A More Competent, Warm, Feminine, and Human Leader: Perceptions and Effectiveness of Democratic Versus Authoritarian Political Leaders

Nowadays, to the detriment of democratic leaders, the emergence of authoritarian leaders has drastically modified the political sphere. This project aims to shed light on this issue by analysing how the perceived effectiveness of democratic and authoritarian political leaders are shaped by the common dimensions of social perception, such as competence/warmth, masculinity/femininity, and human uniqueness/human nature. Accordingly, three experimental studies were conducted. In Study 1 ( n = 1001), we revealed that democratic leaders are perceived as more competent, warm, feminine and human. In Study 2 ( n = 548) and Study 3 ( n = 622), we investigated whether these dimensions of perception mediated the relation ship between leaders and their perceived effectiveness. The results revealed that democratic leaders are perceived as effective in cooperative scenarios due to their competence, femininity, and human nature. Alternatively, democratic leaders are preferred in ambiguous contexts due to their competence and cog nitive flexibility, that is, human nature. In contrast, authoritarian leaders are perceived as effective in competitive scenarios because of their masculinity. In Study 3, we manipulated the (in)stability of socioeconomic contexts. The results revealed that democratic and authoritarian leaders are perceived as more competent, warm, human and more effective in socio-economic contexts that are stable compared with those that are unstable. The implications of the results regarding the emergence of authoritarian leaders are discussed.

seems to be especially relevant to understanding leadership style of political leaders.
Democratic leaders are oriented toward relationships (Yukl, 1994). Furthermore, Bass and Bass (2008) noted that democratic leaders are known to make decisions on the basis of mutual agreement between parts, consider the influence of decisions on individual well-being, endeavour to maintain close relationships with followers and motivate them by reinforcing connection with others and acknowledge the inherent value of individuals and their ideas. In contrast, authoritarian leaders focus on tasks (McClelland, 1961;Wofford, 1970). They make unilateral decisions and mainly target goal realisation rather than the well-being of followers, maintain social distance with followers and employ punishment and threats instead of reinforcement to motivate followers.
The study on the differences between authoritarian versus democratic leadership styles traces its origin to the pioneering research of Lewin et al. (1939) on groups of children. Since then, researchers focused on identifying which leadership style favoured the productivity and performance of subordinates. Several studies illustrated that the democratic style favoured effectiveness and satisfaction (Likert, 1961), whereas others argued that authoritarian and task-oriented leadership provided the best results (Miner, 1968). Fiedler (1967Fiedler ( , 1978 addressed this contrast by demonstrating that the effectiveness of leadership style is dependent on the situation. Recently, a resurgence of interest emerged in investigating the influence of context on leadership and its outcomes (see Oc, 2018, for a review). In this sense, we propose that individuals may value democratic and authoritarian leaders differently in various contexts. Previous research has revealed that strong and dominant leaders are preferred in competitive contexts, such as intergroup conflicts, in contrast to peaceful and cooperative contexts (Laustsen & Petersen, 2017) and during unstable conditions, such as those marked by uncertainty (Rast, Hogg, & Giessner, 2013) and extreme economic inequality (Sprong et al., 2019). We note that the current competitive social environment has been promoted by the growing economic inequality (Sánchez-Rodríguez et al., 2018;Sommet et al., 2019) or uncertainty after the Great Recession of 2008 (Jetten et al., 2017). Thus, citizens may be more likely to perceive authoritarian leaders as better able to solve the difficulties in today's society compared with other leadership styles (Sprong et al., 2019). Despite the established knowledge about the influence of contextual factors on leader preference, less is understood about leader traits that may influence individual decisions about whether democratic or authoritarian leaders are preferable in certain scenarios. The importance of the issue relies on the fact that this preference may lead to long-term consequences, which may exceed the specific setting in which leaders are preferred. For instance, electing an authoritarian leader to address a specific armed conflict and due to their strong traits could have negative consequences, which may spread to other domains, such as economic (e.g., economic relationship with third-world nations) or social (e.g., immigration policies) issues, which could target individuals or groups under this leadership. Therefore, addressing the influence of the perceived traits of leaders in shaping the preferences for certain leaders is important for the well-being of citizens, but also due to the possible negative outcomes of electing leaders on distrust in democracy.
In this sense, the extensive research on individual differences in leadership has identified two broad categories of traits that characterise leaders (Bass & Bass, 2008). The first is task competence, which constitutes traits that influence leaders' ability to cope with the external environment and their followers. The second is socio-emotional competence, which includes empathy, heightened awareness, and consensual solutions to conflict. Task and socioemotional competences are fundamental to successful and effective leadership (Bass & Bass, 2008). Accordingly, compared with authoritarian leaders, democratic leaders may be perceived to exhibit concern about their followers through various means and possess many socio-emotional competences and solve societal problems more effectively. Research has demonstrated that followers view democratic leaders as more effective especially in the long run and more satisfying than authoritarian leaders (Likert, 1977;Miller & Monge, 1986). However, in recent years, democratic leaders seem to be losing popularity in favour of leaders with authoritarian styles. Citizens may perceive that solving the current problems in a competitive and unstable society requires only highly task-competent instead of socio-emotional-competent leadership.
To shed light on these issues, we examined the social perception of political leaders in relation to the different dimensions of social perception. To do so, we rely on a previous research on the prototypes of leaders, which identified several dimensions in this regard (e.g., sensitivity, intelligence, dedication, dynamism, or masculinity; Epitropaki & Martin, 2004). Specifically, this project focuses on analysing the role of certain dimensions of social perception that correspond to leader prototypes, such as competence-warmth (Fiske, 2018;Fiske et al., 2002), masculinity-femininity (Bem, 1974;Ellemers, 2018) and human uniqueness-human nature (Haslam, 2006).

Dimensions of Social Perception: Theoretical Approaches
The first approach employed was the SCM (Fiske et al., 2002), which posits two main dimensions ascribed to individuals or groups, namely, competence and warmth. Competence refers to the capabilities of others to enact an action, while warmth alludes to the social intentions of others that predict behaviours (Fiske, 2018). Previous studies revealed that, in general, although the fact that political leaders want to achieve their goals determines their perceived competence, they are perceived to forget warmth occasionally when they achieve their goals (Fiske & Durante, 2014). Chen, Jing, and Lee (2014) found that trustworthiness, that is, warmth, plays a role in potential electoral success but only when politicians were perceived as competent.
Second, individuals and groups may be also classified according to their ascribed masculinity and femininity (Bem, 1974;Ellemers, 2018). Masculinity is a dimension associated with men and is a cultural construct that constitutes acting as a leader and being self-sufficient, whereas femininity is a characteristic associated with women and includes the value of understanding others (Bem, 1974;Berdahl et al., 2018). In general, leadership has traditionally been associated with masculine attributes (Schein, 1973;Vial & Napel, 2018). Although differences between men and women are considered, female leaders tend to be perceived as democratic, whereas male leaders are considered authoritarian (Eagly & Johnson, 1990;Faiza, Nair, & Haque, 2018).
The third dimension of social comparison refers to the humanity ascribed to individuals and groups. The act of denying humanity, namely, dehumanisation, involves the psychological process through which people perceive a person or social group as if they are not completely human (Haslam & Stratemeyer, 2016). In accordance with Haslam's (2006) dual model of humanity, denial can encompass two dimensions of humanity: human uniqueness (HU), which comprises attributes related only to human beings, such as rationality and culture, and human nature (HN), which constitutes attributes that are lacking in inanimate objects or machines, such as emotionality and cognitive openness. The literature on the political sphere has examined the mutual dehumanisation between nations (Sainz et al., 2020), political parties, and/or political leaders (Cassese, 2019;Martherus et al., 2019;Pacilli et al., 2016).
Although various studies have intended to determine the common features in the three models of social judgement (Abele & Wojciszke, 2014;Haslam et al., 2008;Tipler & Ruscher, 2014;Vaes & Paladino, 2010), such perspectives have been widely and independently implemented to measure perceptions of individuals and groups. On the contrary, previous research has revealed a tendency among individuals to compensate when attributing traits associated with the said dimensions. In other words, in many situations, people typically attribute one dimension while denying others when evaluating individuals and groups (Kervyn, Yzerbyt, & Judd, 2010;Terache, Demoulin, & Yzerbyt, 2020). However, a peculiarity exists in the perception of political leaders: although they are expected to achieve their goals or be competent (Fiske & Durante, 2014), they are required to display concern for the needs of the populace or exhibit warmth (Bruckmüller & Methner, 2018). Out of the leadership styles examined, the democratic style best represents the expectation of ensuring the task is fit and demonstrating concern for the population. Thus, we expect that the perception of democratic leaders includes substantial competence and warmth without producing a compensation effect. Moreover, we expect that the perception of democratic leaders will involve openness to points of view of others and willingness to cooperate. In other words, we expect high scores on SCM, feminine traits, and the capacity to be even more human. On the contrary, we expect that the perception of authoritarian leaders will involve cold and rigid leaders who impose their preferences through force or even engaging in inhumane behaviour without caring for the well-being of others.
Such attribution of the dimensions of social perception may influence the extent to which leaders are perceived as effective when addressing various societal conflicts including those of a cooperative and competitive nature under stable and unstable conditions. Thus, this research aimed to shed light on this issue by analysing the extent to which democratic compared authoritarian leaders are perceived as effective across scenarios as a consequence of their ascribed traits along the dimensions of social perception.

Overview
Three studies were conducted to examine the influence of ascription or denial of traits in relation to the main dimensions of social perception (i.e., competence-warmth, masculinity-femininity, and HU-HN), which vary between democratic and authoritarian leaders. Furthermore, we explored the extent of the influence of these traits on the perceived effectiveness of leaders across contexts. In Study 1, we examined differences in these dimensions between democratic and authoritarian leaders. Study 2 replicates the findings from Study 1. Moreover, we analysed how the dimensions of social perception mediate the relationship between types of leaders and their perceived effectiveness under the competitive and cooperative conditions. In Study 3, we extended the previous findings by exploring the influence of unstable socio-economic contexts on the perception of leaders and their effectiveness. Finally, the current research aimed to differentiate from previous research that analysed the influence of ideological or individual variables on a leader's preferences/perceptions (e.g., Laustsen & Petersen, 2015Van Lange, Bekkers, Chirumbolo, & Leone, 2012). Thus, across the studies, the researchers included ideological variables and the socio-economic status of the participants as the control variables in the main analysis. 1

Study 1
Study 1 aims to analyse how democratic compared with authoritarian political leaders are evaluated according to the aforementioned dimensions of social perception. The previous research on leader prototypes proposed that leaders possess a certain set of traits common among democratic leaders, such as sensitivity, dedication, or intelligence (Epitropaki & Martin, 2004). However, other sets of traits, such as masculinity and tyranny, which are closely related to authoritarianism, are considered non-prototypical. Given the similarities between the main aspects of leader prototypes and dimensions of social perception, we hypothesise that democratic leaders will score higher in competence and warmth (H1 and H2), be perceived as more feminine and less masculine (H3 and H4) and be considered more human (HU and HN; H5 and H6) than authoritarian leaders. Preregistration is online: https://osf.io/xj3tm.

Participants and Procedure
The participants were recruited through an institutional general email list that encompasses students, professors, and other professionals of a university in Southern Spain. Each participant received an email invitation to participate in a study on the social perception of leaders. All participants provided informed written consent at the beginning of the survey and read the information on voluntary participation, anonymity, and confidentiality of responses. Analysis indicated that a minimum of 788 participants was required (G*Power analysis; independent t-test comparison, two-tailed, 80% Power, α = 0.05, d = 0.20; Faul et al., 2009). The final sample comprised 1,001 participants (695 females, M age = 23.13 years, SD = 5.27). Compensation in the form of inclusion in a raffle (200€). The participants were provided with materials.

Manipulating leadership style
The participants were informed that an important published study had examined the profiles of various leaders present in their society. In addition, they were told that a few of the leadership profiles will be presented, and they will be required to share their opinion about these leaders. The participants were shown either the description of a democratic or an authoritarian leader in random order. 1 To enhance the description of the authoritarian leader, we employed the Authoritarian Leader Behavior Scale (De Hoogh & Den Hartog, 2009), which includes items on dominant behaviours and mirrors leaders who are mainly concerned with protecting their position and make independent decisions without considering the suggestions of subordinates. To elaborate the description of the democratic leader, we replaced the characteristics and behaviours of the authoritarian leader with those of the democratic leader following the distinction between both leadership styles (Bass & Bass, 2008).

Measures
After reading the descriptions of the leaders, the participants were given a list of questions. Responses were rated using a 7-point Likert scale ranging from 1 = Not at all to 7 = Completely.

Manipulation checks
We included three questions about leader behaviours (α = 0.95): how the leader distributes his/her power (1 = He/she has all the power to 7 = Power is shared among government members), which objective he/she wants to reach (1 = unique objective important for him/her to 7 = objective important for the entire population) and the process of decision making (1 = exclusively the leader to 7 = distributes among all members of the government) (adapted from De Hoogh & Den Hartog, 2009).

Competence and warmth
Four adjectives related to warmth (e.g., sincere and gentle; α = 0.87) and five adjectives related to competence (e.g., competent and intelligent; Fiske et al., 2002) were included. Three items, namely, self-confident, competitive, and independent, were excluded from the final measure of competence because they decreased the reliability of the measure (α = 0.25). Furthermore, when conducting a Principal Component Analysis we observed that these three items did not adequately reflect the construct ' competence' in the Spanish context (see Supplementary Mate-rial). Thus, the final measure of competence included two items, namely, competent and intelligent (r = 0.641, p < 0.001).

Dehumanisation measure
We implemented a dehumanisation measure (Bastian, Jetten, & Radke, 2012) that comprises four items associated with HU (e.g., 'I think that the leader is refined and cultured'; α = 0.81) and four associated with HN (e.g., 'I think that the leader is superficial, he/she has no depth'; α = 0.88).
Finally, the participants answered the Social Dominance Orientation (SDO) scale, which was validated in a Spanish population by Silván-Ferrero and Bustillos (2007). The 16-item measure assesses opposition to equality (e.g., 'Group equality should be our ideal' [reverse item]; α = 0.80) and group dominance (e.g., 'Inferior groups should stay in their place'; α = 0.74). Responses were rated using a 7-point Likert-type scale ranging from 1 = Totally disagree to 7 = Totally agree. The participants evaluated their political orientation on a single item using a 7-point scale ranging from 1 = Extreme left wing to 7 = Extreme right wing. Furthermore, they assessed their subjective socioeconomic status by employing the 10-step MacArthur ladder (adapted from Adler et al., 2000) and objective socio-economic status by providing indicators, such as annual pre-tax income scope. Responses were rated using a 7-point scale ranging from 1 = Below €500 to 7 = More than €5,000. In addition, they evaluated the mean of the level of education of their parents using a 7-point scale ranging from 1 = Less than a high school degree to 7 = Doctoral degree. A single index of objective socio-economic status was computed by merging both indicators (r = 0.416, p < 0.001; Kraus & Keltner, 2009). Finally, they provided demographic details, such as age, gender, nationality, and language.

Results
First, the manipulation of the type of political leader was successful. Democratic leaders (M = 5.80, SD = 0.99) were perceived as more democratic than authoritarian leaders (M = 1.70, SD = 0.87, t(969.41) = −69.32, p < .001, Hedges' g s = 4.40). 1 Second, to test the hypotheses, MANCOVA was conducted on the main dependent variables (i.e., competence-warmth, masculinity-femininity, and HU-HN) as a within-group factor and leadership manipulation (democratic vs. authoritarian) as a between-group factor. The two SDO factors, participants' political orientation and participants' subjective/objective socio economic status were introduced into the analysis as control variables (Figure 1)

Discussion
The results of Study 1 revealed that democratic leaders are perceived as warm, emotional, open-minded, and feminine as well as competent and civilized. In contrast, authoritarian leaders are regarded as cold, masculine, and insensitive machines as well as less competent. In essence, these findings indicate that both types of leaders are considered different in relation to their ascribed traits. These findings could be significant due to the acknowledgement of behavioural tendencies and/or emotional experiences that may be generated by the two distinct perceptions (Cuddy, Fiske, & Glick, 2007;Ellemers, 2018;Haslam & Loughnan, 2014). In Study 2, we examined the extent to which democratic and authoritarian leaders are perceived as effective under a variety of scenarios.

Study 2
Study 2 explores how political leadership style, specifically, democratic compared with authoritarian, predicts the perceived effectiveness of leaders under cooperative and competitive contexts through the dimensions of social perception. Specifically, apart from replicating the findings of Study 1, we hypothesized that democratic leaders would be perceived as less effective than authoritarian leaders in terms of competitive scenarios, such as armed conflict (H1). The previous literature demonstrated that dominant leaders are preferred under competitive versus cooperative conditions (Laustsen & Petersen, 2017). On the contrary, Gartzia and van Knippenberg (2016) argued that, in terms of cooperation, communal leaders (who are linked to democratic leaders) are more effective than agentic leaders (those linked to dominant leaders). Thus, we hypothesized that democratic leaders will be perceived as more effective than authoritarian leaders during cooperative conflicts, such as a humanitarian crisis (H2). Additionally, we considered that the ascription of traits or characteristics to individuals or groups is fundamental for evaluating individual or group political performance of behavior (Cwalina & Drzewiecka, 2019;Fiske, 2019). In the context of leadership, previous studies associated the styles of leadership not only with perceived effectiveness but also with certain social traits. For example, Hentschel, Braun, Peus, and Frey (2018) established the association between transformational leadership style (i.e., similar to the democratic style) and communal traits, which leads to the association between communality and perceived effectiveness. Thus, we used this literature as a basis for exploring whether the perceived effectiveness of democratic and authoritarian leadership styles is mediated by ascribed social traits (i.e., competence-warmth, masculinity-femininity and humanity). Specifically, we explored if multiple dimensions of social perception could mediate the relationship between leadership and effectiveness under competitive (H3) and cooperative (H4) scenarios. Preregistration was online: https://osf.io/d5yk9.

Participants and Procedure
The same procedure in Study 1 was employed for data collection. In accordance with Schoemann, Boulton, and Short (2017), we calculated a sample size for mediation analysis by employing a Monte Carlo power analysis simulation and testing the indirect effect with a bootstrapped confidence interval. We considered a high correlation (r = 0.60; based on correlations from Study 1) for path a (X*M) and a moderate-high correlation (r = 0.40) for path b (M*Y) and c (X*Y). The standard deviations of X, M, and Y were based on data from Study 1. Analysis revealed that a minimum of 514 participants was required (Power = 0.80, α = 0.05). The sample included 548 participants (403 females, M age = 22.91, SD = 4.81). Compensation was in the form of inclusion in a raffle (100€). 2 After providing consent, the participants were presented with the following information in the following order: They first read about the Mamziba society and and were exposed to the leader manipulation (democratic versus authoritarian). After that, they answered the manipulation check (α = 0.95) and all measures of social perception dimensions. In this sense, given that three of the competence items were inefficient in Study 1, we decided to use a slightly different set of traits to measure competence in a more consistent manner by including the items 'skillful' and 'efficient' (Fiske et al., 2002

Perceived effectiveness of leaders
To assess the perceived effectiveness of leaders, we provided additional information about a fictitious society called Mamziba (see a similar procedure in Jetten et al., 2017). First, to ensure that the experience was realistic, we provided with various details of the society, such as residents and primary source of resources. Subsequently, the participants were informed that a particular political leader that had been described previously as democratic or authoritarian had been elected two years previously as the political leader of Mamziba at a time when the government was facing various critical situations in the country. 1 The participants read two scenarios that the leaders encountered (within-subject, scenarios were counterbalanced) which were adapted from Laustsen and Peteren (2015). In the first scenario, namely, the competitive scenario, the leader was required to protect citizens and drive out an invading army that wanted to take control of the oil reserves. This scenario required the deployment of military force. The second scenario was a cooperative scenario in which the leader was requested to cooperate with other nations to provide humanitarian help to a neighbouring state affected by severe flooding. This scenario required active collaboration to organise humanitarian assistance. After reading each scenario, participants rated the perceived effectiveness of the leaders in solving such conflicts using a 10-point scale ranging from 1 = Not at all effective to 10 = Very effective. As a manipulation check, an item that required participants to assess what kind of solution the leader had to implement for each conflict was included. The item was rated using a 4-point scale ranging from 1 = A solution based on using force to 5 = A solution based on cooperation.

Results
First, the manipulation was successful: the democratic leader was perceived to be more democratic (M = 5.73, SD = 1.18) than the authoritarian leader (M = 1.77, SD = 1.05, t(538.20) = −41.32, p < 0.001, Hedges's g s = 3.54). 2 Second, the results for the between-group comparisons (MANCOVA with dimensions of social perception [withingroup] and leadership manipulation [between-group]) replicated the results of Study 1 (F(6, 532) = 271.68, p < 0.001,  2 p = 0.75). In other words, the democratic leader was perceived as more competent, warm, feminine, less masculine, and with higher levels of HU and HN than the authoritarian leader (Figure 1). 1 Third, analysis of the perceived effectiveness of both leaders under the competitive and cooperative scenarios was conducted. The results revealed that the participants clearly understood the scenarios. Although the use of military force was deemed necessary in the armed scenario (M = 2.67, SD = 1.34), the humanitarian scenario was perceived as requiring cooperation with other countries (M = 4.65, SD = 0.83); t(546) = −32.16, p < 0.001, Hedges' g av = 1.77. Furthermore, MANCOVA was used to test the effect of leadership manipulation on perceived effectiveness in both scenarios. Leadership manipulation (democratic versus authoritarian) was regarded as the between-group variable, whereas the perceived leaders' effectiveness in competitive and cooperative scenarios were the dependent variables. The results revelated that leadership manipulation exerted no effect on perceived effectiveness (F(1, 544) = 0.03, p = 0.858,  2 p = 0.00) under the competitive scenario (democratic: M = 6.52, SD = 2.13; authoritarian: M = 6.56, SD = 2.39). However, a significant effect of leader manipulation on perceived effectiveness was observed (F(1, 544) = 465.42, p < 0.001,  2 p = 0.46) in the cooperative scenario (democratic: M = 8.00, SD = 1.66; authoritarian: M = 4.05, SD = 2.49). Thus, H2 was supported but not H1. 1 Finally, to verify if the dimensions of social perception mediated the relationship between leadership style and perceived effectiveness of leaders, two independent multiple mediational analyses were performed by employing Hayes' (2018) PROCESS (bootstrapping 10,000 interactions with 95% confidence intervals, Model 4; Figure 2). The covariates were included in the model.
A direct effect was not observed in the mediational analysis between leadership manipulation and perceived effectiveness in a competitive scenario (Table 1). However, perceived competence and masculinity mediated this relationship. Moreover, the same mediational analysis between leadership manipulation and perceived effectiveness in the cooperation scenario revealed not only a direct effect but also significant indirect effects. In the humanitarian scenario, competence and HN were deemed suitable mediators, whereas masculinity, femininity, HU and warmth did not play mediating roles. 1

Discussion
In Study 2, we confirmed the previous pattern of results related to the social perception of democratic and authoritarian leaders. Moreover, the results clearly demonstrated that democratic leaders are perceived as the most effective when solving cooperative scenarios during which provid-ing aid to those in need is imperative because they are perceived as competent, emotional, and cognitively flexible, that is, HN. However, the results suggested that in more competitive situations, where leaders are expected to make difficult and unpopular decisions, such as harming others in an armed conflict, authoritarian leaders are and perceived effectiveness (competitive and cooperative contexts), controlling for the subfactors of social dominance orientation, political orientation and participants' socio-economic status (subjective/objective) in Study 2. The total effects are enclosed in brackets; the direct effect is adjacent to the total effect. ** p < 0.001, * p < 0.05, † p < 0.09.  perceived to be as effective as democratic leaders. Compared with authoritarian leaders, democratic leaders are perceived as able to tackle competitive tasks because they are attributed a higher degree of competence. However, authoritarian leaders are believed to be effective in competitive scenarios because they are dominant and aggressive, that is, masculine. However, the perceived effectiveness of leaders is not only evaluated by their traits but also influenced by various contextual factors (Oc, 2018). Furthermore, in real-life situations, conflicts and contexts are frequently ambiguous. Uncertainty in the solution that should be implemented and in the environmental conditions of leaders, such as the stability or instability of their countries (Jetten, et al., 2017) may undermine perceived effectiveness. In Study 3, we analysed whether socio-economic instability influences the interpretation of leaders' effectiveness.

Study 3
Study 3 aims to replicate the effect of democratic and authoritarian political leadership styles on the perceived characteristics and effectiveness of leaders across social conflicts, specifically, cooperative and competitive scenarios, through the dimensions of social perception. We included an ambiguous social conflict, namely, immigration control, to gain an enhanced understanding of the perceived effectiveness of leaders in various social scenarios. We expected that democratic compared with authoritarian leaders will be perceived as more effective in humanitarian crises, namely, cooperative scenarios (H1), and less effective in armed conflicts, that is, in competitive scenarios (H2). Moreover, we expected no clear differences in immigration control conflicts, meaning, ambiguous scenarios (H3). Conversely, the literature shows that uncertain contexts, such as socio-economic instability, favours people's preference for a decisive, authoritative, and dominant leader over a respected, knowledgeable, admired, and permissive leader (Kakkar & Sivanathan, 2017). Regarding perceived social traits, people seemingly prefer agentic versus communal leaders in times of crises (Kulich, Iacoviello, & Lorenzi-Cioldi, 2018). Thus, the second objective of the study was to explore whether socio-economic instability (versus stability) could potentially moderate the ascribed traits of democratic and authoritarian leaders, their perceived effectiveness across scenarios and the mediational model (path a and direct path). Preregistration was online: https://osf.io/xhd57.

Participants and Procedure
The same data collection procedure used in Studies 1 and 2 was employed. The sample size for the mediation analysis was calculated by utilising the Monte Carlo power analysis. The same correlation coefficients for each path and the same power (0.80), alpha (0.05) as in Study 2 were used. A minimum of 514 participants was required. The final sample included 622 participants (443 females, M age =23.13 years, SD = 5.45). Compensation was in the form of inclusion in a raffle (100€). After providing consent, participants were presented with information in the fol-lowing order. First, they were given general information about the Mamziba society similar to Study 2. They were requested to imagine living in that society. Secondly, they were presented with the experimental manipulation of socio-economic context stability.

Manipulating socio-economic stability
To manipulate socio-economic stability (between subject), a procedure similar to that of Jetten et al. (2017) was employed. Participants in the stability condition were presented with a headline from a fictitious newspaper that provided information about the society's socio-economic stability, for example, 'Mamziba: A prosperous and stable economy'. A short text followed the headline describing the stable economic situation of Mamziba society and further explaining that the country was not expecting changes in the economic conditions of the population. On the contrary, participants assigned under the instability condition were presented with a headline that highlighted the unstable economic situation of Mamziba, for example, 'Mamziba: A decadent and unstable economy'. The headline was followed by a short text explaining how the economy of the society was declining and how the population was going to lose purchasing power. 1 After presenting the socio-economic instability manipulation, we presented the same information about leadership styles, namely, democratic and authoritarian, which was employed in Studies 1 and 2 (between-subject). Accordingly, given both manipulations, the participants were randomly assigned to one of four possible scenarios (between-subject), namely, stable context and democratic leader, stable context and authoritarian leader, unstable context and democratic leader, and unstable context and authoritarian leader.
After presenting the manipulations, participants answered a manipulation check question on the socioeconomic situation of Mamziba. The response was rated using a 7-point scale ranging from 1 = Socio-economic instability to 7 = Socio-economic stability. They were required to answer another three items on leadership styles (α = 0.94), similar to Studies 1 and 2. After completing the social dimension measures, participants read about the different scenarios. To measure the perceived effectiveness of leaders in Study 3, scenarios with competitive (armed conflict) and cooperative (humanitarian conflict) social conflicts were provided (within-subject). We modified the description of the armed scenario to highlight that the use of military force was necessary. 1 In addition, we included an ambiguous scenario about the imminent arrival of immigrants to Mamziba society that the government should address. The ambiguity of this scenario was highlighted by telling the participants that the leader should decide to use force and avoid the arrival of immigrants or cooperate with the immigrants' nations to control migratory flow. After reading each scenario, the participants rated leader effectiveness in each scenario using a 10-point scale ranging from 1 = Not at all effective to 10 = Very effective. In addition, we added an item: 'What kind of solution did the leader have to take in each of the conflicts?' to ensure that the participants understood the type of solution required for each scenario. The item was evaluated using a 5-point scale ranging from 1 = A solution based on using force, 3 = Both solutions and 5 = A solution based on cooperation.
In relation to the dimension of social perception, MANCOVA was performed for competence-warmth, masculinity-femininity and HU-HN as within-group factors as well as leadership style (democratic versus authoritarian) and socio-economic stability and instability contexts as between-group factors ( Table 2). The results replicated the previous results. The democratic leader was perceived as more competent, warm, feminine (less masculine), and more human (HU and HN) than the authoritarian leader. Furthermore, the main effects of socio-economic instability were observed on the social dimensions. Specifically, more competence, warmth, and humanity (HU and HN) were perceived in the stable than unstable condition (no differences were found in the M/F dimensions; Figure 3).
In addition, analysis was conducted to test each leader's effectiveness across scenarios. In the preliminary analysis, a repeated-measures ANOVA with type of conflict as the within-group factor and type of leader as a betweengroup factor revealed that the participants identified the solution in each scenario correctly, namely, competitive, ambiguous solution and cooperative (F(2, 620) = 746.72, p < 0.001,  2 p = 0.71). The competitive scenario was perceived as more in need of a solution based on force (M = 2.41, SD = 1.32) than the ambiguous (M = 3.95, SD = 1.18) and cooperative (M = 4.58, SD = 0.88) scenarios. In addition, the main analysis was performed by conducting MANCOVA and employing the competitive, ambiguous, and cooperative scenarios as within-group factors. The democratic and authoritarian leadership styles and socio-economic stability and instability were designated as between-subject factors ( Table 3). The results revealed that leadership manipulation exerted a main effect on perceived effectiveness for the three scenarios of social conflicts. According to H1 to H3, the democratic leader was perceived as more effective under the cooperative (M = 7.85, SD = 1.78) and ambiguous (M = 6.54, SD = 2.13) scenarios compared with the authoritarian leader (cooperative: M = 3.76, SD = 2.20; ambiguous: M = 4.38, SD = 2.43). On the contrary, the authoritarian leader (M = 7.09, SD = 2.35) was perceived as more effective than the democratic leader (M = 5.89, SD = 2.06) in the competitive scenario. In addition, the results of the analysis indicated that socio-economic instability exerted a main effect on perceived effectiveness of the leaders in addressing the cooperative and ambiguous scenarios with a higher perception of effectiveness under the stability condition (cooperative: M = 6.03, SD = 2.90; ambiguous: M = 5.72, SD = 2.52) than the instability condition (cooperative: M = 5.38, SD = 2.80; ambiguous: M = 5.09, SD = 2.50). However, socio-economic instability did not influence the perceived effectiveness of the leaders in the competitive scenario. Furthermore, no interaction effect was observed between leadership style and socio-economic stability. 1 Finally, we tested if the dimensions of social perception mediated the relationship between leadership style manipulation (authoritarian = 0, democratic = 1) and perceived effectiveness in the cooperative, ambiguous and femininity, human uniqueness, and human nature) for the democratic and authoritarian leaders as a function of (in) stability contexts included in Study 3. competitive scenarios. Recall that the MANCOVA analysis did not demonstrate an interaction effect between socioeconomic instability and leadership style on the perceived effectiveness of the leaders. Thus, socio-economic instability was excluded as a moderator in the mediation model. Instead, we conducted three independent multiple mediational analyses by employing Hayes' (2018) PROCESS (bootstrapping 10,000 interactions with 95% confidence intervals, Model 4; Figure 4). The covariates were included in the model. In the competitive scenario, the authoritarian leader significantly predicted perceived effectiveness in the armed conflict (total effect = −1.20, SE = 0.18, 95% CI

Discussion
The study aimed not only to replicate the findings of Studies 1 and 2 but also to explore the possible moderation effect of socio-economic instability on the previously identified relationship. The results revealed that democratic leaders are perceived as more effective in cooperative and ambiguous scenarios, whereas they are less effective in competitive scenarios compared with authoritarian leaders. Furthermore, the results highlighted that leaders' competence is imperative to be perceived as effective in any situation. Other dimensions of social perception play different roles according to each conflict. Although masculine traits appear crucial in competitive scenarios, greater emotional sensitivity, that is, feminine and human nature traits, seem to be essential in cooperative scenarios. Finally, leaders' cognitive flexibility, such as HN, appears imperative in solving conflicts in ambiguous scenarios.
In addition, the second objective of the study was to explore the extent to which the aforementioned relationship could be shaped by perceptions of socio-economic instability. The results revealed that perception of leaders is influenced by context, specifically, socioeconomic instability, especially in relation to competence, warmth and humanity. Moreover, this perceived contextual instability factor seems to reduce, in general, the leaders' effectiveness without differentiation among leaders.

General Discussion
The present research analysed how the ascribed perceived effectiveness of democratic and authoritarian leaders in different settings, namely, competitive, ambiguous, and cooperative scenarios, are mediated by the most common dimensions of social perception, such as SCM (Fiske et al., 2002, Fiske, 2018, masculinity-femininity (Bem, 1974, Ellemers, 2018 and HU-HN (Haslam, 2006).
Firstly, the results of Studies 1 to 3 are consistent, revealing that democratic leaders score higher on most of the dimensions of social perception except for masculinity, which appears to be a primary trait for authoritarian leaders. Thus, two conclusions are drawn. Firstly, strong and authoritarian leaders are perceived as masculine. This view concurs with the previous literature on gender differences in leadership styles (Gipson et al., 2017). Secondly, democratic leaders are perceived as task-competent and socio-emotional-competent. According to Bass and Bass (2008), this view could explain why democratic leaders are perceived more effective and satisfying than authoritarian leaders (Likert, 1977;Miller & Monge, 1986). Therefore, our results contribute to the existing literature on the manipulation (authoritarian = 0; democratic = 1) and perceived effectiveness (competitive, ambiguous and cooperative contexts), controlling for the subfactors of social dominance orientation, political orientation and participants' socioeconomic status (subjective/objective) in Study 3. The total effects are enclosed in brackets, and the direct effect is adjacent to the total effect. ** p < 0.001, * p < 0.05, † p < 0.08. social perception of leaders (Caprara, Barbaranelli, & Zimbardo, 2002) and shed light on the traits that individuals ascribe to politicians with different leadership styles.
Secondly, the findings revealed that although participants considered democratic leaders more effective in managing social conflicts based on cooperation, they rated the authoritarian leaders as more effective in solve social conflicts that require the use of force. The results concur with previous findings about the importance of relational aspects in leadership to stimulate cooperation (Gartzia & van Knippenberg, 2016) as well as the preference for dominant leaders in competitive scenarios (Laustsen & Petersen, 2017). Democratic leaders are even preferred in ambiguous situations characterised by the lack of clear solutions to resolve the conflict, as demonstrated in Study 3. These results indicated a tendency to prefer democratic over authoritarian leaders (Likert, 1977;Miller & Monge, 1986). However, a possible aspect is that the participants believed that solutions based on cooperation were required for ambiguous situations. Although the use of force and cooperation was explicitly stated as possible in ambiguous situations, the mean scores of the manipulation check exceeded the middle point of the scale (M = 3.95 on a 5-point Likert scale), thus highlighting that participants believed that leaders should cooperate. The results suggested that the type of solutions citizens perceive as imperative across contexts influences the type of leaders they prefer.
Importantly, the results revealed that different traits underlie the perceived effectiveness of democratic and authoritarian leaders in competitive, ambiguous, and cooperative scenarios, thus expanding previous findings (Laustsen & Petersen, 2017). Although certain levels of competence appear relevant in any situation, authoritarian leaders are preferred in competitive situations because they are perceived as aggressive and dominant, that is, masculine. In contrast, in cooperative contexts, people prefer understanding and emotional leaders, that is, feminine and with HN traits. Accordingly, democratic leaders are perceived as more effective. Similarly, Hentschel, Braun, Peus, and Frey (2018) found that transformational leaders, who are similar to democratic leaders, were perceived to be more communal and, thus, more effective leaders than autocratic leaders. Finally, in ambiguous conflicts in which different alternatives can be applied, competence and cognitive flexibility, that is, HN mediate the relationship between leaders and perceived effectiveness.
Altogether, we found that the dimensions of social perception differed in terms of the weight of their role in perceived effectiveness. While dimensions, such as competence, were consistently the main drivers for the perception of a leader as effective in solving all presented social conflicts, other dimensions, such as warmth, seemed irrelevant for all scenarios. Moreover, several dimensions of social perception determine the perceived effectiveness of leaders when the required solution is clear -competitive or cooperative -but not in the presence of ambiguity. In this sense, masculinity leads to the perception of a leader as effective when the conflict requires competition as a solution (Studies 2 and 3) but not when cooperation is necessary (Study 3). Notably, femininity seemed to follow the reverse pattern: ascribing more feminine traits to leaders decreases their perception as effective in a competitive context but enhances it when cooperation is a possible solution. Finally, we also observed that human nature seemed to have more weight when a leader is perceived effective (Studies 2 and 3) in all social conflicts compared with human uniqueness (Study 2 in competitive conflict). In summary, the results indicate that although certain dimensions of social perceptions seem required in the evaluation of the perceived effectiveness of a leader (e.g., competence or humanity), other dimensions are deemed to play an unequal role in certain contexts (e.g., masculinity or femininity). Thus, the current research contributes to the previous literature on the influence of contextual factors on leadership and their outcomes (Oc, 2018;for review). Moreover, the research addressed the role of many of the commonly used measures of social perception in the literature as independent dimensions. Although previous research demonstrated that such dimensions have peculiarities, other authors pointed out that they could be reflective of two big fundamental dimensions (Abele & Wojciszke, 2007). Future studies could investigate this issue in detail by providing evidence related to the commonalities of the dimensions of social perception.
Thirdly, complementary to the primary hypothesis, we addressed the role of a more abstract contextual factor, namely, socio-economic instability. Study 3 revealed that instability decreases perceived effectiveness of democratic and authoritarian leaders. However, we observed that instability did not moderate the effect of leadership style on perceived effectiveness. We recommended that future studies should be conducted to confirm this lack of moderation effect by implementing alternative manipulations of instability. In addition, instability exerted an effect on competence, warmth, and humanity, which supports the notion that leaders are perceived as more task-competent and socio-emotional-competent, that is, more effective in socio-economic stability conditions. Accordingly, the results agree with those of other studies that revealed that contexts of economic decline erode public confidence in politics and institutions (Ervasti, Kouvo, & Venetoklis, 2019;Roth, 2009;Tormos, 2019). Furthermore, we recommend that future studies should explore whether instability favours the emergence of authoritarian rather than democratic leaders that may be considered as the best or only possible solution to a country's problems (Sprong et al., 2019).
Although the present study offers novel results that contribute to an enhanced understanding of the perceived effectiveness of authoritarian and democratic leaders, it has limitations. In relation to the different social conflicts, although armed conflict (competitive scenario) and immigration control (ambiguous scenario) influenced the citizens of Mamziba, the humanitarian crisis (cooperative scenario) occurred in another country and did not directly affect the citizens of Mamziba. Future studies may examine this question and analyse the effect of the group affected by the social conflict on the perceived effectiveness of both types of leaders. That is, distinguishing whether the social conflict affects the ingroup or, conversely, the outgroup. Likewise, several possible factors that may influence relative to the content of the used social conflicts can be examined in depth. For instance, we did not consider the types of conflict resolution (i.e., competitive, cooperative or ambiguous) as independent from the type of scenario (i.e., armed conflict, humanitarian crisis or immigration arrival). Further research may explore the possible influence of the strategy to solve problems under the same scenario on a leader's perceived effectiveness as well as compare resolution strategies across scenarios not only on the perceived effectiveness but also on the social perception of leaders.
Moreover, we included short text descriptions in the leadership manipulations to avoid a possible variability in the interpretation of the participants when presented with the label ' democratic' or ' authoritarian'. Future studies can implement real scenarios or examples of these leadership styles to increase ecological validity and reinforce the findings identified across the studies in the current paper. Additionally, future studies may benefit from natural scenarios or a natural manipulation of socio-economic stability to accomplish the result we obtained with the fictitious scenarios. In addition, we note the cognitive burden of the participants, especially in Study 3, as a possible limitation of the research. Participants had to (a) read about the socio-economic situation of a society, (b) read about the leadership style of its political leader, (c) evaluate his/her through several traits, and (d) rate the effectiveness of the leader in three scenarios of social conflicts. Thus, future studies may employ another design that will alleviate the cognitive burden of participants, which can influence the results. Lastly, we carried out the studies with the students and staff of the university (i.e., professors). Future studies can reinforce the current conclusions by exploring the generalisation of the findings to a diverse sample that includes participants from other backgrounds (e.g., non-educated groups and people from deprived backgrounds).
Finally, the study provides various practical implications. The manner in which the media and politicians refer to social problems could shape the citizens' impressions on how problems should be solved and what types of leaders are most suitable in this regard. For example, as cited by Sabucedo, Alzate, and Hur (2020), employing a metaphor of war may be inadvisable in the current COVID-19 crisis.
In accordance with the current results, promoting a perception of a crisis as a competitive situation including a war that should be won and competition between countries to obtain medical supplies rather than appealing for cooperation and solidarity may influence citizens to perceive democratic leaders as less capable of managing the crisis. Thus, we recommend that future research should explore whether people's perceptions in relation to competition and cooperation about crises, such as the COVID-19 pandemic and economic crises, determine preferences for democratic or authoritarian political leaders.
The results revealed that different dimensions of social perception underlie the perceived effectiveness of political leaders with different leadership styles, namely, democratic and authoritarian, in solving different sociopolitical conflicts. Given the current political situation in western countries characterised by the loss of popularity of existing democratic leaders in favour of new authoritarian leaders, these effects provide an enhanced understanding of the manner in which they are perceived and reveal that a democratic style is perceived as more effective in solving conflicts unless competition is the only strategy that can be employed to solve problems.

Notes
1 See supplementary materials for a detailed revision of the manipulations or scenarios, complementary information (descriptive analysis and bivariate correlations) and alternative analysis. Covariates do not exert a consistent effect on the results and do not reverse the main effects. Thus, this information has been omitted from the main text and can be found in the supplementary materials: https://osf.io/64wbc/. 2 We conducted the main analyses and excluded participants with incorrect answers to the attention checks. We observed that the main results did not drastically change, and the significant effects remained significant. Thus, we retained the participants in all reported analyses.