Consumer sensory and hedonic perception of sheep meat coppa under blind and informed conditions

Meat Science 137 (2018) 201–210

Contents lists available at ScienceDirect

Meat Science journal homepage: www.elsevier.com/locate/meatsci

Consumer sensory and hedonic perception of sheep meat coppa under blind and informed conditions

T

⁎

Juliana Cunha de Andradea, , Elen Silveira Nalériob, Citieli Giongob, Marcia Dutra de Barcellosc, Gastón Aresd, Rosires Delizae Institute of Chemistry, Federal University of Rio de Janeiro, Avenida Athos da Silveira Ramos, 149 – bloco A, CEP 21.941-909, Cidade Universitária, Rio de Janeiro, RJ, Brazil b Embrapa Pecuária Sul, P.O. Box 242, BR 153 km 633, CEP 96.401-970, Bagé, RS, Brazil c Post-Graduation Programme in Management (UFRGS/PPGA), Federal University of Rio Grande do Sul, Rua Washington Luiz, 855, 3° andar, CEP, 90010-460 Porto Alegre, RS, Brazil d Sensometrics & Consumer Science, Instituto Polo Tecnológico de Pando, Facultad de Química, Universidad de la República, By Pass de Rutas 8 y 101 s/n. Pando, Canelones, Uruguay e Embrapa Agroindústria de Alimentos, Avenida das Américas, 29501, CEP 23.020-470, Rio de Janeiro, RJ, Brazil a

A R T I C L E I N F O

A B S T R A C T

Keywords: Dry-cured meat product Salt reduction Smoking Sensory evaluation

The development of air-dried cured sheep meat products represents an interesting option to add value to the meat of adult animals. In this context, the aim of the present study was to evaluate consumer sensory and hedonic perception of sheep meat coppa, an innovative product. Four sheep meat coppa samples were formulated by varying smoking (smoked vs. non-smoked) and salt content (4.5% vs. 3.4%), and compared with two commercial samples of regular pork meat coppa. Samples were evaluated under blind or informed conditions by 202 consumers, who had to rate their liking and to answer a check-all-that-apply question. Sheep and pork meat coppa samples did not largely differ in their overall liking in both experimental conditions. Smoking and high salt content significantly increased consumers' hedonic perception of sheep meat coppa. The information included in the labels did not modify consumer hedonic perception but influenced their sensory description, particularly for the terms related to the type of meat used in their manufacture. Results indicate positive market opportunities for sheep meat coppa in the Brazilian market.

1. Introduction Sheep meat quality is directly related to its sensory characteristics, such as tenderness, juiciness, taste, and odor (Paulos et al., 2015). The meat from older animals is generally considered a low-value product due to its stronger flavor and higher toughness compared to those of younger animals (Field, Willians, & Miller, 1983). For this reason, consumers tend to show a strong preference for the meat of young animals and, consequently, carcasses from adult or discarded animals are usually not acceptable for direct consumption, which makes their commercialization difficult (Dutra et al., 2013). In many countries, fresh meat from cull ewes has low value. Nevertheless, several authors have proposed that meat from adult sheep is more suitable to be processed by curing with salts, smoking and drying, or even for the manufacture of fresh sausages after grinding, mixing with salt, spices and other ingredients, and casing (Matos, Menezes, Ramos, Ramos, & Gomide, 2007; Paulos et al., 2015).

⁎

Therefore, production of these processed meat products can be a viable strategy to add value to the meat of adult sheep given that such products are usually commercialized at relatively high prices (Sañudo et al., 2016). In this sense, recent studies have demonstrated the feasibility of using sheep meat in the manufacture of processed meat products, such as Sucuk (Turkish dry-fermented sausage), cooked hamtype pâté, fresh sausage, Stelja sheep ham, and Cecina (dry-cured sheep meat) (Dutra et al., 2013; Paulos et al., 2015; Sañudo et al., 2016; Stajić, Perunović, Stanišić, Žujović, & Živković, 2013; Stojković et al., 2015; Yalınkılıç, Kaban, & Kaya, 2012). The present study focuses on one specific processed meat product, coppa. This product is traditionally manufactured with pork meat (cured neck) and involves two main processes: salting and drying (Toldrá, 2004). Studies investigating the acceptance of sheep coppa have not been found in the literature, therefore representing both a research and market opportunity. In the Brazilian market, 70% of pork consumption is based on

Corresponding author. E-mail addresses: [email protected] (J.C. de Andrade), [email protected] (R. Deliza).

https://doi.org/10.1016/j.meatsci.2017.11.026 Received 4 May 2017; Received in revised form 19 October 2017; Accepted 20 November 2017 Available online 22 November 2017 0309-1740/ © 2017 Elsevier Ltd. All rights reserved.

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

conditions (Asioli et al., 2017). The influence of information on consumer perception is particularly relevant in the case of innovative products, as they do not have previous consumption experiences. Their expectations are mainly based on information and experiences with similar products. Hence, of special interest is the understanding of the acceptance of innovative meat processing technologies and products, since they can satisfy modern consumer demands for convenience, health, eating quality and food safety, ultimately adding value to meat products (De Barcellos et al., 2010). In this context, the objective of the present work was to evaluate consumer sensory and hedonic perception of sheep meat coppa, an innovative dry-cured sheep meat product in the Brazilian market under blind and informed conditions.

processed products (salami, ham, sausages, mortadella, etc.) rather than on fresh products. In 2016, total per capital pork consumption was 14.4 kg/inhabitant, representing 10 kg of processed products. Coppa is considered a niche product, targeted to the high-end market (Guimarães et al., 2017). According to data from Nielsen (2008), cited in Kuo Hue (2011), coppa represented only 0.4% of the total processed pork products evaluated at the time. Increased concern about the relationship between processed meat products and non-communicable diseases has increased demand for healthful products and has motivated the industry to engage in reformulation strategies (Kraus, 2015; Yang et al., 2015). Some of the most relevant trends regarding product reformulation are salt reduction, replacement of artificial additives by natural alternatives, and the addition of ingredients with potential positive effects on health (Jiang & Xiong, 2016; Perry & Grace, 2015; Strijbos et al., 2016; Valenzuela & Pérez, 2016). Salt is an essential ingredient in this type of product as it improves texture, decreases water activity for controlling the growth of pathogens microorganism and controls biochemical and enzymatic reactions during maturation, affecting the final flavor (Inguglia, Zhang, Tiwari, Kerry, & Burgess, 2017). Cured air-dried meat products usually contain high salt content: approximately 4.5% or higher (Feiner, 2006). However, high sodium intake has been linked to increased risk of high blood pressure and cardiovascular disease (Aburto et al., 2013). Considering that consumers are not always willing to compromise the taste of products for potential health benefits, it is necessary to study how consumers perceive salt-reduced products in order to develop feasible strategies for salt reduction (Zandstra, Lion, & Newson, 2015). Smoking is a frequent process in the manufacture of cured air-dried meat products (Toldrá, 2004). It contributes to the flavor, texture and increases the shelf life of the products (Sikorski & Sinkiewicz, 2015). However, smoking has raised concerns about its potential negative effects on health, due to contamination by polycyclic aromatic hydrocarbons, which have carcinogenic, mutagenic and bioaccumulative capacities (Ledesma, Rendueles, & Díaz, 2016). Market success of sheep meat processed products strongly depends on consumer acceptance. For this reason, a comprehensive understanding of consumer sensory and hedonic perception of products is necessary (Hersleth, Monteleone, Segtnan, & Næs, 2015). In the case of dry-cured ham, taste is a key attribute influencing consumer overall liking (Hersleth, Lengard, Verbeke, Guerrero, & Næs, 2011; Issanchou, 1996; Resano et al., 2011). In addition, color, flavor and adequate saltiness are considered important quality signals by consumers (Hersleth et al., 2011; Morales, Guerrero, Claret, Guàrdia, & Gou, 2008; Resano, Sanjuán, & Albisu, 2007). Sensory attributes are one of the main characteristics underlying consumers' overall liking of products (Tuorila, 2007). However, consumer perception does not only depend on sensory characteristics (Köster, 2003). Extrinsic attributes, such as package, brand and price play a key role in consumer perception and food choice (Jaeger, 2006). In particular, the information given by the package is an extrinsic factor and a quality cue that influences consumer behavior and could be an opportunity to motivate consumers to purchase a product (Grunert, 2002). However, when communicating food products in the market, it is desirable that consumers perceive consistency between extrinsic product attributes (i.e., information about price, and processing conditions), and intrinsic attributes (i.e., sensory quality during the subsequent eating experience). The information included in a package can create expectations about the sensory attributes of products or about the level of pleasure generated during consumption (Deliza & MacFie, 1996). After tasting the food, initial expectations of sensory and hedonic characteristics can be confirmed or disconfirmed. Confirmation can lead to high acceptance and repeated consumption and can also confirm the perception of food quality (Deliza, Macfie, & Hedderley, 1999). For this reason, it is important to evaluate consumer perception of products under informed

2. Materials and methods 2.1. Manufacture of sheep meat coppa Sheep meat coppa was produced at laboratory scale using meat cuts from certified Corriedale breed carcasses of approximately 2 years old, raised in grass-fed production system. The meat cuts were randomly taken from a commercial slaughterhouse in a normal day work from chilled carcasses after the neck cut was excised. They were deboned and only muscles from the neck with pH ≤ 5.9 and temperature < 6 °C were selected for processing at the Meat Science & Technology Laboratory (Embrapa Pecuária Sul, Brazil). Then, the cuts were vacuum packaged, frozen and stored at − 20 °C prior to processing. Coppa was produced using the entire deboned neck muscles seasoned with the following powdered spices: black pepper, garlic, clove and nutmeg (5.65 g/kg, total amount of the spices mixture) and matured in natural casings. The manufacturing process was divided into three steps: salting, equalization of the curing additives rubbed in the surface of the meat cut and drying. For some of the samples, smoking was added. In that case, cold smoking process was conducted in a traditional smoke house with controlled parameters (30–40 °C, RH 75–80%) during 6 h. The drying phase continued in a controlled chamber without smoke (12–15 °C, RH 72–75%) until 30% of weight was lost. The production process lasted for 28 days. For each treatment, seven pieces of coppa (approximately 300 g) were produced. Once drying was completed, the casings were removed and the coppa samples were vacuum packaged and stored under refrigeration until started of the sensory analysis. Two processing variables were considered in the formulation of the sheep meat coppa samples, smoking and salt, each of which were varied in two levels, as shown in Table 1. The salt content varied in two levels: 4.5% NaCl and 25% salt reduction, 3.4% NaCl. The highest salt content added in the process, corresponded to the sodium content of commercial products available in the Brazilian marketplace.

2.2. Consumer study 2.2.1. Participants A total of 202 middle/high income Brazilians took part in the study. They were recruited at two supermarkets in two Brazilian cities (Rio de Janeiro and Porto Alegre). To be eligible for participation, consumers had to be regular or occasional consumers of cold meat products (at least once a month) and be interested and available to participate in the study. At recruitment stage, no information about the specific aim of the study was provided. The socio-demographic characteristics of the participants are shown in Table 2. The study was approved by the Ethics Committee of the University Hospital Clementino Fraga Filho at Federal University of Rio de Janeiro, Brazil. Participants completed an informed consent form at the beginning of the study. 202

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

Table 1 Variables considered in the processing sheep meat coppa and information included in the labels.

Process Smoking time (h) Salt (%) Label information Type of coppa Brand Sodium content Smoking

Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

6 4.5

0 4.5

6 3.4

0 3.4

– –

– –

Sheep Fictitious No claim Smoked

Sheep Fictitious No claim No claim

Sheep Fictitious Reduced Smoked

Sheep Fictitious Reduced No claim

Pork Real No claim No claim

Pork Real No claim No claim

and two commercial samples of regular pork meat coppa (Brand A, Brand B). As sheep meat coppa is not available in Brazilian market, only commercial samples of coppa made with pork meat could be included for comparative purposes. The two commercial samples were selected according to the highest market share of the Brazilian market (Datamark, 2009). The six samples were cut into slices of approximately 0.8 mm thick and vacuum-packed the day before the evaluation and stored at 5 °C. For each sample, one slice was presented to consumers at 10 ± 2 °C in a plastic transparent container coded with a 3-digit code.

Table 2 Demographics of the consumers recruited for each of the experimental condition and identified clusters. Experimental condition Blind

n Gender (%) Female Male Age (years) Range Mean Standard deviation Region of birth in Brazil (%) North Northeast Midwest Southeast South Education (%) Primary school Secondary school University Post-graduate Income - minimum Brazilian wage (R$ 880.0)a 1 to 3 > 3 to 5 > 5 to 9 > 9 to 18 > 18 Consumption frequency of cold meat (%) Frequent consumer Occasional consumer Consumption frequency of lamb meat (%) Frequent consumer Occasional consumer Never Consumption frequency of coppa (%) Frequent consumer Occasional consumer Never Consumption frequency of dry-cured ham (%) Frequent consumer Occasional consumer Never a

Total Informed

Cluster 1

Cluster 2

Cluster 1

Cluster 2

53 (52%)

48 (48%)

70 (69%)

31 (31%)

202

47 53

48 52

43 57

58 42

48 52

23–72 49 11

18–87 46 17

18–83 44 16

28–76 48 14

18–87 47 14

0 4 2 49 45

0 2 0 44 54

0 1 7 38 54

0 0 0 52 48

0 2 3 44 51

0 6 40 54

2 10 40 48

0 16 41 43

0 6 36 58

0 10 40 50

6 8 19 26 41

15 8 17 31 29

13 10 13 27 37

6 13 13 23 45

10 10 15 27 38

62 38

75 25

63 37

81 19

68 32

23 68 9

10 75 15

24 67 9

13 77 10

19 71 10

11 85 4

17 77 6

11 82 7

26 68 6

15 79 6

34 64 2

33 61 6

27 67 6

42 55 3

33 63 4

2.2.3. Experimental procedure Participants were randomly divided into two groups of 101 participants, who evaluated samples under two experimental conditions. One of the groups performed a blind evaluation of the samples, in which samples were presented coded with 3-digit numbers without any type of additional information. The second group performed an informed evaluation of the samples. These participants received the same six samples, coded with 3-digit numbers, but accompanied with their corresponding label that included information about the type of coppa (sheep meat vs. pork meat), brand (a fictitious brand for sheep meat coppa and the real brand for the commercial products), sodium content (no claim vs. “reduced sodium content”), and the use of smoking (no claim vs. “smoked”). The same font size, typography, and format were used in all the labels. The information included in the labels of each sample is presented in Table 1, whereas examples of two of the labels are shown in Fig. 1. All compulsory information according to Brazilian legislation was included on the labels, including the list of ingredients and information about the additives. No significant differences between the groups were found in their gender, age, consumption frequency of lamb meat and meat products, region of birth, education and income distribution (p > 0.40). 2.2.4. Evaluation of consumers' sensory and hedonic perception An evoked context was used in the evaluation in an attempt to increase the ecological validity of the results (Jaeger et al., 2017). Consumers were asked to think of a frequent consumption situation for this type of product (Andrade et al., 2017) when evaluating the samples. All participants were first presented with the evoked context and were asked to clearly imagine the situation using the following instructions: “Imagine that you have invited some friends to your house and you will serve some appetizers”. They were instructed to keep in mind the evoked occasion while evaluating each of the six samples. Then, participants were presented with each sample in monadic sequence, following a Williams' Latin Square experimental design to avoid order and carry-over effects (MacFie, Bratchell, Greenhoff, & Vallis, 1989). They were asked to try the samples, to rate their overall liking using a 9-point hedonic scale and to answer a check-all-that-apply (CATA) question composed of 16 sensory terms (pink color, red color, sour, salty, fatty, spicy, meat flavor, lamb flavor, spice flavor, smoked flavor, off-flavor, not juicy, tender, chewy, hard and springy). The sensory terms were selected based on a preliminary test with 10 consumers, who were asked to describe the coppa samples. Those consumers did not participate in the final sensory

In Brazilian currency (Real), approximately R$ 3.35/US$ at the time of the study.

2.2.2. Samples Six samples were included in the present study, as follows: the four sheep meat coppa samples (Smoked, 4.5% salt; Non-smoked, 4.5% salt; Smoked, 3.4% salt; Non-smoked, 3.4% salt) described in Section 2.1 203

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

Fig. 1. Examples of two of the labels presented in the study: (a) smoked product with regular salt content, and (b) non-smoked product with sodium reduction. English translation of the labels: (a) Sliced Smoked Coppa – sheep meat coppa. Ingredients: sheep meat, salt, spices: black pepper, garlic, clove, nutmeg, antioxidant: sodium erythrobate (INS 316), preservative: sodium nitrate (INS 251) and sodium nitrite (INS 250). Packaged on: 24 OCT 2016. Best before date: 24 JAN 2017. (b) Sliced Coppa – sheep meat coppa. Reduced sodium content. Ingredients: sheep meat, salt, spices: black pepper, garlic, clove, nutmeg, antioxidant: sodium erythrobate (INS 316), preservative: sodium nitrate (INS 251) and sodium nitrite (INS 250). Packaged on: 24 OCT 2016. Best before date: 24 JAN 2017.

(a)

(b)

significance level of 5%. An agglomerative hierarchical cluster analysis was used to identify consumers with different hedonic scores for the each experimental condition. Euclidean distances and Ward's aggregation criterion were considered. The optimum number of clusters for each context was determined based on the Calinski and Harabasz index (Milligan & Cooper, 1985).

evaluation. The order in which the terms were presented in the CATA question was balanced between consumers following a Williams' Latin Square experimental design to avoid response bias (Ares & Jaeger, 2013). After evaluating samples, participants were asked to answer a few socio-demographic questions. Plain crackers and still mineral water were used for rinsing between samples. The test was conducted in the supermarkets in which consumers were recruited, in a separated area especially prepared for the test.

2.3.2. Check-all-that-apply (CATA) questions Frequency of use of each one of the terms of the CATA question was determined by counting the number of consumers that used that term to describe each sample. Cochran's Q test (Manoukian, 1986) was carried out to identify significant differences among samples for each of the sensory terms. When differences were significant the sign test was used for pair-wise comparisons. In addition, a generalized linear model was used to evaluate the influence of sample, experimental condition and their interaction on the frequency of use of each of the terms (Cardinal, Zamora, Chambers, Barrachina, & Hough, 2015). Correspondence analysis was used to get a bi-dimensional representation of the samples and the relationship between samples and terms from the CATA question. Confidence ellipses around samples were constructed using a truncated total bootstrapping approach in which only the first two dimensions of the configurations were considered (Cadoret & Husson, 2013). This analysis was also performed separately for each experimental condition and cluster. The RV

2.3. Data analysis All statistical analyses were performed using R language (R Development Core Team, 2014) with the following packages: lmerTest (Kuznetsova, Brockhoff, & Christensen, 2015) and FactoMineR (Lê, Josse, & Husson, 2008). 2.3.1. Overall liking scores Analysis of variance (ANOVA) was performed on hedonic scores. Sample, experimental condition and their interaction were considered as fixed sources of variation and consumer (within experimental condition) as random effect. In addition, an ANOVA was performed separately for each experimental condition considering sample as fixed source of variation and consumer as random effect. Tukey's test was used for post hoc pairwise comparisons of sample means, at a 204

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

describing the non-smoked coppa with salt-reduction (Non-smoked, 3.4% salt) compared to the remaining three sheep meat coppa samples. The commercial pork meat coppa samples were significantly differentiated from the sheep meat coppa in several sensory attributes (Table 4). This difference can be clearly seen in the first two dimensions of the correspondence analysis, which explained 85.5% of the inertia of the experimental data. As shown in Fig. 2a, the first dimension sorted the sheep meat samples from the commercial pork meat samples. The latter samples were associated with the terms pink color, fatty and tender. Meanwhile, sheep meat coppa samples were mainly associated with the terms fatty, lamb flavor, red color, spice flavor, spicy. Regarding differences among the sheep coppa meat samples, the non-smoked coppa with salt-reduction (Non-smoked, 3.4% salt) was clearly separated from the rest due to its association with the terms hard, springy, sour and off-flavor. Information influenced consumers' sensory description of the products. Significant differences in the frequency of use of terms with experimental condition was only found for the term lamb flavor; consumers in the blind condition used this term less frequently than those in the informed condition (16% vs. 24%). In addition, the interaction between sample and experimental condition was significant (p < 0.01) for the terms meat flavor and lamb flavor. As shown in Fig. 3a, consumers in the informed condition used the term meat flavor more frequently for describing the commercial samples than those in the blind condition, and less frequently for describing the sheep meat samples. On the contrary, the term lamb meat was more frequently used for describing the sheep meat samples in the informed condition than in the blind condition, whereas the opposite was found for the commercial pork meat samples (Fig. 3b). Conclusions regarding similarities and differences among samples were not largely affected by information. In the informed evaluation, the first two dimensions of the correspondence analysis explained 89.4% of the inertia of the experimental data. As shown in Fig. 2b, the first dimension sorted the sheep meat samples from the commercial pork meat samples. Sheep meat samples were mainly associated with the terms sour, smoked flavor, spice flavor, lamb flavor and springy. Meanwhile, the commercial pork meat samples were associated with the terms pink color, fatty, meat flavor and tender. The sample configuration of the informed condition was highly similar to the one from the blind evaluation, as evidenced by the high RV coefficient (RV = 0.98). However, the RV coefficient between term configurations was lower (0.79) and indicated moderate agreement, suggesting that although information did not affect perception of differences among samples, it influenced the way in which consumers

Table 3 Mean and standard deviations for overall liking scores (evaluated in a 9-point hedonic scale) in two experimental conditions. Sample

Experimental condition Blind

Informed

ab

6.2 ± 2.0 6.7a ± 2.0 6.2ab ± 2.1 5.8b ± 2.3 6.8a ± 1.8 6.7a ± 2.2

Smoked, 4.5% salt Non-smoked, 4.5% salt Smoked, 3.4% salt Non-smoked, 3.4% salt Brand A Brand B

6.2a 6.8a 6.4a 6.0a 6.5a 6.6a

± ± ± ± ± ±

2.1 1.9 2.0 2.3 2.0 2.0

Mean overall liking scores with different superscripts within a column are significantly different according to Tukey's test for a 5% significance level.

coefficient was used to compare sample and term configurations from each experimental condition and each cluster. 3. Results 3.1. Overall liking scores Overall liking scores were only significantly influenced by sample (F = 3.30, p = 0.006). Experimental condition (F = 0.02, p = 0.868) and the interaction between sample and experimental condition (F = 1.29, p = 0.267) were not significant. Two groups of samples were identified, regardless of the type of experimental condition (Table 3). One of the groups was composed of the commercial samples (Brand A and Brand B) and the non-smoked coppa with regular salt content (Non-smoked, 4.5% salt), which received the highest overall liking scores. The other group was composed of the coppa with smoked (Smoked, 4.5% salt) and the samples with reduced sodium content (Smoked, 3.4% salt and Non-smoked, 3.4% salt). 3.2. Sensory product characterization In the blind evaluation condition, significant differences among samples were established for all the sensory terms except for meat flavor, spice flavor and off-flavor. Process variables significantly (p < 0.05) influenced the frequency of use of the terms springy, hard and salty to describe the sheep meat coppa samples. As expected, frequency of use of the term salty was significantly lower for the salt-reduced samples than for the regular ones (Table 4). In addition, consumers used the terms springy and hard significantly more frequently for

Table 4 Frequency (%) of each term of the CATA question for describing the sheep meat and pork meat coppa samples for consumers who completed the task under blind condition. Term

Sample Smoked, 4.5% salt

Springy Pink color Red color Hard Chewy Fatty Sour Tender Spicy Meat flavor Lamb flavor Spice flavor Smoked flavor Off-flavor Salty Not juicy

b

24 12b 47a 10b 29a 22bc 10ab 36cd 18a 20a 16ab 20a 36b 8a 61ab 8ab

Non-smoked, 4.5% salt b

20 18b 35ab 5b 28a 30ab 8ab 46bc 9ab 25a 15ab 28a 47ab 8a 62ab 4b

Smoked, 3.4% salt bc

Non-smoked, 3.4% salt a

16 14b 39ab 8b 19a 23abc 6b 33cd 16a 23a 16ab 28a 41b 13a 50bc 12a

40 16b 33bc 23a 28a 16c 15a 29d 12a 18a 26a 21a 41b 13a 43c 16a

Percentages with different superscripts within a row indicate significant differences at p < 0.05 according to Sign test.

205

Brand A cd

8 41a 19d 4b 9b 31ab 5b 65a 3b 20a 12b 16a 36b 10a 38c 11ab

Brand B d

3 31a 21cd 6b 7b 35a 4b 57ab 15a 18a 13b 21a 58a 7a 67a 9ab

p-Value 0.000 0.000 0.000 0.000 0.000 0.006 0.039 0.000 0.005 0.712 0.047 0.064 0.003 0.422 0.000 0.037

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

Meat flavor

(a) 40

Blind

Fig. 2. Frequency of use of the terms meat flavor (a) and lamb flavor (b) for describing coppa samples, for consumers who evaluated samples under blind and informed conditions.

Informed

Frequency of use (%)

30

20

10

0 Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

Sample

(b)

Lamb flavor 40 Blind Informed

Frequency of use (%)

30

20

10

0 Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

Sample

samples compared to the samples without smoking (Table 5). Regarding salt reduction, it reduced the overall liking scores of consumers in Cluster 2 only for the non-smoked product. No significant differences between the clusters were found in their gender, age, consumption frequency of lamb meat and meat products, region of birth, education and income distribution in both the blind and informed evaluation (p > 0.15). Sample configuration was similar for both clusters in the blind (RV = 0.93) and informed evaluation (RV = 0.94). However, term configurations tended to differ for both clusters in the blind (RV = 0.19) and informed condition (RV = 0.41). This suggests that although samples were perceived similarly, differences in how they were described were identified. In the blind evaluation, significant differences between clusters in the frequency of use of the terms salty and fatty were found. Consumers in Cluster 1 used these terms more frequently than those in the Cluster 2 (59% vs. 46% and 30% vs. 22%, respectively). In addition, the interaction between sample and cluster was significant (p < 0.05) for the all terms related to texture (tender, chewy, hard, springy, not juicy) and flavor (sour, spicy, lamb and off-flavor), suggesting that clusters differed in the way in which they described samples. Consumers in Cluster 2 used terms related to defects more frequently for describing the sheep

described the products (Fig. 2).

3.3. Consumer segmentation Hierarchical cluster analysis was used to identify consumer segments that differed in their hedonic reaction towards the coppa samples. Two consumer segments with different preference patterns were identified in both experimental conditions, although some differences in their relative size were found (Table 5). In both experimental conditions one of the groups (Cluster 2) clearly disliked the non-smoked product with reduced sodium content and gave the highest overall liking scores for commercial pork meat coppa samples. The relative size of this group was 48% and 31% in the blind and informed evaluation, respectively. Differences among samples were smaller for the other cluster (Cluster 1), but they tended to prefer the sheep meat coppa samples to the commercial ones. This difference was clearer when the evaluation was performed under informed conditions compared to the blind evaluation. In both experimental conditions, consumers in Cluster 1 did not significantly modify their overall liking scores for the four sheep meat coppa samples. On the contrary, consumers in Cluster 2 gave significantly higher overall liking scores for the smoked sheep meat coppa 206

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

(a) Hard Not juicy Pink color

Non-smoked, 3.4% salt

Fig. 3. Representation of coppa samples and terms in the first and second dimensions of the correspondence analysis performed on data from the CATA question comprising 16 sensory terms for consumers who completed the task under blind (a) and informed (b) conditions.

Brand A

Springy

Sour

Off-flavor

Dim 2 (19.72%)

Lamb flavor Tender Smoked flavor Fatty Brand B Smoked, 3.4% salt Meat flavor Chewy Spice flavor Non-smoked, 4.5% salt Red color Smoked, 4.5% salt Salty Spicy

Dim 1 (65.82%)

(b) Pink color Hard

Not juicy

Springy

Brand A Off-flavor

Non-smoked, 3.4% salt

Dim 2 (12.26%)

Meat flavor Chewy Sour Lamb flavor

Fatty

Spice flavor Red color Smoked, 3.4% salt

Tender

Smoked, 4.5% salt Smoked flavor Non-smoked, 4.5% salt Spicy Salty

Brand B

Dim 1 (77.16%)

meat samples than commercial pork meat samples, which explains their lower overall liking scores. Fig. 4 exemplifies this result for the terms off-flavor, chewy, hard and springy. On the contrary, frequency of use of these terms for consumers in Cluster 1 did not largely differ between samples, in agreement with the small differences in their overall liking scores. Similar results were observed in the informed evaluation condition (data not shown).

Table 5 Mean and standard deviations for overall liking scores (evaluated in a 9-point hedonic scale) for each identified clusters in the blind and informed evaluation condition. Sample

Experimental condition Blind

Smoked, 4.5% salt Non-smoked, 4.5% salt Smoked, 3.4% salt Non-smoked, 3.4% salt Brand A Brand B

Informed

Cluster 1 (n = 53)

Cluster 2 (n = 48)

Cluster 1 (n = 70)

Cluster 2 (n = 31)

6.7ab ± 1.9 6.9a ± 1.9

5.5c ± 2.0 6.4b ± 2.1

6.8ab ± 2.1 7.1a ± 1.8

5.2bc ± 1.8 6.1b ± 2.0

6.3ab ± 2.2 6.8ab ± 2.0

6.2bc ± 1.9 4.6d ± 2.1

6.8ab ± 1.9 7.1a ± 1.6

6.0b ± 2.1 4.0c ± 1.9

6.3ab ± 2.0 6.0b ± 2.4

7.4a ± 1.5 7.5a ± 1.6

5.8c ± 2.0 6.1bc ± 2.0

7.5a ± 2.0 7.3a ± 1.2

4. Discussion and conclusions The development of new products requires understanding consumer sensory and hedonic perception, as they can be regarded as pre-requisite for success in the marketplace (Mueller & Szolnoki, 2010). In this context, the present study investigated consumer perception of an innovative product in the Brazilian market, sheep meat coppa under blind and informed conditions, and compared it to the commercial pork meat coppa samples with the largest market share. On average, the sheep meat coppa samples were positively evaluated by consumers and received overall liking scores similar to the commercial pork meat samples, regardless of the experimental condition. This result suggests the existence of potential for the development of new processed sheep meat products, as previously reported by Matos

Mean overall liking ratings with different superscripts within a column are significantly different according to Tukey's test for a 5% significance level.

207

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

Off-flavor

(a) 50

Cluster 1 Cluster 2

Frequency of use (%)

40

30

20

10

0 Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

Sample

(b)

Chewy 50 Cluster 1 Cluster 2

Frequency of use (%)

40

30

20

10

0 Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

Sample Fig. 4. Frequency of use of the terms off-flavor (a), chewy (b), hard (c) and springy (d) for describing coppa samples, for consumers who evaluated samples under blind condition for each identified cluster (Cluster 1 and Cluster 2).

by a segment of the market. Considering that consumer segments did not differ in terms of socio-demographic characteristics, challenges would be faced when trying to target consumers that would like the product. Corral, Salvador, and Flores (2013), Corral, Salvador, Belloch, and Flores (2014) and Aaslyng, Vestergaard, and Koch (2014) pointed out that consumers consider salt reduced sausages an acceptable product although their sensory characteristics are different compared to the conventional product. In the present work, the smoked sample salt reduction did not significantly decrease consumers' overall liking scores. This result can be explained considering the flavor enhancing properties of smoking. Smoking could have acted as a flavor enhancer, contributing to saltiness perception. Flavor enhancers have been reported to be effective for salt reduction in meat products (Inguglia et al., 2017). Information did not affect overall liking scores of the coppa samples, suggesting that consumer hedonic perception was mainly driven by the sensory characteristics of the samples. A similar result has been previously reported when evaluating the influence of label information on the hedonic perception of boar meat (Meier-Dinkel et al., 2013) and fermented sausages made from boar meat (Meier-Dinkel et al., 2013). In

et al. (2007) and Paulos et al. (2015). It should be highlighted that in the present work overall liking scores were collected using a specific evoked context, which referred to serving the product to friends. Although this context is a frequent consumption situation for this type of product in Brazil, further research should be conducted considering other consumption contexts. In particular, consumer perception may differ if contexts involving personal consumption are considered. Market segmentation was observed in both the blind and informed evaluations. One consumer segment preferred commercial pork meat coppa samples over sheep meat coppa, whereas the opposite trend was found for the other segment. Consumers who preferred the pork meat coppa samples used terms related to sensory defects (e.g. off-flavor, chewy, hard) to describe the sheep meat coppa, suggesting that they disliked their sensory characteristics. However, the other segment (which had similar size) positively evaluated the sensory and hedonic characteristics of sheep meat coppa. Market segmentation has been identified as an effective strategy to increase the profit of companies (Amine & Smith, 2009). Therefore, results from the present work suggest that sheep meat coppa represents an opportunity for the Brazilian meat industry to develop innovative products, which would be accepted 208

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

(c)

Hard 50 Cluster 1 Cluster 2

Frequency of use (%)

40

30

20

10

0 Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

Sample

(d)

Springy 50 Cluster 1 Cluster 2

Frequency of use (%)

40

30

20

10

0 Smoked, 4.5% salt

Non-smoked, 4.5% salt

Smoked, 3.4% salt

Non-smoked, 3.4% salt

Brand A

Brand B

Sample Fig. 4. (continued)

trend was found for the rest of the samples. This suggests that consumers did not perceive the lamb meat flavor in the coppa samples and only used the term to describe samples when they were aware of the type of meat used in their manufacture. Henrique, Deliza, and Rosenthal (2015) reported differences in sensory characterization of ham samples between the blind and the informed condition. According to these authors, the frequency of the term “low salt” increased when information about salt reduction was provided to consumers. In practical terms, the study provided important information for the sheep meat chain. An innovative processed sheep meat product was accepted by a segment of consumers. Market opportunities were identified and could be further explored. Changes in the formulation could reduce the perception of off-flavor by one of the consumer segments, which could help to tailor the product as an attractive option to the general market. In addition, it is important to stress that sheep meat coppa and other innovative sheep meat products can stimulate local and regional development in the food supply chain.

contrast, Grasso, Monahan, Hutchings, and Brunton (2017) when examining the effect of information on the overall liking of turkey enriched with plant sterols found a change in how assessors rated the samples before and after receiving information. Therefore, information about meat type did not influence consumer hedonic perception, which suggests that the use of sheep meat in the manufacture of processed products does not generate negative attitudes in consumers. Further research should look at consumers' expectations about sheep meat coppa prior to the evaluation in order to confirm the potential of the product in the Brazilian marketplace. Information had a minor effect on consumer sensory perception of the coppa samples. Perceived similarities and differences among the samples were almost identical in the blind and informed evaluation condition, in agreement with results reported by Vidal, Barreiro, Gómez, Ares, and Giménez (2013) when evaluating the influence of information on consumer sensory characterization of vanilla milk desserts. However, differences in how consumers described the samples were found. Despite the lack of differences in sample configurations, information affected the frequency of use of specific terms, as well as significant differences among samples in specific terms. Consumers used the term lamb flavor for describing sheep meat coppa samples more frequently when information was provided, whereas the opposite

Acknowledgements The authors are indebted to the Brazilian Agricultural Research Corporation (SEG 02.12.01.033.00.00) for the financial support and 209

Meat Science 137 (2018) 201–210

J.C. de Andrade et al.

CAPES - Brazil for the scholarship support. The Design Team from Embrapa Food Technology are thanked for their contribution to this project.

107–117. Köster, E. P. (2003). The psychology of food choice: Some often encountered falacies. Food Quality and Preference, 14, 359–373. Kraus, A. (2015). Development of functional food with the participation of the consumer. Motivators for consumption of functional products. International Journal of Consumer Studies, 39, 2–11. Kuo Hue, C. (2011). O mercado de frios no Brasil: uma estimação da demanda a partir de um modelo em três estágios FGV/EESP. Retrieved from: http://bibliotecadigital.fgv. br/dspace/bitstream/handle/10438/8332/66080100267.pdf. Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B.. Package ‘lmerTest’. (2015). Available at: https://cran.r-project.org/web/packages/lmerTest/lmerTest.pdf (Last accessed January 12th, 2017) . Lê, S., Josse, J., & Husson, F. (2008). FactoMineR: An R package for multivariate analysis. Journal of Statistical Software, 25(1), 1–18. Ledesma, E., Rendueles, M., & Díaz, M. (2016). Contamination of meat products during smoking by polycyclic aromatic hydrocarbons: Processes and prevention. Food Control, 60, 64–87. MacFie, H. J. H., Bratchell, N., Greenhoff, K., & Vallis, L. V. (1989). Designs to balance the effect of order of presentation and first-order carry-over effects in hall tests. Journal of Sensory Studies, 4, 129–148. Manoukian, E. B. (1986). Mathematical nonparametric statistics. New York, NY: Gordon & Breach. Matos, R. A., Menezes, C. M., Ramos, E. M., Ramos, A. L. S., & Gomide, L. A. M. (2007). Effects of fermentation types in final quality of mutton cooked fermented sausages. Boletim Ceppa, 25, 225–234. Meier-Dinkel, L., Sharifi, A. R., Frieden, L., Tholen, E., Fischer, J., Wicke, M., & Mörlein, D. (2013). Consumer acceptance of fermented sausages made from boars is not distracted by respective information. Meat Science, 94, 468–473. Meier-Dinkel, L., Trautmann, J., Frieden, L., Tholen, E., Knorr, C., Sharifi, A., ... Mörlein, D. (2013). Consumer perception of boar meat as affected by labelling information, malodorous compounds and sensitivity to androstenone. Meat Science, 93, 248–256. Milligan, G. W., & Cooper, M. C. (1985). An examination of procedures for determining the number of clusters in a data set. Psychometrika, 50(2), 159–179. Morales, R., Guerrero, L., Claret, A., Guàrdia, M. A., & Gou, P. (2008). Beliefs and attitudes of butchers and consumers towards dry-cured ham. Meat Science, 80, 1005–1012. Mueller, S., & Szolnoki, G. (2010). The relative influence of packaging, labelling, branding and sensory attributes on liking and purchase intent: Consumers differ in their responsiveness. Food Quality and Preference, 21, 774–783. Paulos, K., Rodrigues, S., Oliveira, A. F., Leite, A., Pereira, E., & Teixeira, A. (2015). Sensory characterization and consumer preference mapping of fresh sausages manufactured with goat and sheep meat. Journal of Food Science, 80, S1568–S1573. Perry, B. D., & Grace, D. C. (2015). How growing complexity of consumer choices and drivers of consumption behaviour affect demand for animal source foods. EcoHealth, 12, 703–712. Resano, H., Pérez-Cueto, F. J. A., Sanjuán, A. I., de Barcellos, M. D., Grunert, K. G., & Verbeke, W. (2011). Consumer satisfaction with dry-cured ham in five European countries. Meat Science, 87, 336–343. Resano, H., Sanjuán, A. I., & Albisu, L. M. (2007). Consumers' acceptability of cured ham in Spain and the influence of information. Food Quality and Preference, 18, 1064–1076. Sañudo, C., Monteiro, A. L. G., Valero, M. V., Fugita, C. A., Monge, P., Guerrero, A., & Campo, M. M. (2016). Cross-cultural study of dry-cured sheep meat acceptability by native and immigrant consumers in Spain. Journal of Sensory Studies, 31, 12–21. Sikorski, Z. E., & Sinkiewicz, I. (2015). Principles of smoking. In F. Toldrá (Ed.). Handbook of fermented meat and poultry (pp. 39–45). (2nd ed.). Oxford: Wiley Blackwell. Stajić, S., Perunović, M., Stanišić, N., Žujović, M., & Živković, D. (2013). Sucuk (Turkishstyle dry-fermented sausage) quality as an influence of recipe formulation and inoculation of starter cultures. Journal of Food Processing and Preservation, 37, 870–880. Stojković, S., Grabež, V., Bjelanović, M., Mandić, S., Vučić, G., Martinović, A., ... Egelandsdal, B. (2015). Production process and quality of two different dry-cured sheep hams from Western Balkan countries. LWT - Food Science and Technology, 64, 1217–1224. Strijbos, C., Schluck, M., Bisschop, J., Bui, T., de Jong, I., van Leeuwen, M., et al. (2016). Consumer awareness and credibility factors of health claims on innovative meat products in a cross-sectional population study in the Netherlands. Food Quality and Preference, 54, 13–22. Toldrá, F. (2004). Ethnic meat products: Mediterranean. In W. K. Jensen, C. Devine, & M. Dikeman (Eds.). Encyclopedia of meat sciences (pp. 451–453). London: Elsevier Science Ltd. Tuorila, H. (2007). Sensory perception as a basis of food acceptance and consumption. In H. J. H. MacFie (Ed.). Consumer-led food product development (pp. 34–59). Cambridge: Woodhead Publishing limited. Valenzuela, V. C., & Pérez, M. P. (2016). Natural antioxidants obtained from fruits and vegetables and their effect over the shelf life of meat and meat products: An update. Revista Chilena de Nutricion, 43, 188–195. Vidal, L., Barreiro, C., Gómez, B., Ares, G., & Giménez, A. (2013). Influence of information on consumers' evaluation using check-all-that-apply questions and sorting: A case study with milk desserts. Journal of Sensory Studies, 28, 125–137. Yalınkılıç, B., Kaban, G., & Kaya, M. (2012). The effects of different levels of orange fiber and fat on microbiological, physical, chemical and sensorial properties of sucuk. Food Microbiology, 29, 255–259. Yang, H., Han, M., Wang, X., Han, Y., Wu, J., Xu, X., & Zhou, G. (2015). Effect of high pressure on cooking losses and functional properties of reduced-fat and reduced-salt pork sausage emulsions. Innovative Food Science & Emerging Technologies, 29, 125–133. Zandstra, E. H., Lion, R., & Newson, R. S. (2015). Salt reduction: Moving from consumer awareness to action. Food Quality and Preference, 48, 376–381.

References Aaslyng, M. D., Vestergaard, C., & Koch, A. G. (2014). The effect of salt reduction on sensory quality and microbial growth in hotdog sausages, bacon, ham and salami. Meat Science, 96, 47–55. Aburto, N. J., Ziolkovska, A., Hooper, L., Elliott, P., Cappuccio, F. P., & Meerpohl, J. J. (2013). Effect of lower sodium intake on health: Systematic review and meta-analyses. BMJ, 346, f1326. Amine, L. S., & Smith, J. A. (2009). Challenges to modern consumer segmentation in a changing world: The need for a second step. Multinational Business Review, 17, 71–100. Andrade, J. C., Nalério, E. S., Giongo, C., de Barcellos, M. D., Ares, G., & Deliza, R. (2017). Consumer perception of dry-cured sheep meat products: Influence of process parameters under different evoked contexts. Meat Science, 130, 30–37. Ares, G., & Jaeger, S. R. (2013). Check-all-that-apply questions: Influence of attribute order on sensory product characterization. Food Quality and Preference, 28, 141–153. Asioli, D., Varela, P., Hersleth, M., Almli, V. L., Olsen, N. V., & Næs, T. (2017). A discussion of recent methodologies for combining sensory and extrinsic product properties in consumer studies. Food Quality and Preference, 56, 266–273. Cadoret, M., & Husson, F. (2013). Construction and evaluation of confidence ellipses applied at sensory data. Food Quality and Preference, 28, 106–115. Cardinal, P., Zamora, M. C., Chambers, E., Barrachina, A. C., & Hough, G. (2015). Convenience sampling for acceptability and CATA measurements may provide inaccurate results: A case study with fruit-flavored powdered beverages tested in Argentina, Spain and U.S.A. Journal of Sensory Studies, 30, 295–304. Corral, S., Salvador, A., Belloch, C., & Flores, M. (2014). Effect of fat and salt reduction on the sensory quality of slow fermented sausages inoculated with Debaryomyces hansenii yeast. Food Control, 45, 1–7. Corral, S., Salvador, A., & Flores, M. (2013). Salt reduction in slow fermented sausages affects the generation of aroma active compounds. Meat Science, 93, 776–785. Datamark. Brazil focus. Processed meat, coppa. (2009). Available at: http://brazilfocus. com/newbrazilfocus/ASP/bf/yearypd/yc00771.asp (Last accessed August 9th, 2010) . De Barcellos, M. D., Kügler, M. J. O., Grunert, K. G., Van Wezemael, L., Pérez-Cueto, F. J. A., ... Verbeke, W. (2010). European consumers' acceptance of beef processing technologies: A focus group study. Innovative Food Science & Emerging Technologies, 11(4), 721–732. Deliza, R., & MacFie, H. J. H. (1996). The generation of sensory expectations by external cues and its effect on sensory perception and hedonic ratings: A review. Journal of Sensory Studies, 11, 103–128. Deliza, R., Macfie, H. J. J., & Hedderley, D. (1999). An investigation of the package features affecting consumer perception of fruit juice using repertory grid and focus group methods. Brazilian Journal of Food Technology, 2, 63–71. Development Core Team, R. (2014). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing (ISBN 3-900051-07-0). Dutra, M. P., Palhares, P. C., Silva, J. R. O., Ezequiel, I. P., Ramos, A. L. S., Perez, J. R. O., & Ramos, E. M. (2013). Technological and quality characteristics of cooked ham-type pâté elaborated with sheep meat. Small Ruminant Research, 115, 56–61. Feiner, G. (2006). Cured air-dried meat products. In G. Feiner (Ed.). Meat products handbook: Practical science and technology (pp. 417–441). Cambridge, England: Woodhead Publishing Ltd. Field, R. A., Willians, J. C., & Miller, G. J. (1983). The effects of diet on lamb flavor. Food Technology, 37, 258–263. Grasso, S., Monahan, F. J., Hutchings, S. C., & Brunton, N. P. (2017). The effect of health claim information disclosure on the sensory characteristics of plant sterol-enriched turkey as assessed using the Check-All-That-Apply (CATA) methodology. Food Quality and Preference, 57, 69–78. Grunert, K. (2002). Current issues in the understanding of consumer food choice. Trends in Food Science & Technology, 13, 275–285. Guimarães, D., Amaral, G., Maia, G., Lemos, M., Ito, M., & Custodio, S. (2017). Pig farming: Productive chain structure, panorama of the sector in Brazil and in the world and BNDES' support. Agribusiness BNDES Setorial, 45, 85–136. Henrique, N. A., Deliza, R., & Rosenthal, A. (2015). Consumer sensory characterization of cooked ham using the check-all-that-apply (CATA) methodology. Food Engineering Reviews, 7, 265–273. 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, 31–41. Hersleth, M., Monteleone, E., Segtnan, A., & Næs, T. (2015). Effects of evoked meal contexts on consumers' responses to intrinsic and extrinsic product attributes in drycured ham. Food Quality and Preference, 40, 191–198. Inguglia, E. S., Zhang, Z., Tiwari, B. K., Kerry, J. P., & Burgess, C. M. (2017). Salt reduction strategies in processed meat products – A review. Trends in Food Science & Technology, 59, 70–78. Issanchou, S. (1996). Consumer expectations and perceptions of meat and meat product quality. Meat Science, 43, 5− 19. Jaeger, S. R. (2006). Non-sensory factors in sensory science research. Food Quality and Preference, 17, 132–144. Jaeger, S. R., Hort, J., Porcherot, C., Ares, G., Pecore, S., & MacFie, H. J. H. (2017). Future directions in sensory and consumer science: Four perspectives and audience voting. Food Quality and Preference, 56, 301–309. Jiang, J., & Xiong, Y. L. (2016). Natural antioxidants as food and feed additives to promote health benefits and quality of meat products: A review. Meat Science, 120,

210