Food Research International 50 (2013) 176–188

Contents lists available at SciVerse ScienceDirect

Food Research International

j ourna l homepage: www.e lsev ie r .com/ locate / foodres

Sensory perception and understanding of food uniqueness: From the traditional tothe novel

Sara Favalli ⁎, Thomas Skov, Derek V. ByrneDepartment of Food Science, Faculty of Science, University of Copenhagen, Rolighedsvej 30, DK-1958, Frederiksberg C, Denmark

⁎ Corresponding author. Tel.: +45 35333173; fax: +E-mail address: sfa@life.ku.dk (S. Favalli).

0963-9969/$ – see front matter © 2012 Elsevier Ltd. Allhttp://dx.doi.org/10.1016/j.foodres.2012.10.007

a b s t r a c t

a r t i c l e i n f o

Article history:Received 11 August 2012Accepted 10 October 2012

Keywords:Multisensory perceptionConsumerUniquenessSensory descriptive analysisCrossmodal associationConceptualisation

This study investigated the overall consumer experience of a common complex food product by comparingits traditional and modern form. The aim of the study was to uncover the mechanisms, occurring in the con-sumers' perception, responsible for defining the key elements of the product and therefore its uniqueness.Descriptive analysis was carried out for 59 sensory descriptive attributes covering odour, appearance: colourand shape, basic taste, flavour, texture, mouth feeling and after taste. Moreover, conceptual measurements ofthe descriptors, familiar, common, complexity, harmonious, balance, luxury, elegant, inviting, interesting,feminine and masculine were also measured for all sandwiches. Sensory, conceptual and design relationshipswere analysed by ANOVA Partial Least Squares Regression (APLSR).The conjoint results indicate that the combination of appearance, and texture sensory attributes were able toaffect the overall conceptual understanding of this culturally defined complex food product and were consid-ered key modalities in terms of the product quality assessment. Thus, the results stressed the fact that theoverall consumer experience is a multisensory process in which cross-modal associations play a predominantrole. Further the findings confirmed that cognitive processes, and therefore conceptualisations are responsi-ble for this cross-modal associations and therefore of the overall consumer perception. However, retro nasalperception similar in both traditional and modern versions could be considered as core to the products' im-plicit definition in terms of uniqueness.

© 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Understanding the consumers' food experience means both defin-ing the product's uniqueness and the consumers' food identity.Where for stating a product as unique for a certain consumer, we in-tend addressing the characteristics, or key elements, which differenti-ate it from other products of its category and therefore constitute itsexclusiveness. Since food consumption is strictly related to the con-sumers' identity, while uncovering the uniqueness of a product for aspecific consumer, the consumer food identity is also revealed. “Tellme what you eat, and I will tell you what you are.” Jean AnthelmeBrillat-Savarin (1755–1826), “Man is what he eats.” Ludwig Andreasvon Feuerbach (1804–1872), and accordingly to Fischler “You arewhat you eat, you eat what you are”. In fact when we consume foodwe incorporate, on the real level, the nutrients and, on our personalimaginary level, all of its proprieties. Therefore from this perspective,our identity is represented by what we eat; in other words, we arewhat we eat. By eating the food we imparted a particular meaningto we hope to become more what we are or to become what wewould like to be. We eat what we are (Fischler, 1988). The consumers'food experience and their understanding of food products are strictly

45 3533 3509.

rights reserved.

related to their identity and therefore food culture. In fact, culture maybe one of the most powerful determinants of food attitudes and behav-iour (Rozin, 1990). Thus the product uniqueness is not an intrinsic char-acteristic of the product itself but it is defined by its users (family,regions, countries, social classes and social environments).

What are the unique food products which tell about food identi-ties in the market place? Following James' (1996) observations, wecould group the unique food products found in the market place infour groups. Precisely represented by “multinational fast food chains”(Coca-Cola, McDonald's and other kinds of brands), “expatriate food”(Provence, Tuscany and other foreign cuisines), “nostalgic foods”(traditional food products), and “creolising products” (combinationof various food cultural patterns, as ingredients cooking methods,spices etc., according to local conditions).

As Mattiacci and Vignali (2004) highlighted, for consumers,unique food products seem to have a distinctive and superior qualitywith respect to the general ones and thus producers can position theirunique food products in the premium price range and gain a higherprofit margin. This translated into the increasing spread of uniquefood products in the market place and in the increased business turn-over from them being generated. Apart from the perceived high qual-ity, in the last years the consumers' demand for unique food productshas increased rapidly, as a consequence of food scandals, health con-cerns and excessive homogenisation due to the industrialisation and

177S. Favalli et al. / Food Research International 50 (2013) 176–188

globalisation processes. Therefore, through unique food products theconsumer attempts to differentiate himself with qualitative aspects,for example organic food consumption, veganism, vegetarianism andcultural identification and ethnocentrism (Chambers, Lobb, Butler,Harvey, & Bruce Traill, 2007; Fandos & Flavian, 2006; Fischler, 1988;Parelli, 1996).

Consequently to the increased consumers' demand and spread ofunique food products in the market place raised, both at a governmen-tal, commercial and research levels, the need to characterise and definethe elements constituting unique products and at the same time theneed to understand the consumers' concepts related to unique productsand their consumption behaviour. Thus, an incrementing number ofstudies concerning unique food products, identities and eating arefound in the literature. Studies investigating ‘unique’ products foundin the literature deal with traditional food definition, innovation andimprovement (Almli, Verbeke, Vanhonacker, Næs, & Hersleth, 2011;Guerrero et al., 2009, 2010, 2012; Kühne, Vanhonacker, Gellynck, &Verbeke, 2010), motives of traditional food consumption (Pieniak,Verbeke, Vanhonacker, Guerrero, & Hersleth, 2009), traditional foodand global market (Jordana, 2000), sensory quality of traditional food(Cayot, 2007), objective sensory characteristics of traditional food prod-ucts (Guàrdia, Aguiar, Claret, Arnau, & Guerrero, 2010; Inarejos-Garcia,Santacatterina, Salvador, Fregapane, & Gomez-Alonso, 2010; Rason,Martin, Dufour, & Lebecque, 2007), and local food (Stolzenbach,Byrne, & Bredie, 2011; Trichopoulou, Vasilopoulou, Georga, Soukara, &Dilis, 2006). Studies concerning identities and eating focus on culturalpatterns of food consumption (Askegaard & Madsen, 1998), Europeandiets vs. traditional foods (Behar, 1976), gender (COUNIHAN, 1988;Shafer, 1979), personal traits (Lindeman & Stark, 1999; Sadalla &Burroughs, 1981), ethnicity and region (Bradby, 1997; Devine, Sobal,Bisogni, & Connors, 1999; Kalcik, 1984) vegetarianism (Jabs, Devine, &Sobal, 1998; Jabs, Sobal, & Devine, 2000), organic food use (Shepherd& Raats, 1996; Sparks & Shepherd, 1992), beef eating (Sapp & Harrod,1989), and dietary change for health promotion or illness management(Cohn, 1997; Fries & Croyle, 1993; Lindeman & Stark, 1999). The identi-ty concept has also been applied to the study ofweight, body image, andeating disorders (Cordell & Ronai, 1999; Stein, 1996).

Particularly, the extensive literature in the area of unique foodshighly indicated the necessity of understanding the key elements defin-ing the exclusiveness of a product for its consumers, the consumers'preferences and their buying behaviour (Askegaard & Madsen, 1998;Cayot, 2007; Guerrero et al., 2012), and contrasting the homologationconsequently by the globalisation process perpetuates important ele-ments of our cultural inheritance and thus contributes to the beneficialhealth of theMediterranean diet (Trichopoulou et al., 2006). Avermaeteet al. (2004) suggested that unique foods are essential for the develop-ment and sustainability of rural areas, hence contrasting the actual de-population process. Moreover, unique foods entail substantial productdifferentiation potential for producers and processors. Thus, the devel-opment of unique products can be seen as an opportunity of global seg-mentation for the producers of typical products and thus sustainabilityand development of the small andmedium enterprises (SMEs). Concur-rently it can also affect it indirectly, throughout tourism, which stimu-lates the desire to discover products and modality of consumption byvisiting the specific products site (Mattiacci & Vignali, 2004).

Despite the existing literature researching on uniqueness and foodidentity highlighting the different aspects of it, a holistic perspectivecomprehending all those aspects ismissing. In factwhile some of the in-vestigations assessed and defined the objective unique properties ofspecific foods not taking into account the consumer point of view(Guàrdia et al., 2010; Inarejos-Garcia et al., 2010; Rason et al., 2007),others investigated the consumer concepts defining unique food prod-ucts without linking it with the objective food properties (Almli et al.,2011; Guerrero et al., 2009, 2010, 2012; Kühne et al., 2010).

As Fischler (1988) and Schifferstein (2010) indicated the human re-lation to food is complex. Therefore only through cross-disciplinary and

multidimensional assessments that the overall consumer experienceand the uniqueness concept could be delineated.

The aim of the present study was therefore to investigate and de-fine, through a holistic perspective, the meaning of food uniquenessand the mechanisms responsible for its formulation. In other words,it is understanding which combination of sensory properties andconceptualisations is responsible for the products' uniqueness/authenticity. For doing so, the overall consumer experience of a com-plex food product, in its traditional and novel form, has been assessed.

Because of its holistic perspective (combination of both objectiveproperties and consumer concepts), the study represents a new ap-proach and an advance to the general knowledge in the area of fooduniqueness. Moreover the study constitutes the base for a “prelimi-nary model” of uniqueness, where for “model” we intend a “frame-work for the different components of uniqueness and identity”. The“preliminary model” could be useful as a tool to better understandthe overall consumer experience and thus the factors influencingfood choices and behaviours. In fact, the understanding of the overallconsumer experience is vital for the successful development andcommercialisation of food products. Moreover the findings could sup-port the health promotions.

2. Material and methods

2.1. Sensory descriptive analysis

2.1.1. Experimental design/productsTwelve commercially available Danish open-faced sandwiches were

analysed. Six types of ‘smørrebrød’ having asmain ingredientmeat (pigDanish paté, pork roast, Danish spiced pork meat roll, and roast-beef),or fish (marinated herring and egg and shrimp), were analysed. Eachtype was investigated in its traditional and a novel form. The traditionalsamples are industrial based and prepared as the simplest version of thetraditional ones (normally with higher ingredients amount and decora-tion). The novel samples are artisanal products and prepared as thenovel version of the classical Danish open-faced sandwiches. Details ofthe samples are provided in Table 1.

2.1.2. Panel selectionTen expert judges, four males and six females, nine aged between

twenty and forty-two and one aged sixty-three, have been hired to per-form the descriptive analysis. The judges were selected from the exter-nal sensory panel of Faculty of Sciences, University of Copenhagen,Denmark. The external panel was recruited according to the selectioncriteria of ISO 3972 (1991).

2.1.3. Sample preparationThe samples have been placed on a paper foil (for food use, rectan-

gular, same size), in transparent plastic boxes, and stored at 14.5 °Cfor 1 h before serving. References have been prepared every day be-fore the training session at 14.5 °C before serving.

2.1.4. Vocabulary development and profilingThe sensory descriptive analysis consisted of four training sessions

(9 h in total) and four sensory profiling sessions (2 h each). The aimof the training session was to develop a consensus vocabulary and im-prove the panellists' cognitive clarity toward the descriptors discrim-inating the honeys and the use of the sensory intensity scale. Duringthe training session the judges have been instructed on the procedureto follow when evaluating the samples. Particularly panellists havebeen instructed to get all the ingredients composing the sample inone bite. Initially the panellists individually generated sensory de-scriptors by adding them on a pre-existent list. Subsequently, thejudges discussed in pair the attributes and highlighted the most im-portant ones to describe the sample. The final vocabulary consistedof fifty-nine attributes (Table 2) developed via collaboration between

Table 1The samples in details.

Novel sample (artisanal based) Traditional sample (industrialbased)

Experiment name Experiment nameN-pig paté T-pig paté

Product's name Product's nameDyrlægens natmad (Dyrlægens natmad) Leverpostej (lever paté)

Ingredients IngredientsRye bread, liver paste, veal saltymeat slice,gravy jelly, cress

Rye bread, liver paste,cucumber

Experiment name Experiment nameN-roast beef T-roast beef

Product's name Product's nameRoast beef Roast beef

Ingredients IngredientsRye bread, roast-beef slices, horseradish,remoulade in big chopped pieces, crispyonions, chervil

Rye bread, roast-beef slices,remoulade, crispy onions

Experiment name Experiment nameN-spiced pork meat roll T-spiced pork meat roll

Product's name Product's nameRullepølse (cured pork meat) Rullepølse (cured pork meat)

Ingredients IngredientsRye bread, mustard, rullepølse (cured porkmeat), horseradish, gravy jelly, cress, rawonions

Rye bread, Rullepølse (curedpork meat), raw onions

Experiment name Experiment nameN-pork roast T-pork roast

Product's name Product's nameFlæskesteg (roast pork) Flæskesteg (roast pork)

Ingredients IngredientsRye bread, mustard, roast pork, red cabbage,pickled Persians cucumber, fried pork rind

Rye bread, roast pork, redcabbage

Experiment name Experiment nameN-marinated herring T-herring

Product's name Product's nameOld days Icelandic herring in carry souse Sild (herring)

Ingredients IngredientsOld days Icelandic herring in curry souse, appleand Jerusalem artichoke compost,Crème fraîche, shallot

Marinated herring and currysouse

Experiment name Experiment nameN-egg and shrimp T-egg

Product's name Product's nameEgg and shrimp Egg

Ingredients IngredientsEgg, peeled shrimps, mayonnaise, cress Egg, peeled shrimps,

mayonnaise, tomatoes

Prefix to the experiment name indicate N-novel and T-traditional.

178 S. Favalli et al. / Food Research International 50 (2013) 176–188

the panellists and the panel leader (Byrne, Bak, Bredie, Bertelsen, &Martens, 1999; Byrne, Bredie, & Martens, 1999).

For the sensory descriptive analysis, a 15-cm unstructured linescale was used. The 12 sandwiches were evaluated with three sensoryreplicates served across 4 days following a randomised incompleteblock design. The samples were presented in randomised order tominimise systematic carry-over effects. The data was collected byFizz Acquisition, Version 2.45A, Biosystems, France. The physical eval-uation conditions for the sensory descriptive analysis followed theguidelines in ISO 13299 (2003).

2.2. Conceptual evaluation

In order to add a new perspective of the Danish open-faced sand-wiches and achieve a more complete understanding of the sensory ex-perience of the sandwiches, eleven conceptual descriptors wereincluded in the sensory descriptive analysis. The eleven conceptual de-scriptors were developed throughout a two-step process: a first list ofconceptual attributes was generated from a focus group composed ofDanish people. Subsequently, the attributes of the first list were refinedin order to obtain the final eleven attributes (Table 3). The eleven

attributes were rated without prior training by the same panellist usedin the sensory descriptive analysis. The panellist rated the attributes ona 15-cm unstructured line scale after giving scores for the sensory de-scriptive terms. The descriptive analysis and the conceptual evaluationwere carried out as two consecutive sessions. These data was also col-lected by Fizz Acquisition, Version 2.45A, Biosystems, France.

2.3. Statistical analysis

The progress in panel performance over training sessions was evalu-ated by PanelCheck version 1.3.2 (Nofima, Norway) using the applica-tions Tucker1-plot, profile plots, Eggshell plots and p-MSE plots. All thesubsequent analyses of the sensory data were performed in Unscram-bler X, version 10.1 (CAMO ASA, Norway). First, the sensory raw profil-ing data was pre-processed by means of leveraged correction. Thesensory data was subsequently ready for multivariate data analysis,and precisely for ANOVA Partial Least Square Regression (APLSR). Tofocus on interpretation the fifty-nine descriptive objective attributeswere grouped, following the physiological classification, in the followingthree groups of analysis: “appearance (shapes and colours)”; “odours,basic tastes and flavours”; and “textures, mouth feelings, after tastes”.APLSR visualises anddetermines the structure anddegree of each designvariable's (sandwich samples) contribution to the variation in sensorydata. Thus, APLSR was performed with the X-matrix as 0/1 design vari-ables and the Y-matrix as the sensory descriptors (Byrne, Bredie, &Martens, 1999; Martens & Martens, 2000). The multivariate data analy-sis was based on averaged values over panellists and sensory replicates.The data was mean-centred. To derive significance indications from thesensory data and linkages between the sensory data and conceptual at-tributes, the data was analysed by derived estimated regression coeffi-cients and jack-knife uncertainty testing based on full cross validation(Martens & Martens, 2000).

3. Results

3.1. Data analytical strategy

In order to investigate the uniqueness of the open-faced sandwichesthrough a multisensory approach, the analysis of the data has been di-vided into three parts. Each of the analyses has been performed bymeans of APLSR in order to check the significant relevance of the results.At first, the descriptive sensory profile has been analysed in order to de-fine the objective sensory properties of the twelve products (objectivesensory characterisation of uniqueness). Secondly, the conceptual attri-butes have been assessed (conceptual characterisation of uniqueness).Ultimately, a combined analysis of both sensory and conceptual attri-butes has been performed (sensory perception and understanding ofuniqueness).

3.2. Objective sensory characterisation of uniqueness

The first APLSR analysis of all sensory attributes and samples re-vealed that the traditional and novel pair samples were rather similarbased on the sensory characteristics (visually evaluated from scores —not shown here). In order to have insight into this first result, the anal-ysis has been performed assessing the six pair samples in each of thethree different modalities of visual, retro-nasal and orthonasal, and tac-tile sensations and after tastes. Thus, any significant differences in thesensory properties of the six pair samples according to the differentmo-dalities have been enquired. Details of the just mentioned analyses areillustrated in the three following subparagraphs.

3.2.1. Objective visual assessment: the sample pairs were differentlydescribed

As regard the appearance, the novel and traditional pair sand-wiches were differently described (Fig. 1).

Table 2List of the sensory descriptors developed for the profiling.

Sensory descriptorsa Definition with reference materialsb Scale

Appearance (shapes and colours)Colour Degree of colour by visual evaluation

1 A-Colourful Degree of colourfulness Few→many2 A-Colour intensity Degree of colour intensity Slightly→a lot3 A-Yellow Card colour: blade 8 of DSM Yolk Coulor Fan, http://www.dsm.com None→extremely4 A-Green Card colour: Diana 2122 (Beckers) None→extremely5 A-Violet Card colour: Siv 2113 (Beckers) None→extremely6 A-Brown Card colour: Lukas 2007 (Beckers) None→extremely7 A-White Card colour: Alfred 0401 (Beckers) None→extremely8 A-Silver Argent 800 (Pigments PW6/PW15/PW20/PBK1) Rembrandt acrylic, Royal Talens None→extremely9 A-Red Card colour: Naemi 2110 (Beckers) None→extremely10 A-Dark Degree of dark colours None→extremely11 A-Light Degree of light colours None→extremely12 A-Shiny Degree of shininess None→extremely13 A-Iridescent Salted meat None→extremely14 A-Freshly made Degree of freshly made appearance None→extremely

Shape15 A-Complex Degree of complexity of the sample Not complex→very complex16 A-Structured Degree of structure of the sample Slightly→a lot17 A-Soft curves Degree of soft curved of the sample Slightly→a lot18 A-Sharp curve Degree of sharp curve of the sample Slightly→a lot19 A-Height Degree of highness of the sample Short→height

Odours, basic tastes and flavorsOdour Odour associated with

20 O-Intensity Degree of odour intensity with one sniff None→extremely21 O-Spicy 1 bay leaf (40 g ca.)+4 whole black peppers+2 whole cloves+pinch of allspice powder+30 g of cold water. None→extremely22 O-Green 3 g lettuce leaves+5 g cucumber (small cubes) None→extremely23 O-Sulfur Egg cooked for 20 min and peeled None→extremely24 O-Trigeminal 1 g wasabi+1 g cress None→extremely25 O-Fried Roasted onions None→extremely26 O-Meaty Slices of roastbeef+roast pork None→extremely27 O-Liver 5 g backed liver (oven at 175 °C, 10 min) None→extremely28 O-Ocean 0.25 g seaweed tang extracted in 1 dl boiling water (2 min) served at 4 °C None→extremely29 O-Dry fruit Sun-dried figs None→extremely30 O-Rye bread Mix of two different rye bread without crust (5 g “Hercules–Kohberg”+5 g homemade bread from Aamanns) None→extremely31 O-Butter 4 g butter None→extremely32 O-Vinegar 150 ml vinegar+150 ml tap water None→extremely

Basic taste Taste associated with33 BT-Salty 1.40 g/L Sodium chloride (NaCl) Slightly→a lot34 BT-Bitter 0.27 g/L Caffeine (stand overnight) Slightly→a lot35 BT-Umami 0.7 g/L L-Glutamic acid monosodium salt monohydrate Slightly→a lot36 BT-Sour 0.6 g/L Citric acid monohydrate Slightly→a lot37 BT-Sweet 7.20 g/L sucrose solution Slightly→a lot

Flavour Aromatic taste sensation associated with38 F-Trigeminal 1 g wasabi+1 g cress None→extremely39 F-Fried 0.5 g roasted onion None→extremely40 F-Meaty Slices of roastbeef+roast pork None→extremely41 F-Liver 5 g backed liver (in the oven at 175 °C, 10 min) None→extremely42 F-Ocean 3 g Lettuce leaves+5 g cucumber cut into small cubes. None→extremely43 F-Dry Fruit Sun-dried figs None→extremely44 F-Green 3 g Lettuce leaves+5 g cucumber cut into small cubes. None→extremely45 F-Rye bread Mix of two different rye bread without crust (5 g “Hercules–Kohberg”+5 g homemade bread from Aamanns) None→extremely

Textures, mouth feeling and after tastesAfter taste After taste associated with

46 AT-Butter 4 g butter None→extremely47 AT-Oily Refined sunflower oil (Irma) None→extremely48 AT-Spicy 1 bay leaf (40 g ca.)+4 whole black peppers+2 whole cloves+pinch of allspice powder+30 g of cold water. None→extremely49 AT-Vinegar 150 ml vinegar+150 ml tap water None→extremely50 AT-Rye bread Mix of two different rye bread without crust (5 g “Hercules-Kohberg”+5 g homemade bread from Aamanns ) None→extremely51 AT-Durability Time intensity of the aftertaste Short→ long

Mouth feeling Mouth feeling associated with52 MF-Coating Refined sunflower oil None→extremely

Texture Texture associated with53 T-Crispiness Green salad leaves None→extremely54 T-Crunchiness Wholegrain cracker (knækbrød) — minimum None→extremely55 T-Soft Resistance given when chewing None→extremely56 T-Gritty Quantity of particles left in the mouth after swallowing None→extremely57 T-Juiciness Juiciness while chewing None→extremely58 T-Dry Dryness of the sample None→extremely59 T-Chewing resistant Power used to chew the sample Slightly→a lot

The sensory attributes are grouped here according to the three modalities in which the analyses of data have been performed: ‘appearance (shapes and colours)’, ‘odours, basictastes and flavours’ and ‘textures, mouth feeling and after tastes’.

a Prefix to the sensory descriptors indicates the type of assessment by panellists: A-appearance. O-odour. BT-basic taste. F-flavour. AT-after taste. MF-mouthfeeling. T-texture.b Definition of sensory descriptors as derived during vocabulary development.

179S. Favalli et al. / Food Research International 50 (2013) 176–188

The main variation of the traditional and novel pair sandwicheswas explained by the attributes height, structured, complex, andsoft curves and colourful for the X-9% and Y-22% (PC1) spanning

the pairs in novel and traditional (see Table 4 for the significancevalues). The sandwiches variation was also explained for the X-9%and Y-16% by the light and dark colours (PC2).

Table 3List of conceptual descriptors.

Conceptual descriptors Scale

1 Familiar Not familiar→ familiar2 Inviting Not inviting→ inviting3 Common Not common→common4 Luxurious Not luxurious→ luxurious5 Interesting Not interesting→ interesting6 Complexity Simple→complex7 Harmonious Not harmonious→harmonious8 Balanced Not balanced→balanced9 Elegant Not elegant→elegant10 Feminine Not feminine→ feminine11 Masculine Not masculine→masculine

180 S. Favalli et al. / Food Research International 50 (2013) 176–188

3.2.2. Objective odours, basic tastes and flavours: no variation of thesample pairs

No variation was explained in the traditional and novel pair sand-wiches according to odours, basic tastes and flavours (Fig. 2). The pairsamples were in fact clustered together. The main variation betweensandwiches was explained for the X-9% and Y-21% (PC1) by theodours and flavours, ocean and meaty spanning the sandwiches infishy and meaty (see Table 5 for the significance values).

3.2.3. Objective textures, mouth feeling and after tastes: the sample pairswere differently described

Traditional and novel pair sandwiches were different as regard thetexture, mouth feeling and after taste attributes (Fig. 3). Texture attri-butes of juiciness, dry and gritty and the aftertaste attributes of ryebread and durability were responsible for the X-9% and Y-18% varia-tion of the traditional and novel pair samples (PC1) while the textureattributes of crunchiness and soft were explaining the X-9% and Y-9%of the samples' variation (PC2). All significance values are reported inTable 6.

-1.0 -0.5

-1.0

-0.5

0.0

0.5

1.0

A-Sha

A-Violet

A-Brown

A-D

T-Pork Roast

T-Spiced Pork Meat Roll

T-Roast-beef

T-Pig Paté

N

T-HT

Prin

cipa

l Com

pone

nt 2

(X

-exp

lain

ed v

aria

nce

9%,

Y-e

xpla

ined

var

ianc

e 16

%)

Principal Component Y-explaine

Fig. 1. Visual (colours and shapes): traditional and novel Danish open-faced sandwiches, ANOshowing the differences in apperance of the traditional and novel Danish open-faced sandwY-matrix. The inner and the outer ellipses represent r2=50% and 100%, respectively.A-appearance. Prefix to the samples' name (bold) indicate N-novel and T-traditional.

3.3. Conceptual characterisation of uniqueness

An initial insight into potential overall consumer perception wasgained through investigation of the conceptual attributes.

The visual investigation of the APLSR of the twelve samples revealedthat the pairs, traditional and novel, were differently described. The var-iation between the six pair samples was explained by the common(Pb0.001), familiar (Pb0.001), complexity (Pb0.001), luxurious(Pb0.001), and elegant (Pb0.001) attributes for the X-9% and Y-49%(PC1) spanning respectively the traditional and novel sandwiches(PC1). On PC2 the variation of the samples was explained for the X-9%and Y-5% due to the masculine and feminine attributes (results notshown).

3.4. Linkage of sensory and conceptual attributes: the sensory perceptionand understanding of uniqueness

The influence of cognition on the products' objective uniquenessand therefore the exploration of the mechanism of the overall con-sumer perception were reached through the combined inquiry ofconceptual and the sensory attributes.

The first combined analysis of all conceptual and all sensory attri-butes revealed that the six traditional and novel pair sandwicheswere differently described (Fig. 4). The variation between the sixpair samples was explained by the appearance, texture and flavourrye bread attributes for the X-14% and Y-45% (PC1) spanning respec-tively the traditional and novel sandwiches. The sensory attributes ofappearance, texture and flavour rye bread were strongly correlated tothe conceptual attributes. The traditional sandwiches explained bythe attributes of dry and gritty texture, rye bread after taste and fla-vour, and odour of liver were highly perceived as common and famil-iar. While the novel pair sandwiches described by the appearancesensory attributes of complexity, structure, height and colour andthe texture attribute chewing resistant were perceived as complexity,luxurious, elegant, interesting and inviting. Significant values are

0.0 0.5 1.0

rpe Curves

A-Iridescentark

A-Green

A-ComplexA-Structured

A-Height

A-Soft CurvesA-Colorful

N-Pig Paté

-Spiced Pork Meat Roll

N-Pork Roast

N-Marinated HerringN-Roast-beef

N-Egg and Shrimp

A-Freshly Made

erring-Egg A-White

A-Color Intensity

A-Red

A-Silver

A-Shiny

A-Light

1 (X-explained variance 9%, d variance 22%)

A-Yellow

VA Partial Least Squared Regression (APLSR) correlation loading plot of PC1 versus PC2iches. The sample indicators were in the X-matrix and the sensory descriptors in thePrefix to the sensory descriptors indicates the type of assessment by panellists:

Table 4Visual (colours and shapes): traditional and novel Danish open-faced sandwiches, significance of APLSR derived estimates regression. The symbols used represent respectively theP-values: n.s.=non significative, *=Pb0.05, **=Pb0.01, ***=Pb0.001 and the beta coefficients (raw): +=positive beta coefficient, and −=negative beta coefficient. Prefix tothe sensory descriptors indicates the type of assessment by panelists: A-appearance. Prefix to the experiment name indicate N-novel and T-traditional.

N−Egg and

Shrimp

T−Egg N−Marinated

Herring

T−Herring N−Pig

Pate´

T−Pig Pate´ N−Spiced

Pork Meat

Roll

T−Spiced

Pork Meat

Roll

N−Pork

Roast

T−Pork

Roast

N−Roast−beef

T−Roast−beef

A-Colorful +,*** n.s. n.s. −,*** n.s. −,*** +,** −,** +,* −,*** +,*** −,***

A-Color Intensity +,*** +,*** n.s. n.s. −,** n.s. n.s. n.s. −,** −,*** +,*** −,***

A-Yellow +,*** +,*** +,*** +,*** −,*** n.s. n.s. n.s. −,*** −,*** +,*** −,***

A-Green +,** n.s. n.s. −,** +,*** n.s. n.s. −,* n.s. −,*** +,* n.s.

A-Violet −,*** n.s. n.s. −,*** n.s. −,** n.s. n.s. +,*** +,*** n.s. n.s.

A-Brown −,*** −,*** n.s. −,*** +,*** +,** +,* n.s. n.s. n.s. −,*** +,**

A-White +,*** +,*** −,** n.s. −,*** −,* n.s. n.s. n.s. n.s. +,*** −,**

A-Silver −,* −,** +,*** +,*** n.s. −,*** −,*** −,** −,* −,*** −,*** n.s.

A-Red +,*** +,** −,** −,*** −,** n.s. n.s. n.s. n.s. n.s. +,*** −,***

A-Dark −,*** −,*** +,** −,** +,*** n.s. n.s. −,** +,*** +,* −,** n.s.

A-Light +,*** +,*** −,** +,** −,*** n.s. n.s. +,** −,*** n.s. +,*** −,*

A-Shiny n.s. n.s. +,*** +,*** −,** n.s. −,*** −,* −,*** −,*** −,* −,*

A-Iridescent n.s. −,*** n.s. −,*** n.s. +,* +,*** −,* n.s. −,* n.s. n.s.

A-Freshly Made +,*** n.s. +,*** +,** n.s. −,* n.s. −,*** n.s. −,*** +,** −,***

A-Complex +,*** −,** +,*** −,*** +,*** −,*** +,*** −,*** +,** −,*** +,*** −,***

A-Structured +,*** −,*** +,*** −,*** +,*** −,*** +,*** −,*** +,** −,*** +,*** −,***

A-Soft Curves +,*** n.s. n.s. −,*** +,* −,** +,*** −,*** n.s. −,*** +,*** −,***

A-Sharpe Curves −,** n.s. n.s. +,*** n.s. n.s. −,* n.s. n.s. n.s. −,** n.s.

A-Height +,*** -,*** +,*** n.s. +,*** −,*** +,*** −,*** +,** −,*** +,*** −,***

181S. Favalli et al. / Food Research International 50 (2013) 176–188

reported in Table 7. In Fig. 4 and Table 7 only the sensory attributessignificant correlated to the conceptual attributes have been shown.

In order to reach a deeper insight into the cognitive explanation ofthe samples in the different modalities, the analysis of the sample hasbeen divided into the three modalities of visual, retro-nasal andorthonasal, and tactile sensations and after tastes. Details of the justmen-tioned analyses are illustrated in the three following subparagraphs.

3.4.1. Visual: the pairs were differently describedTraditional and novel pair sandwiches were clustered in two dif-

ferent groups, where the X-25% and Y-48% of variation wereexplained by the PC1 and X-20% and Y-4% of variation, by the PC2.On the PC1, the shape attributes of height, structure, complex, andthe attribute of colourful which were positively strongly correlated(all Pb0.001) to the conceptual attributes of complexity, luxurious,interesting, inviting, elegant balanced and harmonious were mainlyresponsible for the clustering and describing the novel sandwiches.The attribute soft curves were also responsible for the clusteringand strongly positively correlated to the attributes inviting, harmo-nious (Pb0.001) and balanced, elegant (Pb0.01). On the right sideof the PC1 the cluster of the traditional sandwiches described bythe attributes familiar and common strongly negatively correlated(all Pb0.001) to the sensory attributes height, structure and com-plex. The attribute sharp curves describing the traditional sand-wiches were negatively correlated to the conceptual attributeinviting (Pb0.01). The dark and light colours respectively positivelyand negatively correlated (Pb0.001) to the conceptual attributes ofmasculine and feminine were spanning the sandwiches in meatyand fishy along the PC2 (X-20% and Y-4% of variation). All the resultshere described are not shown.

3.4.2. Odours, basic tastes and flavours: the pairs were differentlydescribed

The traditional and novel meaty based pair sandwiches differed intheir odour, basic taste and flavour sensory profiles and the explainedvariance accounted for the X-18% and Y-13% by the first two PrincipalComponents (PCs). The biggest variation between the pair sampleswas shown by the spiced pork meat roll and roast-beef sandwiches.Notmany conceptual and sensory attributeswere significantly correlat-ed. Positively significant correlation was found only between the con-ceptual attributes common and familiar and the sensory attributesflavour rye bread, odour lever (both Pb0.001) and odour intensity (re-spectively Pb0.05 and Pb0.01). All the results here described are notshown.

3.4.3. Textures, mouth feeling and after tastes: the pairs were differentlydescribed

Themain variationwas explained by the textures sensory attributesjuiciness, dry and gritty and the after tastes durability and rye bread ac-counting for the X-24% and Y-23% of variation (PC1) and spanning thesamples into the two different clusters of traditional and novel. Tradi-tional samples were characterised by the conceptual attributes com-mon (positively correlated to textures attributes rye bread Pb0.05,gritty and soft Pb0.001 and dry Pb0.01) and familiar (positively corre-lated to gritty and soft, Pb0.001).The novel pair samples, spanned alongthe PC2 accounting for the X-11% and Y-7% of variation, instead werecharacterised by the conceptual attributes inviting, luxurious and inter-esting (positively correlated to the sensory attributes juiciness andchewing resistant, Pb0.001) and complexity (positively correlated tothe sensory attributes juiciness Pb0.01 and chewing resistant,Pb0.001). All the results here described are not shown.

-1.0 -0.5 0.0 0.5 1.0

-1.0

-0.5

0.0

0.5

1.0

T-Roast-beefBT-Salty F-Trigerminal

O-Trigerminal

N-Pork RoastO-Dry Fruit

BT-Umami

O-Meaty

F-Meaty

O-Fried

F-FriedF-Dry Fruit

BT-BitterBT-Sweet

O-Butter

F-Rye breadO-Rye bread

O-GreenF-Green

F-LiverO-Liver

O-Sulfur

O-IntensityO-Spicy

O-VinegarBT-Sour

O-Ocean

F-OceanPr

inci

pal C

ompo

nent

2 (

X-e

xpla

ined

var

ianc

e 9%

,

Y

-exp

lain

ed v

aria

nce

15%

)

Principal Component 1 (X-explained variance 9%, Y-explained variance 21%)

T-Herring

T-Egg

T-Spiced Pork Meat Roll

T-Pig Paté

T-Pork Roast

N-Spiced Pork Meat Roll

N-Pig Paté

N-Marinated Herring

N-Roast-beef

N-Egg and Shrimp

Fig. 2. Odours, basic tastes and flavours: traditional and novel Danish open-faced sandwiches, ANOVA Partial Least Squared Regression (APLSR) correlation loading plot of PC1 ver-sus PC2 showing the differences in the odours, basic tastes and flavours of the traditional and novel Danish open-faced sandwiches. The sample indicators were in the X-matrix andthe sensory descriptors in the Y-matrix. The inner and the outer ellipses represent r2=50% and 100%, respectively. Prefix to the sensory descriptors indicates the type of assessmentby panellists: O-odour, BT-basic taste, and F-flavour. Prefix to the samples' name (bold) indicate N-novel and T-traditional.

182 S. Favalli et al. / Food Research International 50 (2013) 176–188

4. Discussion

4.1. Introduction to the discussion

Because of its holistic perspective (combination of both objectiveproperties and consumer concepts), the study represents a new ap-proach and an advance to the general knowledge in the area of fooduniqueness and identity. In fact the studyhighlightedwhich combination

Table 5Odours, basic tastes and flavours: traditional and novel Danish open-faced sandwiches, signithe p-values: n.s.=non significative, *=Pb0.05, **=Pb0.01, ***=Pb0.001 and the beta coto the sensory descriptors indicates the type of assessment by panellists: O-odour, BT-basic

N-Egg andshrimp

T-Egg N-Marinatedherring

T-Herring N-Pigpaté

T-pigPaté

N-Sme

O-Intensity n.s. −,*** +,*** +,*** −,* −,*** +,*O-Spicy −,*** −,*** +,*** +,*** n.s. −,** n.s.O-Green n.s. n.s. −,** −,* +,*** +,** n.s.O-Sulfur +,*** +,*** −,*** n.s. −,*** n.s. n.s.O-Trigeminal −,* −,*** n.s. −,* n.s. −,*** +,*O-Fried −,*** −,** −,*** −,*** n.s. −,*** n.s.O-Meaty −,*** −,*** −,*** −,*** +,*** −,* +,*O-Liver −,*** −,*** −,*** −,*** +,*** +,*** n.s.O-Ocean +,*** n.s. +,*** +,*** −,*** −,** −,*O-Dry fruit −,*** −,*** n.s. n.s. n.s. −,** n.s.O-Ryebread +,** +,*** −,** −,* n.s. n.s. −,*O-Butter +,* +,** n.s. n.s. −,** −,* n.s.O-Vinegar −,** −,*** +,*** +,*** −,*** −,*** −,*BT-Salty −,** −,*** +,* n.s. +,*** n.s. +,*BT-Bitter −,* −,*** +,*** +,*** n.s. −,* +,*BT-Umami −,*** −,*** −,*** −,*** +,*** n.s. +,*BT-Sour −,** −,*** +,*** +,*** −,*** −,*** −,*BT-Sweet n.s. n.s. +,*** +,*** −,** −,** n.s.F-Trigeminal −,* −,*** n.s. −,** n.s. −,*** +,*F-Fried −,*** −,*** −,*** −,*** n.s. −,*** n.s.F-Meaty −,*** −,*** −,*** −,*** +,*** −,* +,*F-Liver −,*** −,*** −,*** −,*** +,*** +,*** n.s.F-Ocean +,*** n.s. +,*** +,*** −,*** −,** −,*F-Dry fruit −,** −,*** +,** +,*** n.s. −,** −,*F-Green n.s. n.s. n.s. n.s. +,*** +,*** n.s.F-Ryebread +,** +,*** −,*** −,*** n.s. n.s. n.s.

of sensory properties and conceptualisations is responsible for the prod-ucts' uniqueness/authenticity in the overall consumer perception. Thusour investigation represents a combination of the previous ones whichfocused on the definition of objective unique properties of specificfoods (Guàrdia et al., 2010; Inarejos-Garcia et al., 2010; Rason et al.,2007) and which focused on the understanding of consumer conceptsdefining unique food products (Almli et al., 2011; Guerrero et al., 2009,2010, 2012; Kühne et al., 2010).

ficance of APLSR derived estimates regression. The symbols used represent respectivelyefficients (raw): +=positive beta coefficient, and −=negative beta coefficient. Prefixtaste, and F-flavour. Prefix to the experiment name indicate N-novel and T-traditional.

piced porkat roll

T-Spiced porkmeat roll

N-Porkroast

T-Porkroast

N-Roast-beef T-Roast-beef

* +,*** +,* n.s. +,*** −,*n.s. +,*** n.s. +,*** −,*n.s. −,*** −,** −,*** −,**n.s. −,*** n.s. −,*** n.s.

** +,*** −,** −,*** +,*** −,***−,*** +,*** +,*** +,** +,***

** +,** +,*** +,** +,*** +,***−,*** −,** −,*** −,*** −,***

** −,*** −,* −,** −,*** −,***−,* +,*** +,* +,** +,***−,* n.s. n.s. −,*** +,**n.s. n.s. n.s. n.s. +,*

** −,*** +,*** n.s. n.s. n.s.** +,*** n.s. −,** +,*** −,***

+,** n.s. −,** +,* −,***** +,*** n.s. n.s. +,*** n.s.

n.s. +,*** +,* +,* +,*n.s. n.s. n.s. n.s. n.s.

** +,*** −,** −,*** +,*** −,***−,* +,*** +,*** +,** +,***

** +,*** +,*** +,** +,*** +,***−,*** −,* −,*** −,*** −,***

** −,*** −,* −,** −,*** −,***n.s. +,*** n.s. n.s. n.s.n.s. −,*** −,** −,** −,***n.s. n.s. +,* −,* +,**

-1.0 -0.5 0.0 0.5 1.0

-1.0

-0.5

0.0

0.5

1.0

AT-Durability

AT-Butter

AT-Oily

T-Crispiness T-Chewing resistant

T-Soft

MF-Coating

AT-Rye bread

T-Dry

T-Juiciness

T-Crunchiness

AT-VinegarAT-Spicy

Prin

cipa

l Com

pone

nt 2

(X

-exp

lain

ed v

aria

nce

9%,

Y

-exp

lain

ed v

aria

nce

9%)

Principal Component 1 (X-explained variance 9%, Y-explained variance 18%)

T-HerringT-Egg

T-Spiced Pork Meat Roll

T-Pig PatéT-Pork Roast

T-Roast-beef

N-Spiced Pork Meat Roll

N-Pork Roast

N-Pig Paté

N-Marinated Herring

N-Roast-beef

N-Egg and Shrimp

T-Gritty

Fig. 3. Mouth-feeling, textures and after tastes: traditional and novel Danish open-faced sandwiches, ANOVA Partial Least Squared Regression (APLSR) correlation loading plot ofPC1 versus PC2 showing the differences in the mouth-feeling, textures and after tastes of the traditional and novel Danish open-faced sandwiches. The sample indicators were in theX-matrix and the sensory descriptors in the Y-matrix. The inner and the outer ellipses represent r2=50% and 100%, respectively. Prefix to the sensory descriptors indicates the typeof assessment by panellists: MF-mouthfeeling, T-texture, and AT-after taste. Prefix to the samples' name (bold) indicate N-novel and T-traditional.

183S. Favalli et al. / Food Research International 50 (2013) 176–188

Further it has to be noticed that despite the different findingspresented by the previous investigations, above mentioned, aimingat defining the elements and meaning of unique food products,there is not a consensus definition of uniqueness in the food area asyet. The key elements and concepts found are as follows: specifictaste, quality, appearance (Almli et al., 2011; Guàrdia et al., 2010;Guerrero et al., 2009), concepts related to culture/traditions as heritage,culture history and concepts related to innovation as novelty-change,variety, origin–ethnicity (Guerrero et al., 2010, 2012), ingredients andprocessing methods (Rason et al., 2007),and intrinsic and extrinsic at-tributes for the perception quality (Inarejos-Garcia et al., 2010).

Therefore we believe that with a holistic approach on the overallconsumer perception, a definition of uniqueness could be provided.We believe that defining what food uniqueness means uncoveringthe key elements determining the exclusiveness of a product for itsconsumers and therefore understanding the overall consumer per-ception, preferences and buying behaviour. Thus, the uniqueness def-inition is vital for the successful development and commercialisationof food products and for promoting health behaviours.

As a model for our investigations we utilised a product normallyconsumed by the population in two forms which we have designatedas traditional and novel. Thus the “traditional” form corresponds tothe product consumed by the whole population and for which thereexists, in the society, strict rules regarding the constituent elements,composure and presentation. The “novel” form instead represents amodern rethinking of the product which does not stay within thecommonly accepted guidelines and therefore constituting a new ver-sion of the product widely consumed by the society. Both traditionaland novel forms of the product were used in order to investigatemore in depth the key elements defining the uniqueness of the prod-uct for the population (close to their identity) and at the same timethe space for innovation.

As also indicated from other authors such as Schifferstein (2010) theoverall consumer experience and the uniqueness concepts are holisticmatters which comprehend cross-disciplinary and multidimensionalperspectives. Thus, for a better exemplification of uniqueness, apartfrom the definition, a model would be needed. Moreover, on a more

practical level, the model could be used as a tool in the new develop-ment, promotion and commercialisation of food products and as wellin the promotion of health consumer behaviours.

4.2. Objective uniqueness: first hypothesis of the uniqueness

The results of the objective analysis indicated that the traditionaland novel pair sandwiches are differentiating in the appearance(Fig. 1) and texture (Fig. 3) sensory attributes but not in the odours,basic tastes and flavours (Fig. 2).

This observation implies three considerations presented in the fol-lowing paragraphs. Respectively one concerns the raw ingredientsand processing method of the products, one concerns the objectiveuniqueness definition, and the last one concerns the role of the senso-ry characteristics in the uniqueness perception.

4.2.1. Influence of the raw ingredients origin and processing methodTraditional and novel pair sandwiches were composed of raw in-

gredients differently processed. More precisely, the traditional sand-wiches were industrial based while the novel sandwiches wereartisanal based. Where for artisanal, in opposition to industrial, wehere intend the crafting of handmade food products sold on a smallerscale. According to the European Regulation EC/510/, 2006, ingredi-ents' origin and processing methods were reckon to affect the sensoryprofile and therefore the quality and specific uniqueness of food prod-ucts. Moreover, different authors such as Lanza, Mazzaglia, andPagliarini (2011), Nielsen, Hyldig, Nielsen, and Nielsen (2005),Rason et al. (2007), Rousset-Akrim, Bayle, and Touraille (1995), andStolzenbach et al. (2011) have highlighted that different factorssuch as the industrial versus artisanal processing methods, the originof the raw ingredients, processing methods, and the variations in theingredients proportions were responsible for differences in the senso-ry product characteristics and therefore those parameters were im-portant in the definition of the product uniqueness. In contrast, thepresent findings highlight the non-significant difference betweenthe traditional and novel pair sandwiches in regards of the sensorycharacteristics odours, basic tastes and flavours. Therefore ingredients

Table 6Mouth-feeling, textures and after tastes: traditional and novel Danish open-faced sandwiches, significance of APLSR derived estimates regression. The symbols used represent re-spectively the p-values: n.s.=non significative, *=Pb0.05, **=Pb0.01, ***=Pb0.001 and the beta coefficients (raw): +=positive beta coefficient, and −=negative beta coeffi-cient. Prefix to the sensory descriptors indicates the type of assessment by panellists: MF-mouthfeeling, T-texture, and AT-after taste. Prefix to the experiment name indicateN-novel and T-traditional.

N-Egg andshrimp

T-Egg N-Marinatedherring

T-Herring N-Pigpaté

T-Pigpaté

N-Spiced porkmeat roll

T-Spiced porkmeat roll

N-Porkroast

T-Porkroast

N-Roast-beef T-Roast-beef

AT-Butter n.s. n.s. n.s. −,** n.s. n.s. n.s. n.s. +,* n.s. +,* n.s.AT-Oily +,*** +,** +,* +,*** n.s. n.s. −,** −,*** n.s. n.s. −,* n.s.AT-Spicy −,*** −,*** +,*** +,* +,* n.s. +,*** n.s. n.s. −,*** +,*** −,***AT-Vinegar −,*** −,*** +,*** +,*** −,*** −,*** −,*** −,** +,*** n.s. +,** n.s.AT-Ryebread n.s. +,*** −,*** −,** n.s. n.s. −,* +,** −,** +,*** −,** +,***AT-Durability −,** −,*** +,*** +,*** n.s. −,* n.s. −,** n.s. −,** +,** −,***MF-Coating +,*** n.s. +,** +,*** n.s. n.s. n.s. −,*** n.s. −,** n.s. −,**T-Crispiness −,* −,*** n.s. −,*** n.s. n.s. n.s. n.s. +,*** n.s. +,*** n.s.T-Crunchiness −,** −,*** n.s. −,*** −,*** n.s. −,* n.s. +,*** −,** +,*** +,**T-Soft +,*** n.s. +,*** +,*** +,* n.s. n.s. −,** −,*** −,*** −,* −,***T-Gritty −,** n.s. −,** n.s. −,* n.s. −,* +,** −,** +,*** n.s. +,***T-Juiciness +,* −,*** +,*** +,*** n.s. n.s. +,* −,*** +,*** −,*** +,* −,***T-Dry −,** +,** −,*** −,*** −,** n.s. −,* +,** −,*** +,*** −,* +,***T-Chewingresistant

n.s. −,*** +,** n.s. n.s. n.s. n.s. n.s. +,*** −,*** +,*** n.s.

184 S. Favalli et al. / Food Research International 50 (2013) 176–188

origin and processing methods do not seem to influence those sensorycharacteristics of the food product. This observation headed to two im-portant considerations regarding the product's uniqueness. First, is thissimilarity in the sensory properties, despite the different ingredients or-igin and processing methods (artisanal and industrial), indicating theultimately product's uniqueness? Secondly, is this similarity in the sen-sory properties of the artisanal and industrial product (having differentingredients origin and processing methods) indicating that the percep-tion of foodproducts is not just amatter of pure objectivity? In fact, bothcommon knowledge and scientific evidences (Bower & Baxter, 2003;Resano et al., 2011) indicated that for consumers artisanal food prod-ucts consisted of better sensory properties and therefore better qualitythan the industrial ones. Since our results indicated no difference in thesensory properties (odour, basic taste and flavours) of the artisanal andindustrial pair sandwiches, we could deduce that other factors come

-1.0 -0.5

-1.0

-0.5

0.0

0.5

1.0

O-Liver

FAMILIAR

COMMON

F-Rye bread

Prin

cipa

l Com

pone

nt 1

(X

-exp

lain

ed v

aria

nce

16%

,

Y

-exp

lain

ed v

aria

nce

5%)

Principal Component 1 Y-explaine

T-Egg

AT-Rye bread

T-DryT-Pork Roast

T-GrittyT-Spiced Pork Mea

T-Pig Paté

T-Roast-beef

Fig. 4. Conceptual and sensory attributes and traditional and novel Danish open-faced sandwversus PC2 showing the correlation between the conceptual and sensory attributes and thsensory attributes were in the X-matrix while the conceptual attributes in the Y-matrix. Thsensory descriptors indicates the type of assessment by panellists: A-appearance, O-odour,samples' name (bold) indicate N-novel and T-traditional. The dots in the figure represent thtributes significantly correlated and are reported. Corresponding significant values are repo

into play and influence our perception of food products. Thereforeuniqueness is not just amatter of pure objectivity. In order to further in-vestigate this idea, conceptual attributes have been added to the senso-ry attributes.

4.2.2. Uniqueness and “specific sensory characteristics”As previously mentioned, the fact that both the traditional and

novel pair sandwiches were characterised by the same odours, basictastes and flavours (Fig. 2) regardless the different origin and process-ing of the raw ingredients defined their identity and uniqueness. Infact, we can state that the uniqueness of a food product consists ofspecific sensory characteristics making it different from other foodproducts within the same food category. Further, even if changeswere apported to the food product (as ingredients origin, processingand variations in proportions) in the innovation process, those

0.0 0.5 1.0

ELEGANT

INVITINGINTERESTING

A-HeightA-Colourful

A-Structured

A-ComplexLUXURIOUS

COMPLEXITY

(X-explained variance 14%,d variance 45%)

N-Pork Roast

T-Chewing resistant

N-Roast-beef

N-Marinated Herring

T-Herring

N-Spiced Pork Meat Roll

N-Pig Paté

N-Egg and Shrimp

t Roll

iches: ANOVA Partial Least Squared Regression (APLSR) correlation loading plot of PC1e traditional and novel Danish open-faced sandwiches. The sample indicators and thee inner and the outer ellipses represent r2=50% and 100%, respectively. Prefix to theBT-basic taste, F-flavour, MF-mouthfeeling, T-texture, and AT-after taste. Prefix to thee sensory and conceptual attributes. Only the names of the sensory and conceptual at-rted in Table 7.

Table 7Sensory and conceptual attributes significance of APLSR derived estimates regression. The symbols used represent respectively the P-values: n.s.=non significative, *=Pb0.05,**=Pb0.01, ***=Pb0.001 and the beta coefficients (raw): +=positive beta coefficient, and −=negative beta coefficient. Prefix to the sensory descriptors indicates the typeof assessment by panelists: O-odour, A-appearance, F-flavour, AT-after taste, and T-texture.

Familiar Inviting Common Luxurious Interesting Complexity Harmonious Balance Elegant

O-Liver n.s. n.s. +,* −,* n.s. −,* n.s. n.s. n.s.A-Colourful −,* +,*** −,*** +,*** +,*** +,*** +,*** +,*** +,***A-Complex −,*** +,*** −,*** +,*** +,*** +,*** +,* +,** +,***A-Structured −,*** +,*** −,*** +,*** +,*** +,*** +,*** +,*** +,***A-Height −,*** +,*** −,*** +,*** +,*** +,*** +,*** +,*** +,***F-Ryebread n.s. n.s. +,* +,*** +,** +,*** n.s. n.s. n.s.AT-Ryebread +,* n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.T-Gritty n.s. −,** +,* −,*** −,** −,*** −,** −,** −,***T-Dry −,*** n.s. n.s. n.s. n.s. n.s. −,** −,** n.s.T-Chewing resistant n.s. +,*** −,*** +,*** +,*** +,*** n.s. n.s. +,**

185S. Favalli et al. / Food Research International 50 (2013) 176–188

sensory characteristics because constituting the ultimately identity ofthe food product were not changed. The understanding of uniquenesshere stated is in agreement with the finding that had highlighted theimportance, for consumers, of “the special taste” as one of the maincharacteristics defining the identity of food products (Almli et al.,2011) and further as the most important criteria in their food choices(Glanz, Basil, Maibach, Goldberg, & Snyder, 1998). Besides taste hasbeen considered as one of the most important indicator of perceivedquality (Ophuis & Vantrijp, 1995), overall satisfaction (Resano et al.,2011), acceptability (Hersleth, Lengard, Verbeke, Guerrero, & Næs,2011), and purchasing behaviour (Grunert, Bredahl, & Brunso, 2004).

Moreover the results might indicate the unique national aroma; infact some authors as Pangborn, Guinard, and Davis (1988) hadhighlighted the existence of regional aroma, patterns in the sensorycharacteristics odours, basic tastes and flavours which are character-izing a certain cultural identity (nation and or region of a nation). Infact the similarity of the samples in odours, basic tastes and flavourscould constitute a peculiarity (and therefore uniqueness) of theDanish consumers eating patterns. In other words, we suggestedthat, if considered a similar product belonging to another countryand culture, the patterns of odours, basic tastes and flavours mightbe different.

4.2.3. Uniqueness and “specific sensory characteristics”: considerationregarding its ultimate meaning

So here we formulated the first definition of uniqueness, but are thesensory characteristics by themselves enough to explain the productidentity and uniqueness? In other words, is a product unique just be-cause of particularly sensory or chemical (Cliff &Dever, 1996), geograph-ical (Castro-Vazquez, Diaz-Maroto, de Torres, & Perez-Coello, 2010) orgenetical properties (Rosenfeld, Risvik, Samuelsen, & Rodbotten,1997)? And would the consumers still be able to perceive the similari-ties, in the odours, basic tastes and flavours, of the Traditional andNovel sandwiches or would they be trigger by the differences in the ap-pearance and texture?What is exactly for consumers the “characteristicspecial taste” of a unique food product? Is that “special taste” the result ofthe interaction of other sensory and not-sensory properties of the prod-uct? Is that “special taste” the result of interactions between the differentmodalities through which we perceive the overall food product? Ulti-mately, how much does the objective uniqueness matter? How muchcould uniqueness be understand and explained only by objectiveinvestigations?

4.3. Uniqueness as a result of sensory characteristics and conceptualisations

In order to answer the above research questions, evaluation of theconceptual descriptors was linked to the sensory descriptive analysis(Fig. 4). Thus, the link between sensory and perception sciencewas cre-ated and we moved from description to explanation (Dijksterhuis &Byrne, 2005).

The combined results of sensory and conceptual attributes con-noted that the Traditional and Novel pair sandwiches were perceivedas being significantly different not only as regarding the appearanceand texture attributes but also as regarding odours, basic tastes andflavours. Particularly, it was determined that the more the pairsdifferenced in appearance and texture, the more they were perceivedas being different as well in odours, basic tastes and flavours. The re-sults here mentioned are not shown.

4.3.1. Multisensory process and cross-modal associations as componentsof uniqueness

Consequently the investigation denoted that consumer perceptionis a multisensory process in which cross-modal interactions play animportant role. Specifically, the investigation indicated that the com-bination of appearance and texture sensory attributes were able to af-fect the overall perception of this complex food products (Fig. 4).Particularly, the present results indicated that most attributes ofshapes and colour (height, structured, complex, colourful), texture(dry and gritty), and the flavour rye bread were responsible for affect-ing the orto-nasal and retro-nasal perception of the Danish pair sand-wiches (Fig. 4 and Table 7). Other authors have stated the interactionbetween shapes, colours (Blackwell, 1995; Morrot, Brochet, &Dubourdieu, 2001), texture (Weel et al., 2002) attributes on taste,odour or flavour perception.

4.3.2. Cognitive influences in the multisensory process and cross-modalassociations

Moreover, the fact that the conceptual attributes were related tothe appearance and texture sensory attributes but no correlationwas found with the odours, basic tastes and flavours constituted an-other important result (Fig. 4 and Table 7). In other words, our resultssuggested that the visual and textural cross-modal interaction onodours, basic tastes and flavours was not merely related to the princi-ples of multisensory integration (dominant theoretical frameworkrepresented by the authors mentioned above). The results indicatedthat cognitive factors were instead responsible for the different per-ceptions of the pair sandwiches. In fact, as Lévi-Strauss said “foodmust not only be good to eat, but also good to think”. Meaning that,as also suggested by Fischler (1988), human identification of food in-volves complex cognitive process. Thus, the shapes, colours, and tex-ture attributes proper of the sandwiches carried specific semanticvalues (Levy, 1981). These semantic values are socio-cultural depen-dent since consumers perceive unique food as having a strong distinc-tive character linked to the cultural heritage (Guerrero et al., 2009,2010; Trichopoulou, Soukara, & Vasilopoulou, 2007). For a specificconsumer, the tangible attributes proper of a food product conveyspecific semantic values affecting the overall food perception. Thefood product becomes a symbolic object, which defines the consumerfood identity and therefore the whole society (Fischler, 1988). Thus,those interactions between cognitions and sensory attributes of thefood products are characterizing the uniqueness of the food product

186 S. Favalli et al. / Food Research International 50 (2013) 176–188

and at the same time the identity of their users. The role of cognitivefactors in the cross-modal associations and therefore in the overallperception of food products has been stated by different authors(Allen, Gupta, & Monnier, 2008; Shankar, Levitan, & Spence, 2010;Spence & Gallace, 2011).

Altogether, the present results suggested that the eating and drink-ing experience does not initiate when a food product has been ingested.The interaction between consumer and food products started even be-fore seeing the food product, that is by hearing about it (friends andor family members talking about it, advertisements). Further, the influ-ences of those conceptions have been demonstrated to affect food per-ceptions. The total consumer experience (before, during and after foodconsumption) constitutes the uniqueness of food products and there-fore needs to be investigated (Dijksterhuis, 2012).

4.4. A suggestion for a uniqueness definition

4.4.1. Uniqueness has two interacting sidesAs first hypothesis of uniqueness, based on an objective investiga-

tion, we theorised that the similar sensory characteristics shared bydifferent food products indicated the products identity and thereforetheir uniqueness. Thus, the retro nasal perception similar in both tra-ditional and modern versions (Fig. 2) could be considered as core tothe products' implicit definition in terms of uniqueness.

As a second hypothesis we stated that the uniqueness of a foodproduct corresponds to the overall consumer perception. This last isthe result of a multisensory perception in which cross-modal interac-tion and cognitive process (with cultural-dependent semantic intrin-sic values) play an important role.

The combined results indicate that the combination of appear-ance, and texture sensory attributes were able to affect the overallconceptual understanding of this culturally defined complex foodproduct and were considered key modalities in terms of the productquality assessment (Fig. 4). However, retro nasal perception similarin both traditional and modern versions could be considered as coreto the products' implicit definition in terms of uniqueness (Fig. 2).

4.4.2. Mechanisms of the overall consumer experience and food uniquenessIn this paper the investigation of the overall consumer experience

of a complex food product, in its traditional and novel form, has beenperformed. Consequently a definition of food uniqueness and how toinvestigate it has been presented. So the mechanisms, occurring inthe consumers' perception, responsible for defining the key elementsof the product and therefore its uniqueness have been shown.

The results stressed the fact that the overall consumer experienceis a multisensory process in which cross-modal associations play apredominant role. Further the findings confirmed that cognitive pro-cesses, and therefore conceptualisations are responsible for thiscross-modal associations and therefore for the overall consumer per-ception. Conceptualisations are correlated to specific sensory charac-teristics and have a socio-cultural dimension. Therefore the foodproduct with its specific sensory characteristics has a specific seman-tic meaning and is therefore perceived by consumers as a symbolicobject which defines the consumer identity (as a single individualand as part of a group). Altogether these statements constitute theuniqueness of a food product.

Consequently the study indicated that “food uniqueness” definedhere needs to be investigated both with objective and cognitivemethods. Moreover, in order to define the uniqueness of the overallconsumer experience, a cross-disciplinary and multidimensional ap-proach is required (also highlighted by other authors e.g. Schifferstein(2010).

Finally, with the findings of this study, a preliminary model to in-vestigate and define the “uniqueness of food” for consumers has beendefined. It is the intent of the authors to further investigate the areaand complete the definition of the model.

4.5. Implications of the findings in the uniqueness innovation and qualityconsumer perception

It is a well-known fact that innovation is a major driver of economicgrowth. Thus, continuousmodification is apported to unique food prod-uct. But, what are the implications of the innovations in the consumers'perception of the food uniqueness and quality? Almli et al. (2011)suggested that “the special taste” and the “special appearance” of afood product constitute important criteria in the perception of itsuniqueness. Perceived quality is also an aspect explaining the productuniqueness since it is strictly dependent to the person, place andproduct(Ophuis & Vantrijp, 1995). Our results indicated that innovation in theproduct's appearance has an influence on the perception of its taste. Acrucial question, regarding innovation of unique food products, is nowarising: what is happening to the “special taste”when apporting innova-tions in the “special appearance”? Are the innovations in the product'sappearance influencing the consumer's quality perception? Do con-sumers perceive the novel product, in which appearance and texturecharacteristics have been modified, as being still unique and having abetter/worse quality? Would consumers still recognise the product asrepresenting themselves and therefore close to their identity? Furtherresearch in this area (cognitive investigations) could clarify the relationbetween sensory attributes, its semantic values and the product unique-ness and aswell as its quality perception. On a practical level, by answer-ing the above formulated questions, producers could understand whichtypes of innovations are suitable to be applied to the products. In otherwords, producers could gain insights in the understanding of whichmodifications are to bemade in a product without changing its ultimateuniqueness and identity.

4.6. Implications for producers

The preliminary uniqueness model theorised here and the moreholistic model which will be further delineated with following re-search constitute an achievement in the scientific findings of theoverall food perception and could be an important “user-oriented in-novation” ( Grunert et al., 2008) instrument in the development andcommercialisation and advertisement of food product for specificconsumers. In fact, these findings can help producers in their effortsin the innovation and commercialisations of unique traditional foodproducts (import and export). Thus, when developing a new foodproduct and when communicating it to consumers, the focus has tobe on the appearance, texture and flavour attributes and ensuringthat retro-nasal attributes remain similar.

Moreover, the findings would have important implications forproducers of dietetic food products and for health campaigns and the-oretical models of food and beverage choice. In fact, more attentionon the appearance and texture of the food product could help en-hance the consumer interest and liking for dietetic and or healthyfood products.

5. Conclusions and perspectives on the uniqueness definition

The study demonstrated that the overall consumer perception andtherefore the uniqueness of a food product are a multisensory processin which cross-modal associations, given by cognitive process associo-cultural based conceptualisations, play a predominant role.Therefore the food product with its specific sensory characteristicshas a specific semantic meaning and is therefore perceived by con-sumers as a symbolic object which defines the consumer identity(as a single individual and as part of a group). Particularly the findingsindicated that the combination of appearance and texture attributeswas able to affect the overall conceptual understanding of this cultur-ally defined complex food product. However, retro nasal perceptionsimilar in both traditional and modern versions could be consideredas core to the products' implicit definition in terms of uniqueness.

187S. Favalli et al. / Food Research International 50 (2013) 176–188

More research is needed in the overall consumer perception. Par-ticularly, more research is needed in order to define which are, forconsumers, the key attributes and the related cognitive process con-ferring the uniqueness of a food product. Therefore more insightinto the consumer cognitive area and its relationship with the sensoryfood properties is required. In other words, it would be necessary tofurther investigate the multisensory perception of food productsand cross-modal interactions together with conceptualisations, emo-tions, cognitive–contextual properties, and likings.

Thus the semantic and cultural meanings of the product, whichconstitute its uniqueness to the eyes of the specific user, would beidentified. That implies to gain insight into variables constituting con-sumers' food identity. Due to the holistic nature of the matter, theresearch required to be performed from a multidimensional andmultidisciplinary point of view.

Amodel explaining the different variables and their interactions inthe definition of uniqueness of food products would be delineated.The model, besides scientific achievement, could provide importantinformation to food and beverage marketers developing or innovat-ing products in line with healthy recommendations.

The results achieved with this investigation suggest that by chang-ing appearance and texture and not the ingredients of an existingfood, a new food product could be developed. Moreover the findingscould help nutritional health campaigns to change the preferencesand consumption of a food product by modifying its appearance andtexture attributes.

Acknowledgement

The authors express their gratitude, for their dedicated work, to thetechnical staff at the sensory laboratory, University of Copenhagen:Belinda Nielsen, Charlotte Gottlieb Dandanell, and Signe Gadegaard. TheFoodUnique Project, www.foodunique.eu, to which the PhD project islinked, is also greatly acknowledged. The University of Copenhagen,Faculty of Science PhD fund is thanked for its contribution to the grantthat enabled this study to proceed. The study was also supported byThe Danish Food Industry Agency — The Ministry of Food, Agricul-ture and Fisheries through the project “Local foods in Denmark”(3304-FVFP-08-K-11-01).

References

Allen, M. W., Gupta, R., & Monnier, A. (2008). The interactive effect of cultural symbolsand human values on taste evaluation. Journal of Consumer Research, 35(2),294–308.

Almli, V. L., Verbeke, W., Vanhonacker, F., Næs, T., & Hersleth, M. (2011). General imageand attribute perceptions of traditional food in six European countries. Food Qualityand Preference, 22(1), 129–138.

Askegaard, S., & Madsen, T. K. (1998). The local and the global: Exploring traits ofhomogeneity and heterogeneity in European food cultures. International BusinessReview, 7, 549–568.

Avermaete, T., Viaene, J., Morgan, E. J., Pitts, E., Crawford, N., & Mahon, D. (2004).Determinants of product and process innovation in small food manufacturingfirms. Trends in Food Science & Technology, 15(10), 474–483.

Behar, M. (1976). European diets vs traditional foods. Food Policy, 1(5), 432–435.Blackwell, L. (1995). Visual cues and their effects on odour assessment. Nutrition &

Food Science, 95(5), 24–28.Bower, J., & Baxter, I. (2003). Sensory properties and consumer perception of ‘home-made’

and commercial dairy ice cream. Journal of Sensory Studies, 18(3), 217–234.Bradby, H. (1997). Health, eating and heart attacks: Glaswegian Punjabi women's think-

ing about everyday food. In P. Caplan (Ed.), Food, health and identity (pp. 213–233).London, New York: Routledge.

Byrne, D., Bak, L., Bredie, W., Bertelsen, G., & Martens, M. (1999). Development of a sensoryvocabulary for warmed-over flavor: Part I. In porcine meat. Journal of Sensory Studies,14(1), 47–65.

Byrne, D., Bredie, W., & Martens, M. (1999). Development of a sensory vocabulary forwarmed-over flavor: Part II. In chicken meat. Journal of Sensory Studies, 14(1), 67–78.

Castro-Vazquez, L., Diaz-Maroto, M. C., de Torres, C., & Perez-Coello, M. S. (2010). Effectof geographical origin on the chemical and sensory characteristics of chestnuthoneys. Food Research International, 43(10), 2335–2340.

Cayot, N. (2007). Sensory quality of traditional foods. Food Chemistry, 102(2), 445–453.Chambers, S., Lobb, A., Butler, L., Harvey, K., & Bruce Traill, W. (2007). Local, national

and imported foods: A qualitative study. Appetite, 49(1), 208–213.

Cliff, M. A., & Dever, M. C. (1996). Sensory and compositional profiles of British Columbiachardonnay and pinot noir wines. Food Research International, 29(3–4), 317–323.

Cohn, S. (1997). Being told what to eat: Conversations in a diabetes day centre. In P.Caplan (Ed.), Food, health and identity (pp. 193–212). London: Routledge.

Cordell, G., & Ronai, C. (1999). Identity management among overweight women:Narrative resistance to stigma. In J. Sobal, & D. Maurer (Eds.), interpreting weight:The social management of fatness and thinness (pp. 29–47). Hawthorne, NY: AldineDe Gruyter.

Counihan, C. (1988). Female identity, food, and power in contemporary florence.Anthropological Quarterly, 61(2), 51–62.

Devine, C., Sobal, J., Bisogni, C., & Connors, M. (1999). Food choices in three ethnicgroups: Interactions of ideals, identities, and roles. Journal of Nutrition Education,31(2), 86–93.

Dijksterhuis, G. B. (2012). The total product experience and the position of the sensoryand consumer sciences: More than meets the tongue. New Food, 15, 38–41.

Dijksterhuis, G., & Byrne, D. (2005). Does the mind reflect the mouth? Sensory profilingand the future. Critical Reviews in Food Science and Nutrition, 45(7–8), 527–534.

Fandos, C., & Flavian, C. (2006). Intrinsic and extrinsic quality attributes, loyalty andbuying intention: An analysis for a PDO product. British Food Journal, 108(8),646–662.

Fischler, C. (1988). Food, self and identity. Social Science Information Sur Les SciencesSociales, 27(2), 275–292.

Fries, E., & Croyle, R. (1993). Stereotypes associated with a low-fat diet and their relevanceto nutrition education. Journal of the American Dietetic Association, 93(5), 551–555.

Glanz, K., Basil, M., Maibach, E., Goldberg, J., & Snyder, D. (1998). Why Americans eatwhat they do: Taste, nutrition, cost, convenience, and weight control concerns asinfluences on food consumption. Journal of the American Dietetic Association,98(10), 1118–1126.

Grunert, K. G., Bredahl, L., & Brunso, K. (2004). Consumer perception of meat qualityand implications for product development in the meat sector — A review. MeatScience, 66(2), 259–272.

Grunert, K. G., Jensen, B. B., Sonne, A., Brunsø, K., Byrne, D. V., Clausen, C., et al. (2008).User-oriented innovation in the food sector: Relevant streams of research and anagenda for future work. Trends in Food Science & Technology, 19(11), 590–602.

Guàrdia, M. D., Aguiar, A. P. S., Claret, A., Arnau, J., & Guerrero, L. (2010). Sensorycharacterization of dry-cured ham using free-choice profiling. Food Quality andPreference, 21(1), 148–155.

Guerrero, L., Claret, A., Verbeke, W., Enderli, G., Zakowska-Biemans, S., Vanhonacker, F.,et al. (2010). Perception of traditional food products in six European regions usingfree word association. Food Quality and Preference, 21(2), 225–233.

Guerrero, L., Claret, A., Verbeke, W., Vanhonacker, F., Enderli, G., Sulmont-Rossé, C.,et al. (2012). Cross-cultural conceptualization of thewords traditional and innovationin a food context by means of sorting task and hedonic evaluation. Food Quality andPreference, 25(1), 69–78.

Guerrero, L., Guàrdia, M. D., Xicola, J., Verbeke,W., Vanhonacker, F., Zakowska-Biemans, S.,et al. (2009). Consumer-driven definition of traditional food products and innovationin traditional foods. A qualitative cross-cultural study. Appetite, 52(2), 345–354.

Hersleth, M., Lengard, V., Verbeke, W., Guerrero, L., & Næs, T. (2011). Consumers'acceptance of innovations in dry-cured ham: Impact of reduced salt content,prolonged aging time and new origin. Food Quality and Preference, 22(1), 31–41.

Inarejos-Garcia, A.M., Santacatterina, M., Salvador, M. D., Fregapane, G., & Gomez-Alonso, S.(2010). PDO virgin olive oil quality-minor components and organoleptic evaluation.Food Research International, 43(8), 2138–2146.

ISO 13299 (2003). Sensory analysis —Methodology — General guidance for establishing asensory profile. International Organisation for Standardisation.

ISO 3972 (1991). Sensory analysis—Methodology—Method of investigating sensitivity oftaste. International Organisation for Standardisation.

Jabs, J., Devine, C., & Sobal, J. (1998). Maintaining vegetarian diets — Personal factors,social networks and environmental resources. Canadian Journal of Dietetic Practiceand Research, 59(4), 183–189.

Jabs, J., Sobal, J., & Devine, C. (2000). Managing vegetarianism: Identities, norms andinteractions. Ecology of Food and Nutrition, 39(5), 375–394.

James, A. (1996). Cooking the books: Global or local identities in contemporary Britishfood cultures? In D. Howes (Ed.), Cross-cultural consumption: Global markets, localrealities (pp. 77–92). London and New York: Routledge.

Jordana, J. (2000). Traditional foods: Challenges facing the European food industry.Food Research International, 33(3–4), 147–152.

Kalcik, S. (1984). Ethnic foodways in America: Symbol and the performance of identity.In L. K. Brown, & K. Mussel (Eds.), Ethnic and regional foodways in the united states(pp. 37–65). Knoxville, TN: University of Tennessee Press.

Kühne, B., Vanhonacker, F., Gellynck, X., & Verbeke, W. (2010). Innovation in traditionalfood products in Europe: Do sector innovation activitiesmatch consumers' acceptance?Food Quality and Preference, 21(6), 629–638.

Lanza, C. M., Mazzaglia, A., & Pagliarini, E. (2011). Sensory profile of a specialty Sicilianchocolate. Italian Journal of Food Science, 23(1), 36–44.

Levy, S. J. (1981). Interpreting consumer mythology — A structural approach toconsumer-behavior. Journal of Marketing, 45(3), 49–61.

Lindeman, M., & Stark, K. (1999). Pleasure, pursuit of health or negotiation of identity?Personality correlates of food choice motives among young and middle-agedwomen. Appetite, 33(1), 141–161.

Martens, H., & Martens, M. (2000). Modified jack-knife estimation of parameter uncertaintyin bilinear modelling by partial least squares regression (PLSR). Food Quality andPreference, 11(1–2), 5–16.

Mattiacci, A., & Vignali, C. (2004). The typical products within food “glocalisation”: Themakings of a twenty-first-century industry. British Food Journal, 104(10–11),703–713.

188 S. Favalli et al. / Food Research International 50 (2013) 176–188

Morrot, G., Brochet, F., & Dubourdieu, D. (2001). The color of odors. Brain and Language,79, 309–320.

Nielsen, D., Hyldig, G., Nielsen, J., & Nielsen, H. H. (2005). Sensory properties of marinatedherring (Clupea harengus) processed from raw material from commercial landings.Journal of the Science of Food and Agriculture, 85(1), 127–134.

Ophuis, P. A. M. O., & Vantrijp, H. C. M. (1995). Perceived quality— Amarket driven andconsumer oriented approach. Food Quality and Preference, 6(3), 177–183.

Pangborn, R. M., Guinard, J., & Davis, R. G. (1988). Regional aroma preferences. FoodQuality and Preference, 1(1), 11–19.

Parelli, A. (1996). Insediamenti umani e paesaggi agrari. Milano: Jaca Books.Pieniak, Z., Verbeke, W., Vanhonacker, F., Guerrero, L., & Hersleth, M. (2009). Association

between traditional food consumption and motives for food choice in six Europeancountries. Appetite, 53(1), 101–108.

Rason, J., Martin, J., Dufour, E., & Lebecque, A. (2007). Diversity of the sensory characteristicsof traditional dry sausages from the centre of France: Relation with regional manu-facturing practice. Food Quality and Preference, 18(3), 517–530.

Resano, H., Perez-Cueto, F. J. A., Sanjuan, A. I., de Barcellos, M. D., Grunert, K. G., &Verbeke, W. (2011). Consumer satisfaction with dry-cured ham in five Europeancountries. Meat Science, 87(4), 336–343.

Rosenfeld, H. J., Risvik, E., Samuelsen, R. T., & Rodbotten, M. (1997). Sensory profiling ofcarrots from northern latitudes. Food Research International, 30(8), 593–601.

Rousset-Akrim, S., Bayle, M. C., & Touraille, C. (1995). Influence of preparation onsensory characteristics and fat cooking loss of goose foie gras. Sciences des Aliments,15(2), 151–165.

Rozin, P. (1990). The importance of social factors in understanding the acquisition offood habits. In E. D. Capaldi, & T. L. Powley (Eds.), Taste, experience, and feeding(pp. 255–269). Washington, DC, US: American Psychological Association.

Sadalla, E., & Burroughs, J. (1981). Profiles in eating. Sexy vegetarians and other dietbased social stereotypes. Psychology Today, 15(10), 51–57.

Sapp, S., & Harrod, W. (1989). Social acceptability and intentions to eat beef — Anexpansion of the Fishbein–Ajzenmodel using reference group-theory. Rural Sociology,54(3), 420–438.

Schifferstein, H. N. J. (2010). From salad to bowl: The role of sensory analysis in productexperience research. Food Quality and Preference, 21(8), 1059–1067.

Shafer, R. B. (1979). The self-concept as a factor in diet selection and quality. Journal ofNutrition Education, 11, 37–39.

Shankar, M. U., Levitan, C. A., & Spence, C. (2010). Grape expectations: The role of cognitiveinfluences in color–flavor interactions. Consciousness and Cognition, 19(1), 380–390.

Shepherd, R., & Raats, M. M. (1996). Attitudes and beliefs in food habits. In H. L.Meiselman, & H. J. H. MacFie (Eds.), Food choice acceptance and consumption(pp. 346–364). London: Black Academic and Professional.

Sparks, P., & Shepherd, R. (1992). Self-identity and the theory of planned behavior —Assessing the role of identification with green consumerism. Social PsychologyQuarterly, 55(4), 388–399.

Spence, C., & Gallace, A. (2011). Tasting shapes and words. Food Quality and Preference,22(3), 290–295.

Stein, K. F. (1996). The self-schema model: A theoretical approach to the self-conceptin eating disorders. Archives of Psychiatric Nursing, 10(2), 96–109.

Stolzenbach, S., Byrne, D. V., & Bredie, W. L. P. (2011). Sensory local uniqueness of Danishhoneys. Food Research International, 44(9), 2766–2774.

Trichopoulou, A., Soukara, S., & Vasilopoulou, E. (2007). Traditional foods: A scienceand society perspective. Trends in Food Science & Technology, 18(8), 420–427.

Trichopoulou, A., Vasilopoulou, E., Georga, K., Soukara, S., & Dilis, V. (2006). Traditionalfoods: Why and how to sustain them. Trends in Food Science & Technology, 17(9),498–504.

Weel, K. G. C., Boelrijk, A. E. M., Alting, A. C., van Mil, P. J. J. M., Burger, J. J., Gruppen, H.,et al. (2002). Flavor release and perception of flavored whey protein gels: Perceptionis determined by texture rather than by release. Journal of Agricultural and FoodChemistry, 50(18), 5149–5155.