**1. Introduction**

Choice of food is driven by numerous biological, economic, physical and sociopsychological determinants. The changes in global dietary patterns are influenced by various trends and factors. For example, when eating out, consumers do not just want to meet their physiological needs but are in search of food that brings emotional benefits and gives the pleasure of new flavors and tastes [1–3]. There are many factors that contribute to food satisfaction for consumers' eating out experiences. The most common factors include satiation, sensory experience, food variety and quality (nutritional value, origin, healthiness, visual attractiveness, freshness), financial aspects (price/value, promotion), company of friends or family [4–21], restaurant service, food presentation, taste of the food and the physical details of the restaurant (e.g., interior colors and design, music and mood) [6,8,16,18–21].

**Citation:** Głuchowski, A.; Czarniecka-Skubina, E.; Kostyra, E.; Wasiak-Zys, G.; Bylinka, K. Sensory Features, Liking and Emotions of Consumers towards Classical, Molecular and Note by Note Foods. *Foods* **2021**, *10*, 133. https://doi.org/ 10.3390/foods10010133

Received: 29 November 2020 Accepted: 8 January 2021 Published: 10 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Molecular cuisine and Note by Note cooking (NbN) are novelties that have appeared in recent years. Molecular cooking is typically defined as producing food in professional kitchens using "new" tools (siphons, evaporators, etc.), ingredients (food additives, surprising texturizers) and methods (sous-vide cooking, flash freezing, etc.) [22]. Modern cuisine uses the "food deconstruction" technique, which consists of preparing a classic dish, with its same ingredients and flavors, and presenting it in a very unconventional and nontraditional manner [23]. Note by Note cuisine is the use of pure compounds or their mixtures in the production of dishes. Compounds are used to design the shape, color, taste, odor, temperature, trigeminal stimulation, texture and nutritional aspects of the developed meal [24–27]. The concept of modern cuisines is to create avant-garde, unique dishes that surprise customers. Modern restaurants' dishes should provide a multisensory experience involving emotions, memory, culture [28] and sensory perception, especially by aesthetic visual appearance. It also typically provides rich contrasts of flavors (through, for example, surprising ingredient pairings), textures and temperatures [29,30]. This sensory incongruity and the associated surprise evoke various responses from different people that depend on a consumer's fear and willingness to try new foods [31].

Consumers still do not trust new food technologies, especially when they do not know their effect on human health and the environment [32]. Acceptance of the modification of traditional food depends on their character that results from the changes. Innovations like diversification of shapes and textures and unexpected combinations of ingredients to create new flavors were perceived as the least acceptable and as the most harmful to the traditional character of food [33].

Success of markets for new food technologies depends on consumers' behavioral responses and emotional states [34,35] and is primarily associated with sensory properties [36]. Disgust, food neophobia and related features have been recognized as the main barrier to accepting novel food [37], and negative and suspicious consumer attitudes towards food technologies may lead to product failure [34]. Additionally, the preferences of modern consumers for food prepared at catering establishments are contradictory. On the one hand, consumers appreciate the naturalness of food [38], which is perceived as a crucial feature [39]. On the other hand, consumers accept novel technologies like modern cuisine, which contains food additives [40].

The evaluation of molecular and Note by Note meals is not easy because the idea of creating them is based on the constant novelty and uniqueness of dishes. Our results previously revealed a positive attitude of consumers toward molecular dishes, their moderately high sensory-experience ratings, and openness towards new ideas [41].

According to the bidirectional effect, foods that people consume affect their emotions and vice versa. Most of the choices associated with eating behavior are unarticulated and occur without insight or awareness. Thus, facial expression measurements reflect the dynamic sequence of emotional responses, whereas non-verbal reports correspond to the emotional status at the end of assessment [42].

There is no comprehensive sensory and consumer research in the literature [2,41,43–49] on the molecular and Note by Note meals in both cognitive impression and hedonic terms, which requires a special methodological approach.

It is worth determining the relationship between the expected and the experiencedliking of food products. The sensory expectations are associated with expectations of sensory descriptors such as sweetness and creaminess (at certain intensity), while the hedonic expectations are related to the extent the consumer likes or dislikes the food products [50]. According to Piqueras-Fiszman and Spence [51], the consumer's expectations and perceptions may match or diverge; i.e., their expectations may be confirmed or disconfirmed. This phenomenon likely depends on the consumers and their attitudes, beliefs, personality and product familiarity. Many authors describe new, modernist dishes as a phenomenon in the restaurant business but have not studied sensory quality and emotions associated with it.

The aims of the study were (1) to determine the sensory characteristics of traditional, molecular, and Note by Note dishes; (2) to evaluate liking of key attributes (including expected- and experienced-liking) by consumers and their declared perceptibility of features and emotions in relation to dishes and (3) to observe the type and level of emotions and facial expressions of consumers in response to consumed dishes. As modernist cuisine is a relatively new alternative form of meals, food neophobia and innovativeness level of consumers have been determined. Willingness to try them in the future was also determined.

#### **2. Materials and Methods**

#### *2.1. Material*

A total of 6 dishes were prepared using lemon or tomatoes made in the traditional, molecular and Note by Note versions (Table 1). As a reference point, two traditional dishes were selected: tomato soup with rice and lemon butter cookies. The molecular version of the conventional dishes was prepared by modifying their texture, shape and/or temperature. The NbN version was designed by replacing conventional ingredients with as many technologically possible chemical compounds, distillates and NbN evocations (IQEMUSU SAS, France). A short ingredients list and description of the dishes' preparation are presented in Appendix A (Table A1).



The idea was to obtain completely new dishes that capture the quintessence of classic meals. The dishes (Figure 1) were designed and prepared by a professional chef who has worked in a Michelin-starred restaurant. The plan of the research was to evaluate the new version of foods that were visually different but in congruence with color and taste/flavor of classic products (lemon cakes) alongside dishes that were not in congruence in color (for example of tomato soup). Typical gastronomy equipment, and also Thermomix®VORVERK (Vorwerk Poland, Wrocław), Pacojet®2 (Zug, Switzerland) for ultra-comminution, and Centrifuge 5804 R (Eppendorf, Hamburg, Germany) for centrifugation were used for the preparation of molecular and NbN versions of dishes. The prepared products were complex and involved the presence of a sous-chef during all measurements. The dishes were not suitable for storage, so the chef prepared them immediately before the assessment.

#### *2.2. Experiment Design and Methods*

The study design involved two types of research:


In the sensory analytical test, profiling was performed by trained panelists to provide objective data about products. The consumer tests included a survey aimed to characterize the tested group, as well as hedonic scaling and CATA to explore, respectively, liking of some attributes and declared perception of sensory features and emotions by

consumers toward dishes. To investigate the mimic expressions assessment of consumers, the FaceReader was used.

**Figure 1.** Plating proposition of: (**A**) lemon cookie, (**B**) molecular "lemon cookie", (**C**) Note by Note "lemon cookie", (**D**) traditional tomato soup, (**E**) molecular "tomato soup', (**F**) Note by Note 'tomato soup".

#### 2.2.1. Sensory Analytical Method

*Sensory profiling*. The assessment was carried out using a Quantitative Descriptive Analysis. Sensory descriptors were selected and defined in accordance with the procedure of ISO 13299:2016 [52]. As a result, 16 of them were chosen for lemon-based dishes and 18 for tomato ones. The intensity of sensory attributes was assessed on an unstructured linear graphic scale (100 mm long), with specific word anchors on the edges (none on the left to very intensive on the right). The evaluation was conducted in three repetitions by a group of 8 panelists, who held expert qualifications [53] and have extensive experience in sensory analysis of innovative products.

#### 2.2.2. Consumer Study

Dishes testing included a survey and hedonic assessment performed by consumers (hedonic rating, CATA, FaceReader). Participants were informed that they would assess traditional dishes and their modernist versions.

*Questionnaire* was validated in a group of 4 people and consisted of 9 questions on their familiarity with the molecular and Note by Note cuisines' terms and the frequency of their consumption. To estimate food neophobia level, the Food Neophobia Scale (FNS) by Pilner and Hobden [54] was used, while consumer innovativeness level was measured using the Domain-Specific Innovativeness (DSI) by Goldsmith and Hofacker [55] with modifications of Huotilainen et al. [56]. Ten statements applied from FNS and six from DSI were assessed on a 7-point scale and ranged from 1 = "strongly disagree" to 7 = "strongly agree" (Appendix A, Table A2). Consumers were also tasked with determining their willingness to try similar dishes in the future.

*Hedonic scaling.* A 9-point hedonic scale with expressions ranging from "extremely dislike" to "extremely like" [57] was used to quantify the degree of appearance, odor, taste/flavor, consistency and overall-liking (experienced). Additionally, prior to sensory testing, consumers were asked to observe samples and indicate their expected overallliking of dishes based on their associations related to taste, flavor, and texture of presented dishes. The answer alternatives ranged from "extremely dislike" to "extremely like".

*Check-All-That-Apply (CATA)*. The multiple-choice method was used to evaluate sensoryhedonic sensations and emotions associated with the dish consumption. Consumers had a choice of 22 different emotional attributes (surprising, common, intriguing, typical, disappointing, traditional, dietetic, artificial, delicate, rich, unique, flat flavor, atypical, tasty, interesting, innovative, trendy, boring, fresh, natural, aesthetic and processed).

In our study, attributes were selected by the assessors with a panel leader (after the profiling sessions of the dishes) and on the basis of a literature review made by the authors that was related to for example the use of CATA questions for product evaluation [58,59]. Knowing the sensory characteristics of the dishes was important in proposing the attributes to be identified. The final list of terms was developed by taking into consideration the (1) specificity of the assessed dishes, (2) panel discussion basing on the literature review made by the authors, (3) clarity of meaning terms for consumers (after translation into national language). The coordinator of the consumer assessment explained the principle of CATA questions prior to the evaluation of dishes [60]. Consumers had time to familiarize themselves with the selected attributes and ask questions. No problems with understanding the attributes were reported by consumers.

*Mimic expressions assessment.* The FaceReader 6 analyzing software (Noldus Information Technology, Wageningen, The Netherlands), connected to a web camera facing participants (angle < 40◦), was used to assess mimic expressions in relation to the presented and tested dishes. The program, based on 491 model points, allows the recognition and real-time recording of seven facial reaction patterns: happy, sad, angry, surprised, scared, disgusted and neutral. A more detailed description of the software was published in a previous article [61,62]. The intensity of measured emotion was presented on a numerical scale, wherein 0 means not presented at all and 1 means the maximum value of the fitted model. The procedure for evaluation of the dishes was explained to participants before measurement began. In this study, due to various product textures and moments of swallowing, the emotions of subjects were recorded after the sample was placed in the mouth and for the next 50 s.

#### 2.2.3. Characteristics of Consumers

*The consumer testing* (CATA, hedonic scaling) was conducted in a group of 56 students and academics, who had reported any food allergies. Most were women (88.4%) and under the age of 25 (80.3%) Table 2. The FaceReader mimic expression was evaluated by 15 participants who declared a good mood and agreed to take part in the test.

FNS and DSI scales were used to sort respondents into appropriate groups. The range of possible FNS scores was from 10 to 70 c.u., wherein higher values corresponded to a greater neophobia level. The actual scores ranged from 12 to 50. The respondents were divided into three groups: the most neophilic (10.0–18.3), the most neutral (18.4–33.8) and the most neophobic subjects (33.9–70.0), in which the cutoff points were calculated by adding or subtracting one standard deviation (7.6) from the mean value (25.9). This classification has been applied in many studies, as reported by Vidigal et al. [32]. DSI scores could range between 6 and 42, but the actual scores were between 10 and 42. The mean value in the group was 26.3 (SD ± 5.6). Based on 33rd and 66th percentile points as cutoffs, consumers were classified into three groups: adapters (6–23), neutrals (24–28) and innovators (29–42).

More than half of the group (65.2%) was classified as the most neutral, while 20.5% of subjects were regarded as the most neophilic, that is, eager to try new products (Table 2). This is also supported by a higher percentage of innovators (34.8%) when compared with adapters (27.7%).

The majority (90.2%) of participants were familiar with the molecular cuisine term, but a smaller percentage (only 33.9%) had tasted it before (Table 2). The concept of the Note by Note cuisine is relatively recent, hence, not many consumers (26.8%) participating in the survey had encountered it before.


**Table 2.** Characteristic of participants (*n* = 56).

\* Adapted Food Neophobia Scale (10–70 c.u.) [34]; \*\* Domain-Specific Innovativeness Scale (6–42 c.u.) (Goldsmith and Hofacker [55] with modifications of Huotilainen et al. [56]).

#### 2.2.4. Sample Presentation

Samples of the food (Table 1) were prepared and served immediately. Due to their different textures and consistencies, each was served at the temperature at which it would normally be consumed. They were given as a whole dish for sensory evaluation. Tomato soup and rice soufflé were served at 65 ◦C; lemon sorbet at −10 ◦C; cookie sphere and lemon consommé at 5 ◦C; lemon cookie and puffy rice snack at room temperature. The samples in appropriate quantities (20 g, the only exception was puffy rice snack, which weighed 10 g), were placed in plastic, transparent containers (100 mL) and then covered with a lid, marked with a 3-digit code and given to the assessors in a random order using the sequential monadic test [63]. Natural water was provided as a taste neutralizer between products.

#### 2.2.5. Testing Condition

Both sensory profiling and consumer study were carried out in an accredited sensory laboratory (contract No AB 564) that meets the requirements of ISO 8589:2007 [64]. The products were assessed in the same lighting and conditions to ensure focus on the assessor's perception of sensory characteristics, emotions and liking.

To perform sensory profiling of the products and collect data, the computerized system ANALSENS was used. Panelists that had been trained in sensory analysis were used for profiling in two sessions per day, with sufficient relaxation time (3 h) interval between them. The assessors taking part in the test had about 15 years of experience in profiling of various products, including dishes.

Consumers evaluated the samples during a session in the afternoon of an assessment day that lasted approximately 25 min.

*Mimic expressions* of participants in relation to each food were observed during two sessions that lasted 8 min each and were separated by relaxation time of about 12 min. Lemon-based meals were followed by tomato-based meals.

#### *2.3. Data Analysis*

Statistical analyses were performed using XLStat 2017 (Addinsoft, Paris, France). The Shapiro-Wilk test was used to verify the normality of data distribution. The results were considered to be statistically significant at the level of materiality equal to 0.05.

Analysis of variance (ANOVA) with Fisher's Least Significant Difference (LSD) post hoc test was performed to examine the differences in the intensity of attributes between tomato and lemon-based meals (traditional, molecular, NbN) considering products, assessors, and their interactions as fixed variables (the model two-way ANOVA with interactions).

Pearson correlation coefficients were calculated to relate the liking of given sensory characteristics and results of hedonic scaling of facial expression measurements in consumer tests.

The Kruskal–Wallis test was performed to analyze the differences in degree of liking of examined dishes. The relationships between degree of expected/experienced-liking and liking of particular sensory traits were analyzed by Spearman's rank correlation.

Frequency of use of each one of the terms of the CATA question was determined by counting the number of consumers using a particular term to describe each product. To compare the type of dishes on CATA, results of the correspondence analysis were used. Cochran's Q test was applied to CATA counts to determine whether there was significant difference in consumer perception for a given attribute among the examined dishes in terms of their preparation process. If significant differences were found among the variables, post hoc multiple pairwise comparisons were carried out using McNemar's test with Bonferroni alpha adjustment. Correspondence analysis, based on chi-square distance, was performed to visualize associations between the CATA attributes and the tested products.

To determine the intensity of mimic expressions (FaceReader) in consumption time, the facial expressions of subjects were recorded after the sample was placed in the mouth and for the next 50 s. Then, from 750 records (individual face shots) per person consisting of 7 numbers quantifying emotions, 11 records corresponding to every 5 s of measurement were extracted, and the average for the group was calculated. To determine the intensity of the expressions for swallowing-related time, the moment of swallowing for individuals was determined using video recordings; then, 6 records representing −10, −5, 0, +5, +10, +15 in relation to swallowing were extracted, and the average for the group was computed.

#### **3. Results**

#### *3.1. Sensory Characteristics of Different Dishes*

The modern dishes may differ in sensory properties from traditional counterparts despite similar general assumptions. According to the "food deconstruction" technique, a new dish should preserve the essence of classic one (in our case: tomato soup—tomatoes, rice, roasted meat; lemon cookie—lemon, cookie dough). It becomes essential to recognize the qualitative and quantitative sensory dimensions of the examined dishes.

Sensory characteristics depended mainly on the type of dishes (traditional, molecular or Note by Note) for both lemon cookies and tomato soups (Figure 2A,B). Traditional versions of the dishes, regardless of their type, highlighted the characteristic (typical) attributes. For example, the intense butter note and crispy texture distinguished traditional cookies from others. Among different versions of tomato soup, the highest intensity of typical natural tomato odor and flavor, vegetable note and sour taste were found in traditional products.

**Figure 2.** Sensory properties of (**A**) lemon cookie and (**B**) tomato soup that are prepared in traditional, molecular and Note by Note versions: a, b, c—mean values marked by different letters in versions of dishes differ significantly at *p* ≤ 0.05.

In contrast, modern cuisine caused a statistical increase or decrease in the intensity of some key attributes that are related to odor, flavor, taste and texture. The molecular and NbN versions of cookies showed a greater level of smoothness and meltiness than the traditional ones. The intensity of many odors and taste/flavor attributes were also different. Molecular cookies were characterized by the highest level of natural lemon odor and flavor, sour taste, and astringency, while the Note by Note cookies revealed the most artificial lemon note (Figure 2A). In turn, the lowest sensation of rice odor and flavor and of sweet taste and the greatest fatty and spicy odor and flavor were observed in molecular tomato soup (Figure 2B). In terms of consistency, traditional and NbN soups revealed a very similar level of thickness, but the NbN sample had the greatest smoothness and meltiness. The molecular version had a completely different texture.
