Children’s conceptions of coronavirus

Public communication of COVID-19

Since the emergence of COVID-19 early in 2020, public communication worldwide has been inundated with messages about infection and transmission risks, as well as preventive measures like social distancing, handwashing, and sanitation, or wearing a face mask (Zou and Tang, 2021). Among the general public, children have also been receiving these messages (Kayemba et al., 2020; Mukherjee, 2020; Soma, 2020), which they are expected to handle both cognitively and emotionally (Valadez et al., 2020) and to commit to pertinent prevention practices. In this pandemic context non-experts – among them children – adopt social representations (Assante and Candel, 2020; Idoiaga et al., 2020), that is, shared ideas founded on direct experiences and information they receive through the media and their social interactions. Social representations serve the purpose of transforming scientific knowledge and rendering it familiar to non-experts, thus contributing to a common understanding of controversial, or risky issues (Moscovici, 1988). Inherent in social representations are the mechanisms of anchoring unfamiliar threats into pre-existing knowledge and objectifying abstract ideas by concretising them through images, symbols, or tangible objects (Höijer, 2010; Mayor et al., 2013; Smith and Joffe, 2013).

Visual images about the Sars-CoV-2 coronavirus ¹ have been powerful objectification means in public communication and meaning-creation during the pandemic, as has been the case in previous crises or threatening circumstances (Alcibar, 2018; Höijer, 2010; Stark and Stones, 2019). Such images are especially useful in representing invisible entities. Therefore, due to the emergency caused by COVID-19, visualizations of coronavirus have dominated public discourse, with varying degrees of ‘realism’ and artistic processing. In fact, none of these visual representations is literally realistic, that is, photographic, but result from the application of sophisticated visualization techniques on microscopic imagery (Joubert and Wasserman, 2020).

Visual images addressing the general public convey meanings more easily than verbal communication, by rendering complex information accessible to non-experts (Joubert and Wasserman, 2020). Furthermore, public visual communication goes beyond the representation of scientific information, by addressing social, cultural, emotional, or other aspects of the pandemic crisis (Zou and Tang, 2021). Images are therefore socio-cultural mediating constructs that rely on specific visual rhetorics, thus shaping and echoing public emotions, attitudes, and understandings (Smith and Joffe, 2013).

Visual rhetorics in representations of coronavirus involve two prevalent characteristics. The first concerns the morphology of the virus, with popular imagery emphasizing its spherical shape and the spikes on its surface. These spikes are often depicted with exaggeration, which has been interpreted as a deliberate choice of image creators for drawing attention to the threat posed by the coronavirus (Joubert and Wasserman, 2020). The second prevalent characteristic in coronavirus images is anthropomorphism, that is, the tendency to attribute human abilities, behaviours, and features like a face with eyes, mouth, and facial expression, a body, or human-like limbs to the virus (McGellin et al., 2021). Anthropomorphic depictions are commonly used when representing entities entailing uncertainty and fear, as is the case with Sars-CoV-2, and they are considered as an accessibility aid for non-experts, since they relocate an invisible, unknown entity to a human framework, thus rendering it concrete and familiar (De Rosa and Mannarini, 2020).

Children’s conceptions of microorganisms and infectious diseases

Micro-organisms are interesting objects for studying lay conceptions. They have extraordinary impacts on human lives, since they cause diseases and have historically been responsible for a large variation in human mortality and morbidity (Jee et al., 2015). As COVID-19 is a new global situation, and related research on its public understanding is embryonic, existing studies with microorganisms and infectious diseases can assist in identifying similar or evolved ideas in understanding novel infectious diseases. In the earliest known study, Nagy (1953) exploring children’s graphic representations of and conceptions about germs, concluded that germs are either identified as animal-like worms or insects or depicted in geometric shapes (points, circles, squares, etc.), findings that were duplicated in subsequent studies (Byrne, 2011; Byrne et al., 2009; Byrne and Grace, 2010). In addition, participants believed that all germs were harmful to humans, an idea confirmed by subsequent studies with older students (Simonneaux, 2000; Vasquez, 1985), who additionally associated poor hygiene with germs. Prokop et al. (2016) also found that the typical shape of micro-organisms in 4- to 8-year-old children’s drawings was circle-like. Anthropomorphism has been also reported in several studies (Faccio et al., 2013; Jee et al., 2015; Prokop et al., 2016), as many children attributed both physical (two or more hands, and two legs) and mental human characteristics and intentionality to micro-organisms. Moreover, Byrne (2011) found anthropomorphic ideas about micro-organisms decline with age; 7-year-olds tended to have a highly anthropomorphic perspective, consistent with young children’s naïve biology (Carey, 1985), while anthropomorphic ideas persisted in a substantial minority of 11- and 14-year-olds (Byrne and Grace, 2010). Older students appeared to be aware that anthropomorphisms are 'ways' of explaining rather than literal expressions of reality, but continued to use them to explain particular concepts and ideas (Byrne, 2011; Byrne and Grace, 2010).

In addition, anthropocentric conceptions about micro-organisms were recorded in all age groups. Germs are discussed predominantly in terms of their harmful effects on humans, indicating a negative view of micro-organisms as essentially malevolent (Byrne, 2011; Byrne and Grace, 2010), conceptions possibly deriving from non-formal learning (Ruiz-Gallardo and Paños, 2018).

Furthermore, the distinction between viruses and bacteria seems to be a difficult task for young children, as knowledge concerning viruses, their biology, and their importance as causal agents for infectious diseases distinct from bacterial ones seems rather fragmentary even among university students (Simon et al., 2017). A progression towards more scientifically accurate knowledge has been suggested across different age groups and expertise levels (Jee et al., 2015), but this development does not always follow a predictable, linear developmental trajectory from novice to expert. Thus, participants with varying levels of expertise have been found most knowledgeable about symptoms of microbial illness, treatments of symptoms, and routes of transmission for respiratory illnesses, and this knowledge seems to be tightly linked to their prior experiences with colds and flu (Jones and Rua, 2008).

Children’s understanding of COVID-19

Children are particularly affected by COVID-19, in terms of their physical and mental health, well-being, and education (Garcia de Avila et al., 2020; Waller et al., 2021). Their daily lives have been vastly disrupted and imposed restrictions combined with changes in family routines and practices could have significant psychological repercussions on them (Mukherjee, 2020; Soma, 2020; Valadez et al., 2020).

Therefore, children need to be empowered and supported in understanding and responsibly participating in mitigating the pandemic, taking into account the implications for their everyday activities, such as going to school, playing with peers, or being in contact with elderly or vulnerable family members (Garcia de Avila et al., 2020; Li et al., 2020; Provenzi et al., 2020).

Awareness of public health issues, such as COVID-19, is crucial to instil appropriate hygienic and prevention practices that would help control the pandemic, while strengthening communication between groups of people and improving mental health (Li et al., 2020; Zou and Tang, 2021). To design and implement effective teaching or sustainable communication campaigns about COVID-19 addressing children, it is critical to gain an insight into how they understand this unprecedented condition and handle it emotionally (Assante and Candel, 2020; Garcia de Avila et al., 2020; Idoiaga et al., 2020; Kayemba et al., 2020; Majid et al., 2020).

As with any other conceptual domain, children form their understanding of COVID-19 by combining a large amount of information from different sources, including media (multimodal) messages, explanations, and emotional reactions of family members, or other social contacts (Provenzi et al., 2020). Thus, scaffolding children’s meaning-making of COVID-19 is essential to support them in handling this unpredictable situation and in avoiding traumatic effects (Assante and Candel, 2020; Idoiaga et al., 2020).

However, little is known to date on how and to what extent this abundance of information is understood by children, and thus far related studies reveal different – often contradictory – findings. On the one hand, it has been suggested that children lack an adequate understanding of infectious diseases and appropriate prevention practices (Jee et al., 2015; Li et al., 2020). On the other, children’s awareness of COVID-19 has proven significant by several studies. Thus, children have been found to (a) differentiate between coronavirus (cause) and COVID-19 (result); (b) be aware of symptoms of the disease like cough, or fever, but also of the possibility of asymptomatic infection; (c) recognize the value of COVID-19 prevention measures; and (d) have a preliminary understanding of the virus transmission through droplets (Kayemba et al., 2020) and its high contagiousness (Idoiaga et al., 2020). Despite these findings, none of the relative studies attempted to investigate systematically how children’s understanding of COVID-19 changes during their development.

Furthermore, the above-mentioned studies relied on language-based data collection instruments. However, having participants express their thinking about microorganisms through drawings has several advantages (Byrne et al., 2009; Byrne and Grace, 2010; Nagy, 1953; Prokop et al., 2016). In an earlier study on students’ understanding of the HIV/AIDS, Mutonyi and Kendrick (2011) found that the participants’ drawings reflected a broad understanding of the topic, in terms of causes, effects, ways of transmission, and prevention. Thus, asking students to express their understanding of a complex scientific issue through drawing facilitates them in combining sociocultural with scientific knowledge in their representations. This selection has two crucial advantages. It allows children to express their understanding of scientific knowledge without having to rely on specialized vocabulary, while it provides a more comprehensive picture of their thinking and emotions compared to verbal communication alone (Jolley, 2010; Kress and van Leeuwen, 1996; Mutonyi and Kendrick, 2011).

For the aforementioned reasons, drawings have proved an effective means for exploring children’s conceptions of illness (Mouratidi et al., 2016; Piko and Bak, 2006; Zaloudikova, 2010). Drawing-based studies have confirmed that children identify illness primarily as a biomedical situation (Myant and Williams, 2005; Schmidt and Frohling, 2000), but have also revealed an appreciation of its consequences on one’s social and emotional adjustment (Bonoti et al., 2019; Zaloudikova, 2010).

This is no surprise, since public reactions to illness and viruses are highly emotional, due to the danger and uncertainty they entail (Joubert and Wasserman, 2020; Provenzi et al., 2020). As a consequence of the COVID-19 pandemic and its resulting restrictions, children have been found to report negative emotions – mainly fear (Assante and Candel, 2020; Garcia de Avila et al., 2020; Li et al., 2020), but also sadness, anxiety, anger, boredom, fatigue, loneliness and concern (Idoiaga et al., 2020; Soma, 2020). At the same time, positive emotions, like happiness resulting from spending time with family linked with a sense of security prompted by confinement measures (Idoiaga et al., 2020) have been expressed.

Despite the growing body of research related to the COVID-19 crisis and its consequences on children’s well-being (e.g. Garcia de Avila et al., 2020; Mukherjee, 2020; Soma, 2020; Valadez et al., 2020; Waller et al., 2021), few studies so far have focused on children’s perspective about this phenomenon (Idoiaga et al., 2020; Kayemba et al., 2020). These pioneering studies, although enlightening, have only shed light on some aspects of children’s conceptions of coronavirus and COVID-19, while questions about the development of these conceptions have not been addressed. For example, it would be interesting to investigate whether children’s knowledge of coronavirus follows the major developmental changes occurring in their biological knowledge around the age of 7 years (Duschl et al., 2007). Moreover, none of the existing studies have investigated children’s drawing representations of coronavirus or COVID-19, although visual communication has been a powerful information channel for the general public internationally.

The present study explores children’s conceptions of coronavirus as denoted in their verbal descriptions and drawings.

Specifically, the study addressed the following research questions:

Participants

The sample consisted of 344 children (156 boys and 188 girls) ranging in age from 4 to 10 years who were recruited from different parts of Greece. Participants were distributed in two age groups: Group A, involving 69 boys and 95 girls aged from 4 years, 0 months to 6 years, 11 months (n = 164, M = 68.81 months, SD = 10.62); and Group B, involving 87 boys and 93 girls aged from 7 years, 0 months to 10 years, 11 months (n = 180, M = 103.42 months, SD = 13.05).

Given the lockdown restrictions, a snowball sampling was adopted. Researchers identified families through their social networks, who then nominated other participants to take part in the study. Parents were informed about the aims of the study and were advised not to prepare or orientate their children for the topic under investigation. The procedure of data collection was explained and parents provided their oral consent for their children’s participation, which was reflected in their presence and assistance during the data collection procedure described below. Furthermore, both parents and children were given the choice not to participate or to interrupt the procedure in case they wanted to. This research has obtained the approval of the Ethics Committee of the institution of the first author.

Tasks and procedure

Data were collected between April and May 2020, during lockdown restrictions imposed for COVID-19. Children were tested individually while being at home through video calls on online platforms. Initially, they were asked to answer to the following question: ‘Do you know what coronavirus is?’ and then to use a white page and a set of coloured pencils or crayons and they were prompted ‘Would you like to draw coronavirus?’. There was no time constraint and children were given as much time as needed to complete their drawings. Τhe researchers did not intervene and if there were signs that children felt uncomfortable the procedure was ended. These cases were not included to the final sample. After completion of the drawing, each child was prompted to describe his or her drawing and the researchers, when needed, simply encouraged the children to describe the elements of their drawing. Children’s verbal responses were transcribed verbatim during discussions. Individual communication with children was facilitated by their parents, who ensured their connection to video calls via the links sent to them and subsequently photographed and emailed children’s drawings to the researchers.

Coding of data

Children’s verbal descriptions and drawings were content analysed by adopting an emergent coding approach (Stemler, 2000) to detect the recurring themes they included. Initially, two coders independently inspected the transcripts and drawings to identify verbally described and depicted items, namely any verbal or graphic element denoting coronavirus-related meanings. Next, items identified by the independent coders were compared and differences were reconciled, resulting in a common list. This list was used for independent analysis of the data, and inter-coder agreement calculated as k coefficient (Cohen, 1960) was satisfactory [.85, p < .01 and .89, p < .01 for descriptions and drawings, respectively], signifying that the frequency with which coders agreed was much stronger than by chance alone. Subsequently, all identified items were aggregated to four themes, namely: (a) Coronavirus, (b) Medical, (c) Psychological, and (d) Social Theme. The themes and items that emerged formed the analysis framework presented in Figure 1. Using this framework, the coders reviewed the transcripts of children’s comments and drawings by giving a score of 1 for the presence and a score of 0 for the absence of each theme and each item.

OPEN IN VIEWER

Figure 1.

The analysis framework.

For the first theme, Coronavirus, analysis focused on whether children’s verbal depictions were Realistic, namely whether they described coronavirus as a virus or a germ or illustrated morphological characteristics of coronavirus as commonly depicted in the media, that is, its spherical scheme and/or spikes respectively. Alternatively, when participants verbally identified coronavirus with the resulting disease, or when they added a crown and/or anthropomorphic features to their – otherwise realistic – drawings of the virus, their responses were categorized as Hybrid. Last, when children’s verbal descriptions or graphic representations ascribed supernatural or animal-like characteristics to coronavirus, they were classified as Mythical/Symbolic. Children’s failure to provide any description of or depiction of coronavirus was categorized as No response.

The second theme, namely the Μedical dimension of COVID-19, involved children’s verbal and/or graphic items related to Illness, virus Transmission, Symptoms of the COVID-19 disease, Medical care (such as hospitals, medical staff and equipment, ambulances, medicines, and vaccines), and personal hygiene Preventive measures (e.g. wearing a mask and/or gloves, using sanitizer, handwashing, etc.).

The Psychological theme involved items referring to psychological functions such as emotions or abstract thoughts. Specifically, in this theme were aggregated verbal comments or graphic representations of Emotional expression, or references and graphic signifiers of Danger/Death. Furthermore, analysis of the psychological theme in children’s drawings examined if they included Captions, or depictions of natural or urban Landscapes, which metaphorically denoted mood (e.g. empty streets, deserted woods, etc.).

The last theme, indicating the Social dimensions of COVID-19 included items related to one’s social interactions with others. Therefore, it involved references or depictions of Lockdown conditions; recommended or imposed Restrictions (like avoiding crowded, indoor spaces, or physical distancing); the Globality of the pandemic; and children’s limited possibilities to meet Significant others, such as peers or vulnerable relatives (e.g. grandparents) to protect them from being infected.

To investigate whether the incidence of the Coronavirus, Medical, Psychological, and Social themes and pertinent items varied as a function of age, a series of chi-square tests were conducted using the IBM SPSS Statistics v. 25 package in all cases that the prerequisites of the test were not violated, excluding from further analysis those items whose distribution within cells exhibited frequencies less than 5. Moreover, in order to investigate whether the emergence of themes and items varied between the two data collection techniques, we used Cochran’s Q test since this nonparametric test is suggested (Siegel and Castellan, 1988) for dichotomous variables (presence or absence of themes/items) in repeated measures designs (descriptions and drawings).

Coronavirus theme

Regarding the Coronavirus theme, analysis of participants’ verbal descriptions showed that most children either provided a Realistic description (Example 1), by referring to virus or germ (n = 144) or offered a Hybrid response (Example 2), by identifying the virus with illness (n = 142). Fewer children included Mythical items in their descriptions (n = 37), as in Example 3, or offered No response (n = 21). Quite frequently, though, children combined items from different themes in their responses (Example 4).

Example 1 (9-year-old girl):

I think coronavirus is a germ like other germs [. . .]. I think it’s bad because it kills.

Example 2 (5-year-old boy):

My dad told me that [coronavirus] is a flu, like when our throat aches, but we have to stay at home because the doctors have not yet found a medicine that we can take and stop getting sick and he told me that we are not going to suffer, but we have to be careful with grandmas and grandpas.

Example 3 (4-year-old boy):

I know what it is. It is a ball that you must not touch! It is like a bad snail because it won’t go away quickly. It’s bad because my mom won’t let me go out and she shouts at me whenever I cry because I want to go out.

Example 4 (4-year-old girl):

The bad germ wants to hang out with people but sees that everyone stays at home. [. . .] and this makes [the coronavirus] sad, he wants to cry.

Crosstabulation of these data showed significant differences between age groups [χ²(3, N = 344) = 18.17, p < .001], with younger children providing Mythical/Symbolic responses, or No response more frequently than expected and older ones indicating a preference to Realistic descriptions.

Regarding drawing representations of coronavirus, the majority of children produced Realistic images (n = 177), while a considerable number of drawings included Hybrid representations (n = 122). Few children drew coronavirus as a Mythical/Symbolic agent (n = 29), while 16 failed to depict coronavirus. It should be noted that 283 out of 344 children included spikes in their realistic or hybrid representations. Representative drawings scored as Realistic, Hybrid and Mythical/Symbolic are presented in Figure 2.

OPEN IN VIEWER

Figure 2.

Examples of (a) Realistic, (b), Hybrid, and (c) Mythical/Symbolic drawings of coronavirus produced by an 8-year-old boy, a 9-year-old girl, and a 5-year-old boy, respectively.

Participants’ age does not seem to differentiate drawing representations of coronavirus. This absence of age-related effects can be attributed to children’s tendency in both age groups to depict either a Realistic or a Hybrid coronavirus (see Table 1).

Medical, psychological and social themes

Analysis subsequently focused on the Medical, Psychological, and Social themes identified in children’s descriptions and drawings. The overall frequencies with which each theme and item was included in participants’ verbal descriptions and drawings (Table 1), indicates that in total, 259 participants referred to the Medical dimension of COVID-19 (Example 2), while 124 mentioned its Psychological (Examples 3 and 4) and 115 its Social aspects (Example 5). Prevalent items prompted verbally within these themes concerned Illness (n = 207), as in Example 6, Danger/Death (n = 113), Transmission (n = 105), Globality (n = 44), and Restrictions (n = 43). Children’s drawings primarily emphasized the Psychological theme (n = 162, Figure 3b), while depictions of the Medical (n = 71, Figure 3a) and the Social (n = 61, Figure 3c) themes were less frequent. Also, the items most frequently depicted involved Expression of emotions (n = 134), Captions (n = 53), Lockdown/Stay at home (n = 37), and personal hygiene Prevention measures (n = 34):

Example 5 (4-year-old boy):

I don’t know [what the coronavirus is], it just harms people and I can’t meet my friends, or hug my grandma.

Example 6 (5-year-old girl):

These little things [points at the coronavirus spikes he drew] get inside you . . . they can dig and go inside you. Then they multiply from 1 to 2, from 2 to 3, from 3 to 4 and so on.

Examining the frequencies of themes in children’s verbal responses, analysis indicated statistically significant age differences only for data corresponding to the Psychological theme [χ²(1, N = 344) = 6.25, p < .05], with Group B making relevant comments more frequently than Group A. No statistically significant differences were detected between age groups regarding the introduction of the Medical, Psychological, and Social themes in participants’ drawings.

OPEN IN VIEWER

Figure 3.

Examples of drawings involving items of the (a) Medical, (b) the Psychological, and the (c) Social theme produced by a 7-year-old girl, a 7-year-old boy, and a 10-year-old girl, respectively.

Accordingly, significant age differences were detected for particular items within themes. Specifically, (a) Transmission [χ²(1, N = 344) = 9.20, p < .05] was invoked more frequently by the younger group than by the older one; (b) Signifiers of Danger/Death [χ²(1, N = 344) = 11.68, p < .001], were more frequently alluded to by younger participants than by their older counterparts; (c) Lockdown/Stay at home [χ²(1, Ν = 344) = 5.08, p < .05] was more frequently mentioned by Group A than by Group B; and (d) Restrictions [χ²(1, N = 344) = 4.50, p < .05] were mentioned significantly more frequently within younger children than within older ones; and (e) references to COVID-19 Globality [χ²(1, N = 344) = 12.59, p < .001] were more frequently identified in older than younger participants’ responses. Children’s drawings did not reveal extensive differences concerning the items preferred, with the exemption of Transmission [χ²(1, N = 344) = 6.65, p < .05] and Captions [χ²(1, N = 344) = 6.11, p < .05] which were more frequently introduced in older than in younger children’s depictions.

Ιn a next step, data presented in Table 1 were analysed in order to investigate whether the emergence of themes and items varied between the two data collection techniques. Thus, children indicated more Medical [Cochran’s Q(1, N = 344) = 157.78, p < .001] and Social themes [Cochran’s Q(1, N = 344) = 22.09, p < .001] in their verbal responses than in their drawings, while Psychological themes [Cochran’s Q(1, N = 344) = 8.80, p < .05] were displayed more frequently in drawings than in children’s verbal responses.

Analyses related to specific items showed that children employed more often in their descriptions than in their drawings items of Illness [Cochran’s Q(1, N = 344) = 182.49, p < .001], Transmission [Cochran’s Q(1, N = 344) = 56.98, p < .001], Signifiers of Danger/Death [Cochran’s Q(1, N = 344) = 91.26, p < .001], Restrictions [Cochran’s Q(1, N = 344) = 22.22, p < .001], Globality [Cochran’s Q(1, N = 344) = 13.52, p < .001] and Significant Others [Cochran’s Q(1, N = 344) = 25.97, p < .001]. On the other hand, children displayed more often in their drawings than in their descriptions items of Preventive measures [Cochran’s Q(1, N = 344) = 9.80, p < .05] and Emotional Expression [Cochran’s Q(1, N = 344) = 81.67, p < .001]. The frequencies of items denoting Symptoms, Medical Care and Lockdown/Stay at home did not differ across the two data collection techniques, while Captions and Nature/Landscape items emerged only in children’s drawings.

A separate note should be added for the specific emotions mentioned or depicted by participants. Interestingly, different emotions emerged between descriptions and drawings. Analytically, children mostly referred to fear (n = 12) or sadness (n = 8) while describing coronavirus, whereas they mentioned less frequently anger (n = 3) and happiness (n = 2). Contrarily, anger was the predominant emotion (n = 39) in children’s drawings, followed by happiness (n = 38), sadness (n = 32) and fear (n = 22). Other emotions, like hope, boredom, or hatred were sporadically observed in both descriptions (n = 2) and drawings (n = 3).