Political Extremism and Perceived Anomie: New Evidence of Political Extremes’ Symmetries and Asymmetries Within French Samples

right-wing and left-wing individuals, social and political psychology scholars have recently expressed a growing interest in the study of political extremes’ similarities (e.g., Greenberg & Jonas 2003; Krouwel et al. 2017). Political extremes, despite their obvious ideological differences, might thus share some features regarding how they process information (e.g., Lammers et al. 2017) and how they perceive the social world (e.g., van der Bles et al. 2018). The present research aimed to extend previous work on political extremes’ symmetries and asymmetries (Vasilopoulos & Jost 2020) and addressed the question of political extremes’ (dis)similarities in their perception of societal anomie (i.e., perceptions that current society is disintegrated and disregulated; Teymoori et al. 2017). In line with McClosky & Chong’s findings (1985) showing that far-right and far-left individuals perceive society as ‘growing increasingly degenerate’ (1985: 334) while being ‘deeply estranged from certain features of society’ (1985: 360), we examined whether political extremes would perceive current society as more anomic (i.e., more disintegrated and disregulated) than political moderates. Taking advantage of the 2017 French presidential election, we also investigated whether both far-right and far-left individuals, compared to more moderates, similarly imagined a less anomic society after the election of their candidate.

' deeply estranged from certain features of society ' (1985: 360), we examined whether political extremes would perceive current society as more anomic (i.e., more disintegrated and disregulated) than political moderates. Taking advantage of the 2017 French presidential election, we also investigated whether both far-right and far-left individuals, compared to more moderates, similarly imagined a less anomic society after the election of their candidate.

Political Extremism
For a long time, the rigidity of the right hypothesiswhereby right-wing/conservative people display a less nuanced way of thinking (Eidelman et al. 2012;Tetlock 1983), lower uncertainty tolerance and higher needs for order and structure (e.g., Jost et al. 2007;Jost et al. 2003;van Hiel et al. 2004), thus fostering their appeal of strong authority figures (e.g., Altemeyer 1981;Chirumbolo 2002;Doty et al., 1991)-has been the dominant theory when studying the features of political ideologies. However, recent work suggest that some features of this rigid thought system may actually not be exclusive to conservative or right-wing individuals but could rather be associated with political extremism (e.g., Greenberg & Jonas 2003;Lammers et al. 2017;McClosky & Chong 1985). Indeed, research shows that individuals at both ends of the political spectrum are characterized by a very dichotomous way Ionescu, O., et al. (2021) of thinking (e.g., Fernbach et al. 2013;Lammers et al. 2017) and share a similar 'ideological rigidity' (Greenberg & Jonas 2003) associated with dogmatism, authoritarianism and uncertainty management (e.g., Brandt et al. 2014;).
Beyond an analogous information processing system, political extremes also seem to present similarities in their perception of society (e.g., Krouwel et al. 2017;. For example, recent work revealed that far-right and far-left voters similarly tend to report higher levels of societal pessimism (Steenvoorden & Harteveld 2018) and stronger endorsement of a pessimist Zeitgeist (i.e., ' a shared preconceived opinion that things in society are bad', van Der Bles et al. 2015, p. 5;van der Bles et al. 2018). In the present research, we were interested in a particular perception of society, namely perceived societal anomie. In line with extant data showing that political extremes share negative perceptions of society (e.g., van der Bles et al. 2018), we thus examined whether they would display higher perceptions that society is anomic than more moderates.

Perceived Societal Anomie
According to Durkheim (1897Durkheim ( /2013, anomie arises when moral standards and regulation within society are falling apart, due to rapid social changes and modernization: society then loses its regulatory power over its fundamentally selfish and insatiable members. Since this seminal conceptualization, anomie has been the focus of many theories; ranging from macro-level perspectives that described anomie as a lack of effective regulation ensuing from society (e.g., Merton 1938;Messner & Rosenfeld 2001)-that is, anomie as a state of society (e.g., Bjarnason 2009)-to approaches that mainly focused on the individual (e.g., McClosky & Schaar 1965;Srole 1956) theorizing anomie as state of mind (e.g., Bjarnason 2009).
More recently, it has been suggested that anomie may not be a strictly societal or individual condition, but may instead lie in the interaction between the individual and society (Bjarnason 2009;Teymoori et al. 2017). Based on this idea, Teymoori and colleagues (2017) developed a psychological analysis of anomie, which distinguishes people's perceptions of society from the individual manifestations that can occur in reaction to the perceived societal anomie. In this perspective, anomie is a 'reflection of the societal state in individuals' minds' (Teymoori et al. 2016: 3)-that is, a social perception, thus distinct from objective indicators of economic and social instability (Teymoori et al. 2016)-and refers to the perception that society is disintegrated and disregulated (Teymoori et al. 2017). Perceived societal anomie is therefore composed of two distinct, but dynamically interrelated, dimensions: the perceived disintegration of social fabric, which refers to a perceived 'breakdown in social fabric', illustrated by the disappearance of moral standards and a loss of trust between society members-and the perceived disregulation of leadershipwhich, in turn, refers to the perception that the political system, or 'leadership', is falling apart, being illegitimate (i.e., not well representing the citizen's interests) and ineffective (i.e., failing to preserve the collective well-being). Perceived societal anomie, although close from concepts such as pessimist Zeitgeist (Van Der Bles et al. 2015) or societal pessimism (e.g., Steenvoorden & Harteveld 2018) in the sense that they all converge on negative views of the current state society, therefore allows to go beyond a mere negative perception of society and to determine what this negative view is about; as it distinguishes people's perceptions of society on a vertical dimension (i.e., perceptions of the current political system/leadership) from their perceptions on a more horizontal dimension (i.e., perceptions of the current social fabric).
According to Teymoori and colleagues (2017), perceiving that society is disintegrated and disregulated (i.e., anomic) threatens the fulfillment of fundamental needs (e.g., the need for control or the need for meaning); thus impairing people's personal and social well-being (e.g., Blanco & Díaz 2007;Heydari et al. 2014). Previous data indeed showed that anomie is associated with feelings of meaninglessness and powerlessness (e.g., Thorlindsson & Bernburg 2004) and lack of control (e.g., Bjarnason 2009;Ådnanes 2007). Individuals, in order to overcome these uncomfortable feelings, may notably be tempted to turn to authoritarian ideologies, that would provide them the order and structure threatened by the perception of societal anomie (Teymoori et al. 2017). Perceived societal anomie, since it could indirectly foster the appeal for autocratic ideologies, thus appears like a relevant variable when studying extreme political positioning.

Perceived Societal Anomie and Political Extremism
Incidentally, existing findings suggest that political extremes could indeed perceive greater anomie within society. First, as we just mentioned, Teymoori et al. (2017) suggested that perceiving that society is disintegrated and disregulated could foster the appeal of authoritarian political systems and leaders. Several studies had indeed showed a link between anomie and authoritarianism (e.g., Heydari et al. 2012) and the rise of autocratic parties (Blank 2003;Oesterreich 2005). Moreover, recent empirical data based on Teymoori et al.'s conceptualization (2017) revealed that perceived anomie, and especially its social fabric disintegration dimension, was positively associated with the wish for a strong autocratic leader (Sprong et al. 2019). Knowing that authoritarian tendencies can be a feature of both far-right and far-left ideologies (e.g., Greenberg & Jonas 2003;Sleegers et al. 2015;Van Tilburg & Igou 2016), one can assume that perceived anomie, since not related to linear political orientation (Teymoori et al. 2016) but rather to autocratic stances, could be associated with political extremism. For that matter, recent work showed positive relationships between anomia (a psychological state of normlessness, meaninglessness, powerlessness, and isolation that could result from perceptions of societal anomie; Teymoori et al. 2017) and political extremism (e.g., Troian et al. 2019), as well as support for populism (Spruyt et al. 2016).
Moreover, previous research examining how political extremes perceive the current system shows that both far-right and far-left individuals share suspicious stances toward the political system and the elite as well as higher levels of political distrust (e.g., Akkerman et al. 2017;Inglehart 1987;McClosky & Chong 1985;van Bohemen et al. 2019; see also : Spruyt et al. 2016 for the link between populist attitudes and perceived lack of external political efficacy) and present decreased levels of system justification compared to more moderates (Caricati 2019). This empirical evidence strongly plays in favor of a symmetry between political extremes regarding the leadership disregulation dimension of perceived societal anomie, as the latter refer to perceptions that the political system or leadership is inefficient and illegitimate to regulate the collective well-being. Regarding the disintegration dimension of perceived societal anomie, we must note that the scientific literature offers less evidence to make such a categorical hypothesis about political extremes' similarity. On the one hand, recent results showing that perceived social fabric disintegration is associated with the wish for a strong authoritarian leader (Sprong et al. 2019) while not being linked to linear political orientation (Teymoori et al. 2016) can lead us to assume a symmetry between the extremes. On the other hand, previous work suggesting that rightwing ideology is associated with greater perceived moral disintegration (McClosky and Chong 1985;Parenteau & Parenteau 2008) and lower levels of interpersonal trust ) may lead us to alternatively assume an asymmetry between the extremes, i.e., that individuals would perceive more social fabric disintegration as they position themselves to the political right.

Overview
The present research aimed to extend the existing literature on political extremes similarities and differences, by examining the relationship between perceived anomie and political extremism across three studies conducted within the French population.
In the first two studies, we examined whether participants would report higher levels of perceived societal anomie (i.e., whether they perceive current society as more disintegrated and disregulated) as they positioned themselves toward the political extremes (Studies 1 and 2); and whether perceived societal anomie would predict the support for different kind of political proposals (Study 2). In the final study, we additionally examined whether political extremes, compared to more moderates, similarly imagined reduced levels of societal anomie (i.e., disintegration and disregulation) within society once their candidate is elected.
To do so, we took advantage of the 2017 French presidential elections. Every five years, French citizens directly elect the President of the French Republic in a two-round election. The 2017 presidential election had a particular tone (Rothwell & Samuel 2017) as five candidates representing a wide political spectrum were a priori well-positioned and were all credited with 15-25% of votes: the far-left (i.e., La France Insoumise, 'Untamed France'; Jean Luc Mélenchon), the left-wing (i.e., the Socialist Party; Benoit Hamon), the centrist/moderate (i.e., En Marche!, 'On the move!'; Emmanuel Macron), the right-wing (i.e., the Republicans; François Fillon) and the far-right (i.e., Front National, 'National Front'; Marine Le Pen). The studies were conducted before the first round (i.e., before any candidate had been eliminated).

Studies 1 & 2
The first two studies aimed to examine the relationship between anomie and political orientation. In line with the literature presented above, we assumed a U-shaped relationship between perceived anomie and political orientation. More precisely, we expected political extremes to perceive more anomie within society (i.e., more leadership disregulation and more social fabric disintegration, although our hypothesis regarding social fabric disintegration dimension was less categorical, as noted in the introduction section) than political moderates.
Compared to Study 1, Study 2 additionally controlled participants' perceived social status. As previous studies showed a negative relationship between social status and perceived anomie (e.g., Heydari et al. 2012) and knowing that social class is associated with political attitudes and voting behaviors (e.g., D'Hooge et al. 2018), we aimed to ensure that the relationship between political extremism and anomie existed beyond that of social status.
Moreover, in order to further investigate the relationship between perceived societal anomie and political orientation beyond a mere positioning on a political scale, Study 2 additionally examined whether perceived anomie predicted participants' support for different kind of political proposals-i.e., radical right-wing political proposals vs. proposals that were shared by the far-right and far-left French populist parties in 2017 (e.g., Gougou & Persico 2017;Hewlett 2017) and that we named populist proposals. First, we expected that the support for populist proposals (here, proposals regarding economic redistribution and protectionism) would increase as participants positioned themselves to the political extremes; in line with previous work showing that far-right and far-left parties and voters share populist attitudes (Akkerman et al. 2017) as well as similar stances on economic redistribution (e.g., Gougou & Persico 2017; see also Ivaldi & Mazzoleni 2019 for the stances on economic redistribution of the French far-right party) and protectionism (e.g., Halikiopoulou et al. 2012;Hewlett 2017;Gougou & Persico 2017;van Bohemen et al. 2019). By contrast, we assumed that the support for radical right-wing policies (i.e., with a clear right-wing ideological content; here, right-wing proposals related to immigration, family and economy) would increase as participants positioned themselves to the political right (e.g., Immerzeel et al. 2016). Regarding the link between perceived societal anomie and support for radical proposals, which was of special interest here, we assumed that perceived anomie would be associated with greater support for populist proposals, but would not predict the support for radical right-wing proposals; in line with previous work showing that anomie is associated with authoritarianism (e.g., Heydari et al. 2012;Scheepers et al. 1992;Sprong et al. 2019), while not being related to linear political orientation (Teymoori et al. 2016), as well as work showing links between concepts close to anomie and support for populism (e.g., Akkerman et al. 2017;van Bohemen et al. 2019;Spruyt et al. 2016).

Method Participants and Procedure
Sample sizes were determined according to recommendations of N = 250 (Schönbrodt & Perugini 2013) for correlation studies. 251 French participants (76.2% women, M age = 31.21, SD age = 11.76) fully completed Study 1 in February 2017 (i.e., approximately two months before the first election round); 311 French participants (78.4% women, M age = 31.38, SD age = 11.98) fully completed Study 2 in March 2017. Participants were recruited via social networks to participate voluntarily in an online study about their perception of French society (Study 1) or their perception of socioeconomic issues in France (Study 2). After a message reassuring about anonymity and checking for consent, participants filled in sociodemographic information (e.g., gender, age) before completing the scales described below.

Measures
The same measures were used for both studies to assess participants' perceived anomie and political orientation. Two additional variables were assessed in Study 2: participants' support for radical political proposals and perceived social status.
Perceived Anomie. We used our revised version of the PAS (Teymoori et al. 2016; see the Supplementary Material document on the OSF for further information on the scale's revision and validation). The final version of our measure included 20 items. Participants were asked to think about current French society while answering each of the statements and rated each item on a 7-point scale ranging from 1 = 'Strongly disagree' to 7 = 'Strongly agree'. Ten items referred to Leadership Disregulation (e.g., The political system functions as it should, reversed; α = 0.89 for both studies); and ten items referred to Social Fabric Disintegration (e.g., People think that the end justifies the means; α = 0.85 for Study 1; α = 0.83 for Study 2). The two scores were moderately and significantly correlated to each other (r = 0.40 for Study 1; r = 0.25 for Study 2). Higher scores indicated higher perceived anomie.
Political Orientation. Participants placed themselves on a political scale ranging from 1 = 'Far-left' to 11 = 'Farright' (e.g., van Prooijen et al., 2015; M = 5.12; SD = 2.79; min = 1; max = 11 for Study 1; M = 5.07; SD = 2.86; min = 1; max = 11 for Study 2). Support for Political Proposals (Study 2). Participants indicated their level of agreement with 23 political proposals retained following a pilot study (see the Supplementary Material document on the OSF for further information), using a 7-point scale ranging from 1 = 'Strongly disagree' to 7 = 'Strongly agree'. We created five scores (see the Supplementary Material document for further information on the factorial structure). Three scores were composed of right-wing radical proposals: opposition to immigration (e.g., The banishment of support protests in favor of illegal immigrants; n = 10, α = 0.92), right-wing economy (e.g., The suppression of Wealth Tax; n = 3, α = 0.62), and conservative politics about family (e.g., Marriage and full adoption restricted to heterosexual couples; n = 2, r = 0.44); and two scores referred to populist proposals: protectionism (e.g., The exit of the European Union; n = 3, α = 0.73) and economic redistribution (e.g., The full reimbursement of all medical care; n = 5, α = 0.66).

Anomie and Political Orientation (Studies 1 & 2)
In order to examine the relationship between perceived societal anomie and political orientation we first conducted hierarchical regression analyses, following the recommendations of Cohen et al. (2003). We entered the centered political orientation score (that tested the linear relationship between political orientation and perceived anomie, i.e., an asymmetry between the extremes) and the quadratic term (testing the quadratic relationship between political orientation and perceived anomie, i.e., a symmetry between the extremes) as predictors in two separate steps (e.g., . In both studies, we entered the centered political orientation score as a predictor in step 1 and added the quadratic term in step 2; in Study 2, we additionally added the perceived social status score as a control variable in a last step (step 3). As a significant quadratic term can be consistent with other relationships than U-shaped ones (Cohen et al. 2003;Simonsohn 2018), we subsequently examined the nature of the quadratic effect, by conducting simple slopes analysis (Cohen et al. 2003) at the farleft (-2 SD), the left (-1 SD), the right (+1 SD), and the far-right (+2 SD) of the political spectrum, according to Aiken et al.'s instructions (1991). The simple slopes represent the linear regression of Y (here, perceived anomie) on X (here, political orientation) at particular values of X (here, at four different levels of political orientation) and each simple slope represents 'the slope of a tangent line to the curve at a particular value of X' (Cohen et al. 2003: 207). To conclude that there is a U-shaped relationship, the slopes at the left of the political spectrum (-1 SD and -2 SD) need to be negative and significant, whereas the slopes at the right of the political spectrum (+1 SD and +2 SD) are expected to be positive and significant; and the effect are expected to be stronger at the extremes (-2 SD and +2 SD). Results of hierarchical regression and simple slopes analyses are displayed in Tables 1 (Study 1) and 2 (Study 2; see also: Figures 1 to 4 for the graphic representations of the relationships between anomie and political orientation).
Regarding the leadership disregulation dimension of perceived societal anomie, the linear effect of political orientation was negative and significant in Study 1 (i.e., greater perceived leadership disregulation as participants positioned themselves toward the political left) but was non-significant in Study 2. For both studies, the second step of the analysis highlighted significant quadratic terms; indicating a potential symmetry between the extremes. In both studies, simple slopes analyses subsequently showed the expected negative and significant effects for left-wing (-1 SD) and far-left (-2 SD) participants, which were stronger at the far-left (i.e., the more participants positioned themselves to the far-left, the more they perceived leadership disregulation); and positive and significant effects for far-right participants (+2 SD), indicating that the more participants positioned themselves to the far-right, the more they perceived leadership disregulation. A significant and positive effect for right-wing participants (+1 SD) was found only in Study 2; yet, it was weaker than for farright participants. The simple slopes results therefore offer evidence for the expected U-shaped relationship: as participants positioned themselves toward either Step 1

Anomie -Leadership Disregulation Anomie -Social Fabric Disintegration
Note: * p < 0.05, ** p < 0.01, *** p < 0.001; b (β) stands for the unstandardized (standardized) coefficient. Step 1  political extreme, they perceived greater leadership disregulation (see Figure 1 for Study 1 and Figure 3 for Study 2). In Study 2, the last step of the analysis additionally revealed a negative and significant effect of perceived social status (i.e., greater perceived leadership disregulation as participants' perceived social status decreased) and the quadratic term remained significant.

Anomie -Leadership Disregulation Anomie -Social Fabric Disintegration
Regarding the social fabric disintegration dimension of perceived societal anomie, the results of both studies highlighted significant and positive linear effects of political orientation but the quadratic terms were nonsignificant; indicating, this time, an asymmetry between political extremes. Hence, as participants positioned themselves toward the political right, they perceived stronger social fabric disintegration (see Figure 2 for Study 1 and Figure 4 for Study 2). Finally, the last step of the analysis in Study 2 once again showed a negative and significant effect of perceived social status.

Anomie and Support for Political Proposals (Study 2)
To further investigate the relationship between perceived societal anomie and political extremism, we examined whether perceived anomie predicted participants' support for different kind of political proposals. We expected that the support for populist proposals (here, proposals related to economic redistribution and protectionism) would increase as participants positioned themselves to the political extremes; whereas the support for radical right-wing policies would increase as participants positioned themselves to the political right. Regarding the link between perceived societal anomie and support for radical proposals, we assumed that perceived anomie would be associated with greater support for populist proposals but would not predict the support for radical right-wing proposals.
To do so, we conducted hierarchical regression analyses, in which we entered the centered political orientation score in step 1, the quadratic term in step 2, and the perceived societal anomie scores (leadership disregulation and social fabric disintegration) in a last step. When the quadratic term (i.e., political extremism) was significant, we conducted simple slopes analysis at four levels of political orientation (-2 SD, -1 SD, +1 SD, +2 SD). Results of hierarchical regression analyses are displayed in Tables 3 (for the right-wing radical proposals) and 4 (for the populist proposals).
Regarding the right-wing proposals (i.e., related to immigration, family and right-wing economy; see Table 3) the linear effects of political orientation were positive and significant while the quadratic terms were non-significant; indicating the expected asymmetry between political extremes. Hence, as participants positioned themselves toward the political right, they held more radical conservative positions regarding family matters and reported higher levels of support for radical right-oriented economic proposals and opposition to immigration. As expected, the last steps of the analyses, in which both dimensions of anomie were included, did not reveal any effect of perceived anomie on support for these rightwing radical proposals.
More interestingly, the analysis highlighted a similar pattern regarding the populist proposals (i.e., the protectionist and economic redistribution proposals; see Table 4). The first steps were significant and revealed significant linear effects of political orientation that were positive for the protectionist proposals and negative for the economic redistribution ones: as participants positioned themselves to the political right (left), they reported higher levels of support for protectionist (economic redistribution) proposals. In step 2, the linear effects remained significant, but the quadratic terms were also significant for both kind of proposals. Simple slopes analysis (see Table 4) subsequently showed the expected significant and negative effects at the left (-1 SD) and far-left (-2 SD) of the political spectrum, the effect being stronger at its far-left (i.e., the more participants positioned themselves to the extreme left, the more they supported protectionist and economic redistribution proposals); and significant and positive effects for far-right participants (+2 SD; i.e., the more participants positioned themselves to the right end of the political spectrum, the more they supported protectionist and economic redistribution proposals). Regarding protectionist proposals, the effect was also significant for right-wing participants (+1 SD), yet weaker than for farright respondents (+2 SD). The quadratic effect and the subsequent simple slopes analysis therefore suggest that as participants positioned themselves toward either political extreme (far-left or far-right), they reported higher levels of support for economic redistribution and held more protectionist positions; even though the support for economic redistribution (protectionist) proposals was even more pronounced for far-left (far-right) respondents. Finally, the last steps of the regression analyses were significant and revealed that the leadership disregulation component of anomie, but not the social fabric disintegration one, positively predicted participants' support for protectionist proposals and for proposals regarding economic redistribution.

Discussion
These first two studies highlighted the expected U-shaped relationship between political orientation and the leadership disregulation dimension of perceived societal anomie; indicating that as participants positioned themselves towards either political extreme, they perceived greater leadership disregulation within society. Study 2 additionally showed that the quadratic term remained significant after controlling for participants' perceived social status; thus suggesting that the relationship between political extremism and perceived disregulation existed above and beyond that of social status. In Study 2, perceived leadership disregulation was also associated with greater support for proposals regarding economic redistribution and for protectionist proposals, two kinds of proposals that besides seem to conciliate both ends of the political spectrum. On the other side, the social fabric disintegration dimension of perceived anomie was not associated with political extremism nor with the support for radical political proposals in Study 2, whether right-wing oriented or more general. Instead, perceived breakdown of social fabric increased as participants positioned themselves toward the political right.
The results regarding the support for the political proposals in Study 2 must however be taken with caution, as some of our scores presented low internal reliability. Moreover, the sample of the pilot study in which the perceived radicalness of the proposals was assessed was left-leaning; thus compromising the quality and representativeness of the proposals that were selected. Further research investigating the relationships between anomie and support for radical political proposals using better quality measurements is therefore needed.
In order to extend the study of the symmetries and asymmetries of political extremes regarding their perception of societal anomie, we then took the opportunity of the 2017 Presidential elections to examine whether political extremes, compared to more moderates, would similarly imagine that society would be less anomic after the election of their candidate compared to the current situation. Step 1

Study 3
Previous work showed that projections about the Nation's future are utterly relevant when studying people's current political behaviors. For example, Bain et al. (2013) consistently found across eight studies that present socio-political attitudes were associated with the projection of a better future society. Consistent with this line of thought, this final study examined whether extreme political positioning would be associated with the projection of a less anomic (i.e., less disintegrated and less disregulated) future society. For that matter, joining, supporting and identifying with radical authoritarian ideologies and leaders has been described as a way to reduce the aversive state of uncertainty and perceived societal disorder (e.g., Hogg & Adelman 2013;Kay & Eibach 2013). Extreme groups and ideologies can be perceived as providers of structure and order, through unyielding principles that make the social world more predictable and understandable (e.g., Hogg & Adelman 2013;Kay & Eibach 2013;. Based on Teymoori et al.'s (2017) assumption whereby authoritarian leadership could help coping with the perceived disintegration and disregulation of society, we therefore assumed an inverted U-shaped relationship between political orientation and projected anomie. More precisely, we examined whether political extremes would imagine a less anomic (i.e., less disregulated and disintegrated) society after the election of the candidate they support, compared with the current situation.

Method
Participants and procedure 205 French participants (60.5% women, M age = 39.44, SD age = 14.19) fully completed the present study in March 2017. Participants were recruited via social networks to participate voluntarily in an online study about the French presidential elections. After a short message reassuring about anonymity and checking for consent, participants filled in sociodemographic information. Then, they were invited to project themselves into a near future, in which the candidate they supported was elected as President of the French Republic. Participants were asked to fill in the name of their candidate and were invited to take a few minutes to think about their feelings and the situation of France after their candidate's accession to the Presidency. Participants then gave three reasons why they would vote for their candidate, in order to strengthen the projection. Finally, they completed the measures described below.

Measures
Other measures were included in the study but are not presented here. Nevertheless, the complete set of data is available on the OSF.
Projected Anomie. We adapted our anomie scale to assess participants' perceived anomie after the election, compared with the current situation (an adaptation derived from the measurement of collective futures developed by Bain et al. 2013). To do so, we conjugated Step 1  our twenty items in the future tense (e.g., The political system will be legitimate, reversed). Participants were asked to answer while referring to the following sentence: 'Compared with the current situation, if my candidate is elected president of France …'. They answered using a 7-point scale ranging from -3 = 'Much less than now', through 0 = 'No different than now', to +3 = 'Much more than now'. As for the original version of our scale, we created one score gathering leadership disregulation items (α = 0.93) and one score gathering social fabric disintegration items (α = 0.86). Negative (positive) scores indicated lower (higher) levels of perceived anomie after the election of the participant's candidate compared with the current situation.
Political Orientation was measured using the same item as for the two previous studies (M = 4.98; SD = 2.79; min = 1; max = 11).

Results
Previous research showed that political orientation on a left-right political scale is a stable and consistent predictor of voting preferences in the French context (e.g., Bélanger et al. 2006). Hence, after checking for the correspondence between participants' political orientation scores and voting intentions (by verifying the political orientation mean scores per candidate, see Supplementary Table 5 in the Supplementary Material document hosted on the OSF) we used the political orientation scores to qualify participants' intended vote. We thus followed the same analysis procedure as in the first two studies (Cohen et al. 2003;Aiken et al. 1991): we first conducted a hierarchical regression analysis (in which we entered the centered political orientation score and the quadratic term in two separate steps) and subsequently conducted simple slopes analysis at four levels of political orientation (-2 SD, -1 SD, +1 SD, +2 SD). Compared to the first studies, we were excepting an inverted U-shaped relationship (i.e., lower levels of projected anomie at the extremes); therefore, we excepted significant positive slopes at the left of the political scale and significant negative slopes at its right. Results of hierarchical regression and simple slopes analyses are displayed in Table 5 (see also: Figures 5 and 6 for the graphic representations of the relationships between projected anomie and political orientation).
Step 1 revealed a significant and positive linear effect of political orientation (i.e., as participants positioned towards the political left, they projected lower levels of leadership disregulation after the election of their candidate). In Step 2, the quadratic term was also significant; indicating a potential symmetry between both extremes. Simple slopes analysis showed the expected significant and positive effects at the left (-1 SD) and far-left (-2 SD) of the political spectrum, the effect being stronger for far-left participants (-2 SD: i.e., the more participants positioned themselves to the extreme left, the less they projected leadership disregulation); and a significant and negative effect only for far-right participants (+2 SD; the more participants positioned themselves to the extreme right, the less they projected leadership disregulation). Although the effect was even more pronounced for far-left respondents, as suggested by the linear negative effect of political orientation, the results of the simple slopes analysis were consistent with our inversed U-shaped relationship hypothesis: as participants positioned themselves toward either political extreme (far-left or far-right), they projected lower levels of leadership disregulation after the election of the candidate they supported, compared with the current situation.
Step 1 highlighted a significant and positive effect of political orientation (i.e., as participants positioned towards the political left, they projected lower levels of social fabric disintegration after the election of their candidate).
Step 2 revealed Note: * p < 0.05, ** p < 0.01, *** p < 0.001; b (β) stands for the unstandardized (standardized) coefficient. a significant quadratic term. Simple slopes analysis showed the expected significant and positive effects at the left (-1 SD) and far-left (-2 SD) of the political spectrum, the effect being stronger for far-left participants (-2 SD: i.e., the more participants positioned themselves to the extreme left, the less they projected social fabric disintegration after the election of their candidate). However, it was not significant for the right (+1 SD) or the far-right (+2 SD) of the political spectrum.

Discussion
This study showed that both political extremes, compared to more moderates, projected a less disregulated political system after the election of their candidate compared to the current situation. The latter results further strengthen the idea of a symmetry between political extremes, not only in their perception of disregulation within current society, but also, in their projection of post-election society. However, although the quadratic term significantly predicted projected social fabric disintegration and that the descriptive pattern of results was in line with a symmetry between the extremes, the effect was non-significant for far-right participants. From the first two studies' results (in which perceived social fabric disintegration was higher among far-right participants), we could have expected projected social fabric disintegration to decrease as participants position themselves to the far-right. Instead, our results show that as participants position themselves to the left extreme, they projected lower levels of social fabric disintegration after the election compared to the current situation. We believe that an explanation for these results can be found in the core features of right-wing and left-wing ideologies. Indeed, whereas left-wing ideologies are based on the idea of a continuous progress of mankind and society in general, right-wing ideologies are characterized by both a perception of moral decay and the idea of a morally flawed human nature that cannot be changed (e.g., Parenteau & Parenteau 2008). Taken together, these features might explain, on the one hand, why far-left participants projected lower levels of social fabric disintegration in future post-election society; and on the other hand, why far-right participants perceived stronger moral disintegration within current society (Studies 1 and 2), while they did not project a future society characterized by more moral and trustful people after the election of their candidate.

General Discussion
The present research aimed to examine the relationship between perceived societal anomie and political extremism across three studies conducted within French samples.
In the first two studies, we examined whether political extremes would report higher levels of perceived societal anomie (i.e., whether they perceive current society as more disintegrated and disregulated) than more moderates. The results suggested both symmetries and asymmetries regarding political extremes' perception of societal anomie. On the one hand, they revealed the expected U-shaped relationship between the leadership disregulation component of perceived societal anomie and political orientation: both extremes perceived greater leadership disregulation than political moderates. This result is consistent with previous research emphasizing the symmetries between political extremes regarding their negative perceptions of society (e.g., van der Bles et al. 2018), as well as their lower levels of political trust (e.g., Akkerman et al. 2017) and their ' estrangement' from the political system (McClosky & Chong 1985). On the other hand, the social fabric disintegration component of perceived societal anomie was linearly linked to political    . Study 2 additionally revealed that perceived leadership disregulation, but not perceived social fabric disintegration, was associated with greater support for proposals regarding economic redistribution and protectionism; two kinds of populist proposals that besides conciliated both ends of the political spectrum. These results corroborate Akkerman et al.'s findings (2017) showing that both extremes share populist attitudes while displaying lower levels political trust (see also : Spruyt et al. 2016). Besides, the fact that perceived leadership disregulation predicted people's support for these populist proposals while perceived social fabric disintegration did not, could be explained by the 'representation of the people as a pure and homogeneous group' that characterizes populism (Akkerman et al. 2017, p. 380); a representation that is somehow opposed to the perception of people's immorality that characterizes the perceived social fabric disintegration component of anomie. On the other hand, neither dimension of perceived anomie, nor political extremism, predicted participants' support for radical right-wing proposals regarding immigration, economy and family. Taken together, these results are consistent with work showing that far-right and far-left political parties share a close anti-establishment and 'populist' style, while carrying opposite stances on core ideological issues, such as immigration (e.g., Immerzeel et al. 2016). They also corroborate previous work specific to the 2017 French political landscape, which emphasized that the Untamed France (far-left; Jean-Luc Mélenchon) and the National Front ( far-right: Marine Le Pen) parties, despite their obvious ideological differences, presented some similarities regarding economic distribution and anti-Europe positioning (e.g., Gougou & Persico 2017;Hewlett 2017).
By identifying a split, in which the perceived lack of effective regulation from the political system-that is, perceiving a lack of regulation on a rather vertical dimension-is shared by both political extremes and predicts their support for concrete political measures, while the perceived lack of effective regulation within the social fabric-that is, perceiving a lack of regulation on a more horizontal dimension of society-is more specific to right-wing individuals, the present findings offer some preliminary elements that could potentially help to further refine the scientific knowledge regarding the (a) symmetries between political extremes. In this regard, our results somehow concur with recent work showing that 'upward' conspiracy beliefs (i.e., targeting powerful groups) are associated with political extremism, whereas 'downward' conspiracy beliefs (i.e., targeting powerless groups) are stronger among conservative respondents (Nera et al. 2021). Perceived societal anomie in its social psychology approach (Teymoori et al. 2017) could thus help clarifying the convergences (i.e., the perception of a political system that fails to regulate collective wellbeing effectively and legitimately) and divergences (i.e., the perception of a society whose morals and trust are disintegrating) between political extremes, and therefore, could complement previous work showing that political extremes share mere pessimistic views of society (e.g., van Der Bles et al. 2018;Steenvoorden & Harteveld 2018). However, as we have not measured our participants' endorsement of a pessimist zeitgeist (van Der Bles et al. 2018) nor their levels of societal pessimism (Steenvoorden & Harteveld 2018), further empirical work that will control for these variables is needed to examine whether perceived leadership disregulation is associated with political extremism above and beyond mere pessimistic perceptions of society.
The last study examined whether political extremes, compared to more moderates, similarly projected reduced levels of societal anomie within society once their candidate is elected. We found that both far-right and far-left respondents, compared to more moderates, projected reduced levels of leadership disregulation within postelection society, compared with the current societal situation. However, projected social fabric disintegration was lower as participants positioned themselves towards the left-end of the political spectrum. These findings expand the first studies' results; as they seem to suggest that political extremism is associated with the hope of reaching a less disregulated future society (e.g., Bain et al. 2013). Moreover, by suggesting that the quest for reducing uncomfortable perceptions of disregulation may have a key role in political extremism, these results are, to a certain extent, in line with previous work showing a link between uncertainty (e.g., Hogg & Adelman 2013) or perceived disorder (Kay & Eibach 2013) and extremism.
Despite its contribution, the present work is obviously not free of limitations. The first limitation that must be noted relates to the samples on which the present results are based. Indeed, although post-hoc power analyses indicates that we have sufficient statistical power (see Supplementary document for more detailed information), our results are based on convenience samples, which are not representative of the French population (i.e., overrepresentation of students, left-wing participants 1 and women).
Moreover, the measurement of political orientation we used may have not capture the complexity of political attitudes. Indeed, although we used a more complex measure of political attitudes in Study 2 (i.e., the support for different kind of political proposals), additional studies distinguishing political orientation on social and economic aspects (e.g., Choma et al. 2010) and not mere right/left opposition could be useful to further clarify the relationship between political orientation and perceived anomie.
As perceived leadership disregulation was associated with stronger support for economic measures fostering protectionism and redistribution, one can assume that political orientation on economic aspects would be more strongly associated with the leadership disregulation component of perceived anomie. On the other hand, political orientation on social/cultural aspects might be more sensitive on the issue of social fabric disintegration, as the latter focuses on the breakdown of trust and morality.
Finally, given the present design, we cannot adjudicate on the direction of the relationship between perceived societal anomie-and more precisely, perceived disregulation-and political extremism. Based on previous empirical and theoretical work, it is likely that perceptions of anomie, by the uncertainty and insecurity they generate, could foster radicalism (Kruglanski et al. 2014;Teymoori et al. 2017) and thus, political extremism. However, some elements suggest that political extremism could also contribute to strengthening perceptions of anomie (e.g., Teymoori et al. 2017). Hence, further experimental or longitudinal work is needed to explore and better understand the directions in this relationship.

Note
1 Additional information regarding the distribution of political orientation in our samples and in the general population is available in the Supplementary Material document.