Signals and meaning-context articulations: role of emotions and their salience in the social representations of protective policies in 11 countries

Citizens interpret protective policies through socially shared systems of meaning that integrate cognitive and emotional dimensions. Using Social Representations Theory and Abric's structural approach, we examined the emotions and their articulation with the central elements of citizens' social representations of protective policies with a cross-national comparison of 11 countries. Based on 10,147 valid cases, a prototypical analysis was conducted. Across countries, emotions occupied peripheral and contrast zones. In the pooled structure, only Fear (n = 123), Calmness (n = 79), and Joy (n = 34) met inclusion thresholds, all located in the second periphery. Fear was the only emotion present in all countries, whereas Calmness and Joy varied across countries. Disappointment, Disgust, Shame, and Tranquility were specific to national contexts. Similarity analysis revealed two coherent affective-representational configurations centered on Fear and Calmness, linking emotions with distinct categories despite their peripheral structural position. Therefore, peripherality may not imply relational marginality, and emotions function as flexible representational resources, contextualizing and operationalizing the meaning of protective policies with citizens' contexts. Findings suggest the relevance of emotional literacy to protect the articulation and signaling role of emotions.

In a world shaped by space-time compression (Harvey, 1989), in which technology, transport, and communication advances reduce perceived distances and time, contemporary societies are increasingly characterized by social acceleration (Rosa, 2013), understood as a continuous speeding-up of technological and social change, everyday practices, and decision-making processes. At the same time, citizens' attention has become a contested and exploitable resource, actively competed for by multiple stakeholders (Hayes, 2025; Rose-Stockwell, 2023; Wu, 2016). This competition unfolds in an environment of information abundance, where users are confronted with more content than they can meaningfully process, further straining attentional capacities (Kowalski and Jung, 2021). Threat-related issues and policies framed as responses to them (Albertson and Gadarian, 2015) are presented to citizens in multiple ways and moments. Under such conditions, emotions play an increasingly central role in rapid decision-making (Kahneman, 2011; Slovic et al., 2007). Emotionally charged discourse has become a widespread communicative strategy, particularly in political and policy contexts (Moliner and Piermattéo, 2025). This shift has coincided with the rise of actors who actively seek to evoke, exploit, and shape citizens' emotions within the policy process–emotional policy entrepreneurship (Maor, 2024; Maor and Gross, 2025). Therefore, analyzing how social representations (SRs)-collectively produced meanings that organize how an object is perceived, discussed, and acted upon (Abric, 2001, 2003; Moscovici, 2008)—of protective policies embed emotional components, and the position they occupy within citizens' sociocognitive structures, provides critical insight into how emotions shape the salience of political issues and condition political engagement.

How citizens' emotions are mobilized in response to protective policies—policies framed as protective (Albertson and Gadarian, 2015; Wenzelburger, 2025)—remains insufficiently examined in the literature (Stempel, 2025). Given that SRs can be understood as cognitive-emotional processes (Bouriche, 2022), analyzing SRs offers a valuable lens for uncovering the emotional dynamics through which citizens make sense of such policies. This approach enables examining how emotions stabilize within the representational structure, how they articulate with the central core, and how they ultimately shape citizens' salience of perceived threats and of the policies framed as responses to them.

Through the analyses of SRs of protective policies, this study aims to investigate the emotional components that citizens attribute to protective policies in 11 countries: the Czech Republic, Denmark, France, Germany, Ireland, Israel, Italy, Poland, Portugal, Sweden, and the United Kingdom. After observing the position of emotions in the structure of SRs, the possible articulation with the central elements will also be assessed, considering the position occupied and guided by the literature, the function that the salient emotions have in political engagement and decision-making processes of citizens.

In this age of insecurity, where threats and related anxieties are present in politics (Wenzelburger, 2025), protective policies emerge as technical responses and as political narratives that define what, or who, needs protection from what threats. Therefore, what policies citizens perceive as protective might be impacted by framed or unframed threats (Albertson and Gadarian, 2015). Therefore, the object “protective policies” may be organized around a dialogical thema (Marková, 2003), opposing insecurity (threats to be avoided) and security (the desired state of protection). Rather than representing alternative definitions of the object, these poles constitute complementary dimensions through which its meaning is socially negotiated.

Any policy issue may give rise to multiple possible interpretations and solutions (Kingdon, 2014; Zahariadis, 2003), raising the question of why some alternatives are selected over others (Hoefer, 2022). Given that ambiguity—the coexistence of multiple and competing interpretations—constrains the effective use of rationality (Hoefer, 2022) and inherently features the political context (Zahariadis, 2003), and considering that time, like other resources, is limited, the role of emotions becomes increasingly salient. This is particularly evident in the current accelerated world, marked by advances in communication and transportation technologies that render events more interconnected, proximate, fast-moving, and information-overloaded (Harvey, 1989; Kowalski and Jung, 2021; Rosa, 2013). In this context, several scholars have emphasized the role of emotions in decision-making processes (Damasio, 1994; Kahneman, 2011; Slovic et al., 2007), especially concerning rapid and immediate decisions, which are increasingly used in contemporary policy environments.

Emotions can be understood as contextual phenomena that have evolved to help individuals monitor and manage concerns relevant to survival (Cochrane, 2019; Hartmann, 2024). Their duration is typically short, allowing behavioral flexibility and preventing the unnecessary allocation of the citizens' resources (Scherer, 2005). Although emotions are often triggered by situational stimuli (Dolan, 2018), appraisal theory (Lazarus, 1991) proposes that they arise from individuals' cognitive evaluations of the personal significance of events in relation to their goals, values, and coping resources. Accordingly, different emotions reflect distinct appraisal patterns, which may support the interpretation of the functional meaning of emotional components associated with protective policies. Ahmed (2014) argues that emotions are not private or irrational, but public and political forces that shape identities, belonging, and exclusion. Emotions are naturally connected to politics and can influence the political perceptions and behavior of citizens (Shah, 2024). They are integral to protective policies and central to political life, shaping citizens' responses while also informing how such policies are framed to elicit anticipated emotional reactions from the public (Gadarian and Brader, 2023; Maor, 2024; Penczek and Hamer, 2025). Therefore, emotions play a crucial role in shaping how individuals process political information and make decisions (Capelos, 2011; Webster and Albertson, 2022).

In contemporary politics, emotions such as fear and anger are frequently instrumentalized in public debates and policymaking, particularly in areas like security, surveillance, and immigration (Valdés, 2024). Gadarian and Brader (2023) further argue that emotions—such as fear, anger, and enthusiasm—influence how people engage with political campaigns and form policy preferences, enhancing participation in some contexts while potentially undermining deliberation in others. Similarly, Brader and Marcus (2013) show that fear motivates citizens to seek out additional information, improving factual understanding, whereas anger narrows attention and reinforces partisan biases. Fear and anger can also impact trust in government (Erhardt et al., 2021). Arceneaux (2012) demonstrates that fear-inducing arguments that evoke loss aversion tend to be more persuasive, even when counterarguments are presented. When emotions, including fear and anger, become embedded in policymaking, they create lasting cycles that generate further emotions, as seen in security tensions (Crawford, 2013).

Recent work suggests that emotions can be understood as signals of the tension that individuals and groups experience in relation to an object—real or imagined—when it becomes salient through unexpected or unfamiliar events (Bouriche, 2022). In times of insecurity, the emotions shared within groups take on greater political significance (Hall and Ross, 2019). Those emotions, along with their role in public, political, and international domains, have increasingly been seen as relevant within different fields (Åhäll and Gregory, 2013; Beattie et al., 2019; Gustafsson and Hall, 2021).

Nevertheless, the emotional dimension of citizens' responses to different types of protective policies remains insufficiently examined (Stempel, 2025). While Penczek and Hamer (2025) examine the emotions elicited by a predefined set of protective policy areas, their study did not investigate the broader emotional representations that spontaneously emerge when individuals think about protective policies, nor whether these patterns extend beyond a single national context. The present study addresses these gaps by investigating the emotions associated with SRs of protective policies across 11 countries. Conceived as socially elaborated and shared forms of knowledge, SRs do not merely reflect reality but function as systems of interpretation that guide action and provide a framework through which social objects are understood and pre-decoded. Examining their emotional dimension, in light of the specific national context in which citizen protection policies are developed and implemented, provides a broader understanding of the meanings attached to these policies and how they may orient citizens‘ perceptions and responses to this policy domain.

Social Representations Theory (SRT) and Abric's structural approach of SRs (Abric, 2001, 2003; Moliner and Abric, 2015) provide a robust framework for addressing this objective. According to SRT (Moscovici, 2008), SRs—shared systems of meaning that enable individuals and groups to interpret, communicate about, and act upon social reality (Abric, 2001; Moscovici, 2008; Valentim, 2022)—contribute to the legitimation and naturalization of the worldviews held by citizens or social groups.

Emotions play a crucial role within SRs: they sensitively shape how social objects are perceived and evaluated (Vidrio, 2019). This applies to abstract objects, including protective policies, whose meaning may become more accessible through emotionally charged elements. Emotional sensitivity is central to processes of error detection and (re)negotiation of shared knowledge, contributing to meaning construction through social sharing (Rimé, 2020). Moliner and Piermattéo (2025) argue that emotions function as signals of instability within SRs, prompting individuals to seek interpersonal interactions aimed at restoring representational coherence. Hence, emotions support the identification of representational content and the understanding of how SRs are organized and transformed.

A complementary perspective to the signaling role of emotion is provided by the Affective Intelligence Theory (Marcus et al., 2000), which explains how emotions regulate information processing and political judgment. It argues that citizens rely on two emotional systems: the disposition system—which reinforces habitual behaviors and established preferences through enthusiasm and satisfaction when the outputs match and frustration and depression when they do not match—and the surveillance system—which monitors the environment for novel or threatening situations. When activated, the second interrupts habitual responses, increases attention, and encourages individuals to reassess their prior beliefs and behavioral choices. When no unexpected outputs are detected, the surveillance system sustains a state of calm or tranquility. When novelty or threat is detected, the surveillance system may elicit anxiety, increasing information seeking and deliberative decision-making. Originally, the theory defined anxiety as a broader negative affective state encompassing affective descriptors (e.g., afraid and upset) rather than the emotion anxiety in the strict sense. Subsequent developments suggest that different emotional states may regulate representational dynamics in distinct ways: whereas anxiety and fear facilitate information seeking and the reconsideration of existing interpretations, anger tends to reinforce prior beliefs and established interpretative frameworks (Marcus et al., 2019). Moreover, given that emotions can reshape shared knowledge by stimulating social exchanges around a given object (Piermattéo, 2022), it becomes essential to examine the role of emotional anchoring—a specific modality of anchoring through which a new or unfamiliar object becomes associated with socially recognized emotions, rendering it meaningful and evaluable (Höijer, 2010)—in relation to protective policies.

Within Abric's Structural Approach, SRs are organized as a dual system composed of a central core and peripheral elements (Abric, 2001, 2003; Moliner and Abric, 2015). The central core constitutes the most stable and consensual component of a SR. Closely linked to the collective memories and values of a social group, it provides the representation with overall meaning and coherence. Because it is deeply anchored in shared beliefs, it is particularly resistant to change and defines what is considered acceptable or unacceptable within a given representational field. It fulfills several key functions: a meaning function, organizing and attributing sense to peripheral elements; an organizational function, structuring the internal logic of the representation; and a stabilizing function, reflecting and reinforcing the group's social norms and values, ensuring representational durability.

The peripheral elements surround the central core and are more flexible, context-sensitive, and closely linked to everyday experience (Abric, 2001, 2003; Moliner and Abric, 2015). They allow SRs to adapt to concrete situations without fundamentally challenging the central core's stability. Thus, the periphery acts as a buffer between the representation and its social context, absorbing contradictions and enabling individual variation. Peripheral elements also perform several important functions (Abric, 2001, 2003; Moliner and Abric, 2015): an adaptive function, allowing the representation to respond to situational constraints and social change; a defensive function, protecting the central core from direct challenge; and a regulatory function, guiding everyday practices and behaviors by translating abstract core values into concrete action.

Operationally, prototypical analysis distinguishes four structural zones according to the frequency and order of evocation of representational elements (Abric, 2003; Vergès, 1992). The central core contains highly frequent and readily evoked elements. The first periphery comprises highly frequent but less readily evoked elements that support the stability of the representation. The contrast zone (also known as the mute zone of SRs) includes less frequent but highly salient elements that may characterize specific subgroups or emerging representations. The second periphery contains less frequent and less salient elements that are more sensitive to contextual variation and individual experience.

Research on emotions within representational processes can be broadly divided into two strands (Moliner and Piermattéo, 2025). The first concerns the position and role of emotional components within the structure of SRs. The second focuses on the influence of emotional dynamics on SRs' (trans)formation. Although the transient nature of emotions may appear paradoxical when contrasted with the relative stability typically associated with SRs, this tension can be partly resolved by considering that emotions may take more enduring forms, like sentiments or emotional attitudes (Lheureux and Guimelli, 2009; Moliner and Piermattéo, 2025). This supports the possible mapping of emotional components within Abric's structural approach, while also accounting for the dynamic role emotions may play in shaping representational interaction processes with the contexts. Accordingly, the study of emotional components allows for an assessment of the degree to which emotions are stabilized around a given object—reflected in their positioning within the central core or peripheral zones—and for an understanding of how emotions facilitate adaptive dynamics between representational structures and citizens' or social groups' contextual conditions. This aligns with one of the main aims of SRT: the assessment of SR's sensitivity to citizens' immediate contextual stimuli (Bouriche, 2022). In contemporary societies, these dynamics have become increasingly salient, particularly due to the role of social media and the circulation of emotionally charged threat discourses (Moliner and Piermattéo, 2025). Examining the positioning of emotional components within the structure of SRs thus provides valuable insight into emotions' role in shaping representations of protective policies. Given the present study's cross-national perspective, it may also benefit from the Intergroup Emotions Theory (Mackie et al., 2000). It conceptualizes emotions as arising from collective appraisals of socially relevant contexts, suggesting that variation in emotional profiles across countries may reflect differences in how groups collectively interpret threats and protection within their social and political environments.

Given the theoretical framework outlined above, the present study aims to examine the emotional dimension of SRs of protective policies through a structural and cross-national perspective. Specifically, it addresses the following research question (RQ): What emotional components constitute the SRs of protective policies, and how are these components structurally positioned and distributed, in terms of frequency of occurrence, across countries?

To address this overarching question, the study examines how these emotional components are positioned within the representational structure. By analyzing the structural positioning of emotional components, it aims to assess the degree of stabilization and function of these emotions within the SR of protective policies and explores similarities and discrepancies in the emotional profiles of SR across countries, identifying cross-national patterns and context-specific configurations. Finally, it investigates which emotional components are most salient and recurrent within and across national contexts, based on their frequency of occurrence. As this study considers SRs as cognitive-emotional processes (Bouriche, 2022), it also examines the role of each emotional component in the representational structure.

Participants

The overall sample comprised 12,308 respondents from 11 countries (Czech Republic n = 1,102; Denmark n = 1,088; France n = 1,100; Germany n = 1,081; Ireland n = 1,406; Israel n = 1,031; Italy n = 1,100; Poland n = 1,101; Portugal n = 1,101; Sweden n = 1,102; United Kingdom n = 1,096), of whom 10,147 provided valid responses and were retained for analysis (Czech Republic n = 873; Denmark n = 851; France n = 889; Germany n = 886; Ireland n = 1,150; Israel n = 997; Italy n = 853; Poland n = 875; Portugal n = 995; Sweden n = 867; United Kingdom n = 911). Participants were aged between 18 and 95 years (M = 45.68, SD = 15.34). Gender distribution was balanced, with 50.20% identifying as women (n = 6,174), 46.90% as men (n = 5,771), 2.70% identifying with another gender (n = 330), and 0.30% did not respond (n = 33).

Regarding education, 57.80% had completed basic or secondary education, 39.50% held a higher or university degree, and 2.00% had completed only primary schooling. Most respondents were employed (63.80%, n = 7,847), followed by retirees (14.80%, n = 1,825), unemployed individuals (6.40%, n = 787), students (5.90%, n = 726), and those working from home (4.10%, n = 505). Among respondents with valid job-type data (n = 10,839), 44.50% reported non-qualified occupations (n = 5,472) and 43.60% reported qualified occupations (n = 5,367).

In terms of citizenship (valid responses: n = 11,993), 94.10% were nationals of one of the surveyed countries; 92.10% held a single citizenship (n = 11,335) and 2.00% held multiple (n = 251), while 3.30% were non-nationals residing in the country of residence (n = 407). Most participants identified as heterosexual (83.60%, n = 9,674/11,581), white/Caucasian (96.30%, n = 5,574/5,787), not belonging to a minority group (82.90%, n = 8,821/10,639), and not experiencing discrimination (81.80%, n = 9,062/11,078).

Materials

Data collection was conducted in April 2025 by a certified research company using a self-administered online questionnaire on the Qualtrics XM platform. The questionnaire was administered in the official language of each participating country.

Procedure

Through the first of three survey waves conducted within the PROTEMO project, data were collected by a professional survey company specifically contracted to administer the study across the 11 participating countries. The company used its established online research platform to recruit participants, inviting individuals who had previously registered with the platform provider and met the predefined sampling criteria to take part in the survey. Participation was entirely voluntary, and individuals were provided with comprehensive information beforehand regarding the study's aims, procedures, potential risks, and anticipated benefits. Respondents provided informed consent by confirming that they understood the purpose of the study, that participation was voluntary, that they were at least 18 years old, and that their responses would remain anonymous and confidential. To ensure confidentiality, no personally identifiable information was collected, and all data were analyzed in aggregate form to prevent the identification of individual responses. Participants were further informed of their right to withdraw from the study at any point without any justification or consequences. Completion required approximately 15–20 min. All procedures complied with formal and ethical standards for research in the social sciences, ethical consent was obtained from the Ethics Commission of the University of Saarland—protocol number 24–33, titled “ Emotional dynamics of protective policies in an age of insecurity ”—and adhered to the ethical principles outlined in the Declaration of Helsinki, safeguarding participants' rights, dignity, and overall wellbeing throughout all stages of the study.

The data analyzed in this study derive from a larger dataset, which includes a broad set of variables alongside sociodemographic information. For the present analyses, only responses related to the SRs of protective policies and relevant sociodemographic variables were retained, with a specific focus on emotional components.

Measures

To elicit SRs of protective policies, the study employed the free-evocation technique (e.g., Monaco et al., 2017). Participants were invited to report up to five words or short expressions that spontaneously came to mind when thinking about the protective policies in their country, as stated below: “ States must provide protective policies for people (that is, political actions related to people's protection [partner country]). What comes to your mind when you think about protective policies in [partner country]? Please write down the first five words or short expressions (maximum 5 words per expression) in the order they appear in your mind: ”

Consistent with the SR method (Abric, 2001, 2003), the expression “protective policies” was intentionally left open to participants' interpretation. Rather than seeking a predefined understanding of the object, the free-evocation task aimed to capture the meanings that participants spontaneously associated with it, within their own social, cultural, and political contexts. Consequently, the possibility of variation in the concrete policies or issues evoked was considered an integral feature of the representational object.

This method allows for the examination of the content and structural organization of SRs by capturing immediate associative responses to a given object (Abric, 2001, 2003; Moliner and Abric, 2015). Moreover, it enables the implementation of prototypical analysis, in line with the assumptions of the structural approach (Abric, 2001, 2003; Moliner and Abric, 2015; Wachelke and Wolter, 2011).

Data analysis

Translation and back-translation

All responses were translated into English using Google Translate via its Python API (v. 3.13.5). Semantic equivalence was assessed through a two-stage validation procedure. First, back-translation was performed on a random subset of 100 participants per country who provided one or more responses. In one to five participants per country, at least one minor translation inconsistency was found, which was corrected before the analyses. Second, additional random checks were conducted on specific words and expressions using established online dictionaries (Cambridge University Press Assessment, 2025; Brader and Marcus, 2013), verifying 150 words or expressions. Only six minor translation inconsistencies were identified and corrected before the analyses. All emotion-related responses were subsequently reviewed manually, with support from native speakers when necessary.

Categorization and prototypical analysis procedures

The analytical procedure was conducted in two sequential phases. First, lexical items were standardized by grouping words and expressions sharing identical or closely related meanings, following established practices in content categorization (e.g., Wachelke and Wolter, 2011). This process was implemented using Oxford Learner's Dictionaries (Brader and Marcus, 2013) for support. Category construction followed a hybrid deductive-inductive strategy. Deductively, the coding scheme drew on previously identified classifications of protective policies (Stempel et al., 2024), and on emotional categories derived from the Geneva Emotion Wheel (GEW) 3.0 (Sacharin et al., 2012) and from literature on emotions. Selected emotional needs (e.g., safety and security) were additionally informed by Capelos et al. (2024) conceptualizations. Inductively, the analysis remained open to the emergence of new themes grounded in recurring empirical patterns. All emotional components were retained if verified in the APA Dictionary of Psychology (American Psychological Association, n.d.) or other academic sources. Coding disagreements were resolved through discussion in successive meetings with a third coder, ensuring dialogue, consensus, and contributing to the internal coherence and reliability of the final coding framework.

Before the prototypical analysis, exclusion criteria were applied to the full dataset (n = 12,308): cases were removed if they (a) contained no responses across all five evocation fields (i.e., no meaningful contribution); (b) explicitly reflected a lack of knowledge about protection or protective policies (e.g., “ I don't know ”); or (c) fell into categories whose frequencies did not reach the established minimum thresholds—five occurrences for each national subsample and 33 occurrences for the pooled dataset. Emotions did not have these minimum threshold criteria applied to do a complete assessment of their positions. The literature provides no standardized criteria for defining cut-off points distinguishing high vs. low frequency or high vs. low mean order of evocation (MOE, the mean score of the positions (ranging 1 = first word that appears in respondents' mind to 5 = last word that appears in respondents' mind) of the words/expressions within the category; Wachelke and Wolter, 2011). For the aggregated dataset, the minimum frequency threshold was determined by multiplying the median number of expected evocations (three, given the five available response fields; Wachelke and Wolter, 2011) by the total number of participating countries (11), yielding a cut-off value of 33. At the country level, the default minimum frequency parameter implemented in EVOC (n = 5) was retained. These thresholds were adopted to ensure that the prototypical analysis focused on collectively shared representational content rather than on idiosyncratic and non-representative meanings.

Cut-off values for the MOE were established based on the overall average rank (moyenne générale des rangs) generated by the RANGMOT procedure in EVOC. The intermediary frequency, which served as the frequency threshold, was calculated by dividing the cumulative frequency of all terms that met or exceeded the minimum frequency by the total number of such terms. These calculations were conducted using EVOC2005 to define the MOE threshold and IRaMuTeQ 0.8a7 to compute the intermediary frequency. Applying these predefined criteria allowed the analysis to remain both data-driven and methodologically consistent, ensuring consistency across groups while preserving sensitivity to distributional patterns in the data. After the application of the exclusion criteria, from the sample of 12,308 participants, 10,147 were considered for prototypical analysis. Overall, 443 categories related to protective policies were evoked by the participants (see Appendix A). Of those, only 130 reached or surpassed the aforementioned threshold of 33, which was defined as the minimum frequency for the general results. Therefore, 313 categories were not considered since they did not reach that threshold. From the initial 443 categories, and since this threshold was not applied to emotions for subsequent tests, a total of 21 emotional components were detected (see Table S1, Supplementary material).

In the second phase of the analysis, the final set of categories was examined through prototypical analysis, a method designed to assess the relative salience of representational elements by jointly considering their frequency of occurrence and their MOE. This procedure generates a four-quadrant matrix that offers a structural representation of SRs (Abric, 2001, 2003; Moliner and Abric, 2015; Piçarra et al., 2016).

Following Abric's structural approach (see Table 1; Abric, 2001, 2003; Moliner and Abric, 2015), representational elements are distributed across four zones according to these two criteria. The central core (CC; upper-left quadrant) comprises elements that are both highly frequent and evoked early (lower MOE), reflecting their foundational role in organizing the meaning of the representation and their relative stability over time. The first periphery (1P; upper-right quadrant) includes elements that occur frequently but are evoked later, contextualizing and reinforcing the central core while allowing for greater adaptability. The contrast zone (Contrast; lower-left quadrant) consists of less frequent elements that nonetheless appear early in the evocation order, often signaling the presence of alternative, minority, or emerging representations within specific subgroups. Finally, the second periphery (2P; lower-right quadrant) contains elements characterized by low frequency and higher MOE, capturing the more flexible, context-dependent aspects of the SRs.

Emotional components' analysis

A matrix containing all emotion-related terms was created to allow detailed analyses by country and for the overall dataset. Using magnitude coding—a coding method that captures directionality in the data (Fielding, 2008; Saldaña, 2009)—each positive emotion evocation was assigned a value of 1 and each negative evocation a value of −1. The values reported in Table 2 simultaneously convey the frequency and valence of each emotional component: the absolute value indicates the number of evocations, whereas the sign indicates their valence (positive vs. negative). No neutral or mixed-valence emotion labels were identified during coding. This procedure enabled the calculation of the emotional tone frequencies by country. Although straightforward, this measure allows the identification of specific emotional components within SRs, highlighting which emotions appear most frequently in each country and which are shared across countries regarding protective policies.

Emotional components with frequencies below the minimum were retained for descriptive frequency and MOE reporting, as well as their positioning, but were excluded from subsequent considerations concerning their relevance. This is because MOE estimates based on very low frequencies may be statistically unstable and not reflect shared representational structures.

To examine whether emotional valence (positive vs. negative) differed across countries, a chi-square test of independence was conducted, accompanied by Cramer's V as a measure of effect size and an inspection of the standardized residuals. Correspondence analysis, based on the chi-square distance between country and emotional profiles, was subsequently performed to explore the multivariate associations between countries and emotional components. To further examine the relationships between emotional components and representational elements, a co-occurrence-based similarity analysis was conducted following the structural approach to Social Representations Theory (Flament and Rouquette, 2003). Co-occurrences were computed between emotional components and representational elements that appeared within the same participant's response. A co-occurrence network was performed in which nodes represented emotional components and representational elements, whereas edges represented the frequency with which these categories co-occurred across participants' responses (edge thickness is proportional to co-occurrence frequency). In the total sample, only the 100 strongest edges with co-occurrences consisting of a minimum frequency of three were retained in the pooled network for visual clarity. For the country-specific networks, the 60 strongest edges observed, consisting of co-occurrences (minimum frequency = 1), were retained in the graphic representation. Networks were generated for both the pooled dataset and each participating country. The analyses were implemented in Python using the NetworkX library and visualized using the Fruchterman-Reingold force-directed spring layout, whereby more strongly connected nodes tend to be positioned closer together.

Emotional components in protective policies

The frequencies ranged from 123 (Fear) to two (Despair, Frustration, and Pride). Fear (n = 123; MOE = 2.54), Calmness (n = 79; MOE = 2.80), and Joy (n = 34; MOE = 2.82) were positioned in the 2P of the SRs of protective policies, falling below the intermediate frequency cut-off (n = 237) and above the general mean order of evocation cutoff (MOE = 2.40). While Dissatisfaction, Gratitude, Guilt, Hope, and Sadness are in the contrast zone, the other 13 emotional components are also positioned in the 2P (see Table 3).

Concerning the values from each country with a frequency of five or more (see Supplementary Table S2), it is possible to observe that in the Czech Republic (CZ), only Fear (n = 5; MOE = 3.80) reached the threshold of five, with a position in the second periphery. In Denmark (DK), Calmness (n = 8; MOE = 3.50) reached the second periphery, and Fear (n = 5; MOE = 2.00) was in the contrast zone (Contrast). In France (FR), Fear (n = 8; MOE = 2.88) was the only emotional component reaching the minimum frequency threshold, being positioned in the second periphery. In Germany (DE), Fear (n = 7; MOE = 2.29) reached the frequency threshold again and was in the contrast zone, indicating early salience despite lower overall frequency. In Ireland (IE), several emotional elements reached the threshold and were all positioned in the second periphery: Joy (n = 16; MOE = 2.81), Fear (n = 8; MOE = 2.63), and Calmness (n = 5; MOE = 3.60). In Israel (IL), three emotional components were positioned in the second periphery: Fear (n = 48; MOE = 3.15), Calmness (n = 9; MOE = 3.11), and Disappointment (n = 9; MOE = 2.89). Additionally, Shame (n = 7; MOE = 2.43) was in the contrast zone. In Italy (IT), Calmness (n = 9; MOE = 3.44) and Fear (n = 8; MOE = 2.75) reached the second periphery. Disgust (n = 6; MOE = 1.67) was positioned in the contrast zone. In Poland (PL), three emotional components were in the second periphery: Calmness (n = 34; MOE = 2.79), Fear (n = 10; MOE = 2.00), and Joy (n = 6; MOE = 3.00). In Portugal (PT), Fear (n = 15; MOE = 3.07) and Tranquility (n = 11; MOE = 2.64) reached the second periphery. In Sweden (SE), Fear (n = 7; MOE = 3.29) was the only emotional component reaching the threshold and was positioned in the second periphery. Finally, in the United Kingdom (UK), Calmness (n = 6; MOE = 3.33) and Joy (n = 6; MOE = 3.33) were positioned in the second periphery.

Fear occupied structurally salient positions in the SRs of protective policies in 10 countries, appearing in the second periphery in 8 countries (Czech Republic, France, Ireland, Israel, Italy, Poland, Portugal, and Sweden) and in the contrast zone in 2 countries (Denmark and Germany). Calmness reached a peripheral structural position in 6 countries, being consistently located in the second periphery (Denmark, Ireland, Israel, Italy, Poland, and the United Kingdom). Joy appeared in structurally relevant positions in 3 countries, occupying the second periphery in all cases (Ireland, Poland, and the United Kingdom). Disappointment and Tranquility were in the second periphery in one country each. Disgust reached the contrast zone in one country, as did Shame. All remaining emotions did not reach the minimum frequency threshold required for structural positioning in any country and therefore remained outside the structural zones of the representations.

Emotional components' distribution across countries

The analysis of emotional components across the 11 countries (see Table 3) indicates a clear predominance of negative emotions, both in cumulative frequency and in cross-national distribution. Fear is the most salient emotional component, showing the highest total frequency (n = −123) and appearing in all 11 countries. Other negative emotions with notable cross-national presence include Anxiety (n = −14; 7 countries), Disappointment (n = −19; 7 countries), Disgust (n = −14; 6 countries), Shame (n = −12; 4 countries), and Dissatisfaction (n = −6; 5 countries). Additional negative emotions, such as Sadness, Hate, Frustration, Despair, and Guilt, occur less frequently and are limited to a smaller number of national contexts.

Positive emotional components are comparatively less frequent and unevenly distributed. Calmness emerges as the most prominent positive emotion, with the highest cumulative positive frequency (n = 79) and presence in nine countries. It is followed by Joy (n = 34; 6 countries), Tranquility (n = 17; 5 countries), Love (n = 14; 6 countries), and Hope (n = 11; 6 countries). Other positive emotions, including Enthusiasm, Pride, Satisfaction, and Gratitude, display very low frequencies and appear in only two or three countries.

Marked cross-national variation is observed in the overall emotional tone associated with protective policies. Poland (n = 26), the United Kingdom (n = 17), and Ireland (n = 8) exhibit a predominantly positive emotional tone, largely driven by high levels of Calmness and, in the cases of Ireland and the United Kingdom, Joy. Denmark presents a slightly positive emotional balance (n = 3), reflecting a relative equilibrium between negative emotions—such as Fear and Dissatisfaction —and positive emotions, particularly Calmness and Tranquility.

In contrast, Israel displays a strongly negative emotional tone (n = −62), largely dominated by Fear, Shame, Disappointment, and Anxiety. France (n = −11), Sweden (n = −8), the Czech Republic (n = −7), Germany (n = −6), Italy (n = −4), and Portugal (n = −5) also show predominantly negative emotional profiles, although with lower overall intensity.

Fear is the only emotional component consistently present across all countries, though its magnitude varies substantially—from relatively low frequencies in the United Kingdom (n = −2) to very high levels in Israel (n = −48) and Portugal (n = −15). Among positive emotions, Calmness is particularly salient in Poland (n = 34), Joy is most pronounced in Ireland (n = 16), and Tranquility stands out in Portugal (n = 11). Less frequent emotional components appear sporadically and contribute marginally to the emotional landscape associated with protective policies.

To examine whether emotional valence (positive vs. negative) differed across countries, a chi-square test of independence was conducted. The analysis revealed a statistically significant association between country and emotional valence, χ 2 (10, n = 383) = 80.98, p < 0.001, indicating that the distribution of positive and negative emotions varied across countries. The effect size was moderate (Cramer's V = 0.46).

Inspection of the standardized residuals indicated that the overall association was primarily driven by Israel, Poland, and the United Kingdom. In Israel, significantly more negative emotions were observed than expected (78 observed vs. 66 expected; standardized residual = 3.43). In contrast, Poland showed substantially more positive emotions (44 observed vs. 27 expected; standardized residual = 3.26) and fewer negative emotions than expected (18 observed vs. 35 expected; standardized residual = −2.87). A similar pattern was observed for the United Kingdom, where positive emotions were overrepresented (19 observed vs. 9 expected; standardized residual = 3.25) and negative emotions were underrepresented (2 observed vs. 12 expected; standardized residual = −2.86). These findings indicate that the significant association between country and emotional valence was largely attributable to these three countries, whereas the remaining countries did not differ significantly from the expected distribution.

To provide a multivariate representation of the relationships between countries and emotional components, a correspondence analysis was conducted (Figure 1). The first two dimensions accounted for 44.20% of the total inertia (Dimension 1 = 27.07%; Dimension 2 = 17.13%). The correspondence map revealed clear cross-national variation in emotional profiles. Israel was positioned closer to predominantly negative emotional components, whereas Poland and the United Kingdom were located nearer to positive emotional components. Portugal showed greater proximity to Tranquility, while Germany, Czechia, Italy, France, and Sweden occupied more central positions, indicating comparatively mixed emotional profiles with no strong association with a single emotional component.

To further examine the structural relationships between emotional components and representational elements, similarity (co-occurrence) networks were constructed for the pooled dataset and for each country (Supplementary Figures S1–S12). Within networks, nodes represent emotional components and representational elements, whereas edges indicate their co-occurrence within participants' responses, with thicker edges representing stronger associations.

The overall network revealed two broad representational configurations. One was centered on Calmness, which showed its strongest association with Safety and security (weight = 51), followed by weaker links with Confidence (10), Trust (6), Peace (5), Stability (5), Freedom (5), Caring (5), and institutional elements such as Police and security forces (6), Military (6), Law and order (3), Policies (3), and Healthcare and wellbeing (3). The second configuration was organized around Fear, which was most strongly associated with Insecurity (45) and Criticism (34), followed by Crime (10), Police and security forces (10), Probability and doubt (11), Politics (7), Negligence and abandonment (6), Worry (6), Economy and finances (5), Harm, danger and risk (5), and Military (5). Within this negative configuration, Criticism occupied a central position by connecting Fear (34) with Disappointment (15), Disgust (10), Shame (9), and Hate (3). Positive emotions such as Joy, Love, and Hope were positioned closer to the calmness-centered structure, whereas Anxiety, Disappointment, Disgust, Hate, and Shame clustered around the fear-centered configuration. Notably, Safety and security was associated with both positive and negative emotional components, linking Calmness (51), Joy (19), and Love (4) with Fear (8), suggesting that this representational element constituted an important bridge between the two emotional configurations.

The country-specific networks revealed substantial variation in the organization and density of the associations. Similarity networks were generated for all participating countries to compare their overall representational configurations. For interpretation, however, the analysis focused primarily on networks containing repeated co-occurrences (edge weight ≥ 2). This criterion was adopted because single-occurrence associations reflect the responses of individual participants rather than recurrent representational patterns.

Poland displayed the most coherent positive configuration, centered on Calmness, which was strongly associated with Safety and security (27) and, to a lesser extent, with Stability, Confidence, Caring and taking care, Military, Trust, and Freedom (weights = 3–4). The United Kingdom also showed predominantly positive configurations, characterized by links between Calmness and Safety and security (6), together with associations involving Joy, Hope, and Peace. Ireland presented a comparatively complex but predominantly positive structure, centered on Joy and Calmness, with Joy-Safety and security (8) representing its strongest association.

By contrast, Israel exhibited the densest negative configuration, dominated by Fear-Insecurity (weights = 20), Criticism-Fear (8), Fear-Negligence and abandonment (6), Criticism-Disappointment (6), and Fear linked to Military and Police and security Forces (both 5), reflecting a highly interconnected threat-oriented representational structure. France, Portugal, Sweden, and Italy displayed more heterogeneous configurations in which recurrent associations remained centered on Fear, Insecurity, and Criticism. In France, the strongest associations were Fear-Insecurity (4) and Fear-Safety and security (3), complemented by Fear-Police and security forces, Fear-Upgrade and improve, Fear-Criticism, Disappointment-Criticism, and Fear-Economy and finances (all 2). Portugal showed a predominantly negative configuration centered on Fear-Criticism (8) and Fear-Insecurity (6), whereas positive associations were comparatively weaker. Sweden similarly displayed a predominantly negative configuration, with Fear-Insecurity (4), Fear-Criticism (3), and weaker associations involving Anxiety-Criticism, Fear-Crime, and Shame-Criticism (all 2). Italy presented the most balanced profile among these countries, combining positive associations centered on Calmness-Safety and security (8) with negative links such as Disgust-Criticism (4), Fear-Insecurity (3), Fear-Criticism (3), and Fear-Crime (3). Finally, the networks for Czechia and Germany contained few repeated co-occurrences, with most associations occurring only once.

Overall, the similarity analysis showed that emotional components were linked to representational elements rather than functioning as isolated affective reactions. The observed co-occurrence patterns suggest that emotions played a bridging role by connecting affective responses with broader thematic domains, although the strength and organization of these links varied across national contexts. All findings provide converging evidence that social representations of protective policies differed across countries, both in their overall emotional valence and in the way emotional components are structurally articulated with representational elements.

Although emotional components were located predominantly in the peripheral system of the prototypical analysis, the similarity analysis revealed that they were not randomly distributed. Instead, they formed coherent thematic configurations through systematic associations with representational elements. Positive emotions were consistently linked to themes of safety, trust, peace, and stability, whereas negative emotions clustered around insecurity, military issues, crime, and criticism. These findings suggest that peripheral emotional components contribute to the contextualization, expression, and operationalisation of the representational field, rather than functioning as isolated affective reactions. Despite their predominantly peripheral position in the prototypical analysis, emotional components formed consistent thematic configurations.

General results revealed substantial cross-national variation in emotional profiles, particularly in frequency and emotional tone, while the emotional components were always peripheral. This combination of structural convergence and emotional divergence reinforced the notion that emotions operated as flexible representational resources, allowing national contexts to express distinct affective responses to a shared policy object.

Emotional components and representational stability

Answering the RQ (what emotional components constitute the SRs of protective policies, and how are these components structurally positioned and distributed, in terms of frequency of occurrence, across countries?), we found several emotional components positioned in the second periphery and in the contrast zone of social representations of protective policies. None was present in the central core. This remained the case even when alternative frequency cut-offs were considered (i.e., based on the median), as the MOE of all emotional components exceeded the threshold required for centrality. Fear was the most salient emotion surfacing in the representational field of citizens, followed by Calmness and Joy.

The results that reached the threshold allowed several inputs. From Abric's Structural Approach (Abric, 2001, 2003; Moliner and Abric, 2015) perspective and concerning how emotional components were positioned within the representational structure, the fact that emotional components were not in the central core when they appeared in the representational structure indicated that emotions do not constitute foundational or normative elements of the SRs of protective policies. Rather, they operated as context-sensitive, adaptive, and regulatory components, allowing individuals and groups to engage emotionally with the object without destabilizing its central meaning when in the second periphery, or they represent an alternative, minor, and possibly emerging representation within a subgroup.

The results concerning the localization of emotional components spoke directly to the issue of stabilization and function of the representational structure, answering to what degree of stabilization and what function the emotions have within the SR of protective policies. Abric (2001, 2003) noted that these components are the most flexible in the representational structure, characterized by low consensus and high sensitivity to contextual variation. The presence of emotions in this zone indicated that emotional responses to protective policies were not fully stabilized and remained open to change in response to social, political, and media dynamics. This finding aligned with Moliner and Piermattéo's (2025) argument that emotions often function as signals of representational instability, prompting processes of negotiation and social exchange aimed at restoring coherence. Emotional anchoring appeared here as a mechanism that enhanced evaluability and experiential relevance, rather than one that restructured the representational core. This supported the idea that emotions help make abstract policy objects more tangible and discussable, while remaining secondary to cognitive and normative anchors. This finding underscored the value of a structural approach for comparative research, as it allows for the identification of shared representational logics alongside context-dependent emotional expressions. Emotional anchoring, thus, might facilitate engagement and communication around protective policies without challenging their legitimacy or underlying rationale.

The similarity analysis complemented the prototypical analysis by demonstrating that structural peripherality does not imply relational marginality. Although emotional components remained outside the central core, they occupied meaningful positions within the representational network through stable patterns of association and articulation with other elements. Consequently, emotions seem to contribute less to defining the normative content of the representation than to contextualizing and operationalizing its expression across different national contexts.

Moreover, the two major configurations identified in the similarity analysis could also be interpreted from the perspective of Marková's (2003) dialogical themata. Rather than representing two independent emotional clusters, the Fear -centered and Calmness -centered configurations appear to express the two poles of an underlying antinomic thema opposing security to insecurity (or, more broadly, protection to threat) and related categories. Within this perspective, emotions do not constitute the thema itself. Rather, they function as affective manifestations—or signallers—of meanings through which citizens negotiate and make sense of the underlying tension between protection and threat that structures the SRs of protective policies. Protective policies were, therefore, represented simultaneously as responses to perceived threats and as mechanisms for restoring safety, stability, and social order, with emotions functioning as affective components of this dialogical tension. This interpretation extended Abric's structural approach (2001, 2003) by suggesting that emotional components occupy different structural positions within SRs, while it also revealed the dialogical tensions through which the object itself was organized and negotiated.

This interpretation also supports Bouriche's (2022) proposal that SRs should be understood as cognitive-emotional processes rather than exclusively cognitive structures. Although emotional components remained structurally peripheral, their systematic articulation with representational elements of the central core demonstrates that affective and cognitive dimensions are continuously integrated during the construction and negotiation of meaning. Rather than constituting an independent representational system, emotions contributed to the dynamic articulation of the representational field by contextualizing shared cognitive contents through affective experience.

Concerning the contrast zone, its elements might indicate a strong stabilization, but only for a minority with emergent emotional components within the SRs of protective policies. This information could suggest the transformation of certain emotions, such as shame in Israel, into more consolidated and less transient affective components. These components served as a signal of subgroups' emotional awareness, signaling emerging emotional saliences in the sociocognitive representations. This might indicate a future possible dispute of meanings within the representational structure, but it does not represent a direct influence on the central elements in the present.

Emotional profiles and emotions' role in the SRs of protective policies

Beyond their shared peripheral positioning, it is therefore relevant to consider and compare the emotional components per se, observing what cross-national patterns and context-specific configurations were identified and which emotional components are most salient and recurrent within and across national contexts. The results indicated that Fear constituted the most pervasive emotional component across all countries, appearing systematically and with the highest overall frequency. These values are particularly informative when contrasted with those observed for Anxiety. While Fear and Anxiety share a common origin in the early mobilization to a perceived threat, Fear emerges when effective coping responses are perceived as available, whereas Anxiety arises when the threat remains ambiguous and no clear means of coping can be identified (Öhman, 2008). This pattern suggested that citizens tended to associate protective policies with identifiable threats that are, at least to some extent, perceived as manageable or avoidable through protective action.

However, despite its ubiquity, Fear remained structurally confined to the peripheral zones of the representation, indicating that it functioned primarily as a contextual and experiential response to perceived threats rather than as a stable, normative element defining the meaning of protective policies. This peripheral positioning suggested that threats can be emotionally framed without destabilizing the representational core. In this sense, Fear may have served to activate and give salience to central core elements—such as protection, security, or responsibility—within specific social and political contexts, rather than redefining them. This insight is consistent with Methivier (2010) finding that fear influences the salience of certain elements in the SRs of work and unemployment.

The prominence of Fear was frequently accompanied by related negative emotions: Anxiety, Disappointment, Shame, and Disgust, though with varying combinations and intensities across countries. As previously mentioned, Anxiety emerged in the presence of ambiguous threats (Öhman, 2008). Disappointment was related to agency attribution to others and to the circumstances beyond control and is related to the tendency to do nothing to solve the problem and escape from the situation (Zeelenberg et al., 2000). Shame was a painful emotion directed at the whole citizen's self, and that, similarly to Disappointment, causes the individual to feel worthless and powerless, seeking to escape from the situation (Brader and Marcus, 2013). Disgust was an aversive emotion that can easily be entangled with anger, but that translates into reactions against both perceived physical and moral impurity (Brader and Marcus, 2013).

In Israel, Disappointment clustered in the second periphery with a high frequency of Fear, while Shame was revealed as a contrast element. The only positive emotional component present was Calmness. These results, together with the central elements, suggested that these emotions may signal that protective policy meaning is activated to remove the perceived threats that cause fear to citizens—with external threats being more salient, since the main core element was Military —while calmness might reflect the desired emotional state. Disappointment might be associated with Criticism being a central element present. Shame might reveal a group that feels this emotion while thinking about the protective policies of Israel. Disgust was present in the contrast zone of the SRs of protective policies of Italy, which might be justified by an emotional group detecting impurity in the policies framed as protective by the State. From the perspective of the Intergroup Emotions Theory (Mackie et al., 2000), this emotional profile may have reflect a collective appraisal of persistent threat, whereby emotions emerged from evaluations made on behalf of the group and orient collective responses toward protection against perceived external threats—which may be perceived defined by the presence of Military (the most frequent category), Air security and War in the central core of SRs of protective policies of Israeli inhabitants. This finding suggested that contextual factors (e.g., threat perceptions, political communication, and ingroup-outgroup polarization) may have shaped collective appraisals, thereby conditioning the emergence of specific emotional responses.

Positive emotional components, notably Calmness, Hope, Joy, and Tranquility, also played a significant role in shaping emotional profiles, although their distribution was more uneven and context-dependent. Hope never reached the threshold in any country, not giving signs of stabilization that might give it a role in the articulation with the central core. Calmness emerged as the most salient positive emotion, particularly in countries displaying a more positive or balanced overall emotional tone. This emotion was associated with the absence of threats (Brader and Marcus, 2013). Rather than indicating unqualified approval, its presence could be interpreted as reflecting an emotional orientation toward what protective policies are expected to provide—namely, stability, reassurance, and the normalization of perceived risk. This interpretation is consistent with recent cross-national evidence showing that citizens in Western countries tend to value calmness as a desirable emotional state, even when its experience remains uneven and contingent on broader social conditions (Lomas et al., 2024). Although Calmness and Tranquility were indicated as separate emotional components, based on the literature (Sikka et al., 2023), their overarching function may be considered equivalent. This distinction is further complicated by cross-linguistic variation, as some languages do not clearly differentiate between the two concepts. When the results are examined as a combined Calmness-Tranquility cluster (n = 96, 10 countries), this emotional component becomes even more salient, indicating that citizens attribute considerable importance to protective policies oriented toward generating such states.

Similarly, Joy appeared to reflect forward-looking evaluations and conditional expectations regarding the expected outcome of protective policies. Unlike emotions such as Fear or Calmness, which are functionally aligned with threat perception and risk management, Joy lacked direct congruence with the prospective and preventive logic that characterizes protective policies. Its late evocation and low frequency indicated that Joy operated primarily as a retrospective or episodic evaluation of policy outcomes rather than as an organizing principle of the representation. Although it could articulate with the central element Positive, which captured general compliments and favorable evaluations, it did not activate this core element, as its presence was observed in other countries where Joy did not emerge in the second periphery.

A possible explanation for the presence of this emotion—also applicable to other emotional components—was that emotions enter the cognitive space only when a global organismic state is represented, categorized, and consciously labeled (LeDoux, 2012). However, as suggested by the affect heuristic, emotional responses may already be guiding evaluations and judgements at a pre-reflective level, before conscious articulation (Slovic et al., 2007). This implied that emotional components were explicitly mentioned by citizens primarily when they became cognitively accessible and consciously recognizable, while their influence may have preceded such awareness.

This pattern pointed to the existence of a latent emotional presence, whereby emotions remained operative in shaping perceptions and judgements while remaining unarticulated until sufficiently stimulated to enter the cognitive space and, subsequently, become integrated into the representational structure. From this perspective, Abric's structural approach offered a valuable analytical framework for identifying those emotions that reach the cognitive surface of social representations, thereby enabling the active anchoring of emotional components to the object of protective policies. This highlighted SRs as cognitive—emotional processes (Bouriche, 2022), where both dimensions were critical to the construction of protective policies.

Emotions as signallers and regulators of cognitive polyphasia

These findings concerning the representational structure of protective policies may have provided another insight into the role of emotions in SRs, suggesting an alignment between different concepts within emotions and SR fields. Given that SRs can be understood as cognitive-emotional processes (Bouriche, 2022), and that emotions function as signals of representational instability (Moliner and Piermattéo, 2025), it was analytically relevant to conceptualize emotional components as playing a regulatory role in relation to cognitive polyphasia. Cognitive polyphasia refers to a condition in which multiple forms of knowledge, grounded in distinct rationalities, coexist within the same individual or social group when making sense of a given social object (Jovchelovitch, 2007; Jovchelovitch and Priego-Hernández, 2015; Provencher, 2011).

From this perspective, the presence of emotional repertoires—defined as socially shaped yet adaptable sets of emotional orientations that guide how individuals and groups relate to others (von Poser et al., 2019)—may have contributed to the differential salience of particular forms of knowledge, fostering representational convergence within the majority or contrasting configurations among specific subgroups. Rather than producing displacement, representational dynamics were thus characterized by the coexistence of multiple representational fields (Jovchelovitch and Priego-Hernández, 2015), whose relative salience could vary across time and social contexts.

As mentioned above, emotions play a central role in these dynamics by signaling tensions or instability within the social object, thereby orienting individuals and groups toward particular interpretative frameworks (Bouriche, 2022; Moliner and Piermattéo, 2025). Such processes are often intensified through knowledge encounters—situated moments in which distinct representational or knowledge systems meet and interact through communication (Jovchelovitch and Priego-Hernández, 2015). These encounters can be frequently triggered by emotional prompts to meaning-making (Rimé, 2020), contributing to the negotiation of meaning and to the restoration of representational coherence within SRs (Moliner and Piermattéo, 2025).

From the perspective of the Affective Intelligence Theory (Marcus et al., 2000, 2019), the emotional configurations identified in the present study appeared to reflect primarily the operation of the surveillance system rather than the disposition system. The prominence of Fear —and Anxiety to a certain degree—together with the centrality of Calmness within the positive affective-representational configuration, suggested that citizens predominantly evaluated protective policies in terms of whether they were perceived as effectively reducing uncertainty and threat. Within this framework, Fear and Anxiety could be interpreted as signaling situations in which surveillance remained activated because threats were perceived as unresolved, whereas Calmness reflected situations in which surveillance was our could be satisfied by the perception that protective policies could restore security and stability. Conversely, emotions typically associated with the disposition system, such as Satisfaction, Enthusiasm, Frustration, and Depression, were comparatively infrequently linked to other categories. This pattern suggests that protective policies were represented less as positive or negative evaluated daily experiences and performances and more as responses to ongoing or potential threats requiring continuous monitoring. This interpretation is also compatible with appraisal theory (Lazarus, 1991), according to which emotions arise from individuals' evaluations of the significance of events rather than from the events themselves. The coexistence of distinct emotional configurations surrounding the same category (e.g., Military was associated predominantly with Fear in Israel but with Calmness in Poland, suggesting that the same representational element may acquire different emotional meanings depending on the context in which it is appraised) indicated that protective policies were appraised differently across individuals and national contexts, reflecting variation in perceived threat, coping potential, and expected outcomes rather than differences in the object itself.

Contributions

This research offered several contributions to the fields of cognitive and social psychology. First, it demonstrated how the structural positioning of emotional components could guide articulation within representational structures. The location of specific emotional components in national representational structures—for example, the presence of shame in the contrast zone of SRs of protective policies of Israeli citizens—may have pointed to emerging subgroup rationalities, thereby opening new avenues for future research.

Second, the study proposed a novel approach to assessing the stability of emotional components within representational structures, indicating how firmly embedded particular emotions were in the articulation of the social object. This may have enabled the understanding of emotional consolidation inside SRs. Generally, they were in unstable regions—as expected due to their transient nature—but as it was mentioned in the literature, they could consolidate into more permanent affective forms, such as sentiment or emotional attitudes. Third, this research identified a dual emotional orientation in the SRs of protective policies, structured around both threat-avoidance emotions (e.g., Fear) and desired emotional states (e.g., Calmness), highlighting the complexity of emotional dynamics underpinning these representations. Fourth, cross-national variation in emotional repertoires may reveal systematic patterns in citizens' perceptions, while also highlighting differences in how emotional signals are mapped and interpreted across contexts.

Beyond its theoretical implications, this research also made substantive contributions to debates in political communication and public policy. It suggested that emotional literacy and emotional regulation should be understood as individual competences, and as collective and societal capacities. Given that emotions were embedded in the representational structures of protective policies and could play a meaningful role in articulating policy meanings with lived contexts and signaling threats, citizens need to be aware of potential emotional vulnerabilities. Such vulnerabilities may be deliberately exploited by political actors, including emotional policy entrepreneurs, to frame issues strategically, or may emerge unintentionally through cognitive and communicative biases that foster dysfunctional policy dynamics. In accelerated and information-overloaded societies, where emotionally charged discourses circulate rapidly through digital media, the peripheral positioning of emotions could also indicate a heightened capacity of SRs of protective policies to absorb emotional volatility while maintaining structural coherence.

Accordingly, political and emotional dimensions should be treated as interdependent rather than separate domains, and policies and communicative strategies should aim to empower citizens' reflective political judgement and engagement. The mobilization of fear through populist discourse and fearmongering—often directed at immediate or perceived threats—could be mitigated by the development and application of emotionally informed political communication strategies that enhance citizens' capacity for critical engagement.

Methodologically, this study also demonstrated the complementarity of structural, inferential, and relational approaches for analyzing emotional components within SRs. Whereas, prototypical analysis identified the structural positioning of emotions, correspondence analysis revealed cross-national emotional configurations, and similarity analysis uncovered the relational organization linking emotional components with representational elements. Together, these approaches provide a more comprehensive understanding of the emotional architecture of social representations.

The findings also indicate that occupying a peripheral position within the representational structure does not necessarily correspond to a peripheral role in the network of relationships among representational elements. While emotional components were largely absent from the central core, they were consistently integrated into meaningful affective-representational patterns through their systematic associations with other elements of the representation. This finding contributes to the structural approach by showing that peripheral emotional components play an important role in articulation and signaling without challenging the stability of the representational core.

Limitations and future directions

Some limitations of this study should be acknowledged. First, the use of automated translation tools inevitably introduces risks related to semantic drifting. Though the emotional components were solved, it was still not feasible to verify all the other expressions. Resourcing translators with an adequate level of expertise and knowledge related to the language and context would support future research. Second, despite the use of clearly specified coding criteria and the fact that emotional components were based on direct mentions of the emotions, alternative coding could approach slightly different coding structures, mainly concerning non-emotional content. Third, prototypical analysis is inherently exploratory and provides an initial indication of the representational structure of protective policies in the 11 countries. Consequently, the centrality of the core elements and the peripheral positioning of emotional components could be further examined in future research using centrality-questioning techniques or longitudinal studies on the SRs of protective policies to provide additional evidence for the structural status of the identified elements or to understand their variations through time. Fourth, participants were instructed to report what came immediately to mind when thinking about protective policies. Although this procedure is consistent with the structural approach to SRs, it may have favored the elicitation of cognitive rather than affective content, potentially limiting the spontaneous emergence of emotions and feelings. Future research could address this limitation by inviting participants to report what comes immediately to mind when they think about and feel toward protective policies, thereby facilitating a richer exploration of their emotional dimension. Finally, since this study is spatially and temporally situated, future research should examine other contexts and time periods, characterized by different dynamics, to complement and extend the findings reported here. Future research could also examine whether prolonged exposure to emotionally saturated contexts leads to a gradual reconfiguration of emotional components toward more central zones of the representational structure.