Modern Methods of Sustainable Behaviour Analysis—The Case of Purchasing FMCG

Abstract

In this manuscript, the authors aim to explore sustainable consumer behaviour during shopping at a self-service store with fast-moving consumer goods (FMCG). An innovative combination of virtual reality (VR) equipment and an electroencephalogram (EEG) was used in the study. The objective of the study was to gather information as to how consumers make shopping decisions when buying fast-moving consumer goods (FMCG). The studies conducted so far have used either VR or EEG. To the best knowledge of the authors, no results of a study from the FMCG sector using both these devices simultaneously have ever been published. The results of the pilot studies are presented in the paper. The presented results constitute a part of a wider research project within the scope of which a triangulation of the research methods was used, enabling deeper analyses to be conducted of conscious and non-conscious aspects of the study subjects. The authors analysed primary data indicative of sustainable consumer behaviour. Descriptive statistics, including such measures as a mean value, standard deviation, and correlation analysis, as well as the Valence/Arousal Index, were used. The conducted studies provided knowledge of sustainable behaviour for two types of consumers – non-routine and considerate. Moreover, emotion indicators for FMCG products were defined, out of which the highest satisfaction was recorded for salmon as a product.

1. Introduction

The development of new technologies has created numerous possibilities for the study of consumer behaviour. Nowadays, we can not only retrace consumers’ steps and decisions but also verify whether their declared decisions are indeed true. Such technologies gain particular significance at a time when so much attention is devoted to sustainable activities. The rationality and responsibility of consumer behaviour constitute one of the major elements of sustainable development. Consumer attitudes are expressed through their reaction to a purchased product or service; furthermore, consumer environmental ethics affects the choice of products.

The objective of this paper is to explore and present which clients’ behaviour entails sustainable consumer behaviour during shopping at a self-service store selling fast-moving consumer goods.

The formulated research questions were worded as follows: What types of behaviour demonstrated during shopping have the attributes of sustainable behaviour? What is the relationship between consumer decisions and the shaping of sustainable shopping behaviour?

Currently, no research uses modern technologies to evaluate the balanced behaviours of consumers whilst they are shopping. This research gap can be filled. The multidimensional research is the added value of this article. Therefore, it is worth exploring which clients’ behaviour entail sustainable consumer behaviour, and this constitutes the main focus of this study. Specifically, the emphasis was placed on shopping at a self-service store selling fast-moving consumer goods.

Consumer decisions will be expressed in the basket of purchased goods. Consumer behaviour will depend on the consumer’s character expressed through such attributes as the awareness of a purchase evaluated based on the time during which a product was held and the accompanying emotions (examined with the Index of Arousal and Valence). The obtained research results show that there are differences between women’s basket sizes and men’s basket sizes. The study regarding emotions did not unequivocally show which of the respondents’ emotions are directly related to sustainable behaviour.

In the next sections, the authors present a literature review about sustainable behaviour, characteristics of sustainable consumers and their purchase decisions. In this paper, VR and EEG play a vital role; in this regard, the authors also explain the selected main senses engaged during shopping, including sight and touch. Hence, the explanation shall be broad in nature.

Literature Review

At the age of a developing trend of consumerism [1], encouraging consumers to purchase numerous objects mostly on account of changing fashions and shaping the attitude of possessing numerous things, a new and growing tendency occurs—that of sustainable consumer behaviours [2] which are directly connected with sustainable consumption.

Considering sustainable consumption issues can begin by dividing them into weak and strong varieties. The former is based on a technological factor that will have an impact on, e.g., the improvement in product efficiency. The latter assumes the need for changes in consumption volumes and their patterns. None of these aspects is independently capable of developing sustainable consumer behaviour; however, combining their effects may decidedly increase consumer awareness, thereby resulting in the development of sustainable consumer behaviour [3].

The literature on the subject indicates that one may talk of sustainable behaviour if economic, social and environmental objectives have been fulfilled. A consumer achieves benefits while realizing economic objectives that do not try to maximise or achieve those benefits at any cost—at the cost of society. A consumer is prepared to partly give up on those benefits if such a sacrifice contributes to the balancing of intergenerational needs [4].

Economic objectives are based on attentiveness, paying attention to the brand of products and services, and achieving self-benefit and satisfaction with the product [4].

Social objectives are based to a large extent on ethics, which may be expressed in various aspects during the entire process of product manufacture, e.g., caring for the employees of companies manufacturing sustainable products, promoting environmentally friendly lifestyles, supporting the pro-ecological movement, etc. [5].

Concerning sustainable behaviour, social variables are defined as attitudes, beliefs and subjective norms; furthermore, they are more easily predictable [6].

Declaring the practice of values related to sustainable properties is expressed through purchasing decisions [7]. Altruistic consumer attitudes and biospheric values are linked with pro-environmental intentions and behaviour [8], and they play an important role next to socio-psychological factors [9].

Environmental objectives take into consideration all the elements related to the environment and those that can affect any stage of product creation, e.g., plant fertilisation based on farm gate balance, effective use of water resources, aiming to reduce greenhouse gas emissions [10], the composition of product packaging, waste segregation, etc.

The creation of environmental values by a consumer will arise from the scope of the consumer’s knowledge and sustainable practices. An example of sustainable consumer behaviour concerning such values is consumers’ interest in healthy food, their awareness of how the environment benefits from the use of organic food, and consequently how it affects a positive shopping trend [11].

On the other hand, an individual’s positive attitude toward sustainable behaviour will arise from attitudes and predispositions that are tied to the positive consequences of such behaviour [12,13]. According to Thøgersen, [14] behaviour linked to sustainable consumption constitutes a consumer’s attitude that is expressed through the choice of products and services with a low carbon footprint. Therefore, sustainable behaviour will improve social and environmental efficiency [15]. The trends of implementing the values linked with sustainable characteristics are expressed through shopping choices [7]. In the literature on the subject, various classifications of consumer types exist, and one such interesting classification involves a division into considerate (otherwise defined as limited decision-making), non-routine (defined as broad-range decision-making), habitual and impulsive [16,17,18,19].

Sustainable consumer behaviour, according to Biswas and Roy, involves limiting the use of natural resources, introducing changes into one’s lifestyle and consuming environmentally friendly products to satisfy one’s own needs and the needs of future generations. In line with the approach defined by Martinez et al., what comes to the fore is the aspect of the impact that consumption has on the environment, which in turn affects the choice of services and products [20].

In documents from 2018, the European Union additionally focused on supporting sustainable consumption models, as well as engaging consumers and businesses in initiating new solutions [21].

Sun et al. [22] found that the COVID-19 pandemic has driven consumers towards buying sustainable products and that consumers are now paying more attention to the environment and society. The increase in the environmental concern due to the COVID-19 pandemic was also reported by Jian et al. [23]. Sun et al. [22] and Qi et al. [24] found that the COVID-19 pandemic has driven consumers towards buying sustainable products and that now consumers are paying more attention to the environment and society. The increase in the environmental concern due to the COVID-19 pandemic is also reported by Jian et al. [23].

The aspect of consumer’s emotions, such as fear, tension, stress, awareness, sense of guilt, shame and strength, plays a significant role in forming sustainable behaviour, and these emotions may also cause shopping decisions to be more sustainable [25,26,27].

Who are the sustainable consumers? They are not just individuals purchasing ecological products but also individuals considering social and economic issues [6]. As early as 1994, at a symposium on sustainable consumption held in Oslo, the concept of sustainable consumption was defined as the use of goods and services satisfying one’s needs and leading to a better quality of life in such a way as to simultaneously reduce the consumption of natural resources and toxic materials, as well as to reduce waste and pollution emissions during a product life cycle, thereby not endangering the needs of future generations [3].

A decision-making process constitutes a significant component of research on consumer behaviour. The complexity of the process results in research being undertaken at every stage, which enables predicting, evaluating or supporting consumer behaviour. [28]

Kotler and Keller point out five stages of a decision-making process: need recognition, information search, variant evaluation, deciding on a purchase and post-purchase behaviour. [29]

The attributes of sustainable development are especially important at the stages of choosing a process of making a shopping decision. On the other hand, studies on sustainable products are typically tied to ‘choice models’ with multiple attributes, which constitute a more realistic look at compromise decisions made by clients.

The factors influencing in-store decision-making are also the products themselves, with their attributes such as packaging, brand and label. Food packaging is an essential marketing tool, the importance of which is reflected both in the performed functions and in the role in the process of consumers’ decisions on purchases. A survey shows that the main functions of food product packaging in the opinion of the respondents are its protective function (23.8%) and informative function (23.8%). Labels placed on food product packaging are more and more popular among consumers. This is due to the increasing awareness of the relationship between diet and health, which makes consumers look for nutritional information and helps them make the right choices while buying food products. An analysis of the results of the studies also indicates that the most important packaging features for a consumer include ease of use and durability, while the most important information sought on the packaging is shelf-life, price and composition of a product. Consumers are also interested in information on caloric content, individual nutrients and the origin of the food product [30].

A review conducted by Bangs and Schlegelmilch [31] demonstrated that the types of products with sustainable properties that were most frequently studied included food, clothes, flats and cars.

According to the studies conducted on a group of young people, it was demonstrated that people from one’s closest environment—peers and friends, as well as opinion leaders, perceived by the young as authorities on broadly understood sustainable consumption—play a major role in forming pro-environmental food habits and thereby food shopping habits [32].

Consumers’ sensory experiences generate a significant impulse to make a purchase. The main senses engaged during shopping include sight and touch. The sense of touch plays an important role in the evaluation and decision-making by a consumer, be it on the occasion of purchasing a product, using it or consuming it [33]. Consequently, consumers are more eager to choose products that vendors allow them to touch [34]. The influence of the sense of touch was also examined in VR to verify whether it would play such a major role as it does in the real world [35]. However, a majority of consumers rely on visual stimuli to formulate their first impression of a product, thereby creating a secondary impression regarding a product [36]. On the other hand, tactile stimuli show a two-directional effect concerning product evaluation. Tactile indicators have a positive influence on the assessment of such products that can be best recognized through touch when such products are believed to be of high quality, but those tactile stimuli will hurt the assessment of low-quality products.

Referring to tactile indicators to food products or their packaging, it is the touch experienced through hands or lips that can potentially help the food industry to increase satisfaction with products and the intention to buy [33]. Indeed, product packaging examined through touch is more and more often believed to be an effective marketing tool [37], which is related to a rapidly growing interest in studies tied to product packaging design [38].

Sight is used to look at, compare and gather information that is valid from the point of view of a consumer. Consumers who pay attention to the information placed on product labels must devote more time to learning about a product or comparing it. This was proven, inter alia, in the studies on the impact of time pressure on impulsive buying. It was demonstrated that intensive pressure of time is correlated with affective aspects of impulsive buying, while low pressure of time is linked with cognitive aspects of compulsive buying [39]. The studies regarding health claims placed on products presented the impact of time restrictions occurring in a regular shopping situation on the understanding of such claims [40].

A multitude of factors affect consumers’ attention, including inter alia, label layout, label size, colour scheme and density on a small packaging surface [41,42]. These properties translate to the time a consumer will spend on becoming familiar with the product label.

Therefore, product markings that claim a product is manufactured by a family-owned company increase consumers’ trust [43].

Reliable information found on sustainable development labels has a positive effect on psychological and behavioural variables in consumer behaviour. This translates into a change in their perception and behaviour in terms of environmental, ethical and social aspects [44,45].

The fact that consumers read food labels depends on several factors; however, what is significant is that a majority of consumers do read the information placed on food packaging. It enables avoiding, e.g., health problems related to unsustainable behaviour and unhealthy food habits [46,47,48,49]. In the paper, it was assumed that the time a product is held is directly proportional to conscious shopping decision-making. Longer concentration on a product may result from the fact that a consumer reads product ingredients, looking for information about a producer or quality certificates, a brand, best-before date and packaging type [50,51]. This finding was confirmed by studies in which gender also has an impact on making shopping decisions under the influence of the information found on a label [47].

Apart from the discussed factors that influence decision-making, equally important are demographic factors, such as gender [52,53]. Studies show that men and women look for different products. These studies demonstrate that women, in comparison to men, are more focused on pleasurable aspects of shopping, display a stronger emotional motivation for shopping [54] and more frequently shop under the influence of an impulse [55]. In turn, men seek more utilitarian motivations for shopping [53]. The gender criterion is one of the most frequently applied segmentation criteria used by vendors. It can be assumed that depending on gender, the information process and the decision-making process will differ [56]. Therefore, the authors assume that gender will affect sustainable shopping decisions.

In line with the trend of the experience economy, clients constantly seek new experiences, and emotions, which is why the actual product is not as important as customers’ experiences (educational, entertainment, aesthetic, escapist experiences) [57,58,59,60]. Virtual reality (VR) is one of the tools that can measure the emotions scale and perfectly fit into the experience economy because it provides consumers with new experiences and impressions. Influencing one of the most crucial of the human senses—sight—constitutes an additional stimulant. This sense is considered to be dominant in a majority of consumers. Therefore, perception through sight is crucial, and the effectiveness and efficiency of communication are strengthened by the use of VR technology [61].

Currently, a multitude of studies on consumer behaviour are being conducted in the VR environment [62,63,64,65,66]. An example of such studies involves demonstrating a positive impact of the VR environment on acquiring knowledge by consumers, regarding, among other things, product brand and recognisability and increasing the intention to buy as opposed to other forms of gaining knowledge about products [62,67]. In studies regarding the choice of tourist offers, the attractiveness of offers presented in the VR environment was of particular significance to women [62]. In yet other studies, the focus was on the impact of the VR environment on a consumer and the research opportunities it creates. Among other things, it was observed that study subjects experienced anxiety caused by a mismatch with the real world related to, e.g., the manner of moving [63], hence the need for a prior familiarisation of study subjects with the VR world. It was further demonstrated that study subjects experienced a sense that they can control everything, indicating the presence of typical gamers among study subjects. Additionally, VR found an application in examining the impact of a VR scene on food-related behaviour. It was proven that the VR scene had an impact on the study subjects [64]. In the publication [65], it was demonstrated that VR technology can be used to draw conclusions about the attributes of customers’ personalities based on their behaviour. Furthermore, signals of individuals participating in a VR stimulation were subjected to analysis. Studies demonstrating the similarity of human behaviour in the real and virtual worlds were carried out as well. By using VR, it was shown that clients buy a similar number of their favourite products, in this case fruit and vegetables, irrespective of the manner of their presentation [66]. Virtual reality sometimes deforms an image, which affects the manner of its presentation. Such deformations are related to the parameters at which an image is displayed, which is strictly tied to equipment capabilities. The perception of object images and their quality depends on the degree of irregularities.

VR software provides excellent results in testing store shelves [68,69,70]. Seeing a new packaging of breakfast cereals in a wider context helps assess its originality and aesthetics. During their stay at a store, consumers can look around freely, and their eyes are frequently attracted to random products that they did not include in their shopping list.

By combining eye tracking with VR technology, we can find out much more about the decision-making process itself and natural consumer behaviour. By seating a respondent in front of a computer, we may lose much of its natural responses and reactions [71,72].

When pondering upon the similarity of VR to reality, comparative studies were conducted in both of these environments. No significant differences were noticed when products were selected in both comparative groups [73]. It was demonstrated that consumer behaviour in real and in VR environments is comparable. [74]

Li et al. [75] used virtual reality and EEG to obtain information on how their study subjects perceived the space they found themselves in. The obtained perception results showed that there exists a correlation between EEG and surveys, and the results could be used for optimizing architectural space design.

EEG can also be applied to determining an emotional state of a study subject. For instance, McMahan [76] and Petrantonakisa [77] employed the Index of Arousal–Valence in their studies. Reading these indicators involves defining a plane in which emotions occur, where Arousal constitutes one axis and Valance constitutes another one. Arousal can be defined as a level of emotional intensity, whereas Valence serves for determining whether emotions are “positive” or “negative” [78], as presented in Figure 1.

Emotions constitute the main factor providing valuable implicit or explicit information for a quick and optimal decision [79]. Additionally, being one of the types of influences, it has the capacity of exerting a brief but intensive effect on an individual [80]. This leads to the conclusion that the level of emotions accompanying consumers’ shopping decisions needs to be studied. Additionally, we explore consumer level of product interest in sustainable products.

2. Materials and Methods

2.1. Research Methods

Research aim: The described pilot study aimed to gather primary data and analyse them following the research objectives to draw conclusions.

The following research questions were posed:

• What types of behaviour displayed during shopping have the attributes of sustainable behaviour?
• What is the relationship between consumer decisions and the shaping of sustainable shopping behaviour?

The following variables were used in the study: a group of products, namely fruit, vegetables, dairy, fish, fast food, baked goods, drinks and meat, were adopted as independent variables. In turn, basket size and basket value were treated as dependent variables.

Type of research: The qualitative study method was applied to gather data from 34 respondents.

Research tools: the study was conducted with the use of an EEG apparatus (electroencephalography) and VR (virtual reality).

Research method: descriptive statistics, including mean, standard deviation, minimum and maximum, Z-Score formula, Pearson correlation coefficient, and Index of Arousal and Valence.

2.2. Research Procedure

The research procedure applied in the study was presented in Figure 2.

A sample (n = 34) was chosen specifically on account of the place of residence and active shopping at a self-service store. Then, a research problem was formulated, which was presented in Section 1.

The next step involved creating a simulation of a store using the Unity engine in C# language. It was developed in such a way as to enable the registration of events from the simulation to synchronise them with registered EEG signals. Before commencing the main research, a pilot study was conducted to verify the accuracy of registration, synchronisation, etc. Once this step had been successfully completed, the research was conducted in the target environment. However, before partaking in the research, respondents needed to declare what type of consumer they were (

Table 1. Consumer type by gender.

	Type of Consumer
	Considerate	Non-Routine	Habitual	Impulsive
The number of women	9	2	6	3
The number of men	4	3	5	2

). Research subjects could indicate one of the following four consumer types:

• Considerate—individuals who shop in a planned manner;
• Non-routine—individuals who rarely shop and for whom there is no automatism in their shopping process;
• Habitual—individuals who have their habits and cannot imagine life without those products;
• Impulsive—individuals who assess a product by its appearance or declare that they feel like buying something.

A majority of individuals defined themselves as considerate buyers (13 people) and habitual buyers (11 people). The smallest number of respondents, i.e., 5 people, described themselves as being impulsive or non-routine buyers. Based on a signal registered from the EEG equipment and the registration of events, suitable calculations were carried out, and they included, inter alia, the average time during which a product was held in one’s hand, basket value, average basket value, basket size, creation of trust ranges with Z-Score formula, Pearson correlation coefficient and emotion indices based on Arousal and Valance Index [76].

Eventually, the data were merged, and conclusions were drawn regarding consumer shopping decisions.

2.3. Description of a Store Simulation

A store simulation was created and suitably adapted to virtual reality (VR) with the use of a ready store—Enzone supermarket [81]—in Unity engine (see Figure 3).

The environment was planned in such a way as to reliably recreate the traditional place of FMCG shopping. The main task was to create situations which would closely reflect those encountered in the real world by the study subjects or even to create such situations that in the real world would constitute a threat to them (e.g., an A-Board with information about the wet floor). Before commencing the study, participants watched a short instructional video, which presented how to move around, how to pick objects, and the goal of the game. The main purpose of the visit to the VR store was to perform daily shopping. Research participants, after watching an instructional video and before the study proper, had a chance to try moving around in the virtual environment. This enabled them to become familiar with the VR environment and was intended to reduce anxiety and the sense of being lost, which could have interfered with the study proper [75]. Respondents were given 20 min to carry out their daily shopping. Setting a time limit for completing shopping was determined by showing care for study subjects’ well-being and health. The store comprised 3 aisles (see: Figure 4). The first aisle was located immediately next to the store entry. In the middle of the aisle, islands were placed, inter alia, displaying promotional products. In the second and third aisle, apart from independent figures, A-Boards were placed, blocking passage. The entire store layout was presented in Figure 4.

The store provided clients with the following departments: fruit and vegetables, household goods, frozen foods, dairy products, packaged meat and fish products, pet food, sweets, soft drinks and water, and alcohol. At the end of the store, perpendicular to the aisles, there were three stands—meat, fish and seafood—as well as a baked goods stand. In real life, it rarely happens that a shopper is alone in a self-service store, which is why there were other buyers in the virtual store, who were independent figures.

Nowadays, payment is not always made in cash; hence, respondents had a choice to make at checkout between the most frequently used forms of payment: card, cash, smartphone and contactless payments. Study participants were informed of the course of the study and signed a consent for their voluntary participation in the study (the laboratory had the consent of the local ethics committee for conducting this type of research).

2.4. Description of a Virtual Reality and Electroencephalogram Study

Research participants included individuals who perform their daily shopping independently of a store brand, with an assumption that it is a self-service store. The next step entailed preparing a respondent for the study by putting a cap (Enobio 20, Neuroelectrics, Barcelona, Spain) on the respondent, connecting electrodes to the scalp and putting on a VR device (HTC Vive Pro Eye, New Taipei City, Taiwan).

A cap with 20 electrodes was placed at the P7, P4, Cz, Pz, P3, P8, O1, O2, T8, F8, C4, F4, Fp2, Fz, C3, Fp1, T7, F7, and Fpz points and was used for registering a study session. Channels were placed following the 10–20 system, an international system of EEG electrode placement. Electrodes required a wet setup for proper conductivity. In order to verify whether EEG electrodes had good contact with the scalp, the value of impedance was measured with Neuroelectrics^® Instrument Controller (NIC2) software v2.0.11.1. The sampling frequency was 500 Hz.

Once the above steps were completed, the study was commenced, during which a stay at the store was recorded with OBStudio software v27.2.4. The film footage was used for further analysis. However, before shopping started, a black screen appeared that was displayed for 60 s. The purpose of that step was to calm down the study participants’ emotions and brain waves. Then, the virtual reality store was initiated, where every activity performed by a study subject was recorded (Figure 5). The data were recorded in an Excel file, which was later used for analysis.

2.5. Measures

On the basis of the WHO’s product categorisation, the authors differentiated 8 baskets—fruit, vegetables, dairy, fish, fast food, baked goods, drinks and meat. The purpose of their creation was to enable the segmentation of consumers on account of their behavioural characteristics, i.e., shopping behaviour, with a particular focus on sustainable behaviour.

Then, the events file was used, which contained information about each product taken in hand, the time of holding a product, putting it away into a basket or onto a shelf, and specifying what products were purchased by a respondent. As a result, based on that information, each product was assigned to a suitable basket. In turn, the time during which a product was held in the respondent’s hands allowed the level of concentration on a product to be determined, which became a value necessary for assessing sustainable behaviour. Additionally, a converted Z-Score Formula (1) was used,

$$\overline{X}\pm Z\frac{s}{\surd n} ,$$

(1)

where $\overline{X}$, means the arithmetic mean, Z is a selected Z-Value, s is the standard deviation and n is the number of observations. Trust ranges for individual baskets were thereby calculated.

Another stage involved an analysis of the Pearson correlation coefficient between the average time holding a product from a given basket, basket size and basket value, broken down into consumer type and gender. Then, the Arousal and Valence Indices were calculated, which are presented in

Table 2. Description of the indices used in the test.

Name of the Index	Formula	Counting Method
Arousal [75]	(F3_beta3 + F4_beta3)/(F3_alpha2 + F4_alpha2)	Registration value from electrodes F3 and F4
Valence [75]	(F4_alpha2/F4_beta3) − (F3_alpha2/F3_beta3)	Registration value from electrodes F3 and F4

. The indices were selected for analysis since they can be used to examine consumer decisions in the simulation [4,5].

3. Results

All data were analysed using Matlab R2019a. The EEG signal analysis started with filtering the bandwidth and removing the power-network disturbances, i.e., frequencies above 50 Hz. Furthermore, the signal was detrended and filtered using the Fieldtrip library. The EEG spectral signal was then analysed using a Morse wave, which calculated an average peak frequency of half a second in a frame [82,83,84]. However, to calculate the alpha and beta frequencies, the signal was divided into appropriate bands [85]—alpha2 (7–13 Hz) and beta3 (13–25 Hz).

In the presented research, the indices were used for defining the emotions accompanying respondents while they were holding a specific product. Within that scope, four groups were differentiated—emotions of satisfaction with a given product (positive value of Arousal and Valence indices), emotions of irritation with a given product (negative value of Arousal index and positive value of Valence index), emotions of boredom with a given product (negative value of Arousal and Valence indices) and emotions of indifference towards a given product (negative value of Arousal index and positive value of Valence index).

After conducting the study in the VR environment with the use of cognitive neuroscience methods on 34 people, eight product baskets (see:

Table 3. Assigning the product to the appropriate category: I—fruits, II—vegetables, III—dairy, IV- fish, V- fast food, VI—backed goods, VII—drinks, VIII—meat.

I	II	III	IV	V	VI	VII	VIII
Banana Apple Orange Carrot Plum Strawberry Ananas Melon	Lettuce Zucchini Tomate Peppers Onion Potatoes Garlic Asparagus Cauliflower Avocados Artichoke	Natural yogurt Cheese Milk Butter Eggs	Cod Salmon Panga Shrimps Crab	Pizza Fries	Bakery products (bread, bread-roll, baguettes) Croissant	Water Orange juice White wine Cola Coffee Beer Rose wine Multifruit juice Whiskey Malibu	Beef Sausage

) were created on the grounds of the gathered data. The authors adopted the following types of behaviour, as valid in the process of sustainable shopping: the time during which a product was held in hand, the selected number of products and emotions.

The behaviour related to the time of holding a product was recorded from the moment of it being picked up until it was put away. This enabled calculating the average time during which a product was held, which is presented in

Table 4. Average time (in s) holding a product from the corresponding group.

	I	II	III	IV	V	VI	VII	VIII
Minimum	1.13	1.72	1.51	1.42	2.04	1.79	1.40	2.38
Mean	3.55	4.03	4.16	4.91	3.93	5.03	4.73	4.50
Maximum	8.90	8.25	14.66	10.18	9.35	18.79	18.89	9.22
Standard deviation	1.93	1.86	2.53	2.46	2.03	3.99	3.36	2.46

. The authors assumed that the time holding a product was directly proportional to consciously deciding on purchasing the product. Longer concentration on a product may have resulted from reading product ingredients, looking for information about a producer or quality certificates, a brand, a best-before date and a packaging type.

The respondents held products from the fruit group for the shortest time (1.13 s), and products from the baked goods group (18.79 s) and from the drinks group (18.89 s) for the longest time. Aside from the above, similar results were obtained in the vegetable and fruit groups, which is evidenced by the standard deviation at a level of 1.86 and 1.93, respectively. In turn, the most scattered results were obtained in the baked goods group (3.99) and the drinks group (3.36).

Based on the data obtained in Table 4, trust ranges were calculated, which are presented in

Table 5. Confidence interval by purchasing group (sec).

Fruits		Vegetables		Dairy		Fish		Fast Food		Baked Goods		Drinks		Meat
2.91	4.20	3.40	4.66	3.31	5.01	4.08	5.73	3.68	6.37	3.68	6.37	3.60	5.86	3.67	5.33

. To that end, a converted Z-Score formula was used for a trusted range, where an adopted Z-Value was at 95%, which is equal to 1.960. As a result, we obtained data regarding the range of the average time during which a product was held. For products from the fast food and baked goods categories, the end range is the highest and it is equal to 6.37 s, which means that these types of products were being held the longest. On the other hand, the shortest time of the left-sided range end was determined for the fruit category, and it was equal to 2.91 s, which means that these types of products were held over the shortest time.

Basket value and basket size were the subjects of further analysis. Basket value was assessed based on the amount of money spent by a given individual on shopping, while basket size concerned the amount and type of products that were found in a basket. From the analyses, it arises that on average, respondents spent 83 zlotys on 12 products. Determining the amount spent on shopping and the number of products will enable examining whether there is any correlation between the amount of money spent and the number of products, as well as the type of consumer, which will be elaborated on in the subsequent part of this section.

To verify the type of products in a basket, the authors carried out a classification by creating a tree map of a basket that defines what percentage share a particular group of products constitutes the contents of the entire basket. Each of the consumer types was additionally divided in terms of gender, which is an independent variable that affects a basket size, since men and women pay attention to different details while shopping. This will enable a more accurate examination of the relationships between the consumer type and basket size and basket value.

The first consumer type that was analysed was the considerate type (Figure 6). Considerate consumers are rational individuals who approach shopping by wishing to buy products they already know but at a price acceptable to them and of a recognisable quality (relation of price to quality). In the presented division, distinct differences were found between men’s and women’s baskets. Each of the baskets was of similar size, containing dairy and fruit. However, women’s baskets also contained vegetables, drinks, fast food, baked goods, fish and meat, whereas in men’s baskets, the third group was made up of drinks, followed by baked goods, fish, fast food, vegetables and meat. The obtained results lead to the conclusion that despite various numbers of individual product categories, the examined consumer baskets, both women’s and men’s baskets, contain more products defined by the WHO as being healthy, which is why we may also assume that consumers belong to a health-aware consumer group and that their baskets display the characteristics of sustainable shopping [86].

The second type of consumer is individuals who perform non-routine shopping (Figure 7). Non-routine consumers perform shopping unconventionally. This means, in the case of these individuals, that it is impossible to specify general rules of what they put into their baskets. Several leading products, i.e., most frequently bought products, can be found in the baskets of both men and women shopping in a non-routine fashion. In the case of women, such products include drinks, vegetables and dairy, while in the case of men, they include fruit, dairy and baked goods. In this case, women’s baskets lack meat products or fast food, which can be found in men’s baskets. There is also a distinct difference between fruit and vegetables for men and women. There are more vegetables and less fruit in women’s baskets, while the reverse is true for men’s baskets.

Despite slight differences between men and women in the non-routine consumers’ group and the presence of the fast food type of products in men’s baskets (constituting a small basket size), it can be assumed that baskets are sustainable in that group. In the baskets of both genders, the majority of basket size was made up of healthy products, such as vegetables, dairy and fruit, which in the WHO nutritional pyramid [86] are deemed as being the most important. It is worth noting the drinks type of products, since in the women’s group it was only composed of water. The choice of water instead of other drinks confirms healthy shopping choices.

The third type of consumers comprised individuals shopping habitually (Figure 8). This means that the decisions made at the store are based on routine and habit. Consequently, their decision-making is schematic, while the time devoted to making a decision was short. Typically, basket size and basket value are repeated. In women’s and men’s baskets, similarities could be found; i.e., it is evident that in the first and the second cases, the most typically selected products were fruit and dairy. The main difference was found in the case of meat and vegetables. Despite a distinct difference, it can be concluded that in the case of habitual consumers divided by gender, women do not have a sustainable basket because healthy products such as vegetables constituted only a small percentage of the entire basket size, while drinks and fast-food products constituted nearly 25%. However, in the case of men, a sustainable basket did occur, which was evidenced by finding more healthy products, as recommended by the WHO [86].

The fourth type of consumer is individuals that buy impulsively. The basket of this consumer group contains a high amount of dairy products and fruit. By analysing the examined group in terms of gender, it can be determined that the first two products constituting the highest percentage share are the same. Still, women bought more fish and vegetables, while men bought more baked goods and drinks (Figure 9). In line with the WHO recommendations, such basket sizes can be considered as having the characteristics of sustainable baskets. To obtain more exact answers, a deeper analysis is required, e.g., an analysis taking into account the impact of prices.

Following that, using the Pearson correlation coefficient, it was determined whether there was a correlation between basket size and basket value and the average time holding a product without a division into consumer characteristics but with a division into consumer type and gender.

In the case that does not account for consumer characteristics (

Table 6. No gender breakdown.

Product Groups	Basket Size	Basket Value
Fruits	−0.34	−0.22
Vegetables	0.53	0.20
Dairy	−0.25	−0.16
Fish	0.10	0.36
Fast food	−0.10	0.19
Baked goods	−0.18	0.07
Drinks	−0.09	0.11
Meat	−0.02	−0.05

), a mean and positive relationship between the average time of holding vegetable products and basket size (r = 0.53) becomes noticeable. This means that if we look at vegetables, e.g., in terms of their freshness, there is a significant chance that they will end up in our basket. In the case of other products, such a correlation was weaker or no correlation was found at all. This means that the correlation does not affect the basket size and basket value.

The correlation coefficient assumes a different value when it is differentiated in terms of consumer type and consumer gender.

Table 7. Considerate type of consumer.

	Women		Men
Product Groups	Basket Size	Basket Value	Basket Size	Basket Value
Fruits	–0.18	–0.12	–0.66	–0.83
Vegetables	0.80	0.27	0.95	0.85
Dairy	–0.37	–0.21	–0.43	–0.15
Fish	0.06	0.45	–0.35	–0.06
Fast food	–0.24	0.00	–0.36	–0.06
Baked goods	–0.17	0.04	–0.08	0.17
Drinks	–0.41	–0.04	–0.35	–0.26
Meat	–0.06	0.18	0.95	0.85

presents the results for the considerate consumer type, broken down into men and women. A strong correlation was obtained for females (r = 0.80) between the average time holding vegetables and their basket size. Apart from that, an analogous result was recorded for men, along with strong relationships with a basket size, i.e., if the average time of holding a product from the vegetable group increases, then there is a very high chance that the aforementioned product will be purchased. In turn, an average positive relationship can be observed for women between the average time of holding fish and the basket value (r = 0.45). This means that if the average time holding a fish increases, there is an average chance that the fish will eventually be purchased, which will translate into a growing basket value. In the case of men, the correlation is not significant, as the Pearson correlation coefficient is equal to r = −0.06. For the first and the second consumer groups, the relationship between the average time holding products from the fast food group and their basket sizes and basket values is weak or does not exist at all. However, in the case of the average time holding meat products, a strong positive correlation between basket value and basket size was determined. This may be linked to men’s diet, in which protein typically dominates, being the main muscle-tissue-building material.

In the case of a habitual consumer (

Table 8. Habitual type of consumer.

	Women		Men
Product Groups	Basket Size	Basket Value	Basket Size	Basket Value
Fruits	−0.27	–0.41	0.00	0.59
Vegetables	0.97	0.95	–0.91	–0.58
Dairy	0.08	–0.23	–0.64	–0.55
Fish	0.00	–0.32	–0.18	0.43
Fast food	0.43	0.57	0.66	0.14
Baked goods	–0.17	–0.39	0.52	0.81
Drinks	0.00	0.04	0.70	0.36
Meat	0.66	0.77	–0.81	–0.89

), the results assume different values, with one exception being the relationship between the average time of holding products from the vegetable group and women’s basket size.

In the case of men, we observe a strong negative relation between the average time holding vegetables and the number of products found in a basket. This means that if the average time holding a product among this group increases, then it will not be picked for the basket, thereby decreasing the contents of the basket. In the case of the average time holding a fast food product, for both men and women, there is an average correlation with basket size. This means that the average time holding a product increases on account of it being watched and its “cover” being read, and we succumb to the temptation of buying the product, which consequently translates into a growing number of products in the basket. Furthermore, it is evident that for the group of men, there is an average and strong positive relation between the mean time spent holding baked goods and basket size and the basket value. However, in the case of meat products, there is a strong chance that there is a negative correlation between the average time holding the product and the basket size and value. Therefore, if the average time holding products among that group increases, the products will not be selected for a basket, yet the trend is reversed in the case of female consumers. The studies of other consumer types—non-routine and impulsive ones—demonstrated that the relationship between a group of products and basket size and basket value was weak or there was no relationship at all.

The last stage of the study involved defining emotions with the use of EEG and suitable Arousal and Valence indices. As previously mentioned in Section 2, emotions were divided into four groups.

Table 9. The value of Arousal and Valence Index in terms of positive emotions.

Emotions of Satisfaction with a Given Product
Object	Mean Arousal	Mean Valence
Banana	0.0111	0.0006
Pepper Red	0.1392	0.0021
Tomato	0.2632	0.1527
Salt	0.8417	0.0311
Salmon	1.2590	0.0180
Custard	1.1960	0.0405
Egg Box	0.6009	0.0312
Beef	0.5603	0.0083
Water	0.2478	0.0009
Pizza	0.0325	0.0094

presents the emotions of satisfaction from a given product, where the obtained values of Arousal and Valence index are positive. They were the strongest for products such as salmon and custard, and they were the weakest for bananas and pizza.

In

Table 10. The value of the Arousal and Valence Index in terms of indifference emotions.

Emotions of Annoyance with a Given Product
Object	Mean Arousal	Mean Valence
Potatoes	−0.0798	0.02952
Crab	−0.2221	0.02919
Water Pack	−0.1415	0.0318
Perry	−0.2979	0.19035
Shrimp	−0.4387	0.78240
Milk	−0.4814	0.07494
Carrot	−0.4140	0.04884
Cauliflower	−0.1291	0.07781
Mayo	−0.3145	0.06563

, the emotion of irritation with a given product is presented. The strongest irritation level was found for milk, shrimp and carrot, where the value was, respectively –0,48, −0,44 and −0,41. Additionally, the value was the weakest for potatoes, equalling −0,08.

In

Table 11. The value of the Arousal and Valence Index in terms of annoyance emotions.

Emotions of Boredom with a Given Product
Object	Mean Arousal	Mean Valence
Yogurt	0.0045	−0.0328
Frozen Fish	1.2973	−0.1341
Strawberries	0.3815	−0.2355
Shallot	0.0114	−0.0304
Frozen Chips	0.3941	−0.0247
Cheese	0.4219	−0.0145
Bread	0.0607	−0.0925

, emotions of boredom with a product are presented. The strongest boredom level could be observed for frozen fish, where the value was equal to 1.30, whereas the lowest boredom level was recorded for yoghurt at a value of 0.004.

In

Table 12. The value of the Arousal and Valence Index in terms of bored emotions.

Emotions of Indifference to a Given Product
Object	Mean Arousal	Mean Valence
Cod	−0.2127	−0.1383
Courgette	−0.2883	−0.0346
Salad	−0.2266	−0.0380
Garlic	−0.6832	−0.5223
Apple	−0.3901	−0.0696
Orange	−0.2440	−0.0528
Frozen Carrots	−0.4455	−0.1065
Onion	−0.3565	−0.0520
Frozen Bean	−0.3067	−0.1035
Orange juice	−0.4043	−0.0259
Melon	−0.3280	−0.0977
Asparagus	−0.5180	−0.0529
Leek	−0.3081	−0.0446
Ananas	−0.2538	−0.0030
Fish	−0.0550	−0.0481
Butter	−0.5638	−0.2325
Sausages	−0.2961	−0.3445

, emotions of indifference towards a given product are presented. In comparison to other groups, this group was the most numerous.

The product with regard to which indifference emotions were the strongest was garlic, with an index value equal to −0.68. In turn, the product towards which the least indifference was felt was fish, wan an index value of −0.06, respectively.

4. Discussion

In a discussion of the results of baskets, and independently of consumer type, it can be shown that in theory, consumers can be conscious; however, the obtained results demonstrate that the behaviour types differ.

From the conducted analyses, it arises that there are differences between women’s basket sizes and men’s basket sizes. Thereby, the theory that there is a difference in shopping decisions made by different genders is confirmed [52,53].

The research results showed that the most sustainable basket size is held by the respondents who completed after a certain amount of time. For example, if we let the respondent into a virtual store and the survey lasts 30 min, the survey can be completed after the survey is over and even a little later. This resulted in respondents who were not always able to recall their emotions, especially if we are interested in a lot of specific moments rather than the overall impression of the entire survey. This can lead to false answers. The information given in the survey is subjective (it reflects the respondents’ own opinions about their feelings), while in the case of signal analysis, we determine the feelings based on the recorded signal. There is no problem with the subjective evaluation of the person surveyed. Emotions can be determined at any time.

The use of a survey questionnaire alone in terms of emotion research would give us only the declarative responses of the surveyed group, while through the use of EEG and the corresponding indexes, we see the emotions that occur during each activity performed by the subject.

An analysis of the correlation coefficient delivered additional information on sustainable consumer behaviour. In the “considerate” respondents group, a strong correlation was shown to exist between the average time holding vegetables and the basket size. This means that the decision to make a purchase increases along with the time holding a given product in hand. However, in the case of a majority of products, an analysis of correlation demonstrated that, along with an increase in the time holding a product in hand, basket value and basket size decrease. In such a case, product awareness increased, but there was no shopping decision made. Such behaviour may result from a rejection of a product on the basis of a rational decision [87,88]. An analysis of the correlation demonstrated that differences in gender did not affect, to a large extent, the basket size or basket value for consumers doing non-routine or impulsive shopping. This may result from the approach to shopping, since non-routine behaviour concerns products already purchased [89], while impulsive purchases result from the fact that a consumer is not guided by any objective criteria [90]. The analysis of a correlation coefficient provided additional information on sustainable consumer behaviour. In the “considerate” respondent group, a strong correlation between the average time holding vegetables and basket size was shown. This means that the decision to make a purchase increased along with the time holding such a product in hand. Nevertheless, in the case of a majority of the products, a correlation analysis demonstrated that, along with a lengthened time holding a product in hand, both basket value and basket size decreased. In such a case, product awareness grew, but there was no shopping decision made. Such behaviour may result from a rejection of a product on account of a rational decision. Correlation analysis demonstrated that gender differences did not have a significant effect on basket size or basket value for consumers who did non-routine and impulsive shopping. This may be the effect of the approach to shopping, since non-routine behaviour means that consumers make decisions in the shop and impulsive behaviour makes consumers make shopping decisions with an emotional impulse.

The obtained results do not provide an unequivocal response as to which type of consumer demonstrates sustainable behaviour.

The study showed that the highest values of coefficients confirming satisfaction occurred for salmon. Such an attitude may result from the popularity of this kind of fish among the examined respondents, as well as its being perceived as a healthy product [91].

The study regarding emotions did not unequivocally show which of the respondents’ emotions are directly related to sustainable behaviour.

To the best of our knowledge, the current literature does not provide knowledge on this subject. Understanding what affects consumers’ choice of products is necessary for the propagation of conscious shopping and consequently sustainable consumer attitudes.

Differentiation of various types of consumers enables a better way of reaching groups that demonstrate sustainable attitudes with a suitable message. The assessment of how long a given product was held by consumers may constitute an indicator for individual brands as to what steps need to be taken in order to engage consumers in sustainable shopping. It particularly concerns consumers who hold a product over a longer period of time and do not decide to put it into a basket. Such behaviour may constitute a reason for a company to examine their product packaging and labelling [92,93].

The use of information on emotions accompanying shopping is of particular interest. In a situation when a consumer puts away a product onto a shelf and negative emotions are associated with it, it may constitute a premise for companies for further research on the causes of such emotions [94,95].

5. Conclusions

The authors of this research are of the opinion that it is important to examine the impact of time on making a shopping decision in reference to virtual reality, since respondents’ attitudes may be determined by the behaviour characteristic of gamers in virtual reality. This is because they recognize the environment as a place in which time plays a crucial role.

Once the necessary data were obtained, the average time holding each product was calculated for individual respondents and for the entire collective; furthermore, it was verified whether a product eventually ended up in a basket. Then, using a Z-Score formula, trust ranges were calculated on the basis of which it was determined what the shortest and the longest average times holding a product were in a given product group. However, in order to answer the research question, it was necessary to calculate the Pearson correlation coefficient. Independent variables included product groups, namely fruit, vegetables, dairy, fish, fast food, baked goods, drinks and meat, whereas basket size and basket value were treated as dependent variables. Finally, the emotions accompanying a given product were determined using EEG indices—Arousal and Valence.

It needs to be taken into account that the results obtained are based on a pilot study. In order to achieve a representative research sample, the sample needs to be expanded. Nevertheless, the obtained results confirm the opinion that the non-routine and considerate consumer types demonstrate sustainable behaviour during shopping for FMCG goods. The research authors are planning to expand their study with an eye-tracking analysis.