A THURSTONE-COOMBS MODEL OF CONCURRENT RATINGS WITH SENSORY AND LIKING DIMENSIONS

A popular product testing procedure is to obtain sensory intensity and liking ratings from the same consumers. Consumers are instructed to attend to the sensory attribute, such as sweetness, when generating their liking response. We propose a new model of this concurrent ratings task that conjoins a unidimensional Thurstonian model of the ratings on the sensory dimension with a probabilistic version of Coombs' (1964) unfolding model for the liking dimension. The model assumes that the sensory characteristic of the product has a normal distribution over consumers. An individual consumer selects a sensory rating by comparing the perceived value on the sensory dimension to a set of criteria that partitions the axis into intervals. Each value on the rating scale is associated with a unique interval. To rate liking, the consumer imagines an ideal product, then computes the discrepancy or distance between the product as perceived by the consumer and this imagined ideal. A set of criteria are constructed on this discrepancy dimension that partition the axis into intervals. Each interval is associated with a unique liking rating. The ideal product is assumed to have a univariate normal distribution over consumers on the sensory attribute evaluated. The model is shown to account for 94.2% of the variance in a set of sample data and to fit this data significantly better than a bivariate normal model of the data (concurrent ratings, Thurstonian scaling, Coombs' unfolding model, sensory and liking ratings).     

WHAT SENSORY CHARACTERISTICS DRIVE PRODUCT QUALITY? AN ASSESSMENT OF INDIVIDUAL DIFFERENCES

Consumers differ dramatically in their sensory preferences for products. This paper assesses the relative importance of appearance, taste/flavor and texture as characteristics that drive acceptance for a complex pie product. The paper classifies consumers as being driven by one or more sensory attributes in their judgment of product quality. To do so, each consumer's liking rating is regressed against each separate attribute liking. The coefficient (M) of the linear equation "Overall Liking = M (Attribute Liking) + B" measures the relative importance of the sensory attribute as a driver of overall quality.     

COMPARISON OF ATTRIBUTE LIKING AND JAR SCALES TO EVALUATE THE ADEQUACY OF SENSORY ATTRIBUTES OF MILK DESSERTS

Just-about-right (JAR) scales and attribute liking questions are usually used to study consumer perception of the sensory characteristics of food products. The aim of the present work was to compare the performance of attribute liking and JAR scales to evaluate consumers' perceived adequacy of flavor and texture of milk puddings. Two groups of consumers were asked to evaluate eight milk desserts using (1) overall liking followed by attribute liking for texture and flavor and (2) overall liking followed by JAR scales for thickness, creaminess, sweetness and vanilla flavor. Overall liking scores were significantly different when JAR scales or attribute liking questions were considered. Texture, flavor and overall liking scores were highly correlated to each other, providing the same information. JAR scales correlated better with the intensity of sensory attributes evaluated by a trained sensory panel, being JAR percentages a reliable tool to study the adequacy of sensory attributes.

PRACTICAL APPLICATIONS

Results from the present work showed that consumers might not be able to independently evaluate their liking of different sensory attributes of a product. For this reason, the use of attribute liking questions for studying the adequacy of sensory attributes in complex products would not be recommended. JAR scales were better indicators of the adequacy of sensory attributes. Consumers were able to independently evaluate texture and flavor attributes using JAR scales. However, the influence of JAR scales on overall liking scores should be taken into account when including these scales on consumer studies.     

SENSORY PROPERTIES AFFECTING ACCEPTABILITY OF "BRESAOLA" FROM PODOLIAN YOUNG BULLS

Three muscles ( Semimembranosus, Quadriceps femoris, Adductor ) from Podolian young bulls (aged 18 months) were processed into bresaolas, which were compared with commercial products, in terms of sensory profile and preference expressed by consumers. Slope analysis was conducted to study the preference drivers. The preference of homogeneous groups of consumers and the relationships with sensory properties of bresaola were investigated using partial least squares (PLS) regression analysis. The products differed for several sensory properties, such as sweetness, flavor intensity ( P <  0.001), saltiness ( P <  0.01), peppered, seasoned and tenderness ( P <  0.05). The slope analysis showed that taste/flavor ( k =  0.86) and appearance ( k =  0.76) play an important role in orienting consumer preference, while texture ( k =  0.54) is less important. No general trend was found for product liking. Conversely, PLS regression clustered the consumers in homogeneous groups: one preferred the commercial products characterized by sensory properties such as peppered, saltiness, seasoned and marbling, another one was more oriented toward properties such as sweetness, odor and chewiness, which characterized Podolian bresaolas.

PRACTICAL APPLICATION

Partial least squares regression analysis clustered the consumers in homogeneous groups according to their liking: those preferring peppered, salty, seasoned and marbling bresaolas versus those oriented toward sweet and odorous products. In addition, the relationship between overall liking and attribute liking (appearance, taste/flavor and texture) was analyzed by linear regression analysis. This may be helpful in order to establish meaningful criteria for grading bresaolas and identify the particular attributes that move its acceptance (i.e., taste/flavor and appearance rather than texture).     

SENSORY DIRECTIONALS FOR PIZZA: A DEEPER ANALYSIS

This paper presents a set of analyses on sensory directional attributes used to rate experimentally designed pizza products. Consumers may or may not know the 'optimal' sensory level of attributes for pizza, so that the usefulness of the sensory directional varies by attribute. Furthermore, the sensory magnitude of each sensory directional attribute varies, as shown by the slope (B) relating the two attributes (Sensory Magnitude = A + B (Directional Rating)). The study incorporated sensory directionals into evaluation of products varied according to an experimental design. The optimal product emerging from the design does not necessarily exhibit a sensory directional profile where all attributes are 'on target', nor does a product whose sensory attributes are all on 'target' exhibit the highest level of liking.     

Evolution of heterogeneous perceptual limits and indifference in competitive foraging

The collective behaviour of animal and human groups emerges from the individual decisions and actions of their constituent members. Recent research has revealed many ways in which the behaviour of groups can be influenced by differences amongst their constituent individuals. The existence of individual differences that have implications for collective behaviour raises important questions. How are these differences generated and maintained? Are individual differences driven by exogenous factors, or are they a response to the social dilemmas these groups face? Here I consider the classic case of patch selection by foraging agents under conditions of social competition. I introduce a multilevel model wherein the perceptual sensitivities of agents evolve in response to their foraging success or failure over repeated patch selections. This model reveals a bifurcation in the population, creating a class of agents with no perceptual sensitivity. These agents exploit the social environment to avoid the costs of accurate perception, relying on other agents to make fitness rewards insensitive to the choice of foraging patch. This provides a individual-based evolutionary basis for models incorporating perceptual limits that have been proposed to explain observed deviations from the Ideal Free Distribution (IFD) in empirical studies, while showing that the common assumption in such models that agents share identical sensory limits is likely false. Further analysis of the model shows how agents develop perceptual strategic niches in response to environmental variability. The emergence of agents insensitive to reward differences also has implications for societal resource allocation problems, including the use of financial and prediction markets as mechanisms for aggregating collective wisdom.     

The Importance of Perceptual Experience in the Esthetic Appreciation of the Body

Several studies suggest that sociocultural models conveying extreme thinness as the widespread ideal of beauty exert an important influence on the perceptual and emotional representation of body image. The psychological mechanisms underlying such environmental influences, however, are unclear. Here, we utilized a perceptual adaptation paradigm to investigate how perceptual experience modulates body esthetic appreciation. We found that the liking judgments of round bodies increased or decreased after brief exposure to round or thin bodies, respectively. No change occurred in the liking judgments of thin bodies. The results suggest that perceptual experience may shape our esthetic appreciation to favor more familiar round body figures. Importantly, individuals with more deficits in interoceptive awareness were less prone to increase their liking ratings of round bodies after exposure, suggesting a specific risk factor for the susceptibility to the influence of the extreme thin vs. round body ideals of beauty portrayed by the media.     

Allelic variation in TAS2R bitter receptor genes associates with variation in sensations from and ingestive behaviors toward common bitter beverages in adults

The 25 human bitter receptors and their respective genes (TAS2Rs) contain unusually high levels of allelic variation, which may influence response to bitter compounds in the food supply. Phenotypes based on the perceived bitterness of single bitter compounds were first linked to food preference over 50 years ago. The most studied phenotype is propylthiouracil bitterness, which is mediated primarily by the TAS2R38 gene and possibly others. In a laboratory-based study, we tested for associations between TAS2R variants and sensations, liking, or intake of bitter beverages among healthy adults who were primarily of European ancestry. A haploblock across TAS2R3, TAS2R4, and TAS2R5 explained some variability in the bitterness of espresso coffee. For grapefruit juice, variation at a TAS2R19 single nucleotide polymorphism (SNP) was associated with increased bitterness and decreased liking. An association between a TAS2R16 SNP and alcohol intake was identified, and the putative TAS2R38-alcohol relationship was confirmed, although these polymorphisms did not explain sensory or hedonic responses to sampled scotch whisky. In summary, TAS2R polymorphisms appear to influence the sensations, liking, or intake of common and nutritionally significant beverages. Studying perceptual and behavioral differences in vivo using real foods and beverages may potentially identify polymorphisms related to dietary behavior even in the absence of known ligands.     

MARGARINE: THE DRIVERS OF LIKING AND IMAGE

This paper presents a study of the attributes of margarine, showing the depth of information about consumer perceptions and drivers of liking that emerges from a detailed analysis of relations among attributes. The paper develops three sets of analyses to understand relations among attributes: principal components analysis in order to identify basic dimensions of perception, linear functions relating overall liking to attribute liking or to image ratings in order to identify drivers of liking, and quadratic functions that relate overall liking or image ratings to sensory attribute levels in order to identify optimal sensory levels and to create sensory preference segments. The analyses show how consumer data can generate learning about the consumer perceptions on the one hand, and guidance for product development.     

Cortical Components of Reaction-Time during Perceptual Decisions in Humans

The mechanisms of perceptual decision-making are frequently studied through measurements of reaction time (RT). Classical sequential-sampling models (SSMs) of decision-making posit RT as the sum of non-overlapping sensory, evidence accumulation, and motor delays. In contrast, recent empirical evidence hints at a continuous-flow paradigm in which multiple motor plans evolve concurrently with the accumulation of sensory evidence. Here we employ a trial-to-trial reliability-based component analysis of encephalographic data acquired during a random-dot motion task to directly image continuous flow in the human brain. We identify three topographically distinct neural sources whose dynamics exhibit contemporaneous ramping to time-of-response, with the rate and duration of ramping discriminating fast and slow responses. Only one of these sources, a parietal component, exhibits dependence on strength-of-evidence. The remaining two components possess topographies consistent with origins in the motor system, and their covariation with RT overlaps in time with the evidence accumulation process. After fitting the behavioral data to a popular SSM, we find that the model decision variable is more closely matched to the combined activity of the three components than to their individual activity. Our results emphasize the role of motor variability in shaping RT distributions on perceptual decision tasks, suggesting that physiologically plausible computational accounts of perceptual decision-making must model the concurrent nature of evidence accumulation and motor planning.     

Predicting human perceptual decisions by decoding neuronal information profiles

Perception relies on the response of populations of neurons in sensory cortex. How the response profile of a neuronal population gives rise to perception and perceptual discrimination has been conceptualized in various ways. Here we suggest that neuronal population responses represent information about our environment explicitly as Fisher information (FI), which is a local measure of the variance estimate of the sensory input. We show how this sensory information can be read out and combined to infer from the available information profile which stimulus value is perceived during a fine discrimination task. In particular, we propose that the perceived stimulus corresponds to the stimulus value that leads to the same information for each of the alternative directions, and compare the model prediction to standard models considered in the literature (population vector, maximum likelihood, maximum-a-posteriori Bayesian inference). The models are applied to human performance in a motion discrimination task that induces perceptual misjudgements of a target direction of motion by task irrelevant motion in the spatial surround of the target stimulus (motion repulsion). By using the neurophysiological insight that surround motion suppresses neuronal responses to the target motion in the center, all models predicted the pattern of perceptual misjudgements. The variation of discrimination thresholds (error on the perceived value) was also explained through the changes of the total FI content with varying surround motion directions. The proposed FI decoding scheme incorporates recent neurophysiological evidence from macaque visual cortex showing that perceptual decisions do not rely on the most active neurons, but rather on the most informative neuronal responses. We statistically compare the prediction capability of the FI decoding approach and the standard decoding models. Notably, all models reproduced the variation of the perceived stimulus values for different surrounds, but with different neuronal tuning characteristics underlying perception. Compared to the FI approach the prediction power of the standard models was based on neurons with far wider tuning width and stronger surround suppression. Our study demonstrates that perceptual misjudgements can be based on neuronal populations encoding explicitly the available sensory information, and provides testable neurophysiological predictions on neuronal tuning characteristics underlying human perceptual decisions.     

QUESTIONNAIRE PRACTICE: WHAT HAPPENS WHEN THE JAR SCALE IS PLACED BETWEEN TWO “OVERALL” ACCEPTANCE SCALES?

ABSTRACT

Using the observational analysis of two different foods as test products, this note examined some interesting findings that emerge when the just‐about‐right (JAR) scale was placed after the overall liking rating, and before the preference ranking. The JAR data appeared to be statistically uncorrelated with the overall rating scales. First, this fortuitous observation from several studies suggests that the JAR scale can play a diagnostic role to determine how the consumer feels about changing the product. Second, the statistical analysis of the JAR ratings (e.g., bipolar nature of the JAR scale, its skewness and kurtosis) all suggest that the appropriate measure is %JAR, an incidence metric, rather than scale average, ought to be used both to report the results, and to create models relating to the JAR versus sensory magnitude or versus liking, respectively.

PRACTICAL APPLICATIONS

The results of the present analysis suggest that overall acceptance/liking can be the last question in the questionnaire design. That location is very important for the sensory analysis of foods eaten over time and then rated, as well as personal care and cosmetic studies where the products are used over time periods, and where the true evaluation of performance ought to be made at the end of the trial. In these long‐term uses, overall liking is generally the last question given at the end of the time period of product evaluation. The facts that the just‐about‐right (JAR) scales are independent of this final overall liking rating suggest that the integrity of the overall evaluation and usefulness of JAR as diagnostics will not be compromised.     

MAPPING THE EFFECT OF INFORMATION ABOUT ANIMAL WELFARE ON CONSUMER LIKING AND WILLINGNESS TO PAY FOR YOGURT

Preference mapping identified different groups of consumers on the basis of their disconfirmations (occurring when the product is either better or worse than expected) and assimilations (occurring when actual liking [L] moves toward the expectations). The negative disconfirmation of a group of consumers (Group 1) was based on the information about animal welfare (the products were worse than expected because the information about animal welfare induced high expectations), whereas in Group 2 the sensory properties of the products prevailed in orienting consumer disconfirmation (products were worse than expected because the sensory properties of low-fat yogurt were disliked). The map of assimilation showed that consumers from Group 1 had higher assimilation for plain yogurt associated with high welfare standards as a consequence of the high discrepancy between blind and expected L for these products. A similar behavior was observed for Group 2 (higher discrepancy between blind and expected L corresponding to higher assimilation for low-fat yogurt paired with high welfare standards).

PRACTICAL APPLICATIONS

Consumers based their choices both on sensory properties (plain yogurts were preferred to low-fat yogurts) and on information about animal welfare (products associated with high welfare standards were preferred to the others). Preference mapping was able to identify groups of consumers behaving differently, as compared with the general trend, on the basis of their disconfirmations and assimilations.
Consumer willingness to pay reflected the hedonic behavior, thus validating the auction procedure for food liking evaluation purposes and providing a useful tool to obtain information about the real value (i.e., in monetary terms) consumers give to animal welfare.     

Biochemical measurements of beef are a good predictor of untrained consumer sensory scores across muscles

The ability of the biochemical measurements, haem iron, intramuscular fat (IMF%), moisture content, and total, soluble and insoluble collagen contents, to predict untrained consumer sensory scores both across different muscles and within the same muscle from different carcasses were investigated. Sensory scores from 540 untrained French consumers (tenderness, flavour liking, juiciness and overall liking) were obtained for six muscles; outside (m. biceps femoris), topside (m. semimembranosus), striploin (m. longissimus thoracis), rump (m. gluteus medius), oyster blade (m. infraspinatus) and tenderloin (m. psoas major) from each of 18 French and 18 Australian cattle. The four sensory scores were weighted and combined into a single score termed MQ4, which was also analysed. All sensory scores were highly correlated with each other and with MQ4. This in part reflects the fact that MQ4 is derived from the consumer scores for tenderness, juiciness, flavour and overall liking and also reflects an interrelationship between the sensory scores themselves and in turn validates the use of the MQ4 term to reflect the scope of the consumer eating experience. When evaluated across the six different muscles, all biochemical measurements, except soluble collagen, had a significant effect on all of the sensory scores and MQ4. The average magnitude of impact of IMF%, haem iron, moisture content, total and insoluble collagen contents across the four different sensory scores are 34.9, 5.1, 7.2, 36.3 and 41.3, respectively. When evaluated within the same muscle, only IMF% and moisture content had a significant effect on overall liking (5.9 and 6.2, respectively) and flavour liking (6.1 and 6.4, respectively). These results indicate that in a commercial eating quality prediction model including muscle type, only IMF% or moisture content has the capacity to add any precision. However, all tested biochemical measurements, particularly IMF% and insoluble collagen contents, are strong predictors of eating quality when muscle type is not known. This demonstrates their potential usefulness in extrapolating the sensory data derived from these six muscles to other muscles with no sensory data, but with similar biochemical parameters, and therefore reducing the amount of future sensory testing required.     

THE USE OF FOURIER ANALYSIS TO COMPARE THE SENSORY PROFILES OF PRODUCTS

A consumer panel comprising 60 females evaluated three coffee whiteners and fresh milk for a number of sensory attributes including liking of flavor and appearance, overall liking, natural flavor and odor, creaminess and watery flavor. Fourier analysis was used to produce a plot which readily identified the whiteners with sensory profiles most similar to fresh milk. More importantly, the whitener whose sensory profile was significantly different from that of fresh milk was identified from the Fourier plot, something which could not be done by traditional univariate methods or even by an overall discriminant analysis.     

ON FITTING EQUATIONS TO SENSORY DATA: A POINT OF VIEW, AND A PARADOX IN MODELING AND OPTIMIZING

When fitting equations to data relating ingredients or factor scores to subjective ratings, there are at least two methods to create the equations. One method, (linear) forces in linear terms and allows additional quadratic and interaction terms. The other method, (quadratic) forces in linear and quadratic terms, and then permits cross-terms to enter. The two methods produce contradictory results. The first (and expected) contradiction is that the quadratic model shows optimum levels in the middle range of the levels tested for some, but not all, ingredients. The second (and unexpected) contradiction is that the linear method (which usually does not incorporate many additional terms) generates better validation predictions for " hold-out" samples than does the quadratic method. The differences between the optimum generated by the linear model and the optimum generated by the quadratic model can be quite substantial in terms of expected liking, sensory profile, and image profile.     

CATEGORY APPRAISAL WITHIN A LIMITED PRODUCT RANGE: A METHODS-ORIENTED CASE HISTORY

This paper presents the author's approach to synthesizing useful direction from product testing when the stimuli are not systematically varied. The approach presented here comprises a research design and data analysis strategy, rather than a conventional product optimization with subsequent validation. The design steps comprise stimulus selection, attribute selection, and product evaluation. The data analysis comprises univariate modeling to show how sensory attributes drive overall liking, reduction of the matrix to factor scores for multivariate modeling, and then the creation of an integrated product model. The outcome is a set of factor scores that can be translated to sensory attributes and in turn to target products.     

Perceptual constraints on optimal foraging: The effects of variation among foragers

Summary We present a simple model of habitat selection in which individuals differ in their ability to discriminate between resource sites' profitabilities. The model investigates the effects of violating the ideal assumption of the well-known ideal free distribution (IFD). We show that (1) variability in perceptual limits within a population can significantly change the distribution of foraging animals even though the mean perceptual limit is the same, (2) the direction of this change depends on the proportion of the population that choose randomly between resource sites and (3) better perceivers are more likely to be found at individually more profitable sites, which, because of undermatching with respect to the IFD, are also the absolutely more profitable sites. We note that variability in perceptual limits almost always led to an undermatching of organisms to resources, thereby extending previous workers' results implying that the incorporation of any form of perceptual limits leads to undermatching with respect to the IFD.     

Are Accuracy and Reaction Time Affected via Different Processes?

A recent study by van Ede et al. (2012) shows that the accuracy and reaction time in humans of tactile perceptual decisions are affected by an attentional cue via distinct cognitive and neural processes. These results are controversial as they undermine the notion that accuracy and reaction time are influenced by the same latent process that underlie the decision process. Typically, accumulation-to-bound models (like the drift diffusion model) can explain variability in both accuracy and reaction time by a change of a single parameter. To elaborate the findings of van Ede et al., we fitted the drift diffusion model to their behavioral data. Results show that both changes in accuracy and reaction time can be partly explained by an increase in the accumulation of sensory evidence (drift rate). In addition, a change in non-decision time is necessary to account for reaction time changes as well. These results provide a subtle explanation of how the underlying dynamics of the decision process might give rise to differences in both the speed and accuracy of perceptual tactile decisions. Furthermore, our analyses highlight the importance of applying a model-based approach, as the observed changes in the model parameters might be ecologically more valid, since they have an intuitive relationship with the neuronal processes underlying perceptual decision making.