- Email: [email protected]
Impact of model fat emulsions on sensory perception using repeated spoon to spoon ingestion
Physiology & Behavior 160 (2016) 80–86
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Impact of model fat emulsions on sensory perception using repeated spoon to spoon ingestion I.A.M. Appelqvist ⁎, A.A.M. Poelman, M. Cochet-Broch, C.M. Delahunty CSIRO Food & Nutrition, 11 Julius Avenue, North Ryde, NSW 2113, Australia
H I G H L I G H T S • • • • •
Sensory perception upon repeat ingestion of low and high fat emulsions was measured Fatty mouthfeel and afterfeel increased with repeated ingestion of HF emulsions Lipophilic aroma enhanced in HF indicating cross-modal interaction Sweetness perception did not change with repeated ingestion of HF or LF emulsions Findings of fatty perception might be explained by build-up of fatty oral coatings
a r t i c l e
i n f o
Article history: Received 22 June 2015 Received in revised form 14 March 2016 Accepted 28 March 2016 Available online 7 April 2016 Keywords: Fat Sensory perception Dynamic Repeated spoon ingestion Oral residue Mouth coating
a b s t r a c t Eating is a dynamic behaviour, in which food interacts with the mechanical and physiological environment of the mouth. This dynamic interaction changes the oral surfaces leaving particles of food and building up a film on the oral surfaces, which may impact on the temporal perception during the eating experience. The effect of repeated spoon to spoon ingestion of oil in water emulsion products (2%–50% w/w oil) was evaluated using descriptive inmouth and after swallowing sensory attributes. Descriptive sensory analysis indicated that fatty mouthfeel and afterfeel perception (measured post swallowing) increased with the number of spoonfuls for emulsions containing 50% fat. This effect is likely due to the build-up of oil droplet layers deposited on the mouth surfaces. There was an enhancement of fatty afterfeel intensity for 50% fat emulsions containing the more lipophilic aroma ethylhexanoate compared to ethyl butanoate, indicating a cross-modal interaction. No increase in these attributes from spoon to spoon was observed for the low oil emulsions; since most of the oil in the emulsion was swallowed and very little oil was likely to be left in the mouth. Sweetness perception increased as fat level increased in the emulsion due to an increase in the effective concentration of sugar in the aqueous phase. However, the sweetness perceived did not change from spoon to spoon, suggesting that any oil-droplets deposited on the oral surfaces did not form a complete barrier, restricting access of the sucrose to the taste buds. This study highlights the importance of measuring the dynamic nature of eating and demonstrated change in sensory perception occurring with repeated ingestion of model emulsions, which was likely due to a change in mouth environment. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Eating is a dynamic behaviour and is generally performed as a multiple ingestion process, where each mouthful of food is masticated and swallowed and then immediately repeated. Residues of foods (and beverages) often coat the oral mucosa after consumption, these are termed oral or mouth coatings [1] and as a consequence of their persistence will change the environmental conditions of the mouth. In-mouth sensory perception of food such as an emulsion depends on its microstructure, composition and how it behaves during processing in the
⁎ Corresponding author. E-mail address: [email protected] (I.A.M. Appelqvist).
http://dx.doi.org/10.1016/j.physbeh.2016.03.035 0031-9384/© 2016 Elsevier Inc. All rights reserved.
mouth [2]. When food is brought into the mouth, and depending on the consistency, it is mechanically broken down or manipulated with the tongue, and diluted with saliva to form a lubricated bolus, which can be swallowed safely [3]. The mastication process leads to close interaction between the food and the oral surfaces, where sensory receptors such as taste receptors on the tongue, flavour receptors in the nose and pressure, stress and heat receptors on the oral tissues, are activated by the food where they are integrated by the brain to form an overall perception of sensory texture and flavour attributes [2]. For emulsion based food the sensory perception often occurs after swallowing, due to the persistence of food residues and particles coating the oral surfaces [4]. This mouthcoating impacts on the oral environment changing the composition of saliva, the pH of the mouth and access to the taste and mechano-
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
receptors [5–7], which in turn plays a big role in determining aftertaste and afterfeel [1] and directly influences the sensory perception of the next mouthful of food. The dynamic nature of the eating event upon repeated consumption has also been recognised in relation to hedonic appreciation. In fact, the implicit assumption that consumers do not change has been identified by Köster [8] as one of the fallacies in the study of eating and drinking behaviour. Sensory appreciation can change with repeated consumption of the same food, and can either decrease, potentially due to sensory specific satiety [9], boredom or irritation [10,11], or increase, potentially due to increased familiarity and/or a change in perceived complexity [10,12–15], which in turn can impact on perception. Most sensory perception research is based on single sips or single bites. The dynamic nature of sensory perception is mostly considered within a mouthful, using various sensory methodologies such as Time Intensity [16,17], Progressive Profiling [18] and Temporal Dominance of Sensations [19,20]. However the dynamics of perception between mouthfuls involved in eating a portion of food are usually not considered. Indeed, sensory testing protocols purposely go to great lengths to reduce the carry-over effect from one intake to the next such as in wine-tasting through development of elaborate palate cleansing methodologies [21]. Very few sensory evaluation studies have tried to evaluate the progressive ingestion of food and determined the impact on sensory perception. One notable exception has been the study conducted by Courregelongue [22] who used repeated ingestion to determine composition and rheological factors on the temporal perception of soymilk astringency using time intensity. A more recent study by Zorn [23] used a multisip temporal dominance of sensations methodology to evaluate a range of sweeteners in order to determine changes in the temporal profile of juices. In this study results are reported for the first time on the perception of model emulsion foods throughout eating using a spoon-to-spoon methodology, which mimics closely normal eating repeat ingestion and allows the potential build-up of food emulsion residues in the mouth. The aim of this study was to determine the impact of a spoon to spoon multiple ingestion on selected flavour and texture attributes for a range of emulsions containing different levels of fat. The research was a first step to gain insights on how sensory intensity will change as consumption of the emulsion portion continues, potentially relating to an accumulation of fat emulsion residue in the mouth [24].
81
Table 1 Composition of samples. Sample
Total fat (%)
Whey protein (%)
Xanthan (%)
Sugar (%)
Ethylhexanoate (ppm)
50%-HEX 10%-HEX 2%-HEX 50%-BUT 10%-BUT 2%-BUT
50 10 2 50 10 2
1 1 1 1 1 1
0.05 0.05 0.05 0.05 0.05 0.05
2 2 2 2 2 2
25 25 25
Ethyl butanoate (ppm)
25 25 25
of whey protein in preboiled water at 50 °C using an automatic stirrer. A xanthan/sugar solution was prepared using the same process. The two solutions were combined in appropriate quantities to make a base solution. Canola oil (50% w/w) was then added and the mixture was homogenized for 2 min at 16,000 rpm and then 2 min at 20,000 rpm (Ultra-Turrax, T25 with S25N-25F dispersion tool, IKA, Germany). Separately, a solution of sugar, xanthan and emulsifier was prepared to maintain the concentration of these ingredients constant in all samples, when diluting the 50% fat emulsion in order to obtain emulsions with lower fat contents. Using required quantities of the 50% emulsion and the aqueous solution, the 10% and 2% fat emulsions were prepared. Emulsions were poured into plastic, sterile, screw-top containers. The required quantity of flavour compound, pre-diluted in ethanol, was added to the emulsion, and samples were refrigerated at 5 °C until use the following day. Thirty minutes prior to the start of the sensory evaluations, samples were removed from refrigeration. A 10 mL volume of sample was served in a 30 mL transparent plastic cup with a lid. Twelve cups of the same sample, coded 1 to 12, were placed on a three-digit coded tray and constituted one sample for the purposes of the experiment. 2.2. Particle size distribution Apparent particle size distribution was measured by laser light scattering using a Malvern Mastersizer 2000 Ver 5.60 instrument (Malvern Instruments Ltd, Worcestershire, UK). A differential refractive index of 1.101 (1.465 for particle/1.33 for water) absorption of 0.001 were used as the optical properties of the dispersion. The particle calculation was set for spherical particles. Volume median diameter value d(0.5) was used as the average particle size. Each sample was measured in duplicate.
2. Methods and materials 2.3. Descriptive sensory analysis 2.1. Samples Oil-in-water (O/W) emulsions were used as model food systems. Samples had a range of fat contents (50%, 10% and 2% fat) and contained either ethylhexanoate (HEX) or ethyl butanoate (BUT). Both flavour compounds had a fruity note but they differed in lipophilicity (HEX is more lipophilic than BUT; octanol/water partitioning coefficient (Ko/w) 641 and 80 respectively). All samples contained 2% sucrose (w/w). Thus, the sample set consisted of six samples varying in fat content and added flavour compound (Table 1). Whey protein isolate (1% w/w) was selected due to its good emulsifying capacity and neutral taste. Xanthan gum (0.05% w/w) was added to stabilize the emulsion. Canola oil and sugar were purchased from a local retailer (Woolworths, Homebrand). Whey-protein-isolate (NZMP Whey Protein Isolate 895, Fonterra, Auckland, New Zealand) was ordered via an on-line supplier. Ethylhexanoate and ethyl butanoate were sourced from Sigma Aldrich (Castle Hill, Australia) and xanthan gum was sourced from Kelco International (Chicago, USA). Tap water was used for the aqueous phase. A 50% fat emulsion was processed using a four-step procedure. A whey protein solution was made by dispersing the required quantity
All sensory activities were carried out in a testing laboratory designed in accordance with International Standards on Sensory Analysis (ISO 8589:2007). A trained sensory panel, consisting of ten assessors (2 male, 8 female, age: 44.5 ± 7.2 years) previously screened for normal sensory acuity and with extensive experience in descriptive analysis on a variety of food products, participated to the descriptive analysis of the samples. Three two-hour training sessions were held to familiarise the assessors with the samples, and reach consensus on the method of assessment and the sensory attribute definitions. The assessors were informed that the objective of this study was to gain insights in sensory perception throughout the eating event. No information was provided as to whether the samples on the tray were the same or not, assessors were simply asked to rate the perceived attribute intensities for each sample. Assessors were made aware that familiarisation with the samples would involve a limited number of samples only, and focus group discussion of samples and attributes were based on single-spoon assessments only. To aid in attribute recognition, sucrose solutions and the two pure flavour compounds were used as reference standards. Method of assessment was developed and practiced with emulsion samples and sugar/xanthan
82
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
solutions until all assessors had reached a high level of confidence in conducting procedures according to the timings required. For the evaluation of the samples assessors were presented with a tray containing 12 cups labelled 1 to 12. They were seated in booths and were guided through the procedure using Compusense® five sensory data acquisition software (version 4.6, 2004; Compusense Inc., Guelph, Ontario, Canada). Assessors opened the lid and took one level spoon of the first sample by scooping and pouring. They pressed “next question” while simultaneously putting the sample in their mouth. Assessors kept the sample in their mouth for 3 s (by silently counting 21, 22, 23) while swirling it around with their tongue and then rated the flavour/mouthfeel attributes. The sample was then swallowed while the assessors simultaneously clicked on “next question”. A timer counted down for 5 s, during which assessors threw their cup in the rubbish bin, which ensured they would have the following cup to assess in front of them. After 5 s, aftertaste/ afterfeel attributes were rated. Once attributes were rated, assessors clicked “next question”, which started a timer to count down from 10 s, during which time assessors cleaned their spoon. Assessors continued this procedure with the next cup until all twelve cups were evaluated (assessment of each cup took about 20–25 s). Assessors did not cleanse their palate between cups. The palate was cleansed between different emulsion samples (each consisting of 12 cups); water, bread, cucumber and diluted cold black tea were provided for this purpose. For each cup, assessors rated the same three in-mouth attributes sweet taste, fruity flavour and fatty mouthfeel as well as the same three after swallowing attributes, sweet aftertaste, fruity aftertaste and fatty afterfeel. The panel quantitatively evaluated the sensory properties of the six samples (each sample comprising of 12 cups of the same emulsion) in duplicate. Samples were presented in blocks separately for each flavour compound, starting with HEX. Within blocks order was randomised across assessors and evaluations were carried out over four days. Between samples a ten minute inter-stimulus interval was imposed. All evaluations were carried out in individual sensory booths under white light. Attributes were rated on 100 mm unstructured line scales anchored at 5 and 95%, respectively, with extremes for each descriptive term. Ethical approval for this study was granted by the CSIRO Low Risk Review Panel. Informed consent was obtained from assessors prior to the start of the study. 2.4. Data analysis Panel performance was checked with PanelCheck (v1.4.0, 2008) and was good (data not shown). Data were analysed with SPSS (v20.0.0, 2011). A criterion of p b 0.05 was used for statistical significance. To determine the effect of fat content and flavour compound on sensory perception of attributes measured, univariate analysis of variance was conducted at each of the twelve spoons using fat level, flavour compound, their interaction and assessor as variables in the model. Where significant differences were obtained, Bonferroni post hoc tests were performed. To determine trends throughout eating, repeated measures ANOVA was carried out, using attribute scores for cups 1 to 12 as repeated factor, and samples and assessors as fixed factors. Two-way interactions were included in the model. Area under the curve (AUC) for selected sensory attributes was calculated by summing the total of perceived intensities across all twelve spoonfuls, and differences between samples were tested using univariate ANOVA. 3. Results and discussion 3.1. Particle size distribution of fat emulsions The oil droplet size of the emulsions with different fat content was measured to determine if the particle size distribution of the oil droplets
changed with fat content. The results in Fig. 1 show that the average particle size distribution (major peak) for all the emulsion samples were the same with the majority of the emulsion droplets falling between 3 and 13 μm, with an average size of around 9 μm. The data from the 2% emulsion exhibited a small bimodal distribution with a second peak of larger oil droplets with an average size of around 80 μm, indicating there was a small amount of coalescence with some droplets. The size of the emulsion droplet has been shown to affect the rate of flavour release and therefore the flavour intensity [25]. However, the change in droplet size measured were well above 500 μm to see these effects, which was well outside the size change for this study and therefore the impact on perception would be negligible, especially as this was a small percentage of the total oil emulsion and at the lowest oil concentration used. 3.2. Effect of fat content and aroma lipophilicity on fatty mouthfeel and afterfeel The effect of fat content on the fatty mouthfeel and afterfeel for emulsions containing either ethylhexanoate or ethyl butanoate during spoon to spoon ingestion is shown in Fig. 2a (fatty mouthfeel) and Fig. 2b (fatty afterfeel) respectively. The plots of fatty mouthfeel and afterfeel perception against increasing number of spoonful of emulsion sample show that for the 2% and 10% fat emulsions the fatty mouthfeel perception and afterfeel perception were significantly lower (p b 0.0001) compared to the 50% fat emulsion. The total AUC for fatty mouthfeel and afterfeel was used to give an overall indication of fatty perception over the accumulative repeated eating of 12 spoonfuls of the model emulsion in this study. The results (Table 2) show that with increasing fat level there is an increase in overall fatty perception in mouthfeel and afterfeel (p b 0.001). Perception of fatty mouthfeel (Fig. 2a) increased throughout consumption of the twelve spoons (p b 0.0001) for all the fat levels with the increase most pronounced for the 50% fat emulsion, notably after about six spoons. Fatty afterfeel perception (Fig. 2b), which was measured after swallowing the emulsions, followed a similar trend to the fatty mouthfeel, the perception increasing with repeated spoonful (p b 0.0001). This data suggests that the food emulsions coat the mouth surfaces during mastication and remains as a residue after swallowing, which appears to increase upon repeated ingestion, indicated by the increase in texture perception with increased number of consumption spoonfuls. This observation agrees with work conducted by Camacho [1] who showed an increase in the fatty afterfeel attribute of fatty film intensity (after expectorating) as a function of increasing oil deposition of different fat content emulsions (2 to 20%) on the front and back of the anterior tongue. Previous work from Appelqvist [24] and Adams et al. [4] using fluorescent imaging of emulsion coatings containing 2% and 55% fat also indicated that the level of fat residue increased as a function of repeated ingestion, measured by an increasing intensity of fluorescence. Moreover, their data indicated that more oil mouth coating was left on the surface of the tongue with increasing fat content. It is therefore likely that the increase in sensory perception of fatty mouthfeel and afterfeel with repeated ingestion of oil in water emulsions is explained by an increase in fat residue left on the oral surfaces. Camacho et al., 2015 [26] recently quantified the amount of oil fraction that was deposited on the tongue using in vivo florescence measurements of oil emulsions ranging from 1% to 20% oil. They found a similar trend in which more oil quantified in mg/cm2 was deposited onto the oral surfaces as a function of increasing oil content in the emulsions ingested. In this study, the higher oil content emulsion also had a higher viscosity and it has been shown in previous studies that more residue is left behind as the viscosity of the sample increases [4,27]. The mechanisms of interaction between food emulsions and saliva will also be important in determining the amount of food residue left in the mouth. Positively charged protein stable emulsions as in this case stabilized
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
83
Fig. 1. Particle size distribution of emulsion samples ranging in fat content from 2 to 50%.
with whey protein isolate can interact with charged ions found in saliva so that flocculation in the presence of saliva can potentially occur [5]. This may be largely driven by depletion, van der Waals forces and/or electrostatic interaction between emulsion droplets and salivary proteins, depending on the net charge of the emulsion droplets and the presence of other charged molecules in the saliva. Dresselhuis [28] showed that oral processing of stable emulsions (emulsified with 1% WPI) did not show any in-mouth coalescence suggesting the oil was deposited as individual emulsion droplets, which was indicated in the fluorescence imaging of similar emulsions using WPI in a previous study [4]. We measured fatty mouthfeel and fatty afterfeel in our study, and therefore focused predominantly on the mouthfeel aspect of fattiness perception. However, we recognise that fattiness is a complex and multimodal sensation. In addition to the widely acknowledged contribution of fat to mouthfeel, it also affects odour and appearance [29], and recently perception of free fatty acids is being studied as potentially also being a basic taste [30].
3.3. Effect of mouthcoating on texture perception In order to compare the fatty mouthfeel and afterfeel perception, the data for all 3 fat emulsions containing ethylhexanoate and ethyl butanoate was compared in Fig. 2a and b. It can be seen that fatty afterfeel did not decrease greatly when compared to fatty mouthfeel for all repeated spoonfuls, even though most of the product would have been swallowed between repeated ingestion. Examination of the data indicates that samples having the highest perceived intensities (50% fat emulsions) showed the largest decreases in perception between product in-mouth and after swallowing. This may be explained by the greater difference in amount of fat droplets present during mastication and amount of fat droplets deposited in the oral coating after swallowing, and that perceived intensities will be better discriminated. In the case of the 2% and 10% fat emulsions, the differences between the two texture attributes were small (Fig. 2a and b). For the 10% fat emulsion containing ethylhexanoate no difference was found between the perceived fatty mouthfeel and afterfeel after 6 repeated spoonfuls. In
Fig. 2. Perceived intensity (+SE) of (a) fatty mouthfeel and (b) fatty afterfeel of six emulsions varying in fat content and added aroma ( 50% fat with ethyl hexanoate, 50% fat with ethyl butanoate, 10% fat with ethyl hexanoate, 10% fat with ethyl butanoate, 2% fat with ethyl hexanoate, 2% fat with ethyl butanoate throughout eating twelve spoons.
84
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
Table 2 Area under the curve (AUC) as calculated measure from temporal perception of fatty mouthfeel and afterfeel of twelve spoonfuls of six emulsions varying in fat content and aroma compound added. Sample
Fatty mouthfeel
Fatty afterfeel
50%-HEX 50%-BUT 10%-HEX 10%-BUT 2%-HEX 2%-BUT F value p value
650.80 a⁎
582.30 a 520.57 b 222.83 c 178.07 c 111.95 d 83.64 d 179.04 b0.0001
619.43 a 237.93 b 206.36 b 141.85 c 102.57 c 250.40 b0.0001
⁎ Samples sharing the same letter were not significantly different from each other.
these samples the oil fraction in the samples and that in the deposited oral coatings may not vary greatly and therefore differences in perceived fatty texture would also be small. The overall fatty mouthfeel and afterfeel impression was also scored for emulsions containing ethyl butanoate after twelve spoonfuls. The data (not shown) indicates that the overall intensity appears to be informed by an averaging of perceived intensities over the 12 spoonfuls of product. 3.4. Cross-modal interactions between aroma and texture perception The data in Fig. 2a and b also indicates that the lipophilicity of aroma added to the model fat emulsions had an impact on the perceived fatty mouthfeel and afterfeel for each of the fat levels studied. The data indicate that the 50% fat emulsions containing the higher lipophilic aroma ethylhexanoate was perceived as significantly more fatty in both mouthfeel and afterfeel from spoonful 7 onwards compared to the same emulsion containing ethyl butanoate. These results are supported by a greater number for area under the curve for the 50% fat emulsion containing ethylhexanoate compared with the 50% ethyl butanoate, which was significant in fatty afterfeel, and trending in the same direction (although not significant) for fatty mouthfeel (Table 2). This suggests that there was a cross-modal interaction between aroma and texture, where ethylhexanoate appears to enhance the perceived fatty mouthfeel and afterfeel. One possible explanation is that ethylhexanoate preferentially resides in the oil droplets (as it is more lipophilic) and therefore more aroma molecules remain in the fat emulsion during eating and in the emulsion mouthcoating, enhancing perception of fatty mouthfeel and afterfeel. Aroma has been shown to change texture perception, for instance Bult and co-workers [31] showed that the presence of a ‘creamy’ aroma increased the perceived thickness and creaminess of a range of semi-solid products. 3.5. Impact of spoon to spoon ingestion on taste and flavour perception 3.5.1. Sweet taste Fig. 3 shows the effect of continual food consumption across 12 repeated spoonfuls for the different fat level emulsions, for sweet taste (3a and 3b) and fruity flavour perception (3c and 3d). Fig. 3a and c is the emulsions containing ethylhexanoate and Fig. 3b and d is the emulsions containing ethyl butanoate. The data indicates that for sweet taste the 50% emulsion samples (for both emulsions containing ethyl butanoate and ethylhexanoate) were perceived as sweeter (p b 0.0001), with a 10% difference on average at each time point, compared to the 10% and 2% emulsion samples, and that no perceived significant difference in sweetness was observed between the 10% and 2% fat samples themselves (Fig. 3a and b). The higher sweetness perception for the 50% emulsion samples (based on a constant sugar concentration of 2% on product) was due to an increase in the effective concentration of sugar in the aqueous phase, since sugar is not soluble in fat. Hence the effective sugar concentration in the 50% aqueous phase was 4% w/w, which
compares to 2.04% w/w for the 2% emulsion and 2.2% w/w for the 10% emulsion, explaining why they were perceived to be lower in sweetness compared to the 50% fat emulsion samples. Having similar sugar concentrations of around 2% for the 2% and 10% fat samples also explains the finding that the sweet taste intensity rated was the same. This effect of o/w emulsion phase volume was also shown for salty taste by Malone [25], who observed that using a constant salt concentration on product, perceived saltiness increased with increasing oil content, and was associated with the increase in salt concentration in the aqueous phase. It is noteworthy that the difference in sweetness perception between different fat level emulsions is significant at all spoonfuls with exception of the first spoonful, where the difference in sweetness perception between the samples was not statistically significant. These results indicate that perceptual differences may become more apparent when ingesting more of the same food, and demonstrates the potential risk of relying on single sips or bites in describing a food's sensory properties. It is also clear from the data for sweet taste (Fig. 3a and b) that sweetness does not develop upon spoon to spoon even though mouth coating may increase (statistically there is no significant change in sweet taste or aftertaste). This could be due to the fact that the oil droplets are likely to be preferentially deposited onto the oral surfaces (and composes the bulk of the oral coating) and that most of the aqueous phase is cleared through mixing and dilution with saliva and then swallowed. Therefore the sugar concentrated in the aqueous phase gets washed away during the swallowing and cleansing process. This hypothesis may be reasonable given that the sweet aftertaste for all the fat emulsions superimpose on one another from spoonful to spoonful and are overall lower than the in-mouth sweetness perception (Fig. 3a and b). This indicates that the aqueous phase is quickly diluted with saliva over the 20 second time period between spoonful consumption and the differences in sugar concentration left in mouth after each swallowing are below the just noticeable differences for sweetness [32]. Adaptation may also play a part in which any potential build-up of sugar on the mouth surface is counteracted by decrease in sweetness perception due to adaptation which can take up to 45 to 60 s to recover [33]. Recently, Camacho et al. [26] studied the impact of oil emulsion coatings on the perception of sweetness and saltiness of tastant solutions. It had been shown previously that as the level of oil coating increased a decrease in taste modality was observed [34]. It was suggested that the oil deposited on the oral surfaces create a hydrophobic barrier, reducing the accessibility of the hydrophilic taste molecules to the receptors [35]. Our data on sweet taste did not show a reduction in sweetness perception with additional mouthfuls of emulsion and was independent of the increase in perception of fatty mouthfeel. This indicates that although more oil was probably deposited onto the oral surface, the oil did not form a homogeneous film, but rather is ‘patchy’ in nature and small molecules such as sucrose are freely accessible and can interact with the taste buds. This could be related to the underlying structure of the oral surface but also perhaps indicating the inefficient and irregular behaviour of in-mouth processing in clearing the surface of the buccal mucosa after eating food. Finally, the data also indicates that aroma compounds selected in this study did not affect perceived sweetness. At each fat level, there was no perceived significant difference in sweetness between samples spiked with ethylhexanoate and ethyl butanoate. 3.5.2. Fruity flavour The combined dataset for fruity flavour with emulsions containing either ethyl butanoate or ethylhexanoate (Fig. 3c and d) showed that fat content impacted on the release of the lipophilic aroma compounds (p b 0.01). The lowest oil content (2% emulsion) resulted in a higher perceived intensity of fruity flavour. Generally, in standard o/w emulsions the rate of inter-phase transport of aroma molecules from the oil to the aqueous phase occurs very fast [36]. In the case of low fat o/w emulsions, the lipophilic aroma molecules during eating are rapidly stripped from the oil phase, creating a strong driving force for rapid
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
85
Fig. 3. Perceived intensity (+SE) of (a–b) sweet taste and (c–d) fruity flavour of emulsions varying in fat content and added aroma throughout eating twelve spoons; (a and c) 50% fat with ethyl hexanoate, 10% fat with ethyl hexanoate, and 2% fat with ethyl hexanoate and (b and d) 50% fat with ethyl butanoate, 10% fat with ethyl butanoate, and 2% fat with ethyl butanoate.
mass transfer of aroma from the oil phase to the aqueous phase and headspace [37]. This in turn increases the number of aroma molecules transferred to the olfactory receptors increasing the perceived fruitiness intensity. Previous work by Malone [25,38] and Frank [37] measuring dynamic flavour release demonstrated that the amount of aroma released within a given time decreased with increase in fat content, where fat lowers the vapour pressure of lipophilic aromas (Ko/w ≫ 1) and reduces the rate at which they are released during mastication. Fig. 3c shows a slight decrease in fruity flavour intensity for the 50% fat emulsion containing ethylhexanoate with repeated ingestion. This fits with theory in that as more oil is deposited onto the mouth surface as a mouth coating, the rate of release of this lipophilic aroma will be decreased. Adaptation may also play a role. However, there was no statistically significant trend from spoonful to spoonful in perception of fruity flavour or fruity aftertaste across all emulsions. The area under the curve (AUC) was determined from the fruity flavour data of in-mouth and aftertaste and has been used to give an indication of amount of volatile released over the accumulative repeated eating of 12 spoonfuls (Table 3). The results show that the release of ethyl hexanoate and ethyl butanoate decreased with increasing fat level due to the reasons described in the previous section.
4. Conclusions and practical implications This work clearly demonstrates that spoon to spoon ingestion of a model emulsion impacts on the sensory perception of the food, increasing overall fatty mouthfeel and afterfeel perception over time as a function of number of spoons, especially when the fat content is high (~50% fat w/w). This suggests that with each mouthful, an emulsion coating is deposited on the oral surfaces, which builds up with subsequent intake of food and is persistent after swallowing.
The data also indicates that for the most lipophilic aroma compound ethylhexanoate, there is indication of a cross modal interaction between the fatty texture (mouthfeel and afterfeel) and flavour intensity, resulting in an enhancement of fatty mouthfeel perception, probably because more ethylhexanoate aroma molecules remain in the fat phase compared to ethyl butanoate and is integrated into the perception of fatty texture. Changing the phase volume of fat impacted on the intensity of the sweet taste perception because the sugar level was formulated on product rather than on aqueous phase, so that the effective concentration of sugar in the aqueous phase increased with increasing fat levels. Interestingly, the sweetness perceived did not change with each spoonful of product, suggesting the increase in fatty mouthfeel, probably from deposition of an oral oil coating did not suppress sweetness perception by reducing accessibility of the sugar molecules to the receptors. It was also noted that differences in sweet taste perception of samples of different fat content became apparent only after the first spoonful, demonstrating the value of repeated ingestion for accurate description Table 3 Area under the curve (AUC) as calculated measure from temporal perception of fruity flavour and aftertaste of twelve spoonfuls of six emulsions varying in fat content and aroma compound added. Sample
Fruity flavour
Fruity aftertaste
50%-HEX 50%-BUT 10%-HEX 10%-BUT 2%-HEX 2%-BUT F value p value
402.05 c⁎ 430.50 bc 440.58 bc 485.93 ab 507.00 a 551.82 a 4.60 0.02
356.95 c 361.50 bc 393.65 bc 384.14 bc 419.55 ab 439.64 a 2.90 0.02
⁎ Samples sharing the same letter were not significantly different from each other.
86
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
and differentiation of sensory properties in comparison with single sip/ bite methods. Finally the overall impression of intensity measured after the 12 spoonfuls appears to be related to the average rated responses across all spoonfuls, suggesting that the overall perception is not accumulative and not necessarily biased by the last spoonful taken. 4.1. Practical implications These results have several implications for the development of food products. Fatty mouthfeel is an attribute that contributes to food acceptance in semi-solid food products such as yogurts, creams and mousses. This study shows a build-up of fatty mouthfeel after eating sequential spoonful of product containing high fat (50% fat emulsion) and it is likely that this behaviour is one reason why consumers prefer high fat products such as chocolate or cream, which coats the mouth and can be perceived long after they are swallowed. This means that when food companies are looking to reformulate their products with reduced fat content, they should also consider how they might mimic the same oral mouth coating profile provided by the high fat product counterpart to increase acceptance. Recent research on perception of fat taste has shown differences between individuals in fat taste sensitivity which is in part dependent on fat intake and modifiable [39–41]. Although this study did not look at individual differences in sensory perception, further research to determine individual physiological and genetic differences in taste perception could enhance understanding on how best to reformulate products. In addition, these findings could be useful when developing food products for patients with swallowing difficulties (e.g. dysphagia) to understand how products might behave from one spoonful to the next. As an example, the perception of fatty mouthfeel is an indicator of the amount of fat in a food. The fat content of food is an important functional parameter that provides lubrication and contributes to the formation of a bolus that is safe to swallow. However, the build-up of a mouthcoating beyond a certain level could reverse the effect and make it difficult for some people to swallow. A spoon-to-spoon methodology, such as described in this study, could be applied to gain insights on the safety/suitability of the product for dysphagia patients. Development of methods to measure fat residence time in the mouth and distribution on oral surfaces could be used in food product design to better determine the temporal dynamics of perception, optimise sensory properties and develop foods that are safe to swallow for vulnerable consumer groups. Acknowledgments Tim Wooster is acknowledged for his help in designing the emulsions, and Mi Xu is acknowledged for her help in the measurement of the particle size distribution. References [1] S. Camacho, V. van Riel, C. de Graaf, F. van de Velde, M. Stieger, Physical and sensory characterizations of oral coatings of oil/water emulsions, J. Agric. Food Chem. 62 (2014) 5789–5795. [2] G.A. Van Aken, Relating food microstructure to sensory quality, in: D.J. McClements (Ed.), Understanding and Controlling the Microstructure of Complex Foods, Woodhead Publ Ltd, Abington Hall Abington, Cambridge Cb1 6ah, Cambs, Uk 2007, pp. 449–482. [3] J.B. Hutchings, P.J. Lillford, The perception of food texture - the philosophy of the breakdown path, J. Texture Stud. 19 (1988) 103–115. [4] S. Adams, S. Singleton, R. Juskaitis, T. Wilson, In-vivo visualisation of mouth– material interactions by video rate endoscopy, Food Hydrocoll. 21 (2007) 986–995. [5] G.A. Van Aken, E.H.A.d. Hoog, M.H. Vingerhoeds, Oral Processing and Perception of Food Emulsions: The Relevance for Fat Reduction in Food, Woodhead Publishing Ltd, Cambridge, UK, 2009 481–501. [6] I.A.M. Appelqvist, Measuring the oral behaviour of foods, in: D. Kilcast (Ed.), Designing Functionnal Foods, Woodhead Publishing Limited, 2009.
[7] I.A.M. Appelqvist, In-mouth measurement the oral behaviour of foods, in: D.J.M.a.E.A. Decker (Ed.), Instrumental Assessment of Food Sensory Quality, CRC Press, Woodhead Publishing Limited 2013, pp. 255–277. [8] E.P. Köster, The psychology of food choice: some often encountered fallacies, Food Qual. Prefer. 14 (2003) 359–373. [9] B.J. Rolls, E.T. Rolls, E.A. Rowe, K. Sweeney, Sensory specific satiety in man, Physiol. Behav. 27 (1) (1981) 137–142. [10] E.P. Koster, J. Mojet, Boredom and the reasons why some new food products fail, in: H.J.H. MacFie (Ed.), Consumer-led Food Product Development, Woodhead Publishing Limited 2006, pp. 262–280. [11] E.H. Zandstra, M.F. Weegels, A.A. Van Spronsen, M. Klerk, Scoring or boring? Predicting boredom through repeated in-home consumption, Food Qual. Prefer. 15 (2004) 549–557. [12] C.M. Levy, A. MacRae, E.P. Koester, Perceived stimulus complexity and food preference development, Acta Psychol. 123 (2006) 394–413. [13] P. Pliner, The effects of mere exposure on liking for edible substances, Appetite 3 (1982) 283–290. [14] C. Sulmont-Rosse, C. Chabanet, S. Issanchou, E.P. Koster, Impact of the arousal potential of uncommon drinks on the repeated exposure effect, Food Qual. Prefer. 19 (2008) 412–420. [15] P.L.G. Weijzen, E.H. Zandstra, C. Alfieri, C. de Graaf, Effects of complexity and intensity on sensory specific satiety and food acceptance after repeated consumption, Food Qual. Prefer. 19 (2008) 349–359. [16] M. Cliff, H. Heymann, Development and use of time–intensity methodology for sensory evaluation: a review, Food Res. Int. 26 (1993) 375–385. [17] W. Lee III, M. Pangborn, Time–intensity: the temporal aspects of sensory perception, Food Technol. 40 (1986) 78–82. [18] F.R. Jack, J.R. Piggott, A. Paterson, Analysis of textural changes in hard cheese during mastication by progressive profiling, J. Food Sci. 59 (1994) 539–543. [19] D. Labbe, P. Schlich, N. Pineau, F. Gilbert, N. Martin, Temporal dominance of sensations and sensory profiling: a comparative study, Food Qual. Prefer. 20 (2009) 216–221. [20] N. Pineau, P. Schlich, S. Cordelle, C. Mathonnière, S. Issanchou, A. Imbert, E. Köster, Temporal dominance of sensations: construction of the TDS curves and comparison with time–intensity, Food Qual. Prefer. 20 (2009) 450–455. [21] A.E. Colonna, D.O. Adams, A.C. Noble, Comparison of procedures for reducing astringency carry-over effects in evaluation of red wines, Aust. J. Grape Wine Res. 10 (2004) 26–31. [22] S. Courregelongue, P. Schlich, A.C. Noble, Using repeated ingestion to determine the effect of sweetness, viscosity and oiliness on temporal perception of soymilk astringency, Food Qual. Prefer. 10 (1999) 273–279. [23] S. Zorn, F. Alcaire, L. Vidal, A. Gimenez, G. Ares, Application of multiple-sip temporal dominance of sensations to the evaluation of sweeteners, Food Qual. Prefer. 36 (2014) 135–143. [24] I.A.M. Appelqvist, In-vivo fluorescence microscopy of emulsions in mouth, Food Structure & Quality Conference, Jury's Hotel, Cork, Ireland, 2004 (unpublished data). [25] M.E. Malone, I.A.M. Appelqvist, I.T. Norton, Oral behaviour of food hydrocolloids and emulsions. Part 2. Taste and aroma release, Food Hydrocoll. 17 (2003) 775–784. [26] S. Camacho, A. van Eck, F. van de Velde, M. Steiger, Formation dynamics of oral oil coatings and their effect on subsequent sweetness perception of liquid stimuli, J. Agric. Food Chem. 63 (2015) 8025–8030. [27] D. Goulet, F. Brudevold, The effects of viscosity on oral clearance of glucose, J. Dent. Res. 65 (1986) 283. [28] D.M. Dresselhuis, E.H.A. de Hoog, M.A.C. Stuart, M.H. Vingerhoeds, G.A. van Aken, The occurrence of in-mouth coalescence of emulsion droplets in relation to perception of fat, Food Hydrocoll. 22 (2008) 1170–1183. [29] D. Gaillard, P. Passilly-Degrace, P. Besnard, Molecular mechanisms of fat preference and overeating, Ann. N. Y. Acad. Sci. 1141 (2008) 163–175. [30] R.D. Mattes, Is there a fatty acid taste? Annu. Rev. Nutr. 29 (2009) 305–327. [31] J.H.F. Bult, R.A. de Wijk, T. Hummel, Investigations on multimodal sensory integration: texture, taste, and ortho- and retronasal olfactory stimuli in concert, Neurosci. Lett. 411 (2007) 6–10. [32] R.L. McBride, A JND-scale category-scale convergence in taste, Percept. Psychophys. 34 (1983) 77–83. [33] B. Wark, B.N. Lundstrom, A. Fairhall, Sensory adaptation, Curr. Opin. Neurobiol. 17 (2007) 423–429. [34] H. Valentova, J. Pokorny, Effect of edible oils and oil emulsions on the perception of basic tastes, Nahrung 42 (1998) 406–408. [35] J. Lynch, Y.H. Liu, D.J. Mela, H.J.H. MacFie, A time-intensity study of the effect of oil mouthcoatings on taste perception, Chem. Senses 18 (1993) 121–129. [36] B.L. Wedzicha, C. Couet, Kinetics of transport of benzoic acid in emulsions, Food Chem. 55 (1996) 1–6. [37] D. Frank, I. Appelqvist, U. Piyasiri, T.J. Wooster, C. Delahunty, Proton transfer reaction mass spectrometry and time intensity perceptual measurement of flavor release from lipid emulsions using trained human subjects, J. Agric. Food Chem. 59 (2011) 4891–4903. [38] M.E. Malone, I.A.M. Appelqvist, T. Goff, J.E. Homan, J.P.G. Wilkins, Novel approach to the selective control of lipophilic flavor release in low-fat foods, Abstr. Pap. Am. Chem. Soc. 218 (1999) 120. [39] J.E. Stewart, C. Feinle-Bisset, M. Golding, C. Delahunty, P.M. Clifton, R.S. Keast, Oral sensitivity to fatty acids, food consumption and BMI in human subjects, Br. J. Nutr. 104 (01) (2010) 145–152. [40] J.E. Stewart, L.P. Newman, R.S. Keast, Oral sensitivity to oleic acid is associated with fat intake and body mass index, Clin. Nutr. 30 (6) (2011) 838–844. [41] J.E. Stewart, R.S.J. Keast, Recent fat intake modulates fat taste sensitivity in lean and overweight subjects, Int. J. Obes. 36 (6) (2012) 834–842.
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Impact of model fat emulsions on sensory perception using repeated spoon to spoon ingestion I.A.M. Appelqvist ⁎, A.A.M. Poelman, M. Cochet-Broch, C.M. Delahunty CSIRO Food & Nutrition, 11 Julius Avenue, North Ryde, NSW 2113, Australia
H I G H L I G H T S • • • • •
Sensory perception upon repeat ingestion of low and high fat emulsions was measured Fatty mouthfeel and afterfeel increased with repeated ingestion of HF emulsions Lipophilic aroma enhanced in HF indicating cross-modal interaction Sweetness perception did not change with repeated ingestion of HF or LF emulsions Findings of fatty perception might be explained by build-up of fatty oral coatings
a r t i c l e
i n f o
Article history: Received 22 June 2015 Received in revised form 14 March 2016 Accepted 28 March 2016 Available online 7 April 2016 Keywords: Fat Sensory perception Dynamic Repeated spoon ingestion Oral residue Mouth coating
a b s t r a c t Eating is a dynamic behaviour, in which food interacts with the mechanical and physiological environment of the mouth. This dynamic interaction changes the oral surfaces leaving particles of food and building up a film on the oral surfaces, which may impact on the temporal perception during the eating experience. The effect of repeated spoon to spoon ingestion of oil in water emulsion products (2%–50% w/w oil) was evaluated using descriptive inmouth and after swallowing sensory attributes. Descriptive sensory analysis indicated that fatty mouthfeel and afterfeel perception (measured post swallowing) increased with the number of spoonfuls for emulsions containing 50% fat. This effect is likely due to the build-up of oil droplet layers deposited on the mouth surfaces. There was an enhancement of fatty afterfeel intensity for 50% fat emulsions containing the more lipophilic aroma ethylhexanoate compared to ethyl butanoate, indicating a cross-modal interaction. No increase in these attributes from spoon to spoon was observed for the low oil emulsions; since most of the oil in the emulsion was swallowed and very little oil was likely to be left in the mouth. Sweetness perception increased as fat level increased in the emulsion due to an increase in the effective concentration of sugar in the aqueous phase. However, the sweetness perceived did not change from spoon to spoon, suggesting that any oil-droplets deposited on the oral surfaces did not form a complete barrier, restricting access of the sucrose to the taste buds. This study highlights the importance of measuring the dynamic nature of eating and demonstrated change in sensory perception occurring with repeated ingestion of model emulsions, which was likely due to a change in mouth environment. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Eating is a dynamic behaviour and is generally performed as a multiple ingestion process, where each mouthful of food is masticated and swallowed and then immediately repeated. Residues of foods (and beverages) often coat the oral mucosa after consumption, these are termed oral or mouth coatings [1] and as a consequence of their persistence will change the environmental conditions of the mouth. In-mouth sensory perception of food such as an emulsion depends on its microstructure, composition and how it behaves during processing in the
⁎ Corresponding author. E-mail address: [email protected] (I.A.M. Appelqvist).
http://dx.doi.org/10.1016/j.physbeh.2016.03.035 0031-9384/© 2016 Elsevier Inc. All rights reserved.
mouth [2]. When food is brought into the mouth, and depending on the consistency, it is mechanically broken down or manipulated with the tongue, and diluted with saliva to form a lubricated bolus, which can be swallowed safely [3]. The mastication process leads to close interaction between the food and the oral surfaces, where sensory receptors such as taste receptors on the tongue, flavour receptors in the nose and pressure, stress and heat receptors on the oral tissues, are activated by the food where they are integrated by the brain to form an overall perception of sensory texture and flavour attributes [2]. For emulsion based food the sensory perception often occurs after swallowing, due to the persistence of food residues and particles coating the oral surfaces [4]. This mouthcoating impacts on the oral environment changing the composition of saliva, the pH of the mouth and access to the taste and mechano-
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
receptors [5–7], which in turn plays a big role in determining aftertaste and afterfeel [1] and directly influences the sensory perception of the next mouthful of food. The dynamic nature of the eating event upon repeated consumption has also been recognised in relation to hedonic appreciation. In fact, the implicit assumption that consumers do not change has been identified by Köster [8] as one of the fallacies in the study of eating and drinking behaviour. Sensory appreciation can change with repeated consumption of the same food, and can either decrease, potentially due to sensory specific satiety [9], boredom or irritation [10,11], or increase, potentially due to increased familiarity and/or a change in perceived complexity [10,12–15], which in turn can impact on perception. Most sensory perception research is based on single sips or single bites. The dynamic nature of sensory perception is mostly considered within a mouthful, using various sensory methodologies such as Time Intensity [16,17], Progressive Profiling [18] and Temporal Dominance of Sensations [19,20]. However the dynamics of perception between mouthfuls involved in eating a portion of food are usually not considered. Indeed, sensory testing protocols purposely go to great lengths to reduce the carry-over effect from one intake to the next such as in wine-tasting through development of elaborate palate cleansing methodologies [21]. Very few sensory evaluation studies have tried to evaluate the progressive ingestion of food and determined the impact on sensory perception. One notable exception has been the study conducted by Courregelongue [22] who used repeated ingestion to determine composition and rheological factors on the temporal perception of soymilk astringency using time intensity. A more recent study by Zorn [23] used a multisip temporal dominance of sensations methodology to evaluate a range of sweeteners in order to determine changes in the temporal profile of juices. In this study results are reported for the first time on the perception of model emulsion foods throughout eating using a spoon-to-spoon methodology, which mimics closely normal eating repeat ingestion and allows the potential build-up of food emulsion residues in the mouth. The aim of this study was to determine the impact of a spoon to spoon multiple ingestion on selected flavour and texture attributes for a range of emulsions containing different levels of fat. The research was a first step to gain insights on how sensory intensity will change as consumption of the emulsion portion continues, potentially relating to an accumulation of fat emulsion residue in the mouth [24].
81
Table 1 Composition of samples. Sample
Total fat (%)
Whey protein (%)
Xanthan (%)
Sugar (%)
Ethylhexanoate (ppm)
50%-HEX 10%-HEX 2%-HEX 50%-BUT 10%-BUT 2%-BUT
50 10 2 50 10 2
1 1 1 1 1 1
0.05 0.05 0.05 0.05 0.05 0.05
2 2 2 2 2 2
25 25 25
Ethyl butanoate (ppm)
25 25 25
of whey protein in preboiled water at 50 °C using an automatic stirrer. A xanthan/sugar solution was prepared using the same process. The two solutions were combined in appropriate quantities to make a base solution. Canola oil (50% w/w) was then added and the mixture was homogenized for 2 min at 16,000 rpm and then 2 min at 20,000 rpm (Ultra-Turrax, T25 with S25N-25F dispersion tool, IKA, Germany). Separately, a solution of sugar, xanthan and emulsifier was prepared to maintain the concentration of these ingredients constant in all samples, when diluting the 50% fat emulsion in order to obtain emulsions with lower fat contents. Using required quantities of the 50% emulsion and the aqueous solution, the 10% and 2% fat emulsions were prepared. Emulsions were poured into plastic, sterile, screw-top containers. The required quantity of flavour compound, pre-diluted in ethanol, was added to the emulsion, and samples were refrigerated at 5 °C until use the following day. Thirty minutes prior to the start of the sensory evaluations, samples were removed from refrigeration. A 10 mL volume of sample was served in a 30 mL transparent plastic cup with a lid. Twelve cups of the same sample, coded 1 to 12, were placed on a three-digit coded tray and constituted one sample for the purposes of the experiment. 2.2. Particle size distribution Apparent particle size distribution was measured by laser light scattering using a Malvern Mastersizer 2000 Ver 5.60 instrument (Malvern Instruments Ltd, Worcestershire, UK). A differential refractive index of 1.101 (1.465 for particle/1.33 for water) absorption of 0.001 were used as the optical properties of the dispersion. The particle calculation was set for spherical particles. Volume median diameter value d(0.5) was used as the average particle size. Each sample was measured in duplicate.
2. Methods and materials 2.3. Descriptive sensory analysis 2.1. Samples Oil-in-water (O/W) emulsions were used as model food systems. Samples had a range of fat contents (50%, 10% and 2% fat) and contained either ethylhexanoate (HEX) or ethyl butanoate (BUT). Both flavour compounds had a fruity note but they differed in lipophilicity (HEX is more lipophilic than BUT; octanol/water partitioning coefficient (Ko/w) 641 and 80 respectively). All samples contained 2% sucrose (w/w). Thus, the sample set consisted of six samples varying in fat content and added flavour compound (Table 1). Whey protein isolate (1% w/w) was selected due to its good emulsifying capacity and neutral taste. Xanthan gum (0.05% w/w) was added to stabilize the emulsion. Canola oil and sugar were purchased from a local retailer (Woolworths, Homebrand). Whey-protein-isolate (NZMP Whey Protein Isolate 895, Fonterra, Auckland, New Zealand) was ordered via an on-line supplier. Ethylhexanoate and ethyl butanoate were sourced from Sigma Aldrich (Castle Hill, Australia) and xanthan gum was sourced from Kelco International (Chicago, USA). Tap water was used for the aqueous phase. A 50% fat emulsion was processed using a four-step procedure. A whey protein solution was made by dispersing the required quantity
All sensory activities were carried out in a testing laboratory designed in accordance with International Standards on Sensory Analysis (ISO 8589:2007). A trained sensory panel, consisting of ten assessors (2 male, 8 female, age: 44.5 ± 7.2 years) previously screened for normal sensory acuity and with extensive experience in descriptive analysis on a variety of food products, participated to the descriptive analysis of the samples. Three two-hour training sessions were held to familiarise the assessors with the samples, and reach consensus on the method of assessment and the sensory attribute definitions. The assessors were informed that the objective of this study was to gain insights in sensory perception throughout the eating event. No information was provided as to whether the samples on the tray were the same or not, assessors were simply asked to rate the perceived attribute intensities for each sample. Assessors were made aware that familiarisation with the samples would involve a limited number of samples only, and focus group discussion of samples and attributes were based on single-spoon assessments only. To aid in attribute recognition, sucrose solutions and the two pure flavour compounds were used as reference standards. Method of assessment was developed and practiced with emulsion samples and sugar/xanthan
82
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
solutions until all assessors had reached a high level of confidence in conducting procedures according to the timings required. For the evaluation of the samples assessors were presented with a tray containing 12 cups labelled 1 to 12. They were seated in booths and were guided through the procedure using Compusense® five sensory data acquisition software (version 4.6, 2004; Compusense Inc., Guelph, Ontario, Canada). Assessors opened the lid and took one level spoon of the first sample by scooping and pouring. They pressed “next question” while simultaneously putting the sample in their mouth. Assessors kept the sample in their mouth for 3 s (by silently counting 21, 22, 23) while swirling it around with their tongue and then rated the flavour/mouthfeel attributes. The sample was then swallowed while the assessors simultaneously clicked on “next question”. A timer counted down for 5 s, during which assessors threw their cup in the rubbish bin, which ensured they would have the following cup to assess in front of them. After 5 s, aftertaste/ afterfeel attributes were rated. Once attributes were rated, assessors clicked “next question”, which started a timer to count down from 10 s, during which time assessors cleaned their spoon. Assessors continued this procedure with the next cup until all twelve cups were evaluated (assessment of each cup took about 20–25 s). Assessors did not cleanse their palate between cups. The palate was cleansed between different emulsion samples (each consisting of 12 cups); water, bread, cucumber and diluted cold black tea were provided for this purpose. For each cup, assessors rated the same three in-mouth attributes sweet taste, fruity flavour and fatty mouthfeel as well as the same three after swallowing attributes, sweet aftertaste, fruity aftertaste and fatty afterfeel. The panel quantitatively evaluated the sensory properties of the six samples (each sample comprising of 12 cups of the same emulsion) in duplicate. Samples were presented in blocks separately for each flavour compound, starting with HEX. Within blocks order was randomised across assessors and evaluations were carried out over four days. Between samples a ten minute inter-stimulus interval was imposed. All evaluations were carried out in individual sensory booths under white light. Attributes were rated on 100 mm unstructured line scales anchored at 5 and 95%, respectively, with extremes for each descriptive term. Ethical approval for this study was granted by the CSIRO Low Risk Review Panel. Informed consent was obtained from assessors prior to the start of the study. 2.4. Data analysis Panel performance was checked with PanelCheck (v1.4.0, 2008) and was good (data not shown). Data were analysed with SPSS (v20.0.0, 2011). A criterion of p b 0.05 was used for statistical significance. To determine the effect of fat content and flavour compound on sensory perception of attributes measured, univariate analysis of variance was conducted at each of the twelve spoons using fat level, flavour compound, their interaction and assessor as variables in the model. Where significant differences were obtained, Bonferroni post hoc tests were performed. To determine trends throughout eating, repeated measures ANOVA was carried out, using attribute scores for cups 1 to 12 as repeated factor, and samples and assessors as fixed factors. Two-way interactions were included in the model. Area under the curve (AUC) for selected sensory attributes was calculated by summing the total of perceived intensities across all twelve spoonfuls, and differences between samples were tested using univariate ANOVA. 3. Results and discussion 3.1. Particle size distribution of fat emulsions The oil droplet size of the emulsions with different fat content was measured to determine if the particle size distribution of the oil droplets
changed with fat content. The results in Fig. 1 show that the average particle size distribution (major peak) for all the emulsion samples were the same with the majority of the emulsion droplets falling between 3 and 13 μm, with an average size of around 9 μm. The data from the 2% emulsion exhibited a small bimodal distribution with a second peak of larger oil droplets with an average size of around 80 μm, indicating there was a small amount of coalescence with some droplets. The size of the emulsion droplet has been shown to affect the rate of flavour release and therefore the flavour intensity [25]. However, the change in droplet size measured were well above 500 μm to see these effects, which was well outside the size change for this study and therefore the impact on perception would be negligible, especially as this was a small percentage of the total oil emulsion and at the lowest oil concentration used. 3.2. Effect of fat content and aroma lipophilicity on fatty mouthfeel and afterfeel The effect of fat content on the fatty mouthfeel and afterfeel for emulsions containing either ethylhexanoate or ethyl butanoate during spoon to spoon ingestion is shown in Fig. 2a (fatty mouthfeel) and Fig. 2b (fatty afterfeel) respectively. The plots of fatty mouthfeel and afterfeel perception against increasing number of spoonful of emulsion sample show that for the 2% and 10% fat emulsions the fatty mouthfeel perception and afterfeel perception were significantly lower (p b 0.0001) compared to the 50% fat emulsion. The total AUC for fatty mouthfeel and afterfeel was used to give an overall indication of fatty perception over the accumulative repeated eating of 12 spoonfuls of the model emulsion in this study. The results (Table 2) show that with increasing fat level there is an increase in overall fatty perception in mouthfeel and afterfeel (p b 0.001). Perception of fatty mouthfeel (Fig. 2a) increased throughout consumption of the twelve spoons (p b 0.0001) for all the fat levels with the increase most pronounced for the 50% fat emulsion, notably after about six spoons. Fatty afterfeel perception (Fig. 2b), which was measured after swallowing the emulsions, followed a similar trend to the fatty mouthfeel, the perception increasing with repeated spoonful (p b 0.0001). This data suggests that the food emulsions coat the mouth surfaces during mastication and remains as a residue after swallowing, which appears to increase upon repeated ingestion, indicated by the increase in texture perception with increased number of consumption spoonfuls. This observation agrees with work conducted by Camacho [1] who showed an increase in the fatty afterfeel attribute of fatty film intensity (after expectorating) as a function of increasing oil deposition of different fat content emulsions (2 to 20%) on the front and back of the anterior tongue. Previous work from Appelqvist [24] and Adams et al. [4] using fluorescent imaging of emulsion coatings containing 2% and 55% fat also indicated that the level of fat residue increased as a function of repeated ingestion, measured by an increasing intensity of fluorescence. Moreover, their data indicated that more oil mouth coating was left on the surface of the tongue with increasing fat content. It is therefore likely that the increase in sensory perception of fatty mouthfeel and afterfeel with repeated ingestion of oil in water emulsions is explained by an increase in fat residue left on the oral surfaces. Camacho et al., 2015 [26] recently quantified the amount of oil fraction that was deposited on the tongue using in vivo florescence measurements of oil emulsions ranging from 1% to 20% oil. They found a similar trend in which more oil quantified in mg/cm2 was deposited onto the oral surfaces as a function of increasing oil content in the emulsions ingested. In this study, the higher oil content emulsion also had a higher viscosity and it has been shown in previous studies that more residue is left behind as the viscosity of the sample increases [4,27]. The mechanisms of interaction between food emulsions and saliva will also be important in determining the amount of food residue left in the mouth. Positively charged protein stable emulsions as in this case stabilized
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
83
Fig. 1. Particle size distribution of emulsion samples ranging in fat content from 2 to 50%.
with whey protein isolate can interact with charged ions found in saliva so that flocculation in the presence of saliva can potentially occur [5]. This may be largely driven by depletion, van der Waals forces and/or electrostatic interaction between emulsion droplets and salivary proteins, depending on the net charge of the emulsion droplets and the presence of other charged molecules in the saliva. Dresselhuis [28] showed that oral processing of stable emulsions (emulsified with 1% WPI) did not show any in-mouth coalescence suggesting the oil was deposited as individual emulsion droplets, which was indicated in the fluorescence imaging of similar emulsions using WPI in a previous study [4]. We measured fatty mouthfeel and fatty afterfeel in our study, and therefore focused predominantly on the mouthfeel aspect of fattiness perception. However, we recognise that fattiness is a complex and multimodal sensation. In addition to the widely acknowledged contribution of fat to mouthfeel, it also affects odour and appearance [29], and recently perception of free fatty acids is being studied as potentially also being a basic taste [30].
3.3. Effect of mouthcoating on texture perception In order to compare the fatty mouthfeel and afterfeel perception, the data for all 3 fat emulsions containing ethylhexanoate and ethyl butanoate was compared in Fig. 2a and b. It can be seen that fatty afterfeel did not decrease greatly when compared to fatty mouthfeel for all repeated spoonfuls, even though most of the product would have been swallowed between repeated ingestion. Examination of the data indicates that samples having the highest perceived intensities (50% fat emulsions) showed the largest decreases in perception between product in-mouth and after swallowing. This may be explained by the greater difference in amount of fat droplets present during mastication and amount of fat droplets deposited in the oral coating after swallowing, and that perceived intensities will be better discriminated. In the case of the 2% and 10% fat emulsions, the differences between the two texture attributes were small (Fig. 2a and b). For the 10% fat emulsion containing ethylhexanoate no difference was found between the perceived fatty mouthfeel and afterfeel after 6 repeated spoonfuls. In
Fig. 2. Perceived intensity (+SE) of (a) fatty mouthfeel and (b) fatty afterfeel of six emulsions varying in fat content and added aroma ( 50% fat with ethyl hexanoate, 50% fat with ethyl butanoate, 10% fat with ethyl hexanoate, 10% fat with ethyl butanoate, 2% fat with ethyl hexanoate, 2% fat with ethyl butanoate throughout eating twelve spoons.
84
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
Table 2 Area under the curve (AUC) as calculated measure from temporal perception of fatty mouthfeel and afterfeel of twelve spoonfuls of six emulsions varying in fat content and aroma compound added. Sample
Fatty mouthfeel
Fatty afterfeel
50%-HEX 50%-BUT 10%-HEX 10%-BUT 2%-HEX 2%-BUT F value p value
650.80 a⁎
582.30 a 520.57 b 222.83 c 178.07 c 111.95 d 83.64 d 179.04 b0.0001
619.43 a 237.93 b 206.36 b 141.85 c 102.57 c 250.40 b0.0001
⁎ Samples sharing the same letter were not significantly different from each other.
these samples the oil fraction in the samples and that in the deposited oral coatings may not vary greatly and therefore differences in perceived fatty texture would also be small. The overall fatty mouthfeel and afterfeel impression was also scored for emulsions containing ethyl butanoate after twelve spoonfuls. The data (not shown) indicates that the overall intensity appears to be informed by an averaging of perceived intensities over the 12 spoonfuls of product. 3.4. Cross-modal interactions between aroma and texture perception The data in Fig. 2a and b also indicates that the lipophilicity of aroma added to the model fat emulsions had an impact on the perceived fatty mouthfeel and afterfeel for each of the fat levels studied. The data indicate that the 50% fat emulsions containing the higher lipophilic aroma ethylhexanoate was perceived as significantly more fatty in both mouthfeel and afterfeel from spoonful 7 onwards compared to the same emulsion containing ethyl butanoate. These results are supported by a greater number for area under the curve for the 50% fat emulsion containing ethylhexanoate compared with the 50% ethyl butanoate, which was significant in fatty afterfeel, and trending in the same direction (although not significant) for fatty mouthfeel (Table 2). This suggests that there was a cross-modal interaction between aroma and texture, where ethylhexanoate appears to enhance the perceived fatty mouthfeel and afterfeel. One possible explanation is that ethylhexanoate preferentially resides in the oil droplets (as it is more lipophilic) and therefore more aroma molecules remain in the fat emulsion during eating and in the emulsion mouthcoating, enhancing perception of fatty mouthfeel and afterfeel. Aroma has been shown to change texture perception, for instance Bult and co-workers [31] showed that the presence of a ‘creamy’ aroma increased the perceived thickness and creaminess of a range of semi-solid products. 3.5. Impact of spoon to spoon ingestion on taste and flavour perception 3.5.1. Sweet taste Fig. 3 shows the effect of continual food consumption across 12 repeated spoonfuls for the different fat level emulsions, for sweet taste (3a and 3b) and fruity flavour perception (3c and 3d). Fig. 3a and c is the emulsions containing ethylhexanoate and Fig. 3b and d is the emulsions containing ethyl butanoate. The data indicates that for sweet taste the 50% emulsion samples (for both emulsions containing ethyl butanoate and ethylhexanoate) were perceived as sweeter (p b 0.0001), with a 10% difference on average at each time point, compared to the 10% and 2% emulsion samples, and that no perceived significant difference in sweetness was observed between the 10% and 2% fat samples themselves (Fig. 3a and b). The higher sweetness perception for the 50% emulsion samples (based on a constant sugar concentration of 2% on product) was due to an increase in the effective concentration of sugar in the aqueous phase, since sugar is not soluble in fat. Hence the effective sugar concentration in the 50% aqueous phase was 4% w/w, which
compares to 2.04% w/w for the 2% emulsion and 2.2% w/w for the 10% emulsion, explaining why they were perceived to be lower in sweetness compared to the 50% fat emulsion samples. Having similar sugar concentrations of around 2% for the 2% and 10% fat samples also explains the finding that the sweet taste intensity rated was the same. This effect of o/w emulsion phase volume was also shown for salty taste by Malone [25], who observed that using a constant salt concentration on product, perceived saltiness increased with increasing oil content, and was associated with the increase in salt concentration in the aqueous phase. It is noteworthy that the difference in sweetness perception between different fat level emulsions is significant at all spoonfuls with exception of the first spoonful, where the difference in sweetness perception between the samples was not statistically significant. These results indicate that perceptual differences may become more apparent when ingesting more of the same food, and demonstrates the potential risk of relying on single sips or bites in describing a food's sensory properties. It is also clear from the data for sweet taste (Fig. 3a and b) that sweetness does not develop upon spoon to spoon even though mouth coating may increase (statistically there is no significant change in sweet taste or aftertaste). This could be due to the fact that the oil droplets are likely to be preferentially deposited onto the oral surfaces (and composes the bulk of the oral coating) and that most of the aqueous phase is cleared through mixing and dilution with saliva and then swallowed. Therefore the sugar concentrated in the aqueous phase gets washed away during the swallowing and cleansing process. This hypothesis may be reasonable given that the sweet aftertaste for all the fat emulsions superimpose on one another from spoonful to spoonful and are overall lower than the in-mouth sweetness perception (Fig. 3a and b). This indicates that the aqueous phase is quickly diluted with saliva over the 20 second time period between spoonful consumption and the differences in sugar concentration left in mouth after each swallowing are below the just noticeable differences for sweetness [32]. Adaptation may also play a part in which any potential build-up of sugar on the mouth surface is counteracted by decrease in sweetness perception due to adaptation which can take up to 45 to 60 s to recover [33]. Recently, Camacho et al. [26] studied the impact of oil emulsion coatings on the perception of sweetness and saltiness of tastant solutions. It had been shown previously that as the level of oil coating increased a decrease in taste modality was observed [34]. It was suggested that the oil deposited on the oral surfaces create a hydrophobic barrier, reducing the accessibility of the hydrophilic taste molecules to the receptors [35]. Our data on sweet taste did not show a reduction in sweetness perception with additional mouthfuls of emulsion and was independent of the increase in perception of fatty mouthfeel. This indicates that although more oil was probably deposited onto the oral surface, the oil did not form a homogeneous film, but rather is ‘patchy’ in nature and small molecules such as sucrose are freely accessible and can interact with the taste buds. This could be related to the underlying structure of the oral surface but also perhaps indicating the inefficient and irregular behaviour of in-mouth processing in clearing the surface of the buccal mucosa after eating food. Finally, the data also indicates that aroma compounds selected in this study did not affect perceived sweetness. At each fat level, there was no perceived significant difference in sweetness between samples spiked with ethylhexanoate and ethyl butanoate. 3.5.2. Fruity flavour The combined dataset for fruity flavour with emulsions containing either ethyl butanoate or ethylhexanoate (Fig. 3c and d) showed that fat content impacted on the release of the lipophilic aroma compounds (p b 0.01). The lowest oil content (2% emulsion) resulted in a higher perceived intensity of fruity flavour. Generally, in standard o/w emulsions the rate of inter-phase transport of aroma molecules from the oil to the aqueous phase occurs very fast [36]. In the case of low fat o/w emulsions, the lipophilic aroma molecules during eating are rapidly stripped from the oil phase, creating a strong driving force for rapid
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
85
Fig. 3. Perceived intensity (+SE) of (a–b) sweet taste and (c–d) fruity flavour of emulsions varying in fat content and added aroma throughout eating twelve spoons; (a and c) 50% fat with ethyl hexanoate, 10% fat with ethyl hexanoate, and 2% fat with ethyl hexanoate and (b and d) 50% fat with ethyl butanoate, 10% fat with ethyl butanoate, and 2% fat with ethyl butanoate.
mass transfer of aroma from the oil phase to the aqueous phase and headspace [37]. This in turn increases the number of aroma molecules transferred to the olfactory receptors increasing the perceived fruitiness intensity. Previous work by Malone [25,38] and Frank [37] measuring dynamic flavour release demonstrated that the amount of aroma released within a given time decreased with increase in fat content, where fat lowers the vapour pressure of lipophilic aromas (Ko/w ≫ 1) and reduces the rate at which they are released during mastication. Fig. 3c shows a slight decrease in fruity flavour intensity for the 50% fat emulsion containing ethylhexanoate with repeated ingestion. This fits with theory in that as more oil is deposited onto the mouth surface as a mouth coating, the rate of release of this lipophilic aroma will be decreased. Adaptation may also play a role. However, there was no statistically significant trend from spoonful to spoonful in perception of fruity flavour or fruity aftertaste across all emulsions. The area under the curve (AUC) was determined from the fruity flavour data of in-mouth and aftertaste and has been used to give an indication of amount of volatile released over the accumulative repeated eating of 12 spoonfuls (Table 3). The results show that the release of ethyl hexanoate and ethyl butanoate decreased with increasing fat level due to the reasons described in the previous section.
4. Conclusions and practical implications This work clearly demonstrates that spoon to spoon ingestion of a model emulsion impacts on the sensory perception of the food, increasing overall fatty mouthfeel and afterfeel perception over time as a function of number of spoons, especially when the fat content is high (~50% fat w/w). This suggests that with each mouthful, an emulsion coating is deposited on the oral surfaces, which builds up with subsequent intake of food and is persistent after swallowing.
The data also indicates that for the most lipophilic aroma compound ethylhexanoate, there is indication of a cross modal interaction between the fatty texture (mouthfeel and afterfeel) and flavour intensity, resulting in an enhancement of fatty mouthfeel perception, probably because more ethylhexanoate aroma molecules remain in the fat phase compared to ethyl butanoate and is integrated into the perception of fatty texture. Changing the phase volume of fat impacted on the intensity of the sweet taste perception because the sugar level was formulated on product rather than on aqueous phase, so that the effective concentration of sugar in the aqueous phase increased with increasing fat levels. Interestingly, the sweetness perceived did not change with each spoonful of product, suggesting the increase in fatty mouthfeel, probably from deposition of an oral oil coating did not suppress sweetness perception by reducing accessibility of the sugar molecules to the receptors. It was also noted that differences in sweet taste perception of samples of different fat content became apparent only after the first spoonful, demonstrating the value of repeated ingestion for accurate description Table 3 Area under the curve (AUC) as calculated measure from temporal perception of fruity flavour and aftertaste of twelve spoonfuls of six emulsions varying in fat content and aroma compound added. Sample
Fruity flavour
Fruity aftertaste
50%-HEX 50%-BUT 10%-HEX 10%-BUT 2%-HEX 2%-BUT F value p value
402.05 c⁎ 430.50 bc 440.58 bc 485.93 ab 507.00 a 551.82 a 4.60 0.02
356.95 c 361.50 bc 393.65 bc 384.14 bc 419.55 ab 439.64 a 2.90 0.02
⁎ Samples sharing the same letter were not significantly different from each other.
86
I.A.M. Appelqvist et al. / Physiology & Behavior 160 (2016) 80–86
and differentiation of sensory properties in comparison with single sip/ bite methods. Finally the overall impression of intensity measured after the 12 spoonfuls appears to be related to the average rated responses across all spoonfuls, suggesting that the overall perception is not accumulative and not necessarily biased by the last spoonful taken. 4.1. Practical implications These results have several implications for the development of food products. Fatty mouthfeel is an attribute that contributes to food acceptance in semi-solid food products such as yogurts, creams and mousses. This study shows a build-up of fatty mouthfeel after eating sequential spoonful of product containing high fat (50% fat emulsion) and it is likely that this behaviour is one reason why consumers prefer high fat products such as chocolate or cream, which coats the mouth and can be perceived long after they are swallowed. This means that when food companies are looking to reformulate their products with reduced fat content, they should also consider how they might mimic the same oral mouth coating profile provided by the high fat product counterpart to increase acceptance. Recent research on perception of fat taste has shown differences between individuals in fat taste sensitivity which is in part dependent on fat intake and modifiable [39–41]. Although this study did not look at individual differences in sensory perception, further research to determine individual physiological and genetic differences in taste perception could enhance understanding on how best to reformulate products. In addition, these findings could be useful when developing food products for patients with swallowing difficulties (e.g. dysphagia) to understand how products might behave from one spoonful to the next. As an example, the perception of fatty mouthfeel is an indicator of the amount of fat in a food. The fat content of food is an important functional parameter that provides lubrication and contributes to the formation of a bolus that is safe to swallow. However, the build-up of a mouthcoating beyond a certain level could reverse the effect and make it difficult for some people to swallow. A spoon-to-spoon methodology, such as described in this study, could be applied to gain insights on the safety/suitability of the product for dysphagia patients. Development of methods to measure fat residence time in the mouth and distribution on oral surfaces could be used in food product design to better determine the temporal dynamics of perception, optimise sensory properties and develop foods that are safe to swallow for vulnerable consumer groups. Acknowledgments Tim Wooster is acknowledged for his help in designing the emulsions, and Mi Xu is acknowledged for her help in the measurement of the particle size distribution. References [1] S. Camacho, V. van Riel, C. de Graaf, F. van de Velde, M. Stieger, Physical and sensory characterizations of oral coatings of oil/water emulsions, J. Agric. Food Chem. 62 (2014) 5789–5795. [2] G.A. Van Aken, Relating food microstructure to sensory quality, in: D.J. McClements (Ed.), Understanding and Controlling the Microstructure of Complex Foods, Woodhead Publ Ltd, Abington Hall Abington, Cambridge Cb1 6ah, Cambs, Uk 2007, pp. 449–482. [3] J.B. Hutchings, P.J. Lillford, The perception of food texture - the philosophy of the breakdown path, J. Texture Stud. 19 (1988) 103–115. [4] S. Adams, S. Singleton, R. Juskaitis, T. Wilson, In-vivo visualisation of mouth– material interactions by video rate endoscopy, Food Hydrocoll. 21 (2007) 986–995. [5] G.A. Van Aken, E.H.A.d. Hoog, M.H. Vingerhoeds, Oral Processing and Perception of Food Emulsions: The Relevance for Fat Reduction in Food, Woodhead Publishing Ltd, Cambridge, UK, 2009 481–501. [6] I.A.M. Appelqvist, Measuring the oral behaviour of foods, in: D. Kilcast (Ed.), Designing Functionnal Foods, Woodhead Publishing Limited, 2009.
[7] I.A.M. Appelqvist, In-mouth measurement the oral behaviour of foods, in: D.J.M.a.E.A. Decker (Ed.), Instrumental Assessment of Food Sensory Quality, CRC Press, Woodhead Publishing Limited 2013, pp. 255–277. [8] E.P. Köster, The psychology of food choice: some often encountered fallacies, Food Qual. Prefer. 14 (2003) 359–373. [9] B.J. Rolls, E.T. Rolls, E.A. Rowe, K. Sweeney, Sensory specific satiety in man, Physiol. Behav. 27 (1) (1981) 137–142. [10] E.P. Koster, J. Mojet, Boredom and the reasons why some new food products fail, in: H.J.H. MacFie (Ed.), Consumer-led Food Product Development, Woodhead Publishing Limited 2006, pp. 262–280. [11] E.H. Zandstra, M.F. Weegels, A.A. Van Spronsen, M. Klerk, Scoring or boring? Predicting boredom through repeated in-home consumption, Food Qual. Prefer. 15 (2004) 549–557. [12] C.M. Levy, A. MacRae, E.P. Koester, Perceived stimulus complexity and food preference development, Acta Psychol. 123 (2006) 394–413. [13] P. Pliner, The effects of mere exposure on liking for edible substances, Appetite 3 (1982) 283–290. [14] C. Sulmont-Rosse, C. Chabanet, S. Issanchou, E.P. Koster, Impact of the arousal potential of uncommon drinks on the repeated exposure effect, Food Qual. Prefer. 19 (2008) 412–420. [15] P.L.G. Weijzen, E.H. Zandstra, C. Alfieri, C. de Graaf, Effects of complexity and intensity on sensory specific satiety and food acceptance after repeated consumption, Food Qual. Prefer. 19 (2008) 349–359. [16] M. Cliff, H. Heymann, Development and use of time–intensity methodology for sensory evaluation: a review, Food Res. Int. 26 (1993) 375–385. [17] W. Lee III, M. Pangborn, Time–intensity: the temporal aspects of sensory perception, Food Technol. 40 (1986) 78–82. [18] F.R. Jack, J.R. Piggott, A. Paterson, Analysis of textural changes in hard cheese during mastication by progressive profiling, J. Food Sci. 59 (1994) 539–543. [19] D. Labbe, P. Schlich, N. Pineau, F. Gilbert, N. Martin, Temporal dominance of sensations and sensory profiling: a comparative study, Food Qual. Prefer. 20 (2009) 216–221. [20] N. Pineau, P. Schlich, S. Cordelle, C. Mathonnière, S. Issanchou, A. Imbert, E. Köster, Temporal dominance of sensations: construction of the TDS curves and comparison with time–intensity, Food Qual. Prefer. 20 (2009) 450–455. [21] A.E. Colonna, D.O. Adams, A.C. Noble, Comparison of procedures for reducing astringency carry-over effects in evaluation of red wines, Aust. J. Grape Wine Res. 10 (2004) 26–31. [22] S. Courregelongue, P. Schlich, A.C. Noble, Using repeated ingestion to determine the effect of sweetness, viscosity and oiliness on temporal perception of soymilk astringency, Food Qual. Prefer. 10 (1999) 273–279. [23] S. Zorn, F. Alcaire, L. Vidal, A. Gimenez, G. Ares, Application of multiple-sip temporal dominance of sensations to the evaluation of sweeteners, Food Qual. Prefer. 36 (2014) 135–143. [24] I.A.M. Appelqvist, In-vivo fluorescence microscopy of emulsions in mouth, Food Structure & Quality Conference, Jury's Hotel, Cork, Ireland, 2004 (unpublished data). [25] M.E. Malone, I.A.M. Appelqvist, I.T. Norton, Oral behaviour of food hydrocolloids and emulsions. Part 2. Taste and aroma release, Food Hydrocoll. 17 (2003) 775–784. [26] S. Camacho, A. van Eck, F. van de Velde, M. Steiger, Formation dynamics of oral oil coatings and their effect on subsequent sweetness perception of liquid stimuli, J. Agric. Food Chem. 63 (2015) 8025–8030. [27] D. Goulet, F. Brudevold, The effects of viscosity on oral clearance of glucose, J. Dent. Res. 65 (1986) 283. [28] D.M. Dresselhuis, E.H.A. de Hoog, M.A.C. Stuart, M.H. Vingerhoeds, G.A. van Aken, The occurrence of in-mouth coalescence of emulsion droplets in relation to perception of fat, Food Hydrocoll. 22 (2008) 1170–1183. [29] D. Gaillard, P. Passilly-Degrace, P. Besnard, Molecular mechanisms of fat preference and overeating, Ann. N. Y. Acad. Sci. 1141 (2008) 163–175. [30] R.D. Mattes, Is there a fatty acid taste? Annu. Rev. Nutr. 29 (2009) 305–327. [31] J.H.F. Bult, R.A. de Wijk, T. Hummel, Investigations on multimodal sensory integration: texture, taste, and ortho- and retronasal olfactory stimuli in concert, Neurosci. Lett. 411 (2007) 6–10. [32] R.L. McBride, A JND-scale category-scale convergence in taste, Percept. Psychophys. 34 (1983) 77–83. [33] B. Wark, B.N. Lundstrom, A. Fairhall, Sensory adaptation, Curr. Opin. Neurobiol. 17 (2007) 423–429. [34] H. Valentova, J. Pokorny, Effect of edible oils and oil emulsions on the perception of basic tastes, Nahrung 42 (1998) 406–408. [35] J. Lynch, Y.H. Liu, D.J. Mela, H.J.H. MacFie, A time-intensity study of the effect of oil mouthcoatings on taste perception, Chem. Senses 18 (1993) 121–129. [36] B.L. Wedzicha, C. Couet, Kinetics of transport of benzoic acid in emulsions, Food Chem. 55 (1996) 1–6. [37] D. Frank, I. Appelqvist, U. Piyasiri, T.J. Wooster, C. Delahunty, Proton transfer reaction mass spectrometry and time intensity perceptual measurement of flavor release from lipid emulsions using trained human subjects, J. Agric. Food Chem. 59 (2011) 4891–4903. [38] M.E. Malone, I.A.M. Appelqvist, T. Goff, J.E. Homan, J.P.G. Wilkins, Novel approach to the selective control of lipophilic flavor release in low-fat foods, Abstr. Pap. Am. Chem. Soc. 218 (1999) 120. [39] J.E. Stewart, C. Feinle-Bisset, M. Golding, C. Delahunty, P.M. Clifton, R.S. Keast, Oral sensitivity to fatty acids, food consumption and BMI in human subjects, Br. J. Nutr. 104 (01) (2010) 145–152. [40] J.E. Stewart, L.P. Newman, R.S. Keast, Oral sensitivity to oleic acid is associated with fat intake and body mass index, Clin. Nutr. 30 (6) (2011) 838–844. [41] J.E. Stewart, R.S.J. Keast, Recent fat intake modulates fat taste sensitivity in lean and overweight subjects, Int. J. Obes. 36 (6) (2012) 834–842.