- Email: [email protected]
Aroma-related cross-modal interactions for sugar reduction in milk desserts: Influence on consumer perception
Accepted Manuscript Aroma-related cross-modal interactions for sugar reduction in milk desserts: Influence on consumer perception
Florencia Alcaire, Lucía Antúnez, Leticia Vidal, Ana Giménez, Gastón Ares PII: DOI: Reference:
S0963-9969(17)30081-9 doi: 10.1016/j.foodres.2017.02.019 FRIN 6602
To appear in:
Food Research International
Received date: Revised date: Accepted date:
18 November 2016 24 February 2017 26 February 2017
Please cite this article as: Florencia Alcaire, Lucía Antúnez, Leticia Vidal, Ana Giménez, Gastón Ares , Aroma-related cross-modal interactions for sugar reduction in milk desserts: Influence on consumer perception. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Frin(2017), doi: 10.1016/ j.foodres.2017.02.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Manuscript for submission to Food Research International
Aroma-related cross-modal interactions for sugar reduction in milk desserts:
PT
Influence on consumer perception
RI
Florencia Alcaire*, Lucía Antúnez, Leticia Vidal, Ana Giménez, Gastón Ares
SC
Sensometrics & Consumer Science, Instituto Polo Tecnológico de Pando, Facultad de Química, Universidad de la República. By Pass de Rutas 8 y 101 s/n. C.P. 91000.
MA
NU
Pando, Canelones, Uruguay.
AC
CE
PT E
D
*Corresponding author: Florencia Alcaire [Email: [email protected]]
1
ACCEPTED MANUSCRIPT Research highlights Aroma-related cross modal interactions for sugar reduction were assessed. A 10% sugar reduction did not cause changes in sensory and hedonic perception Vanilla increased sweetness and vanilla flavour in 40% sugar-reduced samples.
PT
The joint increase in vanilla and starch concentration provided better results.
AC
CE
PT E
D
MA
NU
SC
RI
Cross-modal interactions hold potential for product reformulation.
2
ACCEPTED MANUSCRIPT Abstract Reformulation of industrialized products has been regarded as one of the most cost-effective strategies to reduce sugar intake. Although non-nutritive sweeteners have been extensively used to reduce the added sugar content of these products, increasing evidence about the existence of compensatory energy intake mechanisms
PT
makes it necessary to develop alternative strategies to achieve rapid sugar reductions. In this context, the aim of the present work was to evaluate aroma-related cross modal
RI
interactions for sugar reduction in vanilla milk desserts. In particular, the influence of
SC
increasing vanilla concentration and the joint increase of vanilla and starch concentration on consumer sensory and hedonic perception was assessed. Two
NU
studies with 100 consumers each were conducted, in which a total of 15 samples were evaluated. For each sample, consumers rated their overall liking and answered a
MA
check-all-that-apply (CATA) question comprising 12 flavour and texture terms. Sugar reduction caused significant changes in the flavour and texture characteristics of the
D
desserts. An increase in vanilla concentration had a minor effect on their sensory
PT E
characteristics. However, increasing both vanilla and starch concentration led to an increase in vanilla flavour and sweetness perception and reduced changes in consumer hedonic perception. These results showed the potential of aroma-related
CE
cross modal interactions for minimizing the sensory changes caused by sugar
AC
reduction. These strategies could contribute to product reformulation without the need to use non-nutritive sweeteners.
Keywords: sugar reduction; product reformulation; cross-modal interactions; CATA questions; consumer studies
3
ACCEPTED MANUSCRIPT 1. Introduction Excessive sugar consumption has become a significant health problem worldwide as it has linked to the increasing prevalence of obesity (Vio & Uauy, 2007). Added sugar consumption contributes to 500 calories per day, corresponding to 1725% of the recommended calorie intake (Lustig, Schmidt, & Brindis, 2012). In addition,
PT
excessive sugar intake has been linked to several metabolic abnormalities and adverse health problems, such as type 2 diabetes, cardiovascular diseases and dental caries
RI
(Johnson et al., 2009; Moynihan, & Kelly, 2014; Te Morenga, Mallard, & Mann, 2013).
SC
For this reason, it has been recommended to restrict the intake of sugars to less than 5-10% of the total daily energy consumption (World Health Organization & Agriculture
NU
Organization of the United Nations, 2003; Scientific Advisory Committee on Nutrition, 2014). This situation makes it necessary to develop strategies to reduce sugar
MA
consumption worldwide.
Sugar is added to a large proportion of processed foods (Vio & Uauy, 2007).
D
Therefore, one of the possible strategies for rapidly decreasing sugar consumption of
PT E
the population is to reduce the added sugar content of these products (MacGregor & Hashem, 2014). One of the main challenges faced in product reformulation is related to changes in product flavour and texture, which are determinant of product success in
CE
the marketplace (van Raaij, Hendriksen, & Verhagen, 2009). Considering that most
AC
consumers are not willing to compromise on sensory and hedonic aspects of products for their healthfulness (Civille & Oftedal, 2012), it is necessary to develop reformulation strategies that allow to gradually reduce the added sugar content of product without consumer awareness. The addition of non-nutritive sweeteners has been one of the most common strategies to reduce the added sugar content of a wide range of food products (Dubois & Prakash, 2012). However, non-nutritive sweeteners can convey bitter and metallic off-flavours, which can negatively influence consumer hedonic perception (Cardoso & Bolini, 2008; Markey, Lovegrove, & Methven, 2015; Zorn, Alcaire, Vidal, Giménez, &
4
ACCEPTED MANUSCRIPT Ares, 2014). In addition, several studies have suggested that the potential benefits of non-nutritive sweeteners for preventing weight gain may be hindered by compensatory energy intake mechanisms (Gardner et al., 2012; Mattes & Popkin, 2009). For these reasons, reducing the added sugar content of processed products without the addition of non-nutritive has been recommended (Yang, 2010).
PT
An alternative strategy to achieve rapid sugar reductions is to increase sweetness perception during product reformulation using multisensory integration
RI
principles (Stieger & van de Velde, 2013). Taste perception is affected by aroma
SC
compounds through cross-modal aroma-taste interactions, which indicates that the perceived intensity of a taste can be modulated by aroma (Poinot, Arvisenet,
NU
Ledauphin, Gaillard, & Prost, 2013). In particular, sweetness perception has been reported to be enhanced by the addition of aromas related to sweet products, such as
MA
vanilla, caramel or fruity notes, even at subthreshold concentrations, due to the associations formed during previous exposures with complex stimuli (Boakes &
D
Hemberger, 2012; Burseg, Camacho, Knoop, & Bult, 2010; Labbe & Martin, 2009;
PT E
Labbe, Rytz, Morgenegg, Ali, & Martin, 2007; Murphy & Cain, 1980; Tournier et al., 2009). However, cross-modal aroma-taste interactions have been reported to be aroma- and product-specific. Therefore, successful application of aroma-induced
CE
sweetness enhancement in sugar reduced products require careful screening of both
AC
aroma compounds and concentrations (Stieger & van de Velde, 2013). Apart from influencing taste perception, sugar reduction can also influence texture perception (van Raaij et al., 2009), as previously reported by Oliveira, Antúnez, Giménez, Castura, Deliza, & Ares (2015) in probiotic chocolate-flavoured milk, by Drewnowski, Nordensten, & Dwyer (1998) in cookies, and by Biguzzi, Schlich, & Lange (2014) in biscuits. Therefore, it is necessary to compensate the loss in matrix structure due to sugar reduction through the addition of bulking agents (Kaufman & Palzer, 2011). Interactions between texture and aroma perception have been reported in the literature (Poinot et al., 2013). Therefore, research on texturizers on consumer sensory
5
ACCEPTED MANUSCRIPT and hedonic perception of sugar-reduced samples is still necessary, as highlighted by Chollet, Gille, Schmid, Walther, & Piccinali (2013). In this context, the aim of the present work was to evaluate the use of crossmodal aroma-taste interactions and bulking agents for reducing the added-sugar content of vanilla milk desserts. In particular, the influence of vanilla concentration and
PT
the joint increase of vanilla and starch concentration on consumer sensory and hedonic perception was assessed. Milk desserts are dairy products widely consumed worldwide
RI
by several groups of consumers on an almost daily basis, which have a positive
SC
healthful image despite their high added sugar concentration (15-20%) (Ares,
NU
Giménez, & Gámbaro, 2008).
MA
2. Materials and methods
Two studies were carried out to evaluate the influence of an increase of vanilla
D
and a joint increase of vanilla and starch concentration on consumer sensory and
PT E
hedonic perception of sugar-reduced vanilla milk desserts. In a first study, ten milk desserts with different sugar content were formulated by increasing vanilla concentration to enhance sweetness perception and starch to compensate the texture
CE
changes caused by sugar reduction. Based on the results of the first study, a second
AC
study was conducted with different vanilla and starch concentrations.
2.1. Samples
Vanilla milk dessert samples with different sugar, vanilla and starch concentration were formulated. All samples were manufactured using the following base formulation: 12% powdered milk (Conaprole, Uruguay), 0.1% polyphosphate, 0.02% carrageenan (Ticaloid® 710H Stabilizer — Texture Innovation Center, TIC GUMS, Philadelphia, USA), 0.0025% egg yellow colouring and filtered tap water. The first study involved 10 samples, whereas 5 samples were included in the second study.
6
ACCEPTED MANUSCRIPT All samples were obtained by varying the concentration of modified starch (Purity HPC), vanilla flavour (SI 10014 – Waltary SA, Montevideo, Uruguay) and commercial sugar (Alcoholes del Uruguay S.A., Bella Unión, Uruguay), as shown in Table 1. The vanilla and starch concentrations in the first study were selected based on pilot testing.
PT
Insert Table 1 around here
RI
Samples were prepared using Termomix TM 31 (Vorwerk Mexico S. de R.L. de
SC
C.V., Mexico D.F., Mexico). All the ingredients were mixed and heated at 90 ºC for 5 minutes. After that, both vanilla flavour and colouring were added and the dispersion
NU
was mixed again with a powerful agitation (100 rpm) for 1 minute. The desserts were placed in glass containers and stored under refrigeration (4 - 7ºC) for 24 hours prior to
MA
testing. Then, desserts were served (20g) in plastic containers identified with random
D
three-digit codes at 8ºC.
PT E
2.2. Participants
A total of 200 consumers were involved in the present work, 100 in each study. Participants were recruited from the consumer database of Sensometrics & Consumer
CE
Science group (Universidad de la República, Uruguay) based on their consumption of
AC
milk desserts (at least once a month) and availability to participate in the study. They were aged between 17 and 55 and were 71% female. Participants signed an informed consent agreement and received a gift for their participation.
2.3. Experimental Procedure The experimental procedure in both studies was identical. Participants were asked to try the desserts, rate their overall liking using a 9-point hedonic scale (1= dislike very much and 9= like very much) and answer a check-all-that-apply (CATA) question to describe their sensory characteristics. The CATA question was composed
7
ACCEPTED MANUSCRIPT of 12 flavour and texture terms: sweet, very sweet, barely sweet, vanilla flavour, firm, thick, creamy, smooth, liquid, milky flavour, tasteless and gummy. Terms were selected based on previous consumer studies with the same product category (Ares, Giménez, Barreiro, & Gámbaro, 2010; Vidal et al., 2014). Presentation order of terms was balanced between and within participants, as recommended by Ares et al. (2014).
PT
In each study, samples were presented following a presentation order balanced for order and carry over effects (Williams' Latin Square design). Still mineral water was
RI
used for rinsing between samples. Testing took place in a sensory laboratory that was
temperature
control (22ºC). Data
were
SC
designed in accordance with ISO 8589 (ISO 2007), under artificial daylight and collected using Compusense
NU
(Compusense Inc., Guelph, Ontario, Canada).
Cloud
MA
2.4. Data analysis
In each study, overall liking scores were analyzed using a mixed linear model
D
considering sample as fixed effect and consumer as random effect. Tukey’s test was
PT E
used for post-hoc comparison of means. A confidence level of 95% was considered. Data from the CATA question were analyzed by determining the frequency of use of each term for describing each sample. Significant differences among samples
CE
were evaluated using Cochran’s Q Test (Manoukian, 1986). The sign test was used for
AC
pairwise comparisons.
Correspondence analysis was used to obtain a bi-dimensional representation of the samples based on the frequency of use of the terms from the CATA question. Confidence ellipses around the samples for a 95% confidence level were obtained using truncated total bootstrapping on the first two dimensions of the correspondence analysis (Cadoret & Husson, 2013). All statistical analyses were performed with R (R Core Team, 2016). FactoMineR was used to perform correspondence analysis (Lê, Josse, & Husson,
8
ACCEPTED MANUSCRIPT 2008), whereas lmerTest was used to run the mixed linear models (Kuznetsova, Brockhoff, & Christensen, 2016).
PT
3. Results
3.1. Study 1
RI
Table 2 shows the frequency of use of the sensory attributes included in the
SC
CATA question to describe the milk desserts. Significant differences among samples were found in the frequency of use of all the attributes, except for thick and smooth. In
NU
general, sugar reduction caused a decrease in the frequency of use of the terms sweet, very sweet, vanilla flavour and creamy and an increase in the frequency of use of the
MA
terms barely sweet, milk flavour and tasteless.
A reduction of 20% in added sugar caused a decrease in vanilla flavour
D
perception with respect to the control desert, as evidenced by the significant decrease
PT E
in the frequency of use of the term vanilla flavour between samples 0S4.2V0.4 and 20S4.2V0.4 (Table 2). The increase in vanilla concentration did not cause significant changes in the sensory characteristics of the dessert, although a trend towards an
CE
increase in the frequency of use of the term vanilla flavour was observed. Regarding
AC
starch, an increase in its concentration caused an increase in the frequency of use of the terms vanilla flavour and gummy, and a decrease in the frequency of use of the term tasteless and creamy. The reduction of 40% in added sugar (sample 40S4.2V0.4) caused a decrease in the frequency of use of the terms sweet, very sweet, vanilla flavour, creamy and an increase in the frequency of use of the terms barely sweet and milk flavour. When the vanilla concentration of the desserts was increased a tendency towards an increase in sweetness perception was observed, as evidenced by the non-significant difference between sample 40S4.2V0.6 and the control in the term sweet (Table 2). An increase 9
ACCEPTED MANUSCRIPT in the starch concentration of this sample led to an increase in vanilla flavour perception, as well as an increase in thickness and gumminess, as evidenced by the increase in the frequency of use of the terms vanilla flavour and gummy, and the decrease in the frequency of use of the term liquid. Samples with 60% sugar reduction showed the largest differences with respect
PT
to the control dessert, as expected. These samples were characterized by their low sweetness and vanilla flavour intensity, as well as their liquid and not creamy texture
RI
(Table 2). Although an increase in the vanilla concentration did not modify the sensory
SC
characteristics of vanilla milk desserts, the joint increase in the concentration of starch and vanilla lead to an increase in sweetness, vanilla flavour intensity, creaminess, and
NU
gumminess.
MA
Insert Table 2 around here
D
Similarities and differences among samples are summarized in Figure 1. As
PT E
shown, samples were distributed along the first dimension of the CA according to their sugar reduction. Meanwhile, the second dimension sorted samples according to their starch content, sorting those with 5.2% starch apart from the rest due to their gummy
CE
texture. Only samples with 20% sugar reduction had their confidence ellipses
AC
overlapped with the regular dessert, formulated without sugar reduction.
Insert Figure 1 around here
Regarding consumer hedonic perception of desserts, the average overall liking score of the regular dessert did not significantly differ from those of samples with 20% sugar reduction, neither from those with 40% sugar reduction and 0.6% vanilla, regardless of their starch content (Table 2). The rest of the samples had significantly lower overall liking scores. In particular, it is worth highlighting that samples with 60%
10
ACCEPTED MANUSCRIPT sugar reduction had overall liking scores in the dislike range of the 9-point hedonic scale (3.9-4.3). Results from the first study showed that none of the samples was perceived as identical to the control and that the tested starch concentration was excessive to compensate the texture changes caused by sugar reduction. These results suggested
PT
the need to evaluate higher vanilla concentrations and lower starch concentrations to obtain samples with more similar sensory characteristics to the control. In addition to
RI
these changes, a lower sugar reduction was tested in Study 2. Results are presented in
SC
the following section.
NU
3.2. Study 2
Frequency of use of the sensory attributes included in the CATA question to
MA
describe the five vanilla milk desserts is shown in Table 3. Significant differences among samples were found in the frequency of use of seven of the twelve attributes:
D
sweet, very sweet, barely sweet, tasteless, creamy, liquid and gummy.
PT E
As shown, a reduction of 10% in added sugar did not significantly modify the frequency of use of any of the terms when compared to the regular dessert. Similar results were found when sugar was reduced 20% and vanilla concentration increased
CE
to 0.8% (Table 3). Meanwhile, an increase in starch concentration of the 20% sugar
AC
reduced sample from 4.2% to 4.6% and a decrease in vanilla from 0.8% to 0.6% caused an increase in sweetness and thickness perception with respect to the control sample, as evidenced by the significant increase in the frequency of use of the term sweet and the decrease in the frequency of use of the term liquid. Regarding the sample with a reduction of 40% in added sugar, the increase in vanilla and starch concentration did not fully compensate for the decrease in sweetness and creaminess caused by sugar reduction (Table 3).
Insert Table 3 around here
11
ACCEPTED MANUSCRIPT
Figure 2 shows sample and term configurations from the CA performed on the frequency table containing consumer responses to the CATA question. Except for the sample with 40% sugar reduction, the confidence ellipses of the samples overlapped, indicating large similarity in their sensory characteristics. The sample with 40% sugar
PT
reduction was perceived as less sweet and gummier than the rest of the samples,
RI
being described using the terms barely sweet and tasteless (Figure 2).
SC
Insert Figure 2 around here
NU
Significant differences in the average overall liking scores of the samples were found. As shown in Table 3, none of the sugar reduced samples significantly differed
MA
from the regular vanilla milk dessert. However, the samples with 10% and 20% sugar reduction showed significantly higher overall liking scores than the sample with 40%
PT E
D
sugar reduction.
4. Discussion
CE
Reformulation of industrialized products has been recommended as a cost-
AC
effective strategy for reducing sugar intake (MacGregor & Hashem, 2014). One of the main challenges of this approach is related to changes in the sensory characteristics of the products, which can compromise consumer hedonic perception. A decrease in sweetness perception was the most relevant change in the sensory characteristics of samples due to sugar reduction, in agreement with previous studies in different product categories (Biguzzi et al., 2014; Drewnowski et al., 1998; Oliveira et al., 2015; Pineli et al., 2016). Sugar reduction also caused a decrease in the perception of vanilla flavour, which can be explained by aroma-taste interactions (Poinot et al., 2013). In addition, sugar reduction caused a decrease in thickness, which can be explained considering
12
ACCEPTED MANUSCRIPT the bulking properties of this ingredient (Kaufman & Palzer, 2011). Changes in the sensory characteristics of samples due to sugar reduction caused a decrease in consumer overall liking. Consumer hedonic perception was strongly related to sweetness perception, as samples that did not significantly differ in sweetness intensity did not differ in their overall liking scores (Tables 2 and 3). This
PT
result is in agreement with the fact that sweetness has been identified as a key driver of liking in several dairy products (Bayarri, Carbonell, Barrios, & Costell, 2011; Oliveira
RI
et al., 2015; Thompson, Lopetcharat, & Drake, 2007).
SC
When added sugar concentration was reduced by 10%, no relevant changes in the sensory and hedonic characteristics of the samples were detected. This suggests
NU
that this sugar reduction percentage could be implemented without the need to introduce further changes in the formulation. Other studies have reported that sugar
MA
concentration can be reduced by 6-10% without changing sweetness perception in different product categories, such as chocolate-flavoured milk, dairy-based emulsions,
D
cakes and orange nectar (Chang & Chiou, 2006; Hoppert, Zahn, Puschmann, Ullmann,
PT E
& Rohm, 2012; Oliveira et al., 2016; Pineli et al., 2016). In the present study, when sugar reduction was increased to 20%, changes in consumer sensory perception were found but not in their hedonic perception (Tables 2
CE
and 3). This suggests that the definitions of sugar reduction strategies based on
AC
consumer sensory perception are more conservative than those based on hedonic perception. Similar results have been reported by Oliveira et al. (2016) for sugar reduction in chocolate-flavour milk and Antúnez, Giménez, & Ares (2016) for salt reduction in bread. Although sugar can be gradually reduced without consumer awareness, the main disadvantage of this approach is the time needed to achieve benefits in public health (Ma, He, Yin, Hashem, & McGregor, 2016). Therefore, strategies to minimize the sensory changes caused by sugar reduction deserve further exploration. The present work assessed the use of aroma-related cross modal interactions for sugar
13
ACCEPTED MANUSCRIPT reduction in vanilla milk desserts. The effect of increasing vanilla concentration depended on the extent of sugar reduction. In the milk desserts with 40% sugar reduction, vanilla caused a slight increase in sweetness perception, whereas for samples with 20% or 60% sugar reduction differences did not reach significance. Oliveira et al. (2015) also reported that
PT
the addition of artificial vanilla at supra-threshold concentrations to chocolate-flavoured milks only increased sweetness perception of samples reduced in 40-60%. This
RI
reinforces the idea that the efficiency of aroma-taste interactions for increasing
SC
sweetness perception is matrix dependent, as previously highlighted by Poinot et al. (2013).
NU
The effect of vanilla flavour was augmented when starch concentration was increased. As shown in Tables 2 and 3, the joint increase in vanilla and starch
MA
concentration led to an increase in vanilla flavour intensity, regardless of sugar reduction percentage. In addition, in samples with 40 and 60% sugar reduction, this
D
strategy also led to an increase in sweetness perception. Different physico-chemical,
PT E
physiological and psychological mechanisms might explain this result (Poinot et al., 2013). First of all, starch interacts with some volatile compounds, leading to increased retention in the matrix or release due to a “salting out” effect caused by the swollen
CE
starch granules (Arancibia, Jublot, Costell, & Bayarri, 2011; González-Tomás, Bayarri,
AC
Taylor, & Costell, 2007). In particular, Keršiené, Adams, Dubra, De Kimpe, & Leskauskaité (2008) and González-Tomás et al. (2007) reported that an increase in starch concentration in milk desserts led to an increase in flavour release. In the present work, an enhanced release of vanilla flavour from the desserts with higher starch
concentrations
could
have
increased
vanilla
flavour
perception
and
consequently sweetness perception. Secondly, an increase in starch concentration could have enhanced flavour perception by modulating the processing of taste and aroma signals due to the increase in viscosity and/or changes in salivation (Poinot et al., 2013). Finally, texture and aroma interactions could be explained by learned
14
ACCEPTED MANUSCRIPT associations through repeated exposure to similar products (Bult, de Wijk, & Hummel, 2007; Saint-Eve, Lauverjat, Magnan, Déléris, & Souchon, 2009). Further research is needed to better understand the mechanisms that underlie the increase of sweetness and vanilla flavour perception caused by an increase in starch concentration. Cross-modal interactions are strongly dependent on the characteristics of the
PT
food matrix, the volatile compounds and thickening agents (Poinot et al., 2013). Therefore, minor differences in product formulation could largely modify their efficiency
RI
for minimizing the influence of sugar reduction on the sensory characteristics of
SC
products. In addition, a need still exists to deepen our understanding on how sensations interact throughout food consumption to shape our sensory and hedonic
NU
perception of products. Research on the topic is warranted and could largely contribute with recommendations for the development of more healthful food products.
MA
In closing this section it should be acknowledged that one of the limitations of the present study is that the interaction effect between vanilla flavour and starch
D
concentration could not be evaluated. Further research should use a full factorial
PT E
experimental design to study the separate effect of these variables, as well as their interaction, on consumers’ sensory and hedonic perception of vanilla milk desserts with
AC
CE
different sugar content.
5. Conclusions
Results from the present work show the potential of aroma-related cross-modal interactions to increase sweetness perception in the development of sugar-reduced products and, consequently, minimize changes in consumer hedonic perception. This strategy was able to totally and/or partially mitigate the effect of a 20% reduction in added sugar in the sensory characteristics of the desserts, without the need to largely modify the formulation or to add non-nutritive sweeteners. However, further studies are necessary to assess their potential for sugar reduction in different product categories.
15
ACCEPTED MANUSCRIPT
Acknowledgments The authors are indebted to Agencia Nacional de Investigación e Innovación (ANII, grant number
INI_X_2013_1_101098),
and
Comisión
Sectorial
de
Investigación
Científica
RI
PT
(Universidad de la República) for financial support.
SC
References
Antúnez, L., Giménez, A., & Ares, G. (2016). A consumer-based approach to salt
NU
reduction: Case study with bread. Food Research International, 90, 66-72. Arancibia, C., Jublot, L., Costell, E., & Bayarri, S. (2011). Flavor release and sensory
MA
characteristics of o/w emulsions. Influence of composition, microstructure and rheological behavior. Food Research International, 44, 1632–1641.
D
Ares, G., Etchemendy, R., Atúnez, L., Vidal, L., Giménez, A., & Jaeger, S. (2014).
PT E
Visual attention by consumers to check-all-that-apply questions: Insights to support methodological development. Food Quality and Preference, 32, 210-220. Ares, G., Giménez, A., & Gámbaro, A. (2008). Does information about the source of
CE
functional ingredients influence consumer perception of functional milk desserts?
AC
Journal of the Science of Food and Agriculture, 88, 2061-2068. Ares, G., Giménez, A., Barreiro, C. & Gámbaro, A. (2010). Use of an open-ended question to identify drivers of liking of milk desserts. Comparison with preference mapping techniques. Food Quality and Preference, 21, 286–294. Bayarri, S., Carbonell, I., Barrios, E.X., & Costell, E. (2011). Impact of sensory differences on consumer acceptability of yoghurt and yoghurt-like products. International Dairy Journal, 21, 111–118. Biguzzi, C., Schlich, P., & Lange, C. (2014). The impact of sugar and fat reduction on perception and liking of biscuits. Food Quality and Preference, 35, 41-47.
16
ACCEPTED MANUSCRIPT Boakes, R.A., & Hemberger, H. (2012). Odour-modulation of taste ratings by chefs. Food Quality and Preference, 25, 81-86. Bult, J. H. F., de Wijk, R. A., & Hummel, T. (2007). Investigations on multimodal sensory integration: Texture, taste, and ortho- and retronasal olfactory stimuli in concert. Neuroscience Letters, 411, 6–10.
PT
Burseg, K.M.M., Camacho, S., Knoop, J., & Bult, J.H.F. (2010). Sweet taste intensity is enhanced by temporal fluctuation of aroma and taste, and depends on phase shift.
RI
Physiology and Behavior, 101, 726–730.
SC
Cadoret, M., & Husson, F. (2013). Construction and evaluation of confidence ellipses applied at sensory data. Food Quality and Preference, 28, 106–115.
NU
Cardoso, J. M. P., & Bolini, H. M. A. (2008). Descriptive profile of peach nectar sweetened with sucrose and different sweeteners. Journal of Sensory Studies, 23,
MA
804–816.
Chang, M. H., & Chiou, W. B. (2006). Differential threshold and psychophysical power
D
function of sweetness sensation: Applied psychophysics and prospect theory on
PT E
reformulating baking products. Journal of Sensory Studies, 21, 534–551. Chollet, M., Gille, D., Schmid, A., Walther, B. & Piccinali, P. (2013). Acceptance of sugar reduction in flavored yogurt. Journal Dairy Science, 96, 5501-5511.
CE
Civille, C.V., & Oftedal, K.N. (2012). Sensory evaluation techniques — make “good for
AC
you” taste “good”. Physiology & Behavior, 107, 598–605. Drewnowski, A., Nordensten, K., & Dwyer, J. (1998). Replacing sugar and fat in cookies: Impact on product quality and preference. Food Quality and Preference, 9, 13-20. Dubois, G. E., & Prakash, I. (2012). Non-caloric sweeteners, sweetness modulators, and sweetener enhancers. Annual Reviews of Food Science and Technology, 3, 353–380. Gardner, C., Wylie-Rosett, J., Gidding, S. S., Steffen, L. M., Johnson, R. K., Reader, D., et al. (2012). Nonnutritive sweeteners: current use and health perspectives: a
17
ACCEPTED MANUSCRIPT scientific statement from the American Heart Association and the American Diabetes Association. Diabetes Care, 35(8), 1798–1808. González-Tomás, L., Bayarri, S., Taylor, A. J., & Costell, E. (2007). Flavour release and perception from model dairy custards. Food Research International, 40, 520– 528.
PT
Hoppert, K., Zahn, S., Puschmann, A., Ullmann, I., & Rohm, H. (2012). Quantification of sensory difference thresholds for fat and sweetness in dairy-based emulsions.
RI
Food Quality and Preference, 26, 52–57.
Standard
8589.
International
SC
ISO (2007). Sensory analysis: General guidance for the design of test rooms, ISO Organization
Standardization,
Geneva,
NU
Switzerland.
for
Johnson, R.K., Appel, L.J., Brands, M., Howard, B.V., Lefevre, M., Lustig, R.H., et al.
MA
(2009). Dietary sugars intake and cardiovascular health: a scientific statement from the American Heart Association. Circulation, 120, 1011–20.
D
Kaufmann, S.F.M., & Palzer, S. (2011). Food structure engineering for nutrition, health
PT E
and wellness. Procedia Food Science, 1, 1479-1486. Keršiené, M., Adams, A., Dubra, A., De Kimpe, N., & Leskauskaité, D. (2008). Interactions between flavour release and rheological properties in model custard
AC
1191.
CE
desserts: Effect of starch concentration and milk fat. Food Chemistry, 108, 1183–
Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. (2016). lmerTest: Tests in Linear Mixed Effects Models. R package version 2.0-32. Available at: https://CRAN.R-project.org/package=lmerTest. Labbe, D., & Martin, N. (2009). Impact of novel olfactory stimuli at supra and subthreshold concentrations on the perceived sweetness of sucrose after associative learning. Chemical Senses, 34, 645–651. Labbe, D., Rytz, A., Morgenegg, C., Ali, S., & Martin, N. (2007). Subthreshold olfactory stimulation can enhance sweetness. Chemical Senses, 34, 645-651.
18
ACCEPTED MANUSCRIPT Lê, S., Josse, J., & Husson, F. (2008). FactoMineR: An R package for multivariate analysis. Journal of Statistical Software, 25(1), 1–18. Lustig, R.H., Schmidt, L.A., & Brindis, C.D. (2012). The toxic truth about sugar. Nature, 482, 27–29. Ma, Y., He, F.J., Yin, Y., Hashem, K.M., & McGregor, G.A. (2016). Gradual reduction of
PT
sugar in soft drinks without substitution as a strategy to reduce overweight, obesity, and type 2 diabetes: a modelling study. The Lancet Diabetes &
RI
Endocrinology, 4, 105-114.
reduction programme. Lancet, 383, 929–931.
SC
MacGregor, G.A., & Hashem, K.M. (2014). Action on sugar—lessons from UK salt
NU
Manoukian, E. B. (1986). Mathematical nonparametric statistics. New York, NY: Gordon & Breach.
MA
Markey, O, Lovegrove, J.A., & Methven, L. (2015). Sensory profiles and consumer acceptability of a range of sugar-reduced products on the UK market. Food
D
Research International, 72, 133-139.
PT E
Mattes, R. D., & Popkin, B. M. (2009). Nonnutritive sweetener consumption in humans: effects on appetite and food intake and their putative mechanisms. American Journal of Clinical Nutrition, 89(1), 1–14.
CE
Moynihan, P.J., & Kelly, S.A. (2014). Effect on caries of restricting sugars intake
18.
AC
systematic review to inform WHO guidelines. Journal of Dental Research, 93, 8–
Murphy, C., & Cain, W. S. (1980). Taste and olfaction: Independence vs. interaction. Physiology and Behavior, 24, 601–605 Oliveira, D., Antúnez, L., Giménez, A., Castura, J., Deliza, R., & Ares, G. (2015). Sugar reduction in probiotic chocolate-flavored milk: Impact on dynamic sensory profile and liking. Food Research International, 75, 148-156. Oliveira, D., Reis, F., Deliza, R., Rosenthal, A., Giménez, A., Ares, G. (2016). Difference
thresholds
for
added
sugar
in
chocolate-flavoured
milk:
19
ACCEPTED MANUSCRIPT Recommendations for gradual sugar reduction. Food Research International, 89, 448–453. Pineli, L.L.O., de Aguiar, L.A., Fiusa, A., Botelho, R.B.A., Zandonadi, R.P., & Melo, L. (2016). Sensory impact of lowering sugar content in orange nectars to design healthier, low-sugar industrialized beverages. Appetite, 96, 239-244.
PT
Poinot, P., Arvisenet, G., Ledauphin, J., Gaillard, J. L., & Prost, C. (2013). How can aroma-related cross-modal interactions be analysed? A review of current
RI
methodologies. Food Quality and Preference, 28, 304-316.
SC
R Core Team. (2016). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
NU
Saint-Eve, A., Lauverjat, C., Magnan, C., Déléris, I., & Souchon, I. (2009). Reducing salt and fat content: Impact of composition, texture and cognitive interactions on
MA
the perception of flavoured model cheeses. Food Chemistry, 116, 167–175. Scientific Advisory Committee on Nutrition (2014). Draft Carbohydrates and Health
D
report. Scientific consultation: 26 June to 1 September 2014. Available at: http://
PT E
www.sacn.gov.uk/reports_position_statements/reports/scientific_consultation_ draft_sacn_carbohydrates_and_health_report_-_june_2014.html (Accessed 10th August 2016)
CE
Stieger, M., & van de Velde, F. (2013). Microstructure, texture and oral processing:
AC
New ways to reduce sugar and salt in foods. Current Opinion in Colloid & Interface Science, 18, 334-348. Te Morenga, L., Mallard, S., & Mann, J. (2013) Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies. BMJ, 346: e7492. Thompson, J.L., Lopetcharat, K. & Drake, M.A. (2007). Preferences for commercial strawberry drinkable yogurts among African American, Caucasian, and Hispanic consumers in the United States. Journal of Dairy Science, 90,4974–4987.
20
ACCEPTED MANUSCRIPT Tournier, C., Sulmont-Rossé, C., Sémon, E., Vignon, A., Issanchou, S., & Guichard, E. (2009). A study on texture–taste–aroma interactions: Physico-chemical and cognitive mechanisms. International Dairy Journal, 19, 450-458. van Raaij, J., Hendriksen, M., & Verhagen, H. (2009). Potential for improvement of population diet through reformulation of commonly eaten foods. Public Health
PT
Nutrition, 12(3), 325–330. Vidal, L., Cadena, R.S., Correa, S., Ábalos, R.A., Gómez, B., Giménez, A. et al. (2014).
SC
consumers. Journal of Sensory Studies, 29, 74-87.
RI
Assessment of global and individual reproducibility of projective mapping with
Vio, F., & Uauy, R. (2007). Case Study #9-5. The Sugar Controversy. In: P. Pinstrup-
Ithaca, NY: Cornell University Press.
NU
Andersen & F. Cheng (Eds.), Food Policy for Developing Countries: Case Studies.
MA
World Health Organization and Food & Agriculture Organization of the United Nations. (2003). WHO/FAO joint expert consultation on diet, nutrition, and the prevention of
D
chronic diseases. WHO Technical Report Series No. 916. Geneva: World Health
PT E
Organization.
Yang, Q. (2010). Gain weight by “going diet?” Artificial sweeteners and the neurobiology of sugar cravings: Neuroscience 2010. Yale Journal of Biology and
CE
Medicine, 83(2), 101-108
AC
Zorn, S., Alcaire, F., Vidal, L., Giménez, A., & Ares, G. (2014). Application of multiplesip temporal dominance of sensations to the evaluation of sweeteners. Food Quality and Preference, 36, 135-143.
21
ACCEPTED MANUSCRIPT Figure captions
Figure 1. Sample and term configurations of vanilla milk dessert samples with different sugar, vanilla and starch concentrations, obtained from consumer responses to a
PT
check-all-that-apply question in Study 1.
Figure 2. Sample and term configurations of vanilla milk dessert samples with different
RI
sugar, vanilla and starch concentrations, obtained from consumer responses to a
AC
CE
PT E
D
MA
NU
SC
check-all-that-apply question, in Study 2.
22
ACCEPTED MANUSCRIPT Tables
Table 1. Formulation of the vanilla milk desserts included in the two studies.
2
Sugar reduction (%)
Starch (%)
Vanilla (%)
0S4.2V0.4 20S4.2V0.4 20S4.2V0.6 20S5.2V0.6 40S4.2V0.4 40S4.2V0.6 40S5.2V0.6 60S4.2V0.4 60S4.2V0.6 60S5.2V0.6 0S4.2V0.4 10S4.2V0.4 20S4.2V0.8 20S4.6V0.6 40S5.0V0.8
20 16 16 16 12 12 12 8.0 8.0 8.0 20 18 16 16 12
0 20 20 20 40 40 40 60 60 60 0 10 20 20 40
4.2 4.2 4.2 5.2 4.2 4.2 5.2 4.2 4.2 5.2 4.2 4.2 4.2 4.6 5.0
0.4 0.4 0.6 0.6 0.4 0.6 0.6 0.4 0.6 0.6 0.4 0.4 0.8 0.6 0.8
SC
RI
PT
Sugar (%)
NU
1
Sample
MA
Study
Coding of the samples indicates sugar reduction percentage (0, 10, 20, 40 or 60), followed by the starch
AC
CE
PT E
D
concentration (S) and vanilla concentration (V).
23
ACCEPTED MANUSCRIPT
Table 2. Average overall liking score and frequency of use of terms of the CATA question for describing the vanilla milk desserts in Study 1.
0S4.2V0.4
Overall liking a 6.4
20S4.2V0.4
6.1
a
20S4.2V0.6
6.1
a
20S5.2V0.6
5.7
40S4.2V0.4
5.1
Sample
a,b b,c
Sweet*** 62
Barely sweet*** d 4
Vanilla flavour*** a 63
43
a,b
6
c,d
44
b
28
a,b,c
42
a,b
8
c,d
51
b
33
c
71
a
33
c
34
c
59
a,b
40
c
13
30
b,c
24
b
32
b,c
24
b
c
25
b
40S4.2V0.6
5.7
40S5.2V0.6
6.0
a,b
64
a
25
60S4.2V0.4
4.0
d
28
d
10
d
50
a
16
d
60S4.2V0.6
3.9
d
23
d
12
d
53
a
16
d
45
b
10
d
62
a
60S5.2V0.6
4.3
c,d
a,b,c
a,b,c
35
a,b
51
Milk flavour*** f 18
a,b
58 51
a
Very sweet*** a 47
58
35
11
b,c,d
21
39
a,b
17
46
a
c,d
12
c
42
a
a,b
42
a
b,c
a,b,c
37
b
20
c,d,e,f
29
b,c
d,e,f
25
b,c
a
53
a
a,b
56
a
66
a
b,c,d,e
D E
43
a
c
27
44
b,c
14
e
22
41
a
a,b,c
34
a
a,b,c
46
a
a,b
37
a
a
31
a
40
a
A M
27
33
12
c
Creamy***
Smooth
Liquid***
Gummy***
65 61
a,b
29
a
19
a,b
10
a
29
a
20
a,b
9
b,c,d
44
a
14
c,d
68
53
33
a
ns
a,b
T P
I R
SC
U N 22
ns
Thick
d,e
10
53 31
Firm*
e,f
22
c
7
e
c,d,e,f
20 28
Tasteless***
20
a,b
13
b,c b,c
b,c,d
b,c
31
a
31
a
22
a,b
7
c,d
56
a,b,c
33
a
19
a,b
8
c,d
59
a,b,c
43
a
8
f
26
a
26
a
e,f
33
a
27
a
45
25 29 39
d,e
32
a,b
c,d
2
d
18 1 3
b
e
d,e
14
b,c
Coding of the samples indicates sugar reduction percentage (0, 20, 40 or 60), followed by the starch concentration (S) and vanilla concentration (V).
T P E
Overall liking scores with different letters are significantly different according to Tukey’s test (p<0.05). Results of Cochran’s Q test are shown next to each attribute: * indicates significant differences among samples at p<0.05, ** significant differences at p<0.01, *** significant differences at p<0.001, whereas ns indicate no significant differences (p>0.05). Values of frequency of use of CATA terms with different letters within a column are significantly
C C
different according to the sign test (p<0.05).
A
24
ACCEPTED MANUSCRIPT
Table 3. Average overall liking scores and frequency of use of terms of the CATA question for describing the vanilla milk desserts in Study 2. Sample 0S4.2V0.4
Overall Very Barely Vanilla Milk ns ns ns Sweet* Thick Creamy*** Smooth Liquid*** Gummy** ns ns Tasteless*** Firm liking sweet*** sweet*** flavour flavour a,b b,c a b a a b,c a a a a a b,c 6.2 45 40 9 48 19 14 13 44 63 35 29 17
10S4.2V0.4
6.5
a
56
a,b
38
a
9
b
40
a
25
a
12
20S4.2V0.8
6.6
a
56
a,b
39
a
9
b
43
a
19ª
8
7.0
a
43
a
7
b
52
a
15ª
5.6
b
17
b
38
a
a
20S4.6V0.6 40S5.0V0.8
60
a
40
c
35
a
18
10
a
39
a
62
c
20
a
42
a
14
a
39ª
a
a
I R
18
b
52
a
C S U 17
42
a
T P
b,c
74
a
64
a
44
b
30
a
20
a,b
27
a
20
a,b
35
a
14
24
a
c
5
b
15
b,c
10 21
c
a,b
30
a
Coding of the samples indicates sugar reduction percentage (0, 20 or 40), followed by the starch concentration (S) and vanilla concentration (V).
N A
Overall liking scores with different letters are significantly different according to Tukey’s test (p<0.05).
Results of Cochran’s Q test are shown next to each attribute: * indicates significant differences among samples at p<0.05, ** significant differences at p<0.01, *** significant
M
differences at p<0.001, whereas ns indicates no significant differences (p>0.05). Values of frequency of use of CATA terms with different letters within a column are significantly
D E
different according to the sign test (p<0.05).
T P E
C C
A
25
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
Figure 1.
26
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
Figure 2.
27
Florencia Alcaire, Lucía Antúnez, Leticia Vidal, Ana Giménez, Gastón Ares PII: DOI: Reference:
S0963-9969(17)30081-9 doi: 10.1016/j.foodres.2017.02.019 FRIN 6602
To appear in:
Food Research International
Received date: Revised date: Accepted date:
18 November 2016 24 February 2017 26 February 2017
Please cite this article as: Florencia Alcaire, Lucía Antúnez, Leticia Vidal, Ana Giménez, Gastón Ares , Aroma-related cross-modal interactions for sugar reduction in milk desserts: Influence on consumer perception. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Frin(2017), doi: 10.1016/ j.foodres.2017.02.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Manuscript for submission to Food Research International
Aroma-related cross-modal interactions for sugar reduction in milk desserts:
PT
Influence on consumer perception
RI
Florencia Alcaire*, Lucía Antúnez, Leticia Vidal, Ana Giménez, Gastón Ares
SC
Sensometrics & Consumer Science, Instituto Polo Tecnológico de Pando, Facultad de Química, Universidad de la República. By Pass de Rutas 8 y 101 s/n. C.P. 91000.
MA
NU
Pando, Canelones, Uruguay.
AC
CE
PT E
D
*Corresponding author: Florencia Alcaire [Email: [email protected]]
1
ACCEPTED MANUSCRIPT Research highlights Aroma-related cross modal interactions for sugar reduction were assessed. A 10% sugar reduction did not cause changes in sensory and hedonic perception Vanilla increased sweetness and vanilla flavour in 40% sugar-reduced samples.
PT
The joint increase in vanilla and starch concentration provided better results.
AC
CE
PT E
D
MA
NU
SC
RI
Cross-modal interactions hold potential for product reformulation.
2
ACCEPTED MANUSCRIPT Abstract Reformulation of industrialized products has been regarded as one of the most cost-effective strategies to reduce sugar intake. Although non-nutritive sweeteners have been extensively used to reduce the added sugar content of these products, increasing evidence about the existence of compensatory energy intake mechanisms
PT
makes it necessary to develop alternative strategies to achieve rapid sugar reductions. In this context, the aim of the present work was to evaluate aroma-related cross modal
RI
interactions for sugar reduction in vanilla milk desserts. In particular, the influence of
SC
increasing vanilla concentration and the joint increase of vanilla and starch concentration on consumer sensory and hedonic perception was assessed. Two
NU
studies with 100 consumers each were conducted, in which a total of 15 samples were evaluated. For each sample, consumers rated their overall liking and answered a
MA
check-all-that-apply (CATA) question comprising 12 flavour and texture terms. Sugar reduction caused significant changes in the flavour and texture characteristics of the
D
desserts. An increase in vanilla concentration had a minor effect on their sensory
PT E
characteristics. However, increasing both vanilla and starch concentration led to an increase in vanilla flavour and sweetness perception and reduced changes in consumer hedonic perception. These results showed the potential of aroma-related
CE
cross modal interactions for minimizing the sensory changes caused by sugar
AC
reduction. These strategies could contribute to product reformulation without the need to use non-nutritive sweeteners.
Keywords: sugar reduction; product reformulation; cross-modal interactions; CATA questions; consumer studies
3
ACCEPTED MANUSCRIPT 1. Introduction Excessive sugar consumption has become a significant health problem worldwide as it has linked to the increasing prevalence of obesity (Vio & Uauy, 2007). Added sugar consumption contributes to 500 calories per day, corresponding to 1725% of the recommended calorie intake (Lustig, Schmidt, & Brindis, 2012). In addition,
PT
excessive sugar intake has been linked to several metabolic abnormalities and adverse health problems, such as type 2 diabetes, cardiovascular diseases and dental caries
RI
(Johnson et al., 2009; Moynihan, & Kelly, 2014; Te Morenga, Mallard, & Mann, 2013).
SC
For this reason, it has been recommended to restrict the intake of sugars to less than 5-10% of the total daily energy consumption (World Health Organization & Agriculture
NU
Organization of the United Nations, 2003; Scientific Advisory Committee on Nutrition, 2014). This situation makes it necessary to develop strategies to reduce sugar
MA
consumption worldwide.
Sugar is added to a large proportion of processed foods (Vio & Uauy, 2007).
D
Therefore, one of the possible strategies for rapidly decreasing sugar consumption of
PT E
the population is to reduce the added sugar content of these products (MacGregor & Hashem, 2014). One of the main challenges faced in product reformulation is related to changes in product flavour and texture, which are determinant of product success in
CE
the marketplace (van Raaij, Hendriksen, & Verhagen, 2009). Considering that most
AC
consumers are not willing to compromise on sensory and hedonic aspects of products for their healthfulness (Civille & Oftedal, 2012), it is necessary to develop reformulation strategies that allow to gradually reduce the added sugar content of product without consumer awareness. The addition of non-nutritive sweeteners has been one of the most common strategies to reduce the added sugar content of a wide range of food products (Dubois & Prakash, 2012). However, non-nutritive sweeteners can convey bitter and metallic off-flavours, which can negatively influence consumer hedonic perception (Cardoso & Bolini, 2008; Markey, Lovegrove, & Methven, 2015; Zorn, Alcaire, Vidal, Giménez, &
4
ACCEPTED MANUSCRIPT Ares, 2014). In addition, several studies have suggested that the potential benefits of non-nutritive sweeteners for preventing weight gain may be hindered by compensatory energy intake mechanisms (Gardner et al., 2012; Mattes & Popkin, 2009). For these reasons, reducing the added sugar content of processed products without the addition of non-nutritive has been recommended (Yang, 2010).
PT
An alternative strategy to achieve rapid sugar reductions is to increase sweetness perception during product reformulation using multisensory integration
RI
principles (Stieger & van de Velde, 2013). Taste perception is affected by aroma
SC
compounds through cross-modal aroma-taste interactions, which indicates that the perceived intensity of a taste can be modulated by aroma (Poinot, Arvisenet,
NU
Ledauphin, Gaillard, & Prost, 2013). In particular, sweetness perception has been reported to be enhanced by the addition of aromas related to sweet products, such as
MA
vanilla, caramel or fruity notes, even at subthreshold concentrations, due to the associations formed during previous exposures with complex stimuli (Boakes &
D
Hemberger, 2012; Burseg, Camacho, Knoop, & Bult, 2010; Labbe & Martin, 2009;
PT E
Labbe, Rytz, Morgenegg, Ali, & Martin, 2007; Murphy & Cain, 1980; Tournier et al., 2009). However, cross-modal aroma-taste interactions have been reported to be aroma- and product-specific. Therefore, successful application of aroma-induced
CE
sweetness enhancement in sugar reduced products require careful screening of both
AC
aroma compounds and concentrations (Stieger & van de Velde, 2013). Apart from influencing taste perception, sugar reduction can also influence texture perception (van Raaij et al., 2009), as previously reported by Oliveira, Antúnez, Giménez, Castura, Deliza, & Ares (2015) in probiotic chocolate-flavoured milk, by Drewnowski, Nordensten, & Dwyer (1998) in cookies, and by Biguzzi, Schlich, & Lange (2014) in biscuits. Therefore, it is necessary to compensate the loss in matrix structure due to sugar reduction through the addition of bulking agents (Kaufman & Palzer, 2011). Interactions between texture and aroma perception have been reported in the literature (Poinot et al., 2013). Therefore, research on texturizers on consumer sensory
5
ACCEPTED MANUSCRIPT and hedonic perception of sugar-reduced samples is still necessary, as highlighted by Chollet, Gille, Schmid, Walther, & Piccinali (2013). In this context, the aim of the present work was to evaluate the use of crossmodal aroma-taste interactions and bulking agents for reducing the added-sugar content of vanilla milk desserts. In particular, the influence of vanilla concentration and
PT
the joint increase of vanilla and starch concentration on consumer sensory and hedonic perception was assessed. Milk desserts are dairy products widely consumed worldwide
RI
by several groups of consumers on an almost daily basis, which have a positive
SC
healthful image despite their high added sugar concentration (15-20%) (Ares,
NU
Giménez, & Gámbaro, 2008).
MA
2. Materials and methods
Two studies were carried out to evaluate the influence of an increase of vanilla
D
and a joint increase of vanilla and starch concentration on consumer sensory and
PT E
hedonic perception of sugar-reduced vanilla milk desserts. In a first study, ten milk desserts with different sugar content were formulated by increasing vanilla concentration to enhance sweetness perception and starch to compensate the texture
CE
changes caused by sugar reduction. Based on the results of the first study, a second
AC
study was conducted with different vanilla and starch concentrations.
2.1. Samples
Vanilla milk dessert samples with different sugar, vanilla and starch concentration were formulated. All samples were manufactured using the following base formulation: 12% powdered milk (Conaprole, Uruguay), 0.1% polyphosphate, 0.02% carrageenan (Ticaloid® 710H Stabilizer — Texture Innovation Center, TIC GUMS, Philadelphia, USA), 0.0025% egg yellow colouring and filtered tap water. The first study involved 10 samples, whereas 5 samples were included in the second study.
6
ACCEPTED MANUSCRIPT All samples were obtained by varying the concentration of modified starch (Purity HPC), vanilla flavour (SI 10014 – Waltary SA, Montevideo, Uruguay) and commercial sugar (Alcoholes del Uruguay S.A., Bella Unión, Uruguay), as shown in Table 1. The vanilla and starch concentrations in the first study were selected based on pilot testing.
PT
Insert Table 1 around here
RI
Samples were prepared using Termomix TM 31 (Vorwerk Mexico S. de R.L. de
SC
C.V., Mexico D.F., Mexico). All the ingredients were mixed and heated at 90 ºC for 5 minutes. After that, both vanilla flavour and colouring were added and the dispersion
NU
was mixed again with a powerful agitation (100 rpm) for 1 minute. The desserts were placed in glass containers and stored under refrigeration (4 - 7ºC) for 24 hours prior to
MA
testing. Then, desserts were served (20g) in plastic containers identified with random
D
three-digit codes at 8ºC.
PT E
2.2. Participants
A total of 200 consumers were involved in the present work, 100 in each study. Participants were recruited from the consumer database of Sensometrics & Consumer
CE
Science group (Universidad de la República, Uruguay) based on their consumption of
AC
milk desserts (at least once a month) and availability to participate in the study. They were aged between 17 and 55 and were 71% female. Participants signed an informed consent agreement and received a gift for their participation.
2.3. Experimental Procedure The experimental procedure in both studies was identical. Participants were asked to try the desserts, rate their overall liking using a 9-point hedonic scale (1= dislike very much and 9= like very much) and answer a check-all-that-apply (CATA) question to describe their sensory characteristics. The CATA question was composed
7
ACCEPTED MANUSCRIPT of 12 flavour and texture terms: sweet, very sweet, barely sweet, vanilla flavour, firm, thick, creamy, smooth, liquid, milky flavour, tasteless and gummy. Terms were selected based on previous consumer studies with the same product category (Ares, Giménez, Barreiro, & Gámbaro, 2010; Vidal et al., 2014). Presentation order of terms was balanced between and within participants, as recommended by Ares et al. (2014).
PT
In each study, samples were presented following a presentation order balanced for order and carry over effects (Williams' Latin Square design). Still mineral water was
RI
used for rinsing between samples. Testing took place in a sensory laboratory that was
temperature
control (22ºC). Data
were
SC
designed in accordance with ISO 8589 (ISO 2007), under artificial daylight and collected using Compusense
NU
(Compusense Inc., Guelph, Ontario, Canada).
Cloud
MA
2.4. Data analysis
In each study, overall liking scores were analyzed using a mixed linear model
D
considering sample as fixed effect and consumer as random effect. Tukey’s test was
PT E
used for post-hoc comparison of means. A confidence level of 95% was considered. Data from the CATA question were analyzed by determining the frequency of use of each term for describing each sample. Significant differences among samples
CE
were evaluated using Cochran’s Q Test (Manoukian, 1986). The sign test was used for
AC
pairwise comparisons.
Correspondence analysis was used to obtain a bi-dimensional representation of the samples based on the frequency of use of the terms from the CATA question. Confidence ellipses around the samples for a 95% confidence level were obtained using truncated total bootstrapping on the first two dimensions of the correspondence analysis (Cadoret & Husson, 2013). All statistical analyses were performed with R (R Core Team, 2016). FactoMineR was used to perform correspondence analysis (Lê, Josse, & Husson,
8
ACCEPTED MANUSCRIPT 2008), whereas lmerTest was used to run the mixed linear models (Kuznetsova, Brockhoff, & Christensen, 2016).
PT
3. Results
3.1. Study 1
RI
Table 2 shows the frequency of use of the sensory attributes included in the
SC
CATA question to describe the milk desserts. Significant differences among samples were found in the frequency of use of all the attributes, except for thick and smooth. In
NU
general, sugar reduction caused a decrease in the frequency of use of the terms sweet, very sweet, vanilla flavour and creamy and an increase in the frequency of use of the
MA
terms barely sweet, milk flavour and tasteless.
A reduction of 20% in added sugar caused a decrease in vanilla flavour
D
perception with respect to the control desert, as evidenced by the significant decrease
PT E
in the frequency of use of the term vanilla flavour between samples 0S4.2V0.4 and 20S4.2V0.4 (Table 2). The increase in vanilla concentration did not cause significant changes in the sensory characteristics of the dessert, although a trend towards an
CE
increase in the frequency of use of the term vanilla flavour was observed. Regarding
AC
starch, an increase in its concentration caused an increase in the frequency of use of the terms vanilla flavour and gummy, and a decrease in the frequency of use of the term tasteless and creamy. The reduction of 40% in added sugar (sample 40S4.2V0.4) caused a decrease in the frequency of use of the terms sweet, very sweet, vanilla flavour, creamy and an increase in the frequency of use of the terms barely sweet and milk flavour. When the vanilla concentration of the desserts was increased a tendency towards an increase in sweetness perception was observed, as evidenced by the non-significant difference between sample 40S4.2V0.6 and the control in the term sweet (Table 2). An increase 9
ACCEPTED MANUSCRIPT in the starch concentration of this sample led to an increase in vanilla flavour perception, as well as an increase in thickness and gumminess, as evidenced by the increase in the frequency of use of the terms vanilla flavour and gummy, and the decrease in the frequency of use of the term liquid. Samples with 60% sugar reduction showed the largest differences with respect
PT
to the control dessert, as expected. These samples were characterized by their low sweetness and vanilla flavour intensity, as well as their liquid and not creamy texture
RI
(Table 2). Although an increase in the vanilla concentration did not modify the sensory
SC
characteristics of vanilla milk desserts, the joint increase in the concentration of starch and vanilla lead to an increase in sweetness, vanilla flavour intensity, creaminess, and
NU
gumminess.
MA
Insert Table 2 around here
D
Similarities and differences among samples are summarized in Figure 1. As
PT E
shown, samples were distributed along the first dimension of the CA according to their sugar reduction. Meanwhile, the second dimension sorted samples according to their starch content, sorting those with 5.2% starch apart from the rest due to their gummy
CE
texture. Only samples with 20% sugar reduction had their confidence ellipses
AC
overlapped with the regular dessert, formulated without sugar reduction.
Insert Figure 1 around here
Regarding consumer hedonic perception of desserts, the average overall liking score of the regular dessert did not significantly differ from those of samples with 20% sugar reduction, neither from those with 40% sugar reduction and 0.6% vanilla, regardless of their starch content (Table 2). The rest of the samples had significantly lower overall liking scores. In particular, it is worth highlighting that samples with 60%
10
ACCEPTED MANUSCRIPT sugar reduction had overall liking scores in the dislike range of the 9-point hedonic scale (3.9-4.3). Results from the first study showed that none of the samples was perceived as identical to the control and that the tested starch concentration was excessive to compensate the texture changes caused by sugar reduction. These results suggested
PT
the need to evaluate higher vanilla concentrations and lower starch concentrations to obtain samples with more similar sensory characteristics to the control. In addition to
RI
these changes, a lower sugar reduction was tested in Study 2. Results are presented in
SC
the following section.
NU
3.2. Study 2
Frequency of use of the sensory attributes included in the CATA question to
MA
describe the five vanilla milk desserts is shown in Table 3. Significant differences among samples were found in the frequency of use of seven of the twelve attributes:
D
sweet, very sweet, barely sweet, tasteless, creamy, liquid and gummy.
PT E
As shown, a reduction of 10% in added sugar did not significantly modify the frequency of use of any of the terms when compared to the regular dessert. Similar results were found when sugar was reduced 20% and vanilla concentration increased
CE
to 0.8% (Table 3). Meanwhile, an increase in starch concentration of the 20% sugar
AC
reduced sample from 4.2% to 4.6% and a decrease in vanilla from 0.8% to 0.6% caused an increase in sweetness and thickness perception with respect to the control sample, as evidenced by the significant increase in the frequency of use of the term sweet and the decrease in the frequency of use of the term liquid. Regarding the sample with a reduction of 40% in added sugar, the increase in vanilla and starch concentration did not fully compensate for the decrease in sweetness and creaminess caused by sugar reduction (Table 3).
Insert Table 3 around here
11
ACCEPTED MANUSCRIPT
Figure 2 shows sample and term configurations from the CA performed on the frequency table containing consumer responses to the CATA question. Except for the sample with 40% sugar reduction, the confidence ellipses of the samples overlapped, indicating large similarity in their sensory characteristics. The sample with 40% sugar
PT
reduction was perceived as less sweet and gummier than the rest of the samples,
RI
being described using the terms barely sweet and tasteless (Figure 2).
SC
Insert Figure 2 around here
NU
Significant differences in the average overall liking scores of the samples were found. As shown in Table 3, none of the sugar reduced samples significantly differed
MA
from the regular vanilla milk dessert. However, the samples with 10% and 20% sugar reduction showed significantly higher overall liking scores than the sample with 40%
PT E
D
sugar reduction.
4. Discussion
CE
Reformulation of industrialized products has been recommended as a cost-
AC
effective strategy for reducing sugar intake (MacGregor & Hashem, 2014). One of the main challenges of this approach is related to changes in the sensory characteristics of the products, which can compromise consumer hedonic perception. A decrease in sweetness perception was the most relevant change in the sensory characteristics of samples due to sugar reduction, in agreement with previous studies in different product categories (Biguzzi et al., 2014; Drewnowski et al., 1998; Oliveira et al., 2015; Pineli et al., 2016). Sugar reduction also caused a decrease in the perception of vanilla flavour, which can be explained by aroma-taste interactions (Poinot et al., 2013). In addition, sugar reduction caused a decrease in thickness, which can be explained considering
12
ACCEPTED MANUSCRIPT the bulking properties of this ingredient (Kaufman & Palzer, 2011). Changes in the sensory characteristics of samples due to sugar reduction caused a decrease in consumer overall liking. Consumer hedonic perception was strongly related to sweetness perception, as samples that did not significantly differ in sweetness intensity did not differ in their overall liking scores (Tables 2 and 3). This
PT
result is in agreement with the fact that sweetness has been identified as a key driver of liking in several dairy products (Bayarri, Carbonell, Barrios, & Costell, 2011; Oliveira
RI
et al., 2015; Thompson, Lopetcharat, & Drake, 2007).
SC
When added sugar concentration was reduced by 10%, no relevant changes in the sensory and hedonic characteristics of the samples were detected. This suggests
NU
that this sugar reduction percentage could be implemented without the need to introduce further changes in the formulation. Other studies have reported that sugar
MA
concentration can be reduced by 6-10% without changing sweetness perception in different product categories, such as chocolate-flavoured milk, dairy-based emulsions,
D
cakes and orange nectar (Chang & Chiou, 2006; Hoppert, Zahn, Puschmann, Ullmann,
PT E
& Rohm, 2012; Oliveira et al., 2016; Pineli et al., 2016). In the present study, when sugar reduction was increased to 20%, changes in consumer sensory perception were found but not in their hedonic perception (Tables 2
CE
and 3). This suggests that the definitions of sugar reduction strategies based on
AC
consumer sensory perception are more conservative than those based on hedonic perception. Similar results have been reported by Oliveira et al. (2016) for sugar reduction in chocolate-flavour milk and Antúnez, Giménez, & Ares (2016) for salt reduction in bread. Although sugar can be gradually reduced without consumer awareness, the main disadvantage of this approach is the time needed to achieve benefits in public health (Ma, He, Yin, Hashem, & McGregor, 2016). Therefore, strategies to minimize the sensory changes caused by sugar reduction deserve further exploration. The present work assessed the use of aroma-related cross modal interactions for sugar
13
ACCEPTED MANUSCRIPT reduction in vanilla milk desserts. The effect of increasing vanilla concentration depended on the extent of sugar reduction. In the milk desserts with 40% sugar reduction, vanilla caused a slight increase in sweetness perception, whereas for samples with 20% or 60% sugar reduction differences did not reach significance. Oliveira et al. (2015) also reported that
PT
the addition of artificial vanilla at supra-threshold concentrations to chocolate-flavoured milks only increased sweetness perception of samples reduced in 40-60%. This
RI
reinforces the idea that the efficiency of aroma-taste interactions for increasing
SC
sweetness perception is matrix dependent, as previously highlighted by Poinot et al. (2013).
NU
The effect of vanilla flavour was augmented when starch concentration was increased. As shown in Tables 2 and 3, the joint increase in vanilla and starch
MA
concentration led to an increase in vanilla flavour intensity, regardless of sugar reduction percentage. In addition, in samples with 40 and 60% sugar reduction, this
D
strategy also led to an increase in sweetness perception. Different physico-chemical,
PT E
physiological and psychological mechanisms might explain this result (Poinot et al., 2013). First of all, starch interacts with some volatile compounds, leading to increased retention in the matrix or release due to a “salting out” effect caused by the swollen
CE
starch granules (Arancibia, Jublot, Costell, & Bayarri, 2011; González-Tomás, Bayarri,
AC
Taylor, & Costell, 2007). In particular, Keršiené, Adams, Dubra, De Kimpe, & Leskauskaité (2008) and González-Tomás et al. (2007) reported that an increase in starch concentration in milk desserts led to an increase in flavour release. In the present work, an enhanced release of vanilla flavour from the desserts with higher starch
concentrations
could
have
increased
vanilla
flavour
perception
and
consequently sweetness perception. Secondly, an increase in starch concentration could have enhanced flavour perception by modulating the processing of taste and aroma signals due to the increase in viscosity and/or changes in salivation (Poinot et al., 2013). Finally, texture and aroma interactions could be explained by learned
14
ACCEPTED MANUSCRIPT associations through repeated exposure to similar products (Bult, de Wijk, & Hummel, 2007; Saint-Eve, Lauverjat, Magnan, Déléris, & Souchon, 2009). Further research is needed to better understand the mechanisms that underlie the increase of sweetness and vanilla flavour perception caused by an increase in starch concentration. Cross-modal interactions are strongly dependent on the characteristics of the
PT
food matrix, the volatile compounds and thickening agents (Poinot et al., 2013). Therefore, minor differences in product formulation could largely modify their efficiency
RI
for minimizing the influence of sugar reduction on the sensory characteristics of
SC
products. In addition, a need still exists to deepen our understanding on how sensations interact throughout food consumption to shape our sensory and hedonic
NU
perception of products. Research on the topic is warranted and could largely contribute with recommendations for the development of more healthful food products.
MA
In closing this section it should be acknowledged that one of the limitations of the present study is that the interaction effect between vanilla flavour and starch
D
concentration could not be evaluated. Further research should use a full factorial
PT E
experimental design to study the separate effect of these variables, as well as their interaction, on consumers’ sensory and hedonic perception of vanilla milk desserts with
AC
CE
different sugar content.
5. Conclusions
Results from the present work show the potential of aroma-related cross-modal interactions to increase sweetness perception in the development of sugar-reduced products and, consequently, minimize changes in consumer hedonic perception. This strategy was able to totally and/or partially mitigate the effect of a 20% reduction in added sugar in the sensory characteristics of the desserts, without the need to largely modify the formulation or to add non-nutritive sweeteners. However, further studies are necessary to assess their potential for sugar reduction in different product categories.
15
ACCEPTED MANUSCRIPT
Acknowledgments The authors are indebted to Agencia Nacional de Investigación e Innovación (ANII, grant number
INI_X_2013_1_101098),
and
Comisión
Sectorial
de
Investigación
Científica
RI
PT
(Universidad de la República) for financial support.
SC
References
Antúnez, L., Giménez, A., & Ares, G. (2016). A consumer-based approach to salt
NU
reduction: Case study with bread. Food Research International, 90, 66-72. Arancibia, C., Jublot, L., Costell, E., & Bayarri, S. (2011). Flavor release and sensory
MA
characteristics of o/w emulsions. Influence of composition, microstructure and rheological behavior. Food Research International, 44, 1632–1641.
D
Ares, G., Etchemendy, R., Atúnez, L., Vidal, L., Giménez, A., & Jaeger, S. (2014).
PT E
Visual attention by consumers to check-all-that-apply questions: Insights to support methodological development. Food Quality and Preference, 32, 210-220. Ares, G., Giménez, A., & Gámbaro, A. (2008). Does information about the source of
CE
functional ingredients influence consumer perception of functional milk desserts?
AC
Journal of the Science of Food and Agriculture, 88, 2061-2068. Ares, G., Giménez, A., Barreiro, C. & Gámbaro, A. (2010). Use of an open-ended question to identify drivers of liking of milk desserts. Comparison with preference mapping techniques. Food Quality and Preference, 21, 286–294. Bayarri, S., Carbonell, I., Barrios, E.X., & Costell, E. (2011). Impact of sensory differences on consumer acceptability of yoghurt and yoghurt-like products. International Dairy Journal, 21, 111–118. Biguzzi, C., Schlich, P., & Lange, C. (2014). The impact of sugar and fat reduction on perception and liking of biscuits. Food Quality and Preference, 35, 41-47.
16
ACCEPTED MANUSCRIPT Boakes, R.A., & Hemberger, H. (2012). Odour-modulation of taste ratings by chefs. Food Quality and Preference, 25, 81-86. Bult, J. H. F., de Wijk, R. A., & Hummel, T. (2007). Investigations on multimodal sensory integration: Texture, taste, and ortho- and retronasal olfactory stimuli in concert. Neuroscience Letters, 411, 6–10.
PT
Burseg, K.M.M., Camacho, S., Knoop, J., & Bult, J.H.F. (2010). Sweet taste intensity is enhanced by temporal fluctuation of aroma and taste, and depends on phase shift.
RI
Physiology and Behavior, 101, 726–730.
SC
Cadoret, M., & Husson, F. (2013). Construction and evaluation of confidence ellipses applied at sensory data. Food Quality and Preference, 28, 106–115.
NU
Cardoso, J. M. P., & Bolini, H. M. A. (2008). Descriptive profile of peach nectar sweetened with sucrose and different sweeteners. Journal of Sensory Studies, 23,
MA
804–816.
Chang, M. H., & Chiou, W. B. (2006). Differential threshold and psychophysical power
D
function of sweetness sensation: Applied psychophysics and prospect theory on
PT E
reformulating baking products. Journal of Sensory Studies, 21, 534–551. Chollet, M., Gille, D., Schmid, A., Walther, B. & Piccinali, P. (2013). Acceptance of sugar reduction in flavored yogurt. Journal Dairy Science, 96, 5501-5511.
CE
Civille, C.V., & Oftedal, K.N. (2012). Sensory evaluation techniques — make “good for
AC
you” taste “good”. Physiology & Behavior, 107, 598–605. Drewnowski, A., Nordensten, K., & Dwyer, J. (1998). Replacing sugar and fat in cookies: Impact on product quality and preference. Food Quality and Preference, 9, 13-20. Dubois, G. E., & Prakash, I. (2012). Non-caloric sweeteners, sweetness modulators, and sweetener enhancers. Annual Reviews of Food Science and Technology, 3, 353–380. Gardner, C., Wylie-Rosett, J., Gidding, S. S., Steffen, L. M., Johnson, R. K., Reader, D., et al. (2012). Nonnutritive sweeteners: current use and health perspectives: a
17
ACCEPTED MANUSCRIPT scientific statement from the American Heart Association and the American Diabetes Association. Diabetes Care, 35(8), 1798–1808. González-Tomás, L., Bayarri, S., Taylor, A. J., & Costell, E. (2007). Flavour release and perception from model dairy custards. Food Research International, 40, 520– 528.
PT
Hoppert, K., Zahn, S., Puschmann, A., Ullmann, I., & Rohm, H. (2012). Quantification of sensory difference thresholds for fat and sweetness in dairy-based emulsions.
RI
Food Quality and Preference, 26, 52–57.
Standard
8589.
International
SC
ISO (2007). Sensory analysis: General guidance for the design of test rooms, ISO Organization
Standardization,
Geneva,
NU
Switzerland.
for
Johnson, R.K., Appel, L.J., Brands, M., Howard, B.V., Lefevre, M., Lustig, R.H., et al.
MA
(2009). Dietary sugars intake and cardiovascular health: a scientific statement from the American Heart Association. Circulation, 120, 1011–20.
D
Kaufmann, S.F.M., & Palzer, S. (2011). Food structure engineering for nutrition, health
PT E
and wellness. Procedia Food Science, 1, 1479-1486. Keršiené, M., Adams, A., Dubra, A., De Kimpe, N., & Leskauskaité, D. (2008). Interactions between flavour release and rheological properties in model custard
AC
1191.
CE
desserts: Effect of starch concentration and milk fat. Food Chemistry, 108, 1183–
Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. (2016). lmerTest: Tests in Linear Mixed Effects Models. R package version 2.0-32. Available at: https://CRAN.R-project.org/package=lmerTest. Labbe, D., & Martin, N. (2009). Impact of novel olfactory stimuli at supra and subthreshold concentrations on the perceived sweetness of sucrose after associative learning. Chemical Senses, 34, 645–651. Labbe, D., Rytz, A., Morgenegg, C., Ali, S., & Martin, N. (2007). Subthreshold olfactory stimulation can enhance sweetness. Chemical Senses, 34, 645-651.
18
ACCEPTED MANUSCRIPT Lê, S., Josse, J., & Husson, F. (2008). FactoMineR: An R package for multivariate analysis. Journal of Statistical Software, 25(1), 1–18. Lustig, R.H., Schmidt, L.A., & Brindis, C.D. (2012). The toxic truth about sugar. Nature, 482, 27–29. Ma, Y., He, F.J., Yin, Y., Hashem, K.M., & McGregor, G.A. (2016). Gradual reduction of
PT
sugar in soft drinks without substitution as a strategy to reduce overweight, obesity, and type 2 diabetes: a modelling study. The Lancet Diabetes &
RI
Endocrinology, 4, 105-114.
reduction programme. Lancet, 383, 929–931.
SC
MacGregor, G.A., & Hashem, K.M. (2014). Action on sugar—lessons from UK salt
NU
Manoukian, E. B. (1986). Mathematical nonparametric statistics. New York, NY: Gordon & Breach.
MA
Markey, O, Lovegrove, J.A., & Methven, L. (2015). Sensory profiles and consumer acceptability of a range of sugar-reduced products on the UK market. Food
D
Research International, 72, 133-139.
PT E
Mattes, R. D., & Popkin, B. M. (2009). Nonnutritive sweetener consumption in humans: effects on appetite and food intake and their putative mechanisms. American Journal of Clinical Nutrition, 89(1), 1–14.
CE
Moynihan, P.J., & Kelly, S.A. (2014). Effect on caries of restricting sugars intake
18.
AC
systematic review to inform WHO guidelines. Journal of Dental Research, 93, 8–
Murphy, C., & Cain, W. S. (1980). Taste and olfaction: Independence vs. interaction. Physiology and Behavior, 24, 601–605 Oliveira, D., Antúnez, L., Giménez, A., Castura, J., Deliza, R., & Ares, G. (2015). Sugar reduction in probiotic chocolate-flavored milk: Impact on dynamic sensory profile and liking. Food Research International, 75, 148-156. Oliveira, D., Reis, F., Deliza, R., Rosenthal, A., Giménez, A., Ares, G. (2016). Difference
thresholds
for
added
sugar
in
chocolate-flavoured
milk:
19
ACCEPTED MANUSCRIPT Recommendations for gradual sugar reduction. Food Research International, 89, 448–453. Pineli, L.L.O., de Aguiar, L.A., Fiusa, A., Botelho, R.B.A., Zandonadi, R.P., & Melo, L. (2016). Sensory impact of lowering sugar content in orange nectars to design healthier, low-sugar industrialized beverages. Appetite, 96, 239-244.
PT
Poinot, P., Arvisenet, G., Ledauphin, J., Gaillard, J. L., & Prost, C. (2013). How can aroma-related cross-modal interactions be analysed? A review of current
RI
methodologies. Food Quality and Preference, 28, 304-316.
SC
R Core Team. (2016). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
NU
Saint-Eve, A., Lauverjat, C., Magnan, C., Déléris, I., & Souchon, I. (2009). Reducing salt and fat content: Impact of composition, texture and cognitive interactions on
MA
the perception of flavoured model cheeses. Food Chemistry, 116, 167–175. Scientific Advisory Committee on Nutrition (2014). Draft Carbohydrates and Health
D
report. Scientific consultation: 26 June to 1 September 2014. Available at: http://
PT E
www.sacn.gov.uk/reports_position_statements/reports/scientific_consultation_ draft_sacn_carbohydrates_and_health_report_-_june_2014.html (Accessed 10th August 2016)
CE
Stieger, M., & van de Velde, F. (2013). Microstructure, texture and oral processing:
AC
New ways to reduce sugar and salt in foods. Current Opinion in Colloid & Interface Science, 18, 334-348. Te Morenga, L., Mallard, S., & Mann, J. (2013) Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies. BMJ, 346: e7492. Thompson, J.L., Lopetcharat, K. & Drake, M.A. (2007). Preferences for commercial strawberry drinkable yogurts among African American, Caucasian, and Hispanic consumers in the United States. Journal of Dairy Science, 90,4974–4987.
20
ACCEPTED MANUSCRIPT Tournier, C., Sulmont-Rossé, C., Sémon, E., Vignon, A., Issanchou, S., & Guichard, E. (2009). A study on texture–taste–aroma interactions: Physico-chemical and cognitive mechanisms. International Dairy Journal, 19, 450-458. van Raaij, J., Hendriksen, M., & Verhagen, H. (2009). Potential for improvement of population diet through reformulation of commonly eaten foods. Public Health
PT
Nutrition, 12(3), 325–330. Vidal, L., Cadena, R.S., Correa, S., Ábalos, R.A., Gómez, B., Giménez, A. et al. (2014).
SC
consumers. Journal of Sensory Studies, 29, 74-87.
RI
Assessment of global and individual reproducibility of projective mapping with
Vio, F., & Uauy, R. (2007). Case Study #9-5. The Sugar Controversy. In: P. Pinstrup-
Ithaca, NY: Cornell University Press.
NU
Andersen & F. Cheng (Eds.), Food Policy for Developing Countries: Case Studies.
MA
World Health Organization and Food & Agriculture Organization of the United Nations. (2003). WHO/FAO joint expert consultation on diet, nutrition, and the prevention of
D
chronic diseases. WHO Technical Report Series No. 916. Geneva: World Health
PT E
Organization.
Yang, Q. (2010). Gain weight by “going diet?” Artificial sweeteners and the neurobiology of sugar cravings: Neuroscience 2010. Yale Journal of Biology and
CE
Medicine, 83(2), 101-108
AC
Zorn, S., Alcaire, F., Vidal, L., Giménez, A., & Ares, G. (2014). Application of multiplesip temporal dominance of sensations to the evaluation of sweeteners. Food Quality and Preference, 36, 135-143.
21
ACCEPTED MANUSCRIPT Figure captions
Figure 1. Sample and term configurations of vanilla milk dessert samples with different sugar, vanilla and starch concentrations, obtained from consumer responses to a
PT
check-all-that-apply question in Study 1.
Figure 2. Sample and term configurations of vanilla milk dessert samples with different
RI
sugar, vanilla and starch concentrations, obtained from consumer responses to a
AC
CE
PT E
D
MA
NU
SC
check-all-that-apply question, in Study 2.
22
ACCEPTED MANUSCRIPT Tables
Table 1. Formulation of the vanilla milk desserts included in the two studies.
2
Sugar reduction (%)
Starch (%)
Vanilla (%)
0S4.2V0.4 20S4.2V0.4 20S4.2V0.6 20S5.2V0.6 40S4.2V0.4 40S4.2V0.6 40S5.2V0.6 60S4.2V0.4 60S4.2V0.6 60S5.2V0.6 0S4.2V0.4 10S4.2V0.4 20S4.2V0.8 20S4.6V0.6 40S5.0V0.8
20 16 16 16 12 12 12 8.0 8.0 8.0 20 18 16 16 12
0 20 20 20 40 40 40 60 60 60 0 10 20 20 40
4.2 4.2 4.2 5.2 4.2 4.2 5.2 4.2 4.2 5.2 4.2 4.2 4.2 4.6 5.0
0.4 0.4 0.6 0.6 0.4 0.6 0.6 0.4 0.6 0.6 0.4 0.4 0.8 0.6 0.8
SC
RI
PT
Sugar (%)
NU
1
Sample
MA
Study
Coding of the samples indicates sugar reduction percentage (0, 10, 20, 40 or 60), followed by the starch
AC
CE
PT E
D
concentration (S) and vanilla concentration (V).
23
ACCEPTED MANUSCRIPT
Table 2. Average overall liking score and frequency of use of terms of the CATA question for describing the vanilla milk desserts in Study 1.
0S4.2V0.4
Overall liking a 6.4
20S4.2V0.4
6.1
a
20S4.2V0.6
6.1
a
20S5.2V0.6
5.7
40S4.2V0.4
5.1
Sample
a,b b,c
Sweet*** 62
Barely sweet*** d 4
Vanilla flavour*** a 63
43
a,b
6
c,d
44
b
28
a,b,c
42
a,b
8
c,d
51
b
33
c
71
a
33
c
34
c
59
a,b
40
c
13
30
b,c
24
b
32
b,c
24
b
c
25
b
40S4.2V0.6
5.7
40S5.2V0.6
6.0
a,b
64
a
25
60S4.2V0.4
4.0
d
28
d
10
d
50
a
16
d
60S4.2V0.6
3.9
d
23
d
12
d
53
a
16
d
45
b
10
d
62
a
60S5.2V0.6
4.3
c,d
a,b,c
a,b,c
35
a,b
51
Milk flavour*** f 18
a,b
58 51
a
Very sweet*** a 47
58
35
11
b,c,d
21
39
a,b
17
46
a
c,d
12
c
42
a
a,b
42
a
b,c
a,b,c
37
b
20
c,d,e,f
29
b,c
d,e,f
25
b,c
a
53
a
a,b
56
a
66
a
b,c,d,e
D E
43
a
c
27
44
b,c
14
e
22
41
a
a,b,c
34
a
a,b,c
46
a
a,b
37
a
a
31
a
40
a
A M
27
33
12
c
Creamy***
Smooth
Liquid***
Gummy***
65 61
a,b
29
a
19
a,b
10
a
29
a
20
a,b
9
b,c,d
44
a
14
c,d
68
53
33
a
ns
a,b
T P
I R
SC
U N 22
ns
Thick
d,e
10
53 31
Firm*
e,f
22
c
7
e
c,d,e,f
20 28
Tasteless***
20
a,b
13
b,c b,c
b,c,d
b,c
31
a
31
a
22
a,b
7
c,d
56
a,b,c
33
a
19
a,b
8
c,d
59
a,b,c
43
a
8
f
26
a
26
a
e,f
33
a
27
a
45
25 29 39
d,e
32
a,b
c,d
2
d
18 1 3
b
e
d,e
14
b,c
Coding of the samples indicates sugar reduction percentage (0, 20, 40 or 60), followed by the starch concentration (S) and vanilla concentration (V).
T P E
Overall liking scores with different letters are significantly different according to Tukey’s test (p<0.05). Results of Cochran’s Q test are shown next to each attribute: * indicates significant differences among samples at p<0.05, ** significant differences at p<0.01, *** significant differences at p<0.001, whereas ns indicate no significant differences (p>0.05). Values of frequency of use of CATA terms with different letters within a column are significantly
C C
different according to the sign test (p<0.05).
A
24
ACCEPTED MANUSCRIPT
Table 3. Average overall liking scores and frequency of use of terms of the CATA question for describing the vanilla milk desserts in Study 2. Sample 0S4.2V0.4
Overall Very Barely Vanilla Milk ns ns ns Sweet* Thick Creamy*** Smooth Liquid*** Gummy** ns ns Tasteless*** Firm liking sweet*** sweet*** flavour flavour a,b b,c a b a a b,c a a a a a b,c 6.2 45 40 9 48 19 14 13 44 63 35 29 17
10S4.2V0.4
6.5
a
56
a,b
38
a
9
b
40
a
25
a
12
20S4.2V0.8
6.6
a
56
a,b
39
a
9
b
43
a
19ª
8
7.0
a
43
a
7
b
52
a
15ª
5.6
b
17
b
38
a
a
20S4.6V0.6 40S5.0V0.8
60
a
40
c
35
a
18
10
a
39
a
62
c
20
a
42
a
14
a
39ª
a
a
I R
18
b
52
a
C S U 17
42
a
T P
b,c
74
a
64
a
44
b
30
a
20
a,b
27
a
20
a,b
35
a
14
24
a
c
5
b
15
b,c
10 21
c
a,b
30
a
Coding of the samples indicates sugar reduction percentage (0, 20 or 40), followed by the starch concentration (S) and vanilla concentration (V).
N A
Overall liking scores with different letters are significantly different according to Tukey’s test (p<0.05).
Results of Cochran’s Q test are shown next to each attribute: * indicates significant differences among samples at p<0.05, ** significant differences at p<0.01, *** significant
M
differences at p<0.001, whereas ns indicates no significant differences (p>0.05). Values of frequency of use of CATA terms with different letters within a column are significantly
D E
different according to the sign test (p<0.05).
T P E
C C
A
25
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
Figure 1.
26
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
Figure 2.
27