Scholarly article on topic 'A comparison of sensory properties of artisanal style and industrially processed gluten free breads'

A comparison of sensory properties of artisanal style and industrially processed gluten free breads Academic research paper on "Animal and dairy science"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Gluten free" / Napping / Bread / Artisanal}

Abstract of research paper on Animal and dairy science, author of scientific article — Elizabeth M. Muggah, Lisa M. Duizer, Matthew B. McSweeney

Abstract The consumption of gluten free foods is increasing in today’s society and consumers are demanding more from their gluten free products. Consumers are also looking for local alternatives to conventionally produced foods. This study compares the sensory aspects of industrially processed gluten free bread and locally produced, artisanal gluten free bread. Three breads were purchased from grocery stores and three from local bakeries. Global Napping® and ultra flash profiling were used to identify the overall differences between the two different categories of gluten free breads. Partial Napping® and ultra flash profiling were also used to determine if there were any differences between flavor and texture of the breads. Twelve panelists, who regularly consume bread, evaluated all six of the bread samples. Both methods indicated that there are many differences between the industrially processed and artisanal gluten free breads. The main differences being artisanal breads were associated with a wide array of flavors and a dry texture, while the industrially processed breads were found to be bland in flavor and moist. Overall, the artisanal gluten free breads were associated with more negative characteristics than that of industrially processed breads.

Academic research paper on topic "A comparison of sensory properties of artisanal style and industrially processed gluten free breads"

A comparison of sensory properties of artisanal style and industrially

processed gluten free breads

Elizabeth M. Muggaha, Lisa M. Duizerb, Matthew B. McSweeneya,n

aSchool of Nutrition and Dietetics, Acadia University, 15 University Ave, Wolfville, Nova Scotia, Canada B4P 2R6 bDepartment of Food Science, University of Guelph, Guelph, Ontario, Canada

Received 22 June 2015; accepted 28 January 2016

Available online at www.sciencedirect.com

a • g^« __^ International Journal of

bCienCeDireCT Gastronomy and

Food Science

International Journal of Gastronomy and Food Science ■ (■■■■) III III

www.elsevier.com/locate/ijgfs

Scientific paper

Abstract

The consumption of gluten free foods is increasing in today's society and consumers are demanding more from their gluten free products. Consumers are also looking for local alternatives to conventionally produced foods. This study compares the sensory aspects of industrially processed gluten free bread and locally produced, artisanal gluten free bread. Three breads were purchased from grocery stores and three from local bakeries. Global Napping® and ultra flash profiling were used to identify the overall differences between the two different categories of gluten free breads. Partial Napping® and ultra flash profiling were also used to determine if there were any differences between flavor and texture of the breads. Twelve panelists, who regularly consume bread, evaluated all six of the bread samples. Both methods indicated that there are many differences between the industrially processed and artisanal gluten free breads. The main differences being artisanal breads were associated with a wide array of flavors and a dry texture, while the industrially processed breads were found to be bland in flavor and moist. Overall, the artisanal gluten free breads were associated with more negative characteristics than that of industrially processed breads.

© 2016 The Authors. AZTI-Tecnalia. Pubblicato da AZTI-Tecnalia. Tutti i diritti riservati. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Gluten free; Napping; Bread; Artisanal

Introduction

Wheat and other glutinous grains are a staple food in the diet for many of the global population (Mansueto et al., 2014). Humans have been long aware of the unique properties of wheat flours, partrcularly the viscoelastic characteristics of wheat dough that enables the entrapment of carbon dioxide in the process of leavening. These viscoelastic characteristics can be attributed predominantly to the gluten protein complex (Shewry, 2009). Gluten is found in products and flours containing wheat, kamut, spelt, rye and barley. It is formed from the hydrated protein fractions of gliadin and glutenin. This gluten protein complex is an essential factor in forming the crumb and texture of wheat breads that consumers are familiar with (Huttner and Arendt, 2010). A

""Corresponding author. Tel.: +1 902 585 1230; fax: +1 902 585 1637.

E-mail address: matthew.mcsweeney@acadiau.ca (M.B. McSweeney). Peer review under responsibility of AZTI-Tecnalia.

comparison of sensory properties of Artisanal Style and industrially processed gluten free breads

Celiac disease (CD) is an immune-medicated systemic disorder elicited by gluten and related prolamines in genetically susceptible individuals and characterized by the presence of a variable combination of gluten-dependent clinical manifestations (Husby et al., 2012). The disease has a wide variety of manifestations, which can make it difficult to diagnose the condition. Some with the disease may not present gastrointestinal symptoms despite damage to the enterocytes of the small intestine (Cranney et al., 2007). The symptoms of CD can be described as a spectrum that varies from severe mal-absorption symptoms to the absence of symptoms (Mustalahti et al., 2002). The only confirmed treatment of CD is life-long adherence to a gluten-free diet. Celiac disease is the only chronic condition where diet is the exclusive treatment for the disease (Shepherd and Gibson, 2013).

Bread is considered to be a staple food to all of humanity. However, wheat bread consumption has been declining possibly due to changing eating patterns and availability of alternative grain

http://dx.doi.org/10.1016/j.ijgfs.2016.01.001

1878-450X/© 2016 The Authors. AZTI-Tecnalia. Pubblicato da AZTI-Tecnalia. Tutti i diritti riservati. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

products such as breakfast cereals (Gellynck et al., 2009). Due to the growing number of people diagnosed with celiac disease, combined with consumer demand for novel products that are perceived as "healthful", the consumer market for products made from alternative grains is growing (Hüttner and Arendt, 2010).

There has been a recent rise in the market size for gluten-free foods, exceeding the demand from celiac consumers alone. Although, there has been an increase in the prevalence and awareness of CD; this increase in demand for gluten free products could be largely attributed to the recent claims communicated through media and advertising that a gluten-free diet provides health benefits to the general population. The gluten-free market has increased over the past decade climbing from $100 million in 2003 to $1.31 billion in 2011. The gluten-free market is projected to reach $1.68 billion by 2015 (Mansueto et al., 2014).

To satisfy the demand of GF consumers for high quality bread products companies must produce GF bread that has characteristics similar to that of wheat flour bread (Moore et al., 2004). Making bread that is similar to the familiar wheat bread formulations without gluten presents a technological challenge. Gluten-free bread (GFB) is associated with low quality products that have poor crumb and mouth feel. Gluten-free breads and other gluten-free products are also associated with fast staling. Gluten-free doughs are not able to develop a protein network with characteristics like that of gluten. Therefore, the replacement of gluten network in GFB has become a primary goal in the development of new products. GFB formulations have improved in recent years through the addition of various ingredients such as hydrocolloids to the formulation (Hüttner and Arendt, 2010).

In recent years, following a gluten free diet has become easier because of increases in the number and quality of GF products available. A 2013 report by Zarkadas et al. surveyed Canadians living with CD found that, compared to a 2006 survey, consumers have experienced an increase in the ease of finding GF foods, including yeast breads. A study by Laureati et al. (2012) suggested that consumers' preference for GFB is positively affected by softness, crumb, porosity, uniformity and sweet taste. The researchers suggest that these key attributes should be considered in the development of GBF to further improve consumer acceptability.

Most food purchased by consumers is still produced conventionally maintaining the industrial food system (Koury-Hanna, 2014). However, local venues such as farmers' markets have provided outlets for entrepreneurship in local food markets. The goal of these producers tends to coincide with those of the community, to support local economic growth. Local markets are creating a renewed respect for small-scale artisan producers who meet the desires of consumers for good quality food products (Guthrie et al., 2006). Business strategies that relay on niche markets, such as local food are currently experiencing rapid growth (Henryks and Pearson, 2011).

In recent years there has been growth in the U.S. market for unpackaged and artisanal breads, which accounted for 33% of the market share in 2010. These breads are becoming more popular with consumers, as artisanal varieties of breads are perceived as being of premium quality and having more health benefits than the

industrially processed bread (Agriculture and Agri-Food Canada, 2012). Artisan bread, for the purpose of this study, will be defined as produced by small bakeries and sold locally (within 100 km of the location of the study). Like the industrially processed companies, small scale bakeries are beginning to cater to the needs of the consumers that adhere to a gluten free diet. This has led to the production artisanal versions of GFB. Unlike standard wheat bread, each GFB has an individual formulation with a different approach to solving the technological challenge of making high quality bread without the gluten protein complex. This can lead to a wide variation in the sensory characteristics within the gluten free bread category. Sensory research conducted on the niche market of artisan GFB could help small businesses to improve their formulations of GFB. Artisanal GBF is fairly new to the market therefore creating a product description of these products will provide insight into the quality of these breads to help guide future development of such products.

Napping® is a relatively new sensory technique that provides a description of a product and insight into the sensory characteristics of the tested products. Napping® is perceptual mapping technique where the evaluated products are represented in a bi-dimensional (2-D) plane (Lawless and Heymann, 2010). In this test participants are instructed to try the presented samples and position and mark them on a large piece of paper (60 cm by 40 cm), resembling a tablecloth (nappé map), according to the similarity of the products (Valentin et al., 2012). No instruction is given on how to evaluate the products the assessor decides their own criteria. Napping® is often combined with a method called Ultra Flash Profiling, in which case assessors are asked to write a few words describing each sample next to where they positioned them on the map.

This study will examine the differences in the sensory properties of industrially processed GFB and locally sourced artisanal GFB and will identify the descriptive differences between the two subtypes of products.

Materials and methods

The study was conducted at the Acadia Centre for the Sensory Research of Food, Acadia University. Ethics approval for the study was received from the Acadia University Ethic's Board (REB 13-72).

Products

Six gluten-free bread products identified on the package as whole grain or multi-grain were selected to be used as samples for the Napping® trial. Three of the GFB products were industrially processed and were sourced from local chain grocery stores in Atlantic Canada. All of these breads were produced by national companies and are sold across Canada. The other three samples were artisan varieties of whole grain GFB sourced from local bakeries. The sensory center is located in rural area approximately 100 km from the nearest urban area. Because of this a radius of approximately 100 km was used to source the local artisan varieties. The main ingredients found in each bread are shown in

Table 1. These six breads were chosen, as they were the most accessible to consumers living in region surrounding the sensory centre. Furthermore these breads were identified by a small group (n — 3) of consumers, suffering from celiac disease, who were interviewed during the study design.

Most of the GF bread products were purchased frozen, however a couple of the artisan varieties were purchased fresh and frozen upon purchasing to prevent staling. All of the samples were stored in a freezer maintained at — 18 °C. The GFB was stored frozen to maintain the quality of the products. A study by Ronda and Roos (2011) concluded that the frozen storage of GFB is effective to extend the shelf life of GFB without compromising product quality. All breads were stored for one week before the trial sessions.

Participants

The study consisted of twelve participants. The participants had an average age of 38.5( 7 14.9) years. All of the participants lived in the local area and were considered to be experienced panelists, having previously participated in trained sensory panels. Experienced panelists were selected to take part in the study as they have experience in focusing on individual sensory properties. All participants were recruited using e-mail and word of mouth. The participants were not required to be on a gluten free diet nor were they required to be familiar with GFB products.

The participants completed the Napping® in groups from 1 to 6 depending on the availability of the panelists. In total they were three Napping® sessions held.

Sample presentation

Before each Napping® session, the samples were taken out of the freezer, wrapped in plastic wrap and left to warm at room temperature (23 °C). The bread slices were cut into small squares (4 cm by 4 cm) to yield sample portions including the crust. The samples were presented on white standard sized napkins, in rows on a plain white tray. Random three digit sample codes were written on the bottom left corner of the napkins in black permanent marker. The order of sample presentation given to each panelist was randomized.

McIntosh apple slices were provided as a palate cleanser. The panelists were also provided with filtered water served at room temperature. The panelists were instructed and reminded to cleanse their palate and drink water in between trying each sample.

Sensory testing

The panelists were given instruction on how to perform the Napping® method using sandwich cookies as an example food. The moderator demonstrated the procedure for the panelists by placing the sandwich cookies on the large sheet of white paper. Panelists were provided with three large rectangle sheets of white paper 40 cm by 60 cm in the size on which they would create their nappé maps. They were then instructed to perform Napping® a total of three times. First, to assess the overall or global perception of the GFB products followed by partial Napping® (PN) for texture and flavor of the products. The global Napping® involved the panelists evaluating the samples as a whole involving all aspects of the food product. While the partial Napping® asked the panelists to just evaluate the products first for their texture and then for their flavor. The participants were provided with one set of seven samples (6 samples and one replicate) to perform the global Napping® (GN) and another set of samples to perform both the PN for texture and flavor.

During each Napping® exercise, participants were instructed that the more similar two samples were, the closer they were placed together on the paper. Samples, which were different, were placed further apart. Additionally panelists were instructed to write the sample code on the piece of paper and 2-5 words to describe the sample. Writing descriptors of the products is referred to as Ultra Flash Profiling and this method is usually utilized along with Napping® (Perrin et al., 2008). Panelists were requested to avoid using comparison of samples (ex. Sample 1 was whiter than Sample 2). After all three Napping® maps were completed a exit survey was administered at the end of the sessions to collect demographic information, as well as, gain insight into the participant's attitudes and beliefs concerning traditional wheat bread products and GFB products. The panelists answered questions about their perceived notions of what GFB should taste and look like. They also indicated if they believed GFB to be healthier than traditional wheat bread, along with their beliefs concerning local food.

Statistical analysis

For analysis of the nappé maps from each Napping® exercise, the location of coordinates of each product on each assessor's map was measured and recorded. The attributes given to the products by the panelists were also recorded. The terms used were then tallied for each sample. If a panelist expressed an intensity of an

Table 1

Gluten-free bread products used in Napping®.

Gluten-free Bread

Source

Main ingredients (first 5)

Industrially processed

Industrially processed

Industrially processed

Artisan

Artisan

Artisan

Water, modified tapioca starch, corn starch, potato starch, vegetable oil Water, brown rice flour, tapioca flour, whole grain sorghum flour, eggs Water, tapioca starch, brown rice flour, egg whites, canola oil Water, eggs, tapioca flour, millet flour, brown rice flour Potato starch, plain yogurt, millet flour, flax, eggs

Buckwheat flour, sorghum flour, brown rice flour, tapioca flour, potato flour

aWere purchased fresh and frozen upon purchasing.

attribute, each intensity was considered as a separate attribute (ex. dry, medium dry, very dry were all entered as separate attributes). The first table contained the entire consumer x coordinates, the second table contained the consumer y coordinates and, the third table contained the frequency of the identified attributes (Varela and Ares, 2012).

Napping® is a perceptual mapping method, because of this the multivariate statistical method of Multiple Factor Analysis (MFA) was used to analyze the data. Pages (2005) proposed that using MFA is the best method for analyzing Napping® because it accounts best for the differences between participants. This method accounted for both qualitative and quantitative variables within the data set collected. The MFA analysis of the data was performed using XLSTAT© software Version 2014.6 (Addinsoft, New York, NY).

Results and discussion

The goal of the study was to identify differences between industrially processed and artisan GFB. Napping®. Along with Napping®, ultra-flash profiling was used to provide specific attributes that differentiated the products.

The use of Napping® to describe the sensory attributes of the gluten free breads provides an opportunity to observe the attributes most often perceived by individuals consuming these products. Often when describing food product, conventional trained panels are used to characterize products. With this approach, however, there is no insight into which attributes are most often used to describe the products. This makes it difficult for product developers to select which attributes they should focus on during development and reformulation. Napping® is a projective mapping method that uses an untrained panel to create a more complete description of a product as well as insight into the weight of each attribute (Grygorczyk, 2012). The Napping® method has been described as simple and user friendly, making it an easy method to utilize with inexperienced panelists (Nestrud and Lawless, 2008).

Data from the Napping® sessions were analyzed using multiple factor analysis (with the attributes considered to be supplementary variables. The results of the MFA created a visual representation of the products in graphical form. The three sessions, Global Napping®, and Partial Napping® for flavor and texture coupled with ultra flash profiling will be discussed separately.

Global napping of gluten free bread samples

Table 2 lists the eigenvalues and inertia for the first four axes of the MFA for the global Napping® trial examining the overall differences in the GFB and the partial Napping® trials examining the flavor and texture of the samples. These four axes of the MFA for the global Napping® accounted for just over 88% of the variance. The graphical description of the GFB following MFA can be seen in Fig. 1. Dimension 1 separates the artisan GFB and the industrially processed samples while dimension 2 separates B3C from the remaining industrially processed breads and B6A from the two other artisanal breads.

To identify the sensory attributes differentiating the products, the terms used to describe the products were examined. The panelists

generated 84 terms to describe the six GFB samples. The frequency with which each attribute was mentioned and the correlations of the attributes are listed in Table 3. Only attributes that were used by panelists more than once to describe a GFB are included in the table.

The first dimension appears to be governed by attributes that describe the texture of the bread samples. The positive side of the dimension is associated with terms describing a desirable texture of bread (light crumb, chewy, soft) (Hager et al., 2012; Szcesniak, 2002; Torbica et al., 2010). The negative side of the first dimension is correlated with undesirable texture components of the bread, for instance dense, mealy, crumbly and dissolving (Schober, 2009). Furthermore the first dimension also appears to be governed by the color of the bread sample with light color associated positively with dimension 1 and dark, brown and gray associated negatively with dimension 1.

Comparing the sensory terms with the product map in Fig. 1, the industrially processed bread samples have a texture that is described as chewy, soft and with a light crumb. These breads are also light in color. The three artisanal breads are dense, crumbly and dissolving with a dark or brown color.

The second dimension in Fig. 1 describes the moistness of the product with moist falling on the positive end of the dimension and

Table 2

Eigenvalues and percent of variance accounted for by each dimension of the multiple factor analysis of the Global Napping® trial and the two Partial Napping® trials.

Factor 1 Factor 2 Factor 3 Factor 4

Global Napping

Eigenvalue 3.78 1.81 1.101 1.0

Percent of variance 36.64 23.89 14.52 13.15

Cumulative percent of variance 36.64 60.53 75.04 88.2

Partial Napping® for flavor

Eigenvalue 2.53 1.77 1.49 1.02

Percent of variance 34.27 23.91 20.11 13.86

Cumulative percent of variance 36.3 58.18 78.28 92.14

Partial Napping® for texture

Eigenvalue 2.84 1.63 1.33 0.81

Percent of variance 38.60 22.1 18.017 10.98

Cumulative percent of variance 38.60 60.70 78.72 89.70

B5A. B4A. B2C • . B1C

B6A* ■ B3C

-2 -1 0 F1 (36.64 %)

Fig. 1. Product map for Global Napping® of gluten free bread samples

Table 3

Correlations of attributes with the first four dimensions of multiple factor analysis for Global Napping® data and the frequency of usage of terms. On those terms that were used more than once by panelists are included.

Attribute Usage frequency Correlation with dimensions'1

Dimension 1 Dimension 2 Dimension 3 Dimension 4

Yellow 10 0.385 0.744 0.369 - 0.025

Moist 9 0.481 0.725 0.222 - 0.352

Dense 13 - 0.722 0.119 0.290 0.167

Smooth 5 0.577 0.754 - 0.282 - 0.059

Sweet 20 0.482 0.372 0.007 0.390

Seedy 11 - 0.329 0.510 0.251 - 0.655

Fluffy 2 0.618 0.650 0.220 - 0.374

Golden 9 0.001 0.687 - 0.210 0.457

Light crumb 8 0.819 0.078 - 0.304 0.228

Chewy 13 0.642 0.303 - 0.594 0.144

Soft 6 0.618 0.650 - 0.220 - 0.374

Light color 9 0.901 - 0.054 - 0.194 - 0.384

Brown crust 3 0.051 - 0.520 - 0.713 - 0.319

Dry 28 - 0.353 - 0.593 - 0.519 0.482

Bland 11 0.422 - 0.645 - 0.384 - 0.344

Nutty 6 - 0.437 0.601 0.487 0.140

Grainy flavor 2 0.139 0.290 0.148 0.198

Chemical after taste 2 - 0.284 0.156 - 0.770 - 0.305

Gluten free taste 4 0.283 0.814 0.293 0.399

Cake like 4 0.283 0.814 0.293 0.399

Crunchy 4 - 0.162 0.441 0.500 - 0.694

Brown bread 2 0.144 0.523 0.146 0.201

Grainy texture 16 0.331 0.172 0.428 - 0.540

Light texture 2 0.619 - 0.320 0.257 - 0.468

White 6 0.427 - 0.895 0.017 - 0.042

Rough 2 0.427 - 0.895 0.017 - 0.042

Uniform texture 4 0.215 - 0.791 0.341 0.447

Tough crust 4 - 0.200 - 0.398 0.561 - 0.696

Mealy 3 - 0.901 0.054 0.194 0.384

Loaf appearance 2 - 0.335 0.164 0.513 0.773

Crumbly 11 - 0.940 - 0.236 - 0.186 - 0.158

Corn 2 - 0.819 - 0.078 0.304 - 0.228

Short bread texture 3 - 0.901 0.054 0.194 0.384

Dark 6 - 0.939 - 0.087 - 0.238 - 0.230

Bitter 2 - 0.819 - 0.078 0.304 - 0.228

Brown 12 - 0.901 0.054 0.194 0.384

Dissolving 3 - 0.851 - 0.047 - 0.439 - 0.060

Gray 3 - 0.785 - 0.134 - 0.388 - 0.463

Spongey 4 0.818 - 0.484 - 0.122 - 0.279

Pastey 2 0.200 0.398 - 0.561 0.696

Flavorful 3 - 0.901 0.054 0.194 0.384

aCorrelations > 0.6 are highlighted in bold. Correlations just under 0.6 are in bold and italicized.

dry on the negative. Brown crust, and rough were also on the negative end of the dimension while smooth golden, fluffy, gluten free taste and cake like fell on the positive end. Artisanal breads B4A and B5A as well as industrially processed breads B2C and B1C are described as a moist, smooth golden and fluffy, while B6A and B3C are dry and rough with a brown crust. These two breads included tapioca starch and brown rice flour in their ingredient list (Table 1) and the mixture of these two grains may have led to the products being associated with these undesirable attributes. Dimensions 3 and 4 were much more difficult to interpret and did not distinguish between the industrially processed and artisanal products.

Overall the Global Napping® indicated that texture is very important to consumers when they are evaluating GFBs. Furthermore consumers differentiate breads based on the perceived moisture contents (moist vs. dry). The trial also indicated that a range of sensory attributes, many of which are texture terms, describe GFBs.

Flavor Partial Napping®

In addition to providing a global overview of the sensory attributes present in the breads, the panelists were also asked to determine how different the samples were when only examining

the flavor of the GFB samples. Once again the results were analyzed using MFA and the first four axes accounted for 92.1% of variability. Table 2 lists the eigenvalues for the first four axes of the MFA. The graphical description of the flavor of GFB following MFA can be seen in Fig. 2.

Three groupings of bread are apparent based on results shown in Fig. 2. B3C and B1C load high on both dimension 1 and dimension 2 while B2C and B5A load high on dimension 1 but low for dimension 2. B6A and B4A are low on both dimension 1 and 2 (Fig. 2).

The attributes used to describe the products are listed in Table 4. The first dimension is dominated by sweetness terms on the positive side and bitter, burnt and grainy terms on the negative side. Looking along the first dimension, all industrially processed breads as well as B5A were associated with sweetness. This may be due to differing ingredients in the bread, with the industrially processed breads mainly using tapioca starch or flour and B5A showing potato starch as the first ingredient. The artisanal breads use a wide range of flour including millet, brown rice, sorghum and buckwheat flour. B6A and B4A are both associated with bitter, burnt and grainy notes and this may be due to consumer being unfamiliar with the taste of buckwheat, sorghum or millet flour.

Looking at dimension 2, it seems to be governed by a lack of taste on the positive side versus a wide range of flavors on the negative. The positive side is dominated by terms like bland and yeasty, while the negative has a multitude of flavor terms, including nutty, sweet, grainy, fruity and cinnamon. B3C and B1C as associated with yeasty while the remaining breads are associated with grainy, fruity notes.

The Flavor Partial Napping® indicated there is a wide range of flavors when examining gluten free breads and that the industrially processed breads tend to have a bland but sweet flavor, while the artisanal breads have a wide range of flavors. However included in these wide range of flavors is bitter, burnt and other negative attributes that may be off putting to some consumers.

Texture Partial Napping®

The global Napping® trial indicated that texture attributes are commonly used by individuals to differentiate GFBs. The texture partial napping trial was conducted to more closely examine the different attributes consumers use to describe texture properties of

. B3C • B1C

B6A* B4A* * B2C * B5A

-3-2-10123 F1 (34.27 %)

Fig. 2. Product map for Flavor Partial Napping® for the gluten free bread samples.

GFBs. Table 2 lists the eigenvalues and inertia for the first four axes of the data analysis. The first four axes accounted for just 89% of variance. The representation of the GFBs following MFA are presented in Fig. 3.

As in the other two trials the industrially processed GFBs and artisanal GFBs are differentiated in the product map with all of the artisanal breads being positioned on the positive side of dimension 1 and the industrially processed breads falling on the negative side. Once again indicating there is a difference in the industrially processed and artisanal breads in terms of texture.

The panelists created 53 terms to describe the texture of the bread. That is more terms than were created to describe the flavor of the bread samples, further confirming the importance of texture to consumers when evaluating bread. This agrees with the work by many different authors, that texture is a driving force when consumers are evaluating bread (Cardello et al., 1982; Challacombe et al., 2012; Kihlberg et al., 2005; Bakke and Vickers, 2007). The frequency with which each term was mentioned by panelists and the correlations of the terms to the dimensions are listed in Table 5.

In dimension 1, there is a clear divide in the bread products, which follows a similar trend to the differentiation observed with global napping (Fig. 1). All of the industrially processed bread products are associated with chewy, soft, moist and fluffy. These are all deemed to be desirable attributes in bread (Crowley et al., 2002). However the industrially processed breads were also associated with spongy, which may be undesirable to consumers. All of the artisanal breads were associated with undesirable terms such as pasty, quick dissolving, crumbly, dry and gritty. This could be due to difference in ingredients used as industrially processed products mainly use tapioca and potato starch and a industrially processed baking process while the artisanal breads use a wide variety of starches and flours and predominantly make the breads by hand.

The second dimension shown in Fig. 3 was described by words such as cake, moist, dense, soft and heavy on the negative side and granular, medium dry, fluffy, tough crust and mushy on the positive side. B3C was the sample most obviously different to the other samples for this dimension. Results suggest that this bread was not as dense as the other breads. Being dense is a common trait to gluten free products and has been shown to be a detractor from consumer liking of these products (Huttner and Arendt, 2010). Furthermore this bread is drier than those found on the negative side, which was described as moist.

The results concerning the artisanal breads were unanticipated in this study as the recipes (Table 1) listed ingredients that had been found in previous studies to be suitable for the production of gluten free products. These ingredients include buckwheat (Torbica et al., 2010, millet (Eneche 1999), and brown rice flour (Watanabe et al., 2004). However the studies listed above use a percentage (1050%) of these flours to produce GFBs and do not produce breads from strictly 100% of one of these listed flours. Also these studies do not compare these breads to those produced with other grains, but rather indicate that these breads were found to be acceptable to consumers. If they were compared to industrially processed GFBs that may have also been found to be linked to unacceptable

Table 4

Correlations of attributes with the first four dimensions of multiple factor analysis for the Flavor Partial Napping® data and the frequency of usage of terms. On those terms that were used more than once by panelists are included.

Attribute Usage frequency Correlation with dimensionsa

Dimension 1 Dimension 2 Dimension 3 Dimension 4

Flax 2 - 0.424 - 0.059 0.500 0.396

Nutty 20 - 0.229 - 0.335 - 0.313 - 0.708

Sweet 28 0.430 - 0.392 - 0.059 - 0.322

Grainy 14 - 0.436 - 0.351 - 0.066 - 0.405

Bland 14 - 0.210 0.606 - 0.382 0.089

Molasses 2 0.263 0.233 0.379 - 0.687

Bitter aftertaste 3 - 0.517 - 0.229 - 0.274 - 0.245

Yeasty 3 0.129 0.639 - 0.275 0.313

Rye 4 - 0.424 - 0.059 0.500 0.396

Bitter 7 - 0.417 0.016 - 0.419 - 0.538

Salty 9 0.362 0.347 0.343 - 0.108

Burnt 4 - 0.314 0.387 0.416 0.455

Sweet bread 2 0.367 - 0.450 - 0.053 0.399

Fruity 2 0.412 - 0.496 0.158 - 0.052

Cinnamon 3 0.367 - 0.450 - 0.053 0.399

Buttery 2 0.263 0.233 0.379 - 0.687

Chemical aftertaste 3 - 0.439 - 0.329 0.165 - 0.386

Stale 4 - 0.412 0.496 - 0.158 0.052

Mild flavor 3 0.280 0.532 0.062 - 0.258

Low sweet 5 0.426 0.371 - 0.197 0.312

White bread 2 0.175 0.705 - 0.133 0.094

Sweet aftertaste 2 0.412 - 0.496 0.158 - 0.052

aCorrelations > 0.6 are highlighted in bold. Correlations just under 0.6 are in bold and italicized.

B2C" * B5A

-3-2-10123 F1 (38.60 %)

Fig. 3. Product map for texture partial Napping® for the gluten free bread samples.

characteristics. Furthermore this result may have been due to the fact that panelists were screened for the consumption of bread, but not their consumption of artisanal bread. These panelists may be accustom to industrially processed breads and were not expecting different flavors and texture imparted in the bread by artisanal production.

Using non-consumers of gluten free products to evaluate the sensory properties of products they are not familiar with may be a weakness of this research. Laureati et al. (2012), however, compared sensory and hedonic preference perception of GFB between celiac and non-celiac consumers. They concluded that both celiac and non-celiac consumers evaluated GFB in a similar way and that the choice of bread was based upon the sensory descriptors. Indicating that non-celiac consumer data collected in

this study should translate to celiac consumers despite the fact there were no participants who were adhering to a gluten free diet. Another limitation of this study, was the freezing of the fresh GFB, this may have imparted different sensory properties within the product. However Ronda and Roos (2011) indicated that freezing of GFB should not alter the product quality.

Conclusions

Results of the Napping® and MFA analysis shows individuals perceive gluten free artisanal breads differently to industrially processed gluten free breads. The global Napping® trial generated a large number of terms and indicated that texture and perceived moistness is an important factor when creating GFBs. When products were assessed based on texture or flavor alone in the partial Napping® trials, a clear divide were seen between the industrially processed breads and artisanal breads. The artisanal breads had much greater amplitude than the industrially processed breads and consisted of many different flavors. However these artisanal breads were also associated with a crumbly or dry texture. While the industrially processed products were found to be bland, they were perceived to be moist with a soft crust. It is apparent that flavors and textures of gluten free products are very dependent on the ingredients used in their manufacture. Overall, this study has found that consumers associate artisanal GFBs with more negative attributes, however since they are perceived as local or healthier the general public may continue to buy them. Although this research indicates how individuals perceive gluten free breads, more work

Table 5

Correlations of attributes with the first four dimensions of multiple factor analysis for the texture partial Napping® data and the frequency of usage of terms. On those terms that were used more than once by panelists are included.

Attribute Usage frequency Correlation with dimensions"

Dimension 1 Dimension 2 Dimension 3 Dimension 4

Crumbly 18 0.921 - 0.110 0.077 - 0.304

Granular 5 0.536 0.604 0.528 0.189

Dry 37 0.816 0.076 - 0.560 0.002

Medium dry 3 - 0.388 0.537 0.290 0.588

Chewy 12 - 0.959 - 0.148 - 0.225 - 0.089

Seedy 10 0.116 0.100 0.907 - 0.329

Moist 11 - 0.710 - 0.536 0.427 0.052

Dense 9 0.022 - 0.419 - 0.775 0.164

Smooth 2 - 0.659 0.212 - 0.660 - 0.291

Cake 2 - 0.536 - 0.604 - 0.528 - 0.189

Grainy 14 0.250 - 0.067 0.854 0.078

Crunchy 8 0.334 - 0.188 0.922 - 0.053

Fluffy 2 - 0.591 0.621 0.454 - 0.111

Soft 4 - 0.779 - 0.546 0.278 - 0.120

Light 5 - 0.817 0.017 0.402 0.346

Rough crust 2 - 0.244 - 0.529 - 0.564 0.585

Coarse 3 - 0.107 0.418 0.434 - 0.484

Tough crust 3 - 0.510 0.809 0.131 - 0.155

Medium dense crumb 3 0.565 0.563 - 0.124 - 0.580

Light crumb 3 - 0.565 - 0.563 0.124 0.580

Not crunchy 2 - 0.056 0.638 - 0.389 0.594

Powdery 5 0.972 - 0.124 0.039 - 0.112

Heavy 4 0.591 - 0.621 - 0.454 0.111

Solid 2 0.537 0.034 - 0.043 0.663

Porous 2 - 0.450 - 0.087 0.881 0.038

Soft crust 2 - 0.659 0.212 - 0.660 - 0.291

Spongey 7 - 0.506 - 0.426 - 0.711 0.235

Wheat bread 2 - 0.450 - 0.087 0.881 0.038

Nutty 3 0.452 - 0.152 0.760 - 0.419

Gritty 4 0.529 - 0.048 0.576 - 0.326

Cohesive 3 - 0.957 0.136 0.034 -0.246

Pastey 2 0.658 - 0.230 - 0.140 0.333

Gluey 4 - 0.123 0.247 - 0.702 0.372

Mushy 2 - 0.056 0.638 - 0.389 0.594

Quick dissolving 2 0.658 - 0.230 - 0.140 0.333

aCorrelations > 0.6 are highlighted in bold. Correlations just under 0.6 are in bold and italicize.

must be undertaken to determine consumer liking and sensory attributes, which drive the purchase of GFB.

Acknowledgments

This research was supported by the Government of Nova Scotia's Strategic Cooperative Education Incentive (201400361).

References

Agriculture and Agri-Food Canada, 2012. International markets bureau American eating trends report: bread. Retrieved from: (http://www5.agr. gc.ca/resources/prod/Internet-Internet/MISB-DGSIM/ATS-SEA/PDF/ 6257-eng.pdf) (10.07.14). Bakke, A., Vickers, Z., 2007. Consumer liking of refined and whole wheat

breads. J. Food Sci. 72 (7), S473-S480. Cardello, A.V., Maller, O., Kapsalis, J.G., Segars, R.A., Sawyer, F.M., Murphy, C., Moskowitz, H.R., 1982. Perception of texture by trained and consumer panellists. J. Food Sci. 47 (4), 1186-1197.

Challacombe, C.A., Abdel-Aal, E.M., Seetharaman, K., Duizer, L.M., 2012. Influence of phenolic acid content on sensory perception of bread and crackers made from red or white wheat. J. Cereal Sci. 56 (2), 181-188.

Cranney, A., Zarkadas, M., Graham, I.D., Butzner, J.D., Rashid, M., Warren, R., Molloy, M., Case, S., Burrows, V., Switzer, C., 2007. The Canadian celiac health survey. Dig. Dis. Sci. 52 (4), 1087-1095.

Crowley, P., Schober, T.J., Clarke, C.I., Arendt, E.K., 2002. The effect of storage time on textural and crumb grain characteristics of sourdough wheat bread. Eur. Food Res. Technol. 214 (6), 489-496.

Eneche, E.H., 1999. Biscuit-making potential of millet/pigeon pea flour blends. Plant Food Hum. Nutr. 54 (1), 21-27.

Gellynck, X., Kühne, B., Van Bockstaele, F., Van de Walle, D., Dewettinck, K., 2009. Consumer perception of bread quality. Appetite 53 (1), 16-23.

Grygorczyk, A., 2012. A novel approach to structure generation for texture improvement in a soymilk-dariy gel (doctoral dissertation). The University of Guelph, Guelph, Canada (No. 3933).

Guthrie, J., Guthrie, A., Lawson, R., Cameron, A., 2006. Farmers' markets: the small business counter-revolution in food production and retailing. Br. Food J. 108 (7), 560-573.

Hüttner, E.K., Arendt, E.K., 2010. Recent advances in gluten-free baking and the current status of oats. Trends Food Sci. Technol. 21 (6), 303-312.

Hager, A., Wolter, A., Czerny, M., Bez, J., Zannini, E., Arendt, E.K., Czerny, M., 2012. Investigation of product quality, sensory profile and ultrastructure of breads made from a range of industrially processed gluten-free flours compared to their wheat counterparts. Eur. Food Res. Technol. 235 (2), 333-334.

Henryks, J., Pearson, D., 2011. Retail outlets: nurturing organic food consumers. Org. Agric. 1 (4), 247-259.

Husby, S., Koleizko, S., Korponay-Szabo, I.R., Mearin, M.L., Phillips, A., Shamir, R., Troncome, R., Giersiepen, K., Branski, D., Catassi, C., Leigeman, M., Maki, M., Ribes-Koninckx, C., Ventura, A., Zimmer, K. P., 2012. European Society for pediatric gastroenterology, heptalogy, and nutrition guidelines for the diagnosis of coeliac disease. J. Pediatr. Gastroenterol. Nutr. 56 (1), 136.

Kihlberg, I., Johansson, L., Langsrud, O., Risvik, E., 2005. Effects of information on liking of bread. Food Qual. Preference 16 (1), 25-35.

Koury-Hanna, J., 2014. Participatory Food Costing and its Impacts on Local and/or Sustainable Food Purchases. Acadia University, Wolfville, Canada (No. LE3.A272014).

Laureati, M., Giussani, B., Pagliarini, E., 2012. Sensory and hedonic perception of gluten-free bread: comparison between celiac and non-celiac subjects. Food Res. Int. 46 (1), 326-333.

Lawless, H.T., Heymann, H., 2010. In: Harry, T., Lawless, H.H. (Eds.), Sensory Evaluation of Food Principles and Practices 2nd ed. Springer, New York c2010.

Mansueto, P., Seidita, A., D'Alcamo, A., Carroccio, A., 2014. Non-celiac gluten sensitivity: literature review. J. Am. Coll. Nutr. 33 (1), 39-54.

Moore, M.M., Schober, T.J., Dockery, P., Arendt, E.K., 2004. Textural comparisons of gluten-free and wheat-based doughs, batters, and breads. Cereal Chem. 81 (5), 567-575.

Mustalahti, K., Lohiniemi, S., Collin, C., Vuolteeho, N., Laippala, P., Maki, M., 2002. Gluten-free diet and quality of life in patients with screen-detected celiac disease. Eff. Clin. Pract. 5 (3), 105-113.

Nestrud, M.A., Lawless, H.T., 2008. Perceptual mapping of citrus juices using projective mapping and profiling data from culinary professionals and consumers. Food Qual. Preference 19 (4), 431-438.

Pages, J., 2005. Collection and analysis of perceived product inter-distances using multiple factor analysis: Application to the study of 10 white wines from the Loire Valley. Food Qual. Preference 16 (7), 642-649.

Perrin, L., Symoneaux, R., Maitre, I., Asselin, C., Jourjon, F., Pages, J., 2008. Comparison of three sensory methods for use with the Napping® procedure: case of ten wines from Loire valley. Food Qual. Preference 19 (1), 1-11.

Ronda, F., Roos, Y.H., 2011. Staling of fresh and frozen gluten-free bread. J. Cereal Sci. 53 (3), 340-346.

Schober, T.J., 2009. Manufacture of gluten-free specialty breads and confectionary products. Gluten Free Food Sci. Technol., 130-180.

Shepherd, S., Gibson, P.R., 2013. Nutritional inadequacies of the gluten-free diet in both recently-diagnosed and long-term patients with coeliac disease. J. Hum. Nutr. Diet. 26 (4), 349-358.

Shewry, P.R., 2009. Wheat. J. Exp. Bot. 60 (6), 1537-1553.

Szcesniak, A.S., 2002. Texture is a sensory property. Food Qual. Preference 13 (4), 215-225.

Torbica, A., Hadnadev, M., Dapcevic, T., 2010. Rheological, texture and sensory properties of gluten-free based formulations based on rice and buckwheat flour. Food Hydrocolloids 24 (6), 628-632.

Valentin, D., Chollet, S., Lelievre, M., Abdi, H., 2012. Quick and dirty but still pretty good: a review of new descriptive methods in food science. Int. J. Food Sci. Technol. 47 (8), 1563-1578.

Varela, P., Ares, G., 2012. Sensory profiling, the blurred line between sensory and consumer science. A review of novel methods for product characterization. Food Res. Int. 48 (2), 893-908.

Watanabe, M., Maeda, T., Tsukahara, K., Kayahara, H., Morita, N., 2004. Cereal Chem. 81 (4), 450-455.