity test. (c) 2005 Elsevier Ltd. All rights reserved.

Sensitivity to 6-n-propylthiouracil (PROP) for a group of 180 naïve consumers was related to their perception of 16 commercially available vanilla custard desserts. Rated intensities of taste and texture attributes varied moderately and inconsistently with PROP sensitivity. In contrast, discriminative ability increased with PROP sensitivity resulting in higher numbers of significant differences between pairs of custards. In terms of signal/noise theory, the results indicate that PROP sensitivity enhances the separation of the response signals but does not reduce their noise. The naïve consumers were also compared with highly trained panelists to test whether effects of PROP sensitivity resemble the effects of experience and training. Naïve consumers and trained panelists responded similarly with respect to taste and texture sensations such as creaminess and thickness, but were clearly different with respect to others such as heterogeneity and fattiness. Trained panelists demonstrated even stronger discriminative abilities than consumers with high PROP sensitivities for some attributes but weaker abilities for others. A practical implication of these findings is that selection criteria for participation in sensory panels should include PROP sensitivity, if the panel is aimed at maximum discriminative performance. © 2006 Elsevier Ltd. All rights reserved.

Flavour compounds vary in physicochemical properties and therefore behave differently in foods with different fat content. The objective was to investigate differences in dynamic flavour intensity in a realistic food system and relate them to two relevant molecular descriptors for flavour compounds. Ice creams with different fat levels (3%, 6% and 12% milk fat) and flavouring (β-ionone (berry), δ-nonalactone (coconut), isopentyl acetate (banana), vanillin (vanilla)) were examined. Approximately iso-intense concentrations (in 12% fat) were selected. Samples were analysed with time-intensity methodology, evaluating perceived melt rate and flavour intensity (trained panel N=12, 3 replicates). Data were analysed by ANOVA, principal component analysis (PCA) and ANOVA partial least squares regression (APLSR). Analyses of data showed faster perceived melt rates and faster increases and decreases in dynamic flavour perception with lower fat levels. Individual flavour compounds were not affected similarly by changes in fat level. Increase and decrease rates of dynamic flavour perception were related to boiling points and hydrophobicity of the flavour compounds. Isopentyl acetate with the lowest boiling point had the fastest increase rate and the slowest decrease rate in intensity. Vanillin with the lowest hydrophobicity index (logP) had the fastest decrease rate. δ-nonalactone and β-ionone with high boiling points and high logP had similar increase rates as vanillin, but had a slower decrease rate in flavour intensity.

We present a phenomenological parametric model for time-intensity (T-I) curves. The rising and decaying parts of the curve are described by logistic curves. The logarithm of their product models the sensory response. A crucial feature of the model is its ability to fit the long straight tails often seen in T-I data, as well as plateaus of arbitrary length. The parameters of the model have a straightforward correspondence to the shape of the curve. We discuss fitting of the model, show results for experimental data and show how systematic patterns of deviations from the prototypical T-I curve shape lead to directions for extension of the model.

A sensory analytical test, viz. descriptive analysis, and the authenticity test, which relies on affective responses, were used to evaluate subtle flavour differences in milk. We hypothesise that the affective test is more sensitive to small differences than the analytical test. Consumers, who drink milk habitually, have an expectation as to what constitutes the typical taste of milk. These expectations are encoded without the subjects' awareness and, hence, are implicit, which means that they are not consciously available to the consumers. In the present authenticity test an upsetting story concerning foreign milk on the Danish market was told, and the consumers were asked to identify foreign and Danish milk, among samples of milk which differed with regard to the feed given to the cows and the length of time the milk had been stored. A standard descriptive analysis was carried out with the same types of milk used for the authenticity test. The results suggest that subjects in the authenticity test and in descriptive analysis make use of different evaluative processes, and that the ability to perceive subtle flavour differences was sharpened in the authenticity test.

The effect of lateral presentation of odours on the hedonic evaluation is reported using a range of different substances. The hypothesis that hedonic evaluation of odours depends on stimulated nostril and on gender and handedness is tested using psychophysical methodology. A total of 51 untrained subjects evaluated 16 substances with different hedonic valences. Each odour substance was presented to the subjects four times, twice at each nostril using a balanced experimental design. Effects of gender and handedness, and interactions, are observed. Some parallels with the perception of visual emotional stimuli are suggested. Hedonic processing of odour stimuli is concluded to be an emotional, rather than an analytical task.

Time–intensity (TI) is an increasingly used sensory method, however, no proper guidelines for training panellists with TI seem to have been set up, hence comparison of results from different TI studies is difficult. In this paper a three steps approach is proposed: (1) introduction to the method; (2) training for the TI task through a simple product assessment: basic taste beverages; (3) running a TI pilot experiment. The loadings from a principal components analysis (PCA) on TI data have proved to be useful in assessing panel agreement and providing information about individual differences. Pilot profiling proved useful for choosing the attributes to be used in the TI study. This minimises the chance of performing a TI study with irrelevant attributes. It is concluded that, following the steps in this study, most panellists are able to learn to perform a TI task reliably.

What is the optimal size of an experiment? How should the practical experimenter determine this optimal experimental size? The paper presents a conceptually simple method for estimating the statistical power of experimental designs, based on Monte Carlo simulation. In the planning stage of a project, several alternative experimental designs may thereby be compared with respect to their ability to balance the risk of committing Type I and Type II errors against the cost. The Monte Carlo power estimation of a design is based on the following main steps: generate artificial data in a number of (say 5000) hypothetical experiments, based on the design and on certain assumptions; analyse each artificial data set in the same way that the future, real data set is intended to be analysed; from the distributions of the obtained parameter estimates, study the risks associated with the given experimental design. The method is illustrated for a factorial design in sensory analysis concerning warmed-over flavour development in meat. Here the Monte Carlo simulations indicated that four replicates were needed, given certain assumptions. The real experiment was performed independently twice, each time in four replicates. The resulting analysis of effects gave satisfactory results, indicating that four replicates had given the necessary and sufficient power in each of the two experiments. Copyright © 2000 John Wiley & Sons, Ltd.

Understanding the relationship between food texture perception and food structure is of increasing importance for companies wishing to produce texturally attractive food products. The perception of texture is a complex process involving the senses of vision, hearing, somesthesis and kinesthesis. Texture perception takes place partly during the dynamic process of food breakdown in the mouth and is affected by oral processes, such as motility, saliva production and temperature. To take account of these factors, a multidisciplinary approach is proposed for studying the relationship between food structure and texture perception, combining sensory research, physiology studies and research into food physicochemical characteristics. Recent developments in these three areas that give potential for a better understanding of texture perception and its relationship with food structure are discussed. © 2001 Elsevier Science Ltd. All rights reserved.

From the ‘ESN coffee data set’ the data from the two attributes that were used in all countries, bitter and burnt, were selected for analysis. The consistency of use of these attributes, over all countries, was studied using (Q-mode) Principal Component Analysis. In addition, the result from this analysis was used to study the agreement between the replicate coffees. The conclusion was that there were some deviant assessors in France, and that there might exist systematic differences between the replicates.

Usually time-intensity curves are summarised using a number of parameters, such as the maximum intensity, the time of maximum intensity, the area under the curve, etc. These parameters are derived from a TI curve averaged over assessors. This averaging has been the subject of some debate and some alternative methods to averaging the TI curves have been proposed. Recently a new approach, ‘the projected prototype curve model’, is suggested based on the assumption of an underlying smooth curve which is projected onto the data. This projection takes place in both the time and intensity domains separately. The model is applied to a set of TI curves and it is shown to provide a good fit to the data.