Universalists or Utilitarianists? The Social Representation of COVID-19 Pandemic in Brazil

: A ﬁerce debate arose in Brazil on how to manage and mitigate the coronavirus disease 2019 (COVID-19) pandemic—a debate derived from a dissonant perception by society about the actual signiﬁcance of the COVID-19 pandemic. The Brazilian population has divided into two contrasting philosophical approaches: the universalism—understanding life as an asset of inﬁnite value and, therefore, more important than the country’s economic preservation—and the utilitarianism—where the focus is on the mitigation of the COVID 19 pandemic-enabled economic crisis, due to its potential devastating e ﬀ ect on people’s lives, thereby leaving health issues in the background. The main cause for these di ﬀ erent sensemakings is associated with the lack of a monosemic deﬁnition for the “COVID-19 pandemic” construct. Thus, the objective of this article is to investigate the social representation of the COVID-19 pandemic in Brazil through the Social Representation Theory operationalized by the word’s evocation technique. The results show that Brazilian society privileged prophylaxis and health, via social isolation, to the detriment of the country’s economic preservation. In addition, trends emphasized by experts, such as a post-pandemic “new-normal” and the digital transformation of society, played a peripheral role in the social representation of the COVID-19 pandemic in Brazil.


Introduction
The evolution of a pandemic is one of the most dangerous and complex problems for society, and its management and mitigation by governments are therefore challenging [1,2]. In December 2019, the city of Wuhan-capital of China's Hubei province-became the hub of an unknown cause pneumonia outbreak. On 7 January 2020, Chinese scientists isolated a new virus-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-from patients with symptoms of this disease. This virus was named coronavirus disease 2019 (COVID- 19), in February 2020, by the World Health Organization (WHO), which decreed, on 11 March 2020, COVID-19 as a pandemic to be confronted by humanity. After the initial outbreak, with more than 80,000 cases and 3000 deaths in China, COVID-19 has spread to 216 countries and territories [3].
The aim of this paper is to analyze the perception of the Brazilian population about the real significance of the COVID-19 pandemic, since in Brazil, society, media, medical associations, and municipal, state and federal governments have been divided between two contrasting philosophical approaches: utilitarianism and universalism [4,5]. The utilitarianist philosophical approach argues that, in order to mitigate the COVID-19 pandemic, the country must apply a selective social isolation (from now on named vertical lockdown) [1,6,7] in order to mitigate the economic damage resulting from combating the COVID-19 pandemic. Conversely, the universalist philosophical approach argues that human life is priceless and a full social isolation (with the exception of absolutely essential activities) Sustainability 2020, 12 (from now on named horizontal lockdown) must be implemented at any cost for the preservation of lives-whether through non-contagion or the preservation of a health system that can help the victims of this pandemic. In Brazil, these opposing sensemakings (sensemaking is the process of individuals collectively creating reality in their everyday life; it is an ongoing accomplishment that involves assigning meaning to experiences and creating order out of events by making sense of them [1,6,7]) seem to be associated with the lack of a monosemic definition for the "COVID-19 pandemic" construct. Thus, the objective of this article is to investigate the social representation of the COVID-19 pandemic, according to the Brazilian population, in order to understand how society has understood and incorporated the concept of "COVID-19 pandemic". In other words, the research question that this article intends to answer is: What is the social representation of the COVID-19 pandemic for Brazilian society?
The focus on Brazil is due to the very fact that the number of death tolls in the country was one of the biggest in the world [8] and a tough dispute between horizontal lockdown and vertical lockdown supporters took place in the nation as is described further in this article.

The Importance of Constructs Definition
One of the significant reasons for the rejection of papers submitted to academic journals is the absence of clearness about the construct under scrutiny [9,10]. A construct is characterized as an idea intentionally and deliberately developed for a particular scientific reason, being it a concept that cannot be noticed straightforwardly [11]. Constructs are, accordingly, the bedrock of any theory [10,12], which reveals the significance of clearness in their definitions.
As such, to improve the understanding of the COVID-19 pandemic phenomenon, this study aims to identify how Brazilian citizens perceive the same by using the Social Representation Theory [13] as this is an efficient approach to better understand constructs [14,15].

Social Representation Theory
The Social Representation Theory (SRT) was created in the 1960s in France by Serge Moscovici to explore how a common sense is shaped, coordinated, organized and diffused inside particular human groups. At that time, the social psychology research was applied in an individual manner. Indeed, Moscovici was the first scholar to challenge this predominant domineering practice, focusing on the common sense as a property of social investigation [16]. Social representations are more than mentalities, feelings, or qualities; they grasp the disclosure and association of social reality [17]. A motivational guideline determines the content of social representations, namely the fear of the stranger and the inspiration to transform what is hard to comprehend and threatening into what is familiar and comforting [18,19].
Social representations are created to understand, restrict and change the world [20], being validated through the conviction that they are significant in the day to day life, as there is consistently the need of understanding what an individual or an artifact has to do with the world that encompasses the same [21].
When a social representation is defined, an identity is created collectively [10]. As one cannot live independently but rather together with different people with whom the world is shared, this representation becomes a social representation, through which a social reality is portrayed [22,23].
The research on social representation focused at first on individuals and their associations with society, this being understandable as the SRT has its inception in the psychoanalysis. Nonetheless, in the long run social representations began to be applied in non-human artifacts. One supports this new path, as a social representation is a speculation by which a subject relates to an object, this object being either an individual, a thing, a material, a social or mental event, a natural phenomenon, a thought, a theory, or a hypothesis [21,22]. The central nucleus theory was proposed by Jean-Claude Abric, emerging as a supplement to the SRT [24][25][26]. Through it, the most clear and cohesive qualities and observations shared by a group about the object whose social representation is sought are characterized. Consequently, the central nucleus highlights the consensual aspects associated with the social representation of an object [27].
Thus, the central kernel of a social representation includes values that, as a rule, a subject is unaware of, or values that are not disclosed but guide the subject's activities and conduct. The central kernel represents the permanent portion of a social representation, being stable and impervious to changes, which guarantees stability to the social representation. Thus, within a specific social context, the central nucleus is critical to the meaning that an object holds for a group [26].
Moreover, the social representation requires a more adaptable peripheral system around the central nucleus. This peripheral system houses the current relevant divergences of the group, subsequently lodging the specific and particular view of the members of the group to permit the social representation to adjust to the everyday schedule without influencing the central kernel [26,28]. Hence, the peripheral system is less steady than the central nucleus, assuming the role of an individual mediator so as not to put the significance of the central kernel in danger [27].

Covid-19 Pandemic: The Brazilian Scenario
The official start of the contagion of COVID-19 in Brazil occurred on 26 February 2020, after a man from São Paulo, who returned from Italy, tested positive for COVID-19. On 17 March 2020, São Paulo confirmed the first death in Brazil-a 62-year-old man. On 20 March 2020, community transmission (in local transmission, one knows that a person became infected through contact with another, who in turn contracted the virus after being in a region where there is contagion. In community or sustained transmission, it is no longer possible to track the infection chain and know who was responsible for the contamination of the others) in Brazil is confirmed by the Ministry of Health, which started to impact several aspects of Brazilian society. Thus, since the first confirmed and official contamination, the COVID-19 pandemic has evolved exponentially in Brazil, taking the lives of mainly the elderly and/or those with immunodeficiencies, having reached, at the end of July 2020, more than 2 million cases and 90,000 deaths.
With the evolution of the pandemic, at the end of November 2020, there were more than 172,000 victims and 6.3 million people infected in Brazil. As such, in late November 2020, Brazil was the third country in the world with the most infections by COVID-19, behind only the United States, which added up to more than 12.3 million cases, and India, with 9.1 million. However, in November 2020, Brazil was the second country in number of deaths, after the USA, where more than 257 thousand people died (sources: https://www.dw.com/pt-br/brasil-tem-mais-302-mortes-ligadas-%C3%A0-covid-19/a-55705668 and https://www.statista.com/statistics/1104709/ coronavirus-deaths-worldwide-per-million-inhabitants/ (accessed on 29 November 2020)).

The Sensemaking Dispute in Brazil: Vertical Lockdown vs. Horizontal Lockdown
The management and mitigation of the COVID-19 pandemic is a serious public health challenge for any country for several reasons: (1) Without social isolation, there is a very high rate of contagion (R0 = 2.5). This is reflected in an exponential expansion of contagion. Thus, a patient would infect 2.5 people in 5 days and 406 in 30 days. However, with 50% social isolation, a patient would infect 1.25 people in 5 days and 15 in 30 days [29]; (2) The absence of a specific vaccine or antiviral during the pandemic peak [30]; (3) Uncertainty regarding the real degree of lethality of COVID-19, due to the lack of knowledge about the number of asymptomatic infected people, especially in countries with a large population, such as Brazil, without mass testing of the population. Estimates based on mathematical models suggest that 86% of infected individuals are asymptomatic, which may lead mortality estimates to be inaccurate [31]; (4) Doubt about COVID-19 recurrence. That is, there is no certainty whether a patient infected with COVID-19 would create immunity to it. Furthermore, it is not known exactly when a patient, after being considered healed and if there is no recurrence, stops infecting another person [32].
Thus, to face these challenges, the World Health Organization protocol [33] imposed social distancing, justifying that this is the most appropriate way to prevent the spread of the disease. In this way, an attempt is made to adjust the demand for treatment to the capacity of nations' health systems (in terms of human resources, facilities, equipment, and supplies).
Social distancing, however, can be of two types: horizontal lockdown (most of the population remains at home, with the exception of essential services) or vertical lockdown (only the elderly and/or immunodeficient people stay at home; healthy young people and adults return to the normal activities). In the horizontal lockdown, the hospitalization pace of infected people becomes slower [34]. Thus, the reduction in the rate of contagion-by "flattening the curve"-saves time to deal with the pandemic.
However, in life there is no one-sided coin. The horizontal lockdown has a huge negative impact on the economy in the short and medium terms, generating unemployment and a drastic reduction in GDP in all countries where it is implemented [35]. For each 1% increase in the unemployment rate, there is an increase of 0.5% in the mortality rate in the country [35]. In Brazil, there was a tough confrontation between supporters of vertical lockdown (utilitarianists) and horizontal lockdown (universalists). The first perspective was clearly supported by the President of the Republic, while the second was supported by the then Minister of Health, the mainstream media and most medical associations and state governors. As a result, two Ministers of Health were dismissed by the President and, in September 2020, a general of the Brazilian Army took over as Minister of Health.

Methodological Approach
This research uses a qualitative-quantitative methodological approach, the data being collected via the words evocation technique and analyzed by means of the four-quadrant technique (Vergès' quadrant), as well as implicative statistics and content analyses [24,26,36,37].
The sample of respondents was obtained by means of contacts in social networks. Data collection took place from 1 June to 17 June 2020 through a questionnaire sent both by e-mail (about 175,000 e-mails delivered) and direct message via social networks (about 1500 direct messages). In total, the study sample included 1780 participants who fully responded the questionnaire.
The questionnaire had two parts, totaling 17 questions, namely: part one with 7 questions, including the word evocation test and complementary information about the COVID-19 pandemic; part two with 10 questions related to the sample profile, including age, gender, educational background and political preference. The questionnaire was validated by two specialists in Social Representation Theory, being available at the Appendix A of this work.
There are distinct techniques to identify social representations, such as the words evocation technique herein adopted [38]. The words evocation strategy depends on gathering words communicated by the respondents when a particular interesting word or expression is introduced to them orally or in writing [39]. Thus, in this work, the participants were asked to list the five words or expressions that immediately came to their minds [10] when they were faced with the expression "COVID-19 pandemic".
Following the words evocation, ancillary questions were posed to perform implicative statistics and content analyses, which in turn were used to support the understanding, eventual modifications, and sensemaking of the central nucleus [37].
The words evoked were analyzed using the four-quadrant technique developed by Pierre Vergès, as well as implicative statistics and content analyses, which thus were utilized to help the identification, possible changes, and sensemaking of the central nucleus of the social representation [37].
As was said below, the words evoked were examined utilizing the four-quadrant method, by means of which the words evoked are split and assembled into classifications or clusters related to the social representation under investigation [24,36]. The four-quadrant technique cross-checks the evocation frequency of the categories-of a quantitative nature-with the order of evocation of same-of a qualitative nature [25,37,40]. Figure 1 below depicts the Vergès' four-quadrant technique.
Sustainability 2020, 12, x FOR PEER REVIEW 5 of 17 the social representation under investigation [24,36]. The four-quadrant technique cross-checks the evocation frequency of the categories-of a quantitative nature-with the order of evocation of same-of a qualitative nature [25,37,40]. Figure 1 below depicts the Vergès' four-quadrant technique. The average frequency of evocation (AFE) is calculated by the total number of evocations over the total number of distinct evoked words [15]. Besides which, the average order of evocation (AOE) of a category is obtained by considering the average order in which the words belonging to this category were evoked by the respondents, namely first, second, third, fourth or fifth place. The average of the AOE is then obtained by dividing the sum of all AOEs calculated by the number of distinct categories [10].
In this research, categories were sought to satisfy the Vergès quadrant, with special emphasis on the categories being located in the upper left quadrant, referred to as the central nucleus, and in the lower right quadrant, referred to as the second peripheral or peripheral system [25,38,40]. The lower left quadrant (contrast zone) and the upper right quadrant (first periphery) allow only an indirect understanding of social representation, as they speak of insights that are not so close to the central kernel [10,25,40].
After assembling the Vergès' quadrants, one carried out the implicative statistics and content analyses.

Results
In this section, a concise and accurate description of the results obtained is made, which will be discussed in more detail in the next section.

Sample Analysis
The characteristics of the sample of this study is presented in Table 1. The average frequency of evocation (AFE) is calculated by the total number of evocations over the total number of distinct evoked words [15]. Besides which, the average order of evocation (AOE) of a category is obtained by considering the average order in which the words belonging to this category were evoked by the respondents, namely first, second, third, fourth or fifth place. The average of the AOE is then obtained by dividing the sum of all AOEs calculated by the number of distinct categories [10].
In this research, categories were sought to satisfy the Vergès quadrant, with special emphasis on the categories being located in the upper left quadrant, referred to as the central nucleus, and in the lower right quadrant, referred to as the second peripheral or peripheral system [25,38,40]. The lower left quadrant (contrast zone) and the upper right quadrant (first periphery) allow only an indirect understanding of social representation, as they speak of insights that are not so close to the central kernel [10,25,40].
After assembling the Vergès' quadrants, one carried out the implicative statistics and content analyses.

Results
In this section, a concise and accurate description of the results obtained is made, which will be discussed in more detail in the next section.

Sample Analysis
The characteristics of the sample of this study is presented in Table 1. When the numbers in Table 1 are compared with the last demographic survey in Brazil, developed by the Brazilian Institute of Geography and Statistics (IBGE) (more information at https://educa.ibge.gov.br/jovens/conheca-o-brasil/populacao/18318-piramide-etaria.html (accessed on 30 November 2020)), it is clear that although the sample covers all Brazilian states, the majority of respondents (66%) live in the states of São Paulo and Rio de Janeiro-the most developed in Brazil. In addition, the level of education of the respondents in the sample is quite high by Brazilian standards, which means that the sample is composed mainly of people with a level of education above the Brazilian average. The reason for this lies in the fact that, due to the pandemic, the survey was carried out online, making it difficult for people without reliable access to the Internet to participate (in Brazil, at the end of 2020, almost 23% of the population had no access to the Internet, with the majority of the digitally excluded people being afro-descendant and / or the poorest. More information at https://dowbor.org/wp-content/uploads/2020/11/OCDE-O-Caminho-da-Era-Digital-no-Brasil.pdf (accessed on 30 November 2020)). In addition, most respondents are in the 26 to 45 age bracket. In fact, the percentage of young population (0-15 years old) is very high in Brazil and obtaining answers from young people of this age bracket was not expected.
Thus, although one has obtained a very robust sample with respondents from all Brazilian states, it has a bias in favor of more educated people, with access to the Internet and aged between 26 and 45 years. However, as this work intends to investigate the social representation of the COVID-19 pandemic for the Brazilian population in order to assess the Brazilian public policies to mitigate the COVID-19 pandemic, it is not a big issue that the research has privileged the population segment that has greater influence on public policies defined by governments at all levels.

Central Nucleus and Peripheral System
The central nucleus and the peripheral system can be determined by the participants' answers for the following question: "When you think of Coronavirus (or COVID- 19), what are the first five expressions that come immediately to your mind?" From the 1780 questionnaires fully answered, 8900 terms were obtained. Then, a semantic analysis of each evocation was provided to create the categories. Finally, 8772 terms were converted into Sustainability 2020, 12, 10434 7 of 18 17 different categories. The following step was to allocate the categories into the four quadrants of Vergès. To compose the quadrants, it was necessary to calculate the minimum average value of the evocation frequency [41]. In this research, the minimum average found was 495 and the maximum was 935, representing a total of 6535 evocations from 9 different categories, or 74.5% of the total 17 categories.
The last part of the process to allocate the categories within the quadrants requires the calculation of the average order of evocation (AOE), by means of the weighted average of the AOE of each category, in order to identify the cutoff point for the distribution of the categories [42,43]. Table 2 illustrates the frequencies of evocation and the AOE for each of the categories. This calculation enabled the proper arrangement of nine categories that comply with the minimum frequency of evocation (495) and the mean value of the AOE (3.26) within the Vergès' quadrant, as shown in Figure 2.
It is then recommended to carry out a similitude analysis to produce cognitive maps that allow the interpretation of the representation through implicative statistics, in order to validate the central nucleus and the peripheral system proposed in the Vergès' quadrants [40]. Therefore, this is done in the next section.

Similitude Analysis
The similitude analysis allows the evaluation of the connection level of the evoked elements by measuring the associations and the connectivity levels among the categories [44,45]. The similitude tree was generated for all the 17 categories of Table 2 ( Figure 3) and one found two main categories with greater relevance in the number of co-occurrences (connections), edges (lines) and frequency (circle size), namely Social Distancing and Fear. They have the strongest connectivity of all, with the largest number of edges and co-occurrences (see the numbers shown in Figure 3) with other categories. It is then recommended to carry out a similitude analysis to produce cognitive maps that allow the interpretation of the representation through implicative statistics, in order to validate the central nucleus and the peripheral system proposed in the Vergès' quadrants [40]. Therefore, this is done in the next section.

Similitude Analysis
The similitude analysis allows the evaluation of the connection level of the evoked elements by measuring the associations and the connectivity levels among the categories [44,45]. The similitude tree was generated for all the 17 categories of Table 2 ( Figure 3) and one found two main categories with greater relevance in the number of co-occurrences (connections), edges (lines) and frequency (circle size), namely Social Distancing and Fear. They have the strongest connectivity of all, with the largest number of edges and co-occurrences with other categories. When comparing the similitude tree with the Vergès' quadrant, those categories are confirmed as significant components of the central nucleus (in both frequency and AOE), being the greatest link among several other main categories that represent the COVID-19 pandemic.
Politics and Government, Disease, and Health and Prophylaxis also appear in the similitude tree When comparing the similitude tree with the Vergès' quadrant, those categories are confirmed as significant components of the central nucleus (in both frequency and AOE), being the greatest link among several other main categories that represent the COVID-19 pandemic.
Politics and Government, Disease, and Health and Prophylaxis also appear in the similitude tree as the three next higher categories in co-occurrence. However, while Disease and Health and Prophylaxis appear with lower AOEs-confirming their position in the central nucleus-Politics and Government has the highest AOE of all the five categories in the central nucleus. Thus, as its AOE is very close to the mean figure for the AOE, which separates the central nucleus from the first periphery, Politics and Government was moved to the first periphery [46], thereby leading to the final social representation of the COVID-19 pandemic as presented in Figure 4.

Content Analysis
A content analysis was also performed via the respondents' open responses (phrases-also called textual corpus), addressing the reasons for having chosen the words evoked about the coronavirus pandemic, in order to allow triangulation via comparative analysis with the results of the social representation obtained. The text corpus with open responses were used to build word clouds. This technique provides an easy visualization, by displaying the words in different font sizes, proportionally to their simple frequencies as depicted in Figure 5, which shows all the words evoked at least twenty-five times from over 1300 phrases analyzed.

Content Analysis
A content analysis was also performed via the respondents' open responses (phrases-also called textual corpus), addressing the reasons for having chosen the words evoked about the coronavirus pandemic, in order to allow triangulation via comparative analysis with the results of the social representation obtained. The text corpus with open responses were used to build word clouds. This technique provides an easy visualization, by displaying the words in different font sizes, proportionally to their simple frequencies as depicted in Figure 5, which shows all the words evoked at least twenty-five times from over 1300 phrases analyzed. Figure 5 shows that social distancing is more relevant than economy and employment. That is, one may realize that the universalism is more important than the utilitarianism for the Brazilian population surveyed, which supports via triangulation what have been already mentioned in this article. coronavirus pandemic, in order to allow triangulation via comparative analysis with the results of the social representation obtained. The text corpus with open responses were used to build word clouds. This technique provides an easy visualization, by displaying the words in different font sizes, proportionally to their simple frequencies as depicted in Figure 5, which shows all the words evoked at least twenty-five times from over 1300 phrases analyzed.  Figure 5 shows that social distancing is more relevant than economy and employment. That is, one may realize that the universalism is more important than the utilitarianism for the Brazilian population surveyed, which supports via triangulation what have been already mentioned in this article.

Discussion
As can be seen from the final Vergès' quadrant ( Figure 4), the following categories comprise the central nucleus of the social representation of the COVID-19 pandemic, namely Fear, Social Distancing, Health and Prophylaxis, and Disease, as explained below.
-Social Distancing: associated with the expressions quarantine, social isolation, feelings of confinement and loss of freedom, as well as the "stay at home" jargon. The following transcripts obtained from content analysis further explain the meaning of this category: -I am afraid of contracting the disease and that my family and friends can also be infected, which forces me to do the isolation.

-
Isolation is the only action we should take in the absence of knowledge for new treatments.

-
The disease can lead to death and social isolation is the best way to avoid it, because there is a chaos in the health system.
-Health and Prophylaxis: refers to human life and care associated with the pandemic, including expressions such as alcohol gel, hand washing, mask, among others. The excerpts below help to better understand this category: - The pandemic requires personal care with direct consequences for society.

-
One requires care during this period so that it is possible for us to return to the "new normal". -Alcohol gel (70 • ) to protect us, otherwise we will die. -Mask for everyone to wear it daily-it's boring. -I believe that through prevention we can go through this delicate moment, with the minimum of deaths.
Finally, the categories Fear and Disease refer, respectively, to the bad feelings associated with the pandemic (containing words such as fear, panic, anguish, and dread) and the virus itself and its associated symptoms. These are two categories whose names are confused with the construct analyzed, since the pandemic is an illness and, since there is yet no cure, it causes fear in people. The evocation of words clearly associated with the analyzed expression is common in word evocation tests [10,26].
On the other hand, the categories Economy and Employment and Changes and New Normal were positioned in the peripheral system of the Vergès quadrant, that is, they are less cited categories than those belonging to the central nucleus and in a higher evocation order-namely, they are not the first expressions that come to the respondents' minds during the process of evoking words. As they are in the peripheral system, issues related to the economy and employment, as well as the perception of change and a new post-pandemic normal, address the differences between the interviewees, supporting the heterogeneity of the group and accommodating the contradictions brought by specific contexts [10,24,43]. Thus, these categories-depending on circumstances-are subject to negotiation by the interviewees, regarding their relationship with the social representation under analysis. In other words, they are not rigidly and immutably associated with the COVID-19 pandemic.
The Economy and Employment category refers to expressions related to the economic impact of social isolation, as well as unemployment, crisis and poverty, as can be seen in the transcripts below: -Concern about our salaries and the poorest people. - We will have tragic consequences for our economy. -It is stopping the country and increasingly harming the poorest people.
Thus, it seems that, for the Brazilian population, although the economic issue is important and recalled by the interviewees, universalism (social isolation, health and prophylaxis) supersedes utilitarianism (economy and employment)-that is, health and life are worth more for Brazilians than preserving the economy and jobs in the pandemic. In summary and in line with [47,48], for the majority of the population, the COVID-19 pandemic is anchored in the health issue and in the necessary social isolation, being objectified by its association with disease and death.
Likewise, the Changes and New Normal category was also placed in the peripheral system. This category concerns future changes resulting from the pandemic, new work routines, a new daily life and the digital transformation of society, encompassing words such as digitization, future, adaptation, reinvention, learning, technology, among others.
Thus, Changes and New Normal, being a category subject to negotiation for respondents [46] and depending on circumstances and context, may not even establish itself over time, in disagreement with what has been foreseen about a new normal in a post-COVID-19 world [49][50][51][52].
Additionally deserving special attention is the fact that the sub-category digital transformation is embedded into the Changes and New Normal category-positioned in the peripheral system of the Vergès' quadrant. Thus, the Brazilian population interviewed do not understand the digitalization of society as an inevitable trend due to the COVID-19 pandemic-a fact taken for granted in some studies [51,53,54]. It is true that this finding may be typical of a developing country like Brazil. Indeed, Brazil as well as other emerging economies present further significant challenges regarding digital transformation due to inefficient ICT infrastructure and higher digital exclusion [55][56][57]. In this regard, it is worth recalling the serious difficulty that the Federal Government-through Caixa Econômica Federal (CAIXA) (CAIXA is the Brazilian public bank in charge of social programs funded by the Federal Government (www.caixa.gov.br))-had in operationalizing the release of financial aid to the needy via mobile apps during the pandemic [58].
Finally, the Politics and Government category was positioned in the first periphery of the Vergès' quadrant. That is, it is a category that is often cited, but does not come to mind immediately-thus presenting a lower priority for respondents. It refers to aspects related to both criticism and support to national and international political actors, involving words such as politics, president, government, corruption, opportunism, etc.
The categories located in the first periphery of the Vergès quadrant do not allow further conclusions [15,26]. However, due to the high frequency of evocation associated with this category, there is undoubtedly a political view of the COVID-19 pandemic [59][60][61], which is, however, less relevant for the population than the health problems associated with the pandemic. In other words, this category, as it is located in the first periphery-an intermediate quadrant between the central nucleus (hosting a category related to health) and the peripheral system (hosting a category related to the economy)-seems to function as a moderating variable in the utilitarianism vs. universalism debate that exists today in Brazil [62,63].
After discussing the results, the conclusions of this research can be presented in a consolidated manner in the next section.

Conclusions
The first conclusion of this research is that, in Brazil, during the COVID-19 pandemic, the universalist paradigm triumphed over the utilitarian one. That is, Brazilian society associated the social representation of the COVID-19 pandemic primarily with health issues (placing it in the central nucleus), leaving in the background (i.e., in the peripheral system) the economic impact derived from its management and mitigation. This does not mean that the economic crisis associated with the pandemic was not perceived by the population. However, the fact that it was situated in the peripheral system vis-à-vis the social isolation measures adopted shows that, unequivocally, Brazilian society mostly understood that health and life are more important than the country's economic recovery.
On the other hand, surprisingly, it appears that changes and a "new normal"-especially an accelerated digital transformation of Brazilian society-may simply not occur, since this category (Changes and New Normal) was positioned in the peripheral system of social representation of the COVID-19 pandemic. Interestingly, this scenario has been forecast by several companies and gurus around the world (see CIFS, 2020, for example), although there are already signs that this "new normal" may not be so different from the old "normal" [64].
Finally, the political variable, positioned in the first periphery of Vergès' quadrant of the social representation of the COVID-19 pandemic, although relevant in a number of citations, is not a priority for Brazilian society, functioning more as a moderating variable between the health debate versus economy. It is also interesting to note that this category is intrinsically linked to the Media category, as can be seen in the similitude tree ( Figure 3). In other words, for the Brazilian society, the political division installed in Brazil today [65] is largely associated with the media-whether traditional or social.
It is important to highlight that the theory of social representation, operationalized by the words' evocation technique, allows Brazilian society to become conscious of judgments, prejudices and perceptions, of which-in most cases-they are not even aware. In addition, the triangulation of several methodological approaches in this work-Vergès' quadrant, implicative and content analyses-reinforces the conclusions presented here, which is an additional contribution of this research.
As the authors of this research are Brazilian citizens and witnessed how the country managed and tried to mitigate the COVID-19 pandemic, a personal interpretation is that the task of dealing with a pandemic in a country with so many social, technological, economic and political disparities, such as Brazil, is a very difficult task. Indeed, there are so many inequalities in Brazil that it is a challenge for governments at all levels to establish a unique COVID-19 mitigation public policy-based on the utilitarianist or universalist paradigm-that is well accepted by all of society. It was realized that for the wealthiest citizens in Brazil, closing the economy and staying at home was a wonderful solution (the universalist perspective). On the other hand, for the poorest citizens, the vertical lockdown was preferred (the utilitarianist perspective), as they belong to a portion of the society that needs to move in cities, not to mention the great difficulty of having adequate infrastructure to work at home-if that option were given. This personal interpretation shows how difficult it is to make comparisons on how, for better or for worse, different countries managed and mitigated the COVID-19 pandemic in their territories and, consequently, how flawed it is to imitate the solutions taken by other countries. Indeed, context is important-and in this case, it is king.

Research Limitations
The first limitation of this study refers to the clustering of the words evoked. Although rules had been previously set up from the bibliographical references, errors of inference may have been made during the categorization of the words evoked. However, an attempt was made to mitigate this potential flaw by applying implicative and content analyses. Furthermore, although a very robust sample was obtained with respondents from all Brazilian states, there was a bias in favor of more educated people, with access to the Internet and aged between 26 and 45 years. In other words, although the survey covered practically all segments of the Brazilian population, a substantial part of the respondents is in the middle class or higher income bracket. It is, therefore, a portion of the population that has a greater capacity to defend itself from the perverse economic effects of the COVID-19 pandemic. This fact may have influenced the prevalence of universalism over utilitarianism. In other words, the work had not the opportunity to hear the digitally excluded people in Brazil, namely the ones who most of the time may not be working in home office and by consequence tend not to opt for the universalist paradigm.
Finally, the survey data was collected about three months after the COVID-19 pandemic outbreak in Brazil. Thus, it is possible that the disastrous economic effects associated with it were not yet clearly visible to the respondents. Thus, further longitudinal approach research may evaluate the possible modification of the social representation of the COVID-19 pandemic in Brazil over time, comparing the results to be obtained with those presented in this work.
The abovementioned limitations notwithstanding, this work sought to contribute to a better understanding of how Brazil has managed and mitigated the COVID-19 pandemic. Hopefully, this knowledge may help governments to better deal with potential further pandemics.