The Role of Trust and Attitudes toward Democracy in the Dissemination of Disinformation—a Comparative Analysis of Six Democracies

Abstract

Trust lies at the heart of the disinformation crisis, as citizens must decide which narratives to follow and whether to accept “alternative truths.” Therefore, trust in institutions that publish reliable information can act as a shield against disinformation. This comparative study investigates the role of trust in news media and political actors and general attitudes toward democracy in the willingness to spread disinformation (i.e., likes, shares, or comments). Findings of this study show that news media trust plays a subordinate role but that trust in social media news has a strong relationship with willingness to spread disinformation. In 2020, citizens in the United States and United Kingdom who had high trust in their governments were more willing to spread disinformation, whereas in France and Belgium, citizens who trusted opposition leaders were more likely to do so. Moreover, citizens who were satisfied with democracy appeared to be less vulnerable to disinformation, with the exception of those in the United States. Therefore, political actors bear great responsibility for the current (dis)information crisis because they can exploit citizens’ trust to their advantage.

Keywords:

• Trust
• media
• social media
• disinformation
• politics

Introduction

With the advent of social media, the spread of disinformation has emerged as a major societal concern. Disinformation is generally understood as false or misleading information that is intentionally spread to destabilize political institutions and delegitimize media organizations by undermining trust in these information sources (Ognyanova et al. 2020). This can lead to a decline in media consumption, as people turn to alternative sources of information, such as social media or partisan outlets, which can further contribute to the spread of disinformation. In previous research, the term misinformation has been frequently used because of the difficulty of determining whether false information was spread intentionally or unintentionally (Hameleers et al. 2021; Morosoli et al. 2022). However, while both misinformation and disinformation are problematic, disinformation is more concerning because of its intentional nature and potential to cause harm. Disinformation during the coronavirus disease (COVID-19) pandemic lowered the compliance with public health guidelines such as mask wearing and social distancing (Hameleers, van der Meer, and Brosius 2020). Disinformation competes with more credible news sources such as news media, government officials, or scientists. By contrast, trust in these institutions could build resilience to disinformation by limiting the exposure to and sharing of it (Humprecht, Esser, and Van Aelst 2020, Humprecht et al. 2021). Disinformation on social media is particularly problematic because it is massively disseminated and thus gains high visibility (Metzger et al. 2021). Furthermore, political beliefs are an important driver of the spread of disinformation (Van Bavel et al. 2021). Against this background, the question arises as to what role democratic attitudes and trust in politics and the news media play in the further spread of disinformation.

This study investigates the willingness to react to and spread disinformation (like, share, or comment), attitudes toward democracy, and trust in news media and political actors by using survey data collected in 2020 (n = 7,006) from the United States, United Kingdom, France, Germany, Belgium, and Switzerland. This paper makes several contributions to the literature by examining the relationship between trust, democratic attitudes, and willingness to spread disinformation in a cross-national context. The underlying research interest is examining how different political and media environments shape individual behaviors and attitudes during the pandemic. Finally, the normative value of trust in news media and politics in democratic societies is discussed, as well as the responsibility of political actors, as disinformation often comes from political elites. The findings of this study illuminate the relationship between willingness to spread disinformation and patterns of democratic attitudes and institutional trust, which reveal the long-term and global implications of disinformation. Understanding these implications is important in times of crisis because of the salience of ideological debates and the high stakes involved in overcoming the crisis. Research on the relationship between disinformation and trust has focused primarily on two perspectives on trust: news media and politics. These two perspectives will be discussed in the next section.

Disinformation and Trust

Trust in the News Media

The first perspective emphasizes that the use of high-quality news media makes citizens less susceptible to disinformation because they recognize it more quickly (Hopp, Ferrucci, and Vargo 2020; Stier et al. 2020; Valenzuela, Halpern, and Araneda 2022). Citizens may also have less contact with disinformation because they are more likely to seek information from trusted sources (Humprecht et al. 2021; Sterrett et al. 2019). Disinformation that contradicts credible news sources may raise doubts about their accuracy and undermine trust in them (Stubenvoll, Heiss, and Matthes 2021; Valenzuela, Halpern, and Araneda 2022). Alternative news outlets capitalize on distrust by promising to provide information that has been “concealed” by official sources (Holt, Ustad Figenschou, and Frischlich 2019). Disinformation often targets legacy news media sources with accusations that they disseminate biased or inaccurate information (Hameleers 2020; Marwick and Partin 2022). Research has claimed that exposure to this false information can impact attitudes and behaviors (Hameleers, van der Meer, and Brosius 2020; Loomba et al. 2021). Overall, it has been argued that decreasing trust in news can lead citizens to turn to dubious content (e.g., alternative media that publish disinformation). This development weakens the established news media because of loss of revenue, creating a self-reinforcing cycle due to the fewer resources to fact check. Media-skeptical citizens turn to content that reinforces their misperceptions and become increasingly difficult to reach with factual information (Frischlich and Humprecht 2021). This cycle is pronounced particularly in countries where trust in the media is generally comparatively low, such as the United States (Humprecht et al. 2021; Newman et al. 2021). Therefore, trust in the media has been identified as an important factor in resilience to disinformation (Humprecht, Esser, and Van Aelst 2020). News media outlets that enjoy higher levels of trust are often public media. Research has shown that in countries with strong public media, the population’s knowledge of socially relevant topics is greater than in other countries (Aalberg and Curran 2012).

On the basis of these considerations, it is hypothesized that individuals with higher trust in news media are less willing to like, share, or comment on disinformation (H1a) and assumed that the negative effect of media trust occurs mainly for news media that the respondents themselves use regularly (H1b). Trust in individual actors is more pronounced than trust in institutions (Strömbäck et al. 2020). Therefore, it is postulated that individuals with high trust in journalists are less likely to spread disinformation (H1c). Finally, disinformation is mainly disseminated via social media, so it is assumed that trust in news on social media has a positive relationship with users’ reactions to disinformation because trusting users are likely to be less critical of social media content (H1d).

Trust in Politics

Trust in the news media is closely linked to trust in political institutions. Hanitzsch, Van Dalen, and Steindl (2018) introduced the so-called trust nexus, which states that the link between trust in the media and trust in politics is particularly strong in politically polarized societies. Ognyanova et al. (2020) found that in the United States, this ideology plays a significant role in the effect of disinformation on political trust. Among conservatives and moderates, disinformation increased political trust, whereas among liberals, it decreased political trust. The authors argued that both low and blind political trust in political actors can be harmful. If citizens have an unrealistically optimistic view of the government, this can impact their willingness to mobilize and participate (Ognyanova et al. 2020). In less polarized societies, on the other hand, political trust can have a protective effect against disinformation. Zimmermann and Kohring (2020) observed that in Germany, political trust decreases the likelihood of believing in disinformation, but disinformation does not affect political trust. Their findings illustrate the need to consider the political situation in each country. This particularly applies to sharing disinformation because it is often ideologically motivated (Osmundsen et al. 2021; Shin and Thorson 2017). In summary, distrust toward democratic institutions and actors, particularly in the extreme form of a general conspiracy mentality (Halpern et al. 2019), can make individuals more susceptible to disinformation (Stier et al. 2020).

Finally, the COVID-19 pandemic represents a special case for the study of disinformation and political trust. On one hand, disinformation was disseminated to a high degree, and on the other hand, there was a great need for trust in societal institutions (Frischlich and Humprecht 2021; Ognyanova et al. 2020). Citizens must trust governments to take public health measures to reduce the spread of the virus (Petersen et al. 2021) and the news media to provide up-to-date and accurate information about the spread of the virus and its consequences (Lewandowsky et al. 2021). Furthermore, disinformation may have different relationships with media and political trust, depending on the political ideology (Ognyanova et al. 2020). Populist leaders can fuel distrust in the news media and the political elite (Hameleers 2020), undermining their pandemic advice (Petersen et al. 2021). Hence, the relationship between trust in political actors and disinformation sharing can be twofold, depending on whether populist actors constitute the government or are part of the opposition. It can be assumed that during the crisis, traditional democratic politicians were primarily interested in containing the pandemic for the benefit of the people, whereas populists might have pursued self-interest and criticize the elites. Depending on whether populists or traditional politicians control the government in each respective country, I assumed that both trust in government politicians (H2a) and trust in opposition politicians (H2b) are negatively related to the spread of disinformation.

General Democratic Attitudes

Trust in politics has been found to be closely related to general attitudes toward politics and democracy (Bromme and Rothmund 2021; van der Meer and Hakhverdian 2017). Research has argued that low levels of trust in politics and news media stem from a general political malaise, particularly in polarized societies (Frischlich and Humprecht 2021). van der Meer and Hakhverdian (2017) indicated that political trust is closely related to satisfaction with democracy and confidence in political institutions. The relationship between satisfaction with democracy and willingness to spread disinformation is still underexplored, but research has indicated a relationship between certain motives and susceptibility to disinformation (Sternisko, Cichocka, and Van Bavel 2020). Furthermore, populist attitudes are often associated with an antidemocratic worldview (Dekeyser and Roose, 2021), and individuals with populist attitudes are also more likely to believe in disinformation (Hameleers 2020; Stecula and Pickup 2021). Petersen, Osmundsen, and Arceneaux (2018) found a connection between the “need for chaos” and the spread of political disinformation. Such extreme motivations include a desire to “burn down” the entire democratic sphere. According to the authors, the underlying intent is to harm unpopular elites who play a central role in most democracies. Similarly, individuals who state that they hate their political opponents are more likely to share disinformation on social media (Petersen, Osmundsen, and Arceneaux 2018). Even though this type of extreme partisanship does not directly translate to dissatisfaction with democracy, it can be interpreted as a sign of a general disregard for democratic values. Thus, I assumed that lower levels of satisfaction with democracy (H3a) and lower perceived importance of living in a democracy (H3b) are positively related to the willingness to spread disinformation.

Comparative research has shown that trust in democratic institutions, including the media, and democratic attitudes depend strongly on a country’s current political situation. Given that political disinformation is often shared out of partisan political beliefs, it is likely that such country-based differences are relevant to the spread of disinformation and the relationship between trust and democratic attitudes. Moreover, I am interested in whether previous findings about ideology, disinformation, and trust in news media and politics have been replicated elsewhere or are distinctive to the highly polarized U.S.-American context. Therefore, I examined country differences in willingness to disseminate disinformation (1), trust in legacy news media and journalists (2), trust in politicians in the government and opposition (3), trust in news on social media (4), and democratic attitudes (5).

In summary, trust in news media and politics and democratic attitudes are closely related. Moreover, media and political trust are related to belief in disinformation (Zimmermann and Kohring 2020), but findings on trust and sharing disinformation are mixed (Halpern et al. 2019; Valenzuela, Halpern, and Araneda 2022). Finally, research has suggested that those who deliberately spread disinformation are more likely to hold antidemocratic attitudes. However, little research has been done on how these relationships differ between countries. This study filled this research gap by examining both trust in various actors and institutions and trust in democratic attitudes related to the spread of disinformation across six countries.

Methods

Sample

The survey was conducted by Respondi, the polling company contracted to survey representative population samples in six countries using equivalent methods (Belgium [Flanders] = 1,063, France = 1,255, Germany = 1,019, Switzerland = 1,251, the United Kingdom = 1,380, and the United States = 1,038). These countries are characterized by different political and media systems (Brüggemann et al. 2014; Hallin and Mancini 2004; Lijphart 2012). As Table 1 shows, the countries also differ systematically in terms of trust in news media and politics. In the United Kingdom, the United States, and France, general trust in the news media is lower, whereas in Switzerland and Germany, trust in the government is comparatively high. The United States also differ from other countries in terms of higher levels of polarization of society, prevalence of populist communication, and marginalization of public service broadcasting.

Table 1. Sample Description.

	Belgium	France	Germany	Switzerland	United Kingdom	United States	Source
Population	11.6 m	65.5 m	82.4m	8.6m	67.0 m	331m	Digital News Report Newman et al. (2022)
Internet penetration	94%	92%	96%	94%	95%	90%	Digital News Report Newman et al. (2022)
Political organizations	Party centered	Candidate centered	Party centered	Party centered	Party centered	Candidate centered	Lijphart (2012)
Mass media system	Democratic-corporatist	Polarized-pluralist	Democratic-corporatist	Democratic-corporatist	Liberal	Liberal	Hallin and Mancini (2004)
Trust in news media	51%	29%	50%	46%	34%	26%	Digital News Report Newman et al. (2022)
Trust in government	32%	31%	43%	69%	30%	32%	Brezzi et al. (2021)
Prevalence of populist communication (index)	1.06	0.02	0.60	−0.29	0.72	−1.20	Humprecht, Esser, and Van Aelst (2020)*
Polarization of society (index)	1.12	−0.89	1.81	−0.32	−0.43	−0.58	Coppedge et al. (2022)*
Share of PBS and public revenue (index)	−0.42	−0.02	1.37	0.44	0.89	−2.80	Brüggemann et al. (2014)

Note: *Higher values indicate lower polarization/populist communication and vice versa.

When investigating the relationships between the independent and dependent variables (willingness to like, share, or comment on disinformation), I used country comparisons as a robustness check to determine which relationships hold in different political information environments (Esser et al. 2012).

The survey was conducted in April and May 2020, when lockdowns were imposed in most countries included in this study and pandemic-related disinformation was being widely disseminated (Brennen et al. 2020). Respondi drew quota samples from its online access samples according to gender, age, and education. As the study focuses on the spread of disinformation on social media, only participants who used social media (Twitter, Facebook, Instagram, or YouTube) at least once a month were included. The participants used Facebook most often, with daily usage rates ranging from 72.8% in France to 48.4% in Switzerland. Twitter was used least frequently, with 26.2% of the participants from the United States using the platform daily and 5.3% of the respondents from Switzerland (see Table 2).

Table 2. Country Means of Main Dependent and Independent Variables.

	Belgium		France		Germany		Switzerland		United Kingdom		United States
	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	F
Disinformation index	2.39^c	1.44	2.39^c	1.43c	2.18^b	1.47	1.89^a	1.22	2.17^b	1.41	2.75^d	1.67	F(5,7000) = 42.433, p < .001
General trust in news media	4.59^d	1.40	3.80^a	1.43	4.50^d	1.56	4.30^c	1.45	4.20^bc	1.52	4.07^b	1.68	F(5,6999) = 42.555, p < .001
Trust in self-used news	4.86^c	1.30	4.00^a	1.43	4.89^c	1.45	4.75^c	1.36	4.53^b	1.44	4.49^b	1.57	F(5,6998) = 64.064, p < .001
Trust in news in social media	3.28^bc	1.56	3.14^ab	1.48	3.37^c	1.60	3.21^bc	1.48	2.99^a	1.58	3.30^bc	1.73	F(5,6999) = 9.350, p < .001
Trust in journalists	4.25^c	1.45	3.59^d	1.57	4.21^c	1.53	4.00^b	1.45	3.35^a	1.55	3.95^b	1.71	F(5,7000) = 64.268, p < .001
Trust in politicians in government	3.24^b	1.62	2.72^a	1.63	3.79^d	1.71	4.07^e	1.53	3.46^c	1.64	3.12^b	1.61	F(5,7000) = 106.224, p < .001
Trust in politicians of the opposition	3.36^c	1.57	2.79^a	1.51	3.73^d	1.63	3.77^d	1.42	3.15^b	1.59	2.78^a	1.64	F(5,7000) = 90.353, p < .001
Satisfaction with democracy	3.87^a	1.65	3.77^a	1.73	4.56^c	1.69	5.31^d	1.35	4.60^c	1.46	4.35^b	1.58	F(5,6999) = 154.121, p < .001
Importance of democracy	5.52^ab	1.50	5.36^a	1.64	6.08^d	1.34	6.23^d	1.15	5.85^c	1.31	5.55^b		F(5,7000) = 71.338, p < .001
N		1063		1255		1019		1251		1380		1037	7005

Note: Univariate ANOVA at the national level. Means with the same superscript letter within a row are not statistically significantly different (based on post hoc Bonferroni’s test, p = .05).

In the first step, the participants completed a section that included standard demographics and several measures for news media trust, political trust, and democratic attitudes. In the second step, each participant was presented with three news articles that included false claims. To make the user experience as realistic as possible, the articles were presented as social media posts by a fictional source (“news.com”). The false claims were taken from fact-checking websites in the countries under study and manipulated to match the research interests. The article on COVID-19 claimed that it was a bioweapon. The article on immigration claimed that the legacy media concealed outbreaks of violence in refugee camps. The third article, which was on climate change, claimed that environmentalists left trash in public parks after their protests. After reading the posts, the participants were asked about their willingness to like, share, or comment on the posts. The last step was debriefing. The mean time to complete the survey was 18.6 min (SD = 7.09). After removing careless respondents identified on the basis of a quality failure question, their response times, and response patterns, an adjusted sample of 7,006 respondents was obtained.

Measures

The dependent variable was the willingness to further spread disinformation on social media (e.g., in the form of liking, sharing, or commenting). Owing to the algorithms used by social media platforms such as Facebook, these reactions led to the post having a higher probability of being displayed in other users’ news feeds, thus increasing its reach.

Willingness to Spread Disinformation

Willingness to spread disinformation was measured using three different items. On a 7-point Likert scale (1 = very unlikely, 7 = very likely), the participants indicated how they would usually react to the manipulated post: “like the post,” “share the post,” or “leave a comment.” The three reactions highly correlated (Cronbach’s α = .90), so they were combined into a mean index (M = 2.28, SD = 1.46). Willingness to comment could include both positive and negative comments. Negative comments were also considered as dissemination, as any form of interaction with such content contributes to greater visibility because of the algorithms used by social media platforms.

Research has established that trust is a complex concept that requires complex measurement (Strömbäck et al. 2020). For example, citizens appeared to have a deeper trust in individuals than in institutions, even if they belong to these institutions (Frischlich and Humprecht 2021; Holmberg, Lindberg, and Svensson 2017). To account for these differences, I draw on a range of measures of trust across actors and institutions and of attitudes toward democracy.

General Trust in News Media

I used established scales to measure news trust on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree) and asked the respondents whether they thought they could trust most news in their country most of the time (M = 4.28; SD = 1.53).

Trust in Self-Used News Sources

The participants were asked whether they thought they could trust most of the news they consumed most of the time (1 = strongly disagree, 7 = strongly agree; M = 4.47, SD = 1.46).

Trust in News on Social Media

The participants indicated their level of agreement with the following statement: “I think I can trust news in social media most of the time” (1 = strongly disagree, 7 = strongly agree; M = 3.20, SD = 1.57).

Trust in Journalists

To explore trust in individual actors, the participants were asked, “How much do you trust journalists?” (1 = I don’t trust at all, 7 = I fully trust; M = 3.86, SD = 1.58).

Trust in Government Politicians

The participants were asked, “How much do you trust politicians in government?” (1 = I don’t trust at all, 7 = I fully trust; M = 3.40; SD = 1.69).

Trust in Opposition Politicians

The participants were asked, “How much do you trust politicians of the opposition?” (1 = I don’t trust at all, 7 = I fully trust, M = 3.226; SD = 1.61).

Satisfaction with Democracy

The survey included the following question: “Overall, how satisfied are you with the way democracy works in your country?” (1 = not satisfied at all, 7 = fully satisfied; M = 4.43; SD = 1.66).

Importance of Democracy

To measure how important democratic values are for individuals, the participants were asked, “Overall, how important do you think it is to live in a country governed democratically?” (1 = not important at all, 7 = very important; M = 5.77; SD = 1.44).

Findings

I begin by analyzing country differences in the main dependent and independent variables. Analyses of variance (ANOVAs) revealed large country variations (see Table 2). Swiss citizens were least likely to react to disinformation (M = 1.89, SD = 1.22), and U.S. citizens showed the highest levels of willingness to like, share, or comment on it (M = 2.75, SD = 1.67). Regarding the trust indicators, a uniform pattern emerged: In Belgium, Germany, and Switzerland, I found generally higher trust levels in news media and political actors than in France, the United Kingdom, and the United States. Satisfaction with democracy was highest in Switzerland (M = 5.31, SD = 1.35) and lowest in Belgium (M = 3.87, SD = 1.65) and France (M = 3.77, SD = 1.73). Living in a democracy was rated as highly important in all countries; it was most important in Switzerland (M = 6.23, SD = 1.15) and Germany (M = 6.08, SD = 1.34) and least important in France (M = 5.85, SD = 1.64) and Belgium (M = 5.52, SD = 1.50).

To analyze the relationships between the dependent variable (willingness to like, share, or comment on disinformation) and the independent variables (trust and democratic attitudes), I conducted ordinary least squares (OLS) regressions with cluster-robust standard errors for each country. Table 3 presents the results of the various models for all six countries.

Table 3. Trust and Attitudes Predicting the Willingness to Spread Disinformation (Like, Share, or Comment).

	Belgium		France		Germany		Switzerland		United Kingdom		United States
	SE (B)	Beta	SE (B)	Beta	SE (B)	Beta	SE (B)	Beta	SE (B)	Beta	SE (B)	Beta
Constant	.21		.17		.24		.21		.18		.22
General news trust	.04	−.10*	.05	−.04	.05	−.03	.04	.03	.04	−.04	.05	−.02
Trust in news used	.05	−.003	.05	.07	.05	−.07	.04	−.02	.04	−.15	.04	.03
Trust in news on social media	.03	.41***	.03	.24***	.03	.37***	.03	.26***	.03	.46***	.32	.32***
Trust in journalists	.04	.07	.03	−.06	.04	.01	.03	.01	.03	.06	.04	−.09
Trust in government politicians	.04	−.04	.04	−.11*	.05	−.02	.04	−.10*	.03	.17***	.04	.17***
Trust in opposition politicians	.04	.12**	.04	.16***	.05	.10	.04	.07	.03	−.07*	.04	.11
Satisfaction with democracy	.03	−.08*	.03	−.06	.04	−.14***	.03	−.10**	.03	−.02	.03	.11***
Importance of democracy	.03	.04	.03	−.06*	.04	.00	.03	−.11***	.03	.01	.03	.03
Adjusted R^2		.18		.09		.14		.11		.20		.24

Note: N = 7006. OLS regression models at country level. The cells include robust standard errors (clustered on respondents) and betas. *p < .05; **p < .01; and ***p < .001.

Hypotheses H1a–c postulate a negative relationship between trust in news media and willingness to spread disinformation. To account for the complexity of the concept of media trust, four different measures were used: general trust in news media, trust in news media that the respondents used regularly, trust in news on social media, and trust in journalists. The results showed that general trust in news media was not related to willingness to spread disinformation, except for Belgium, where a negative relationship was found, as expected (β = −.12, p ⩽ .01). Hence, H1a was only partially accepted.

For trust in self-used news outlets, the analysis revealed a weak but positive effect for France (β = .11, p ⩽ .05) but no effects for other countries. Again, these results did not correspond to the theoretical assumptions, so H1b was partially rejected. Similarly, trust in journalists did not yield any significant results, so H1c was rejected.

By contrast, trust in news on social media was positively associated with the willingness to like, share, or comment on disinformation in all countries under study (Belgium: β = .35, p ⩽ .001; France: β = .18, p ⩽ .001; Germany: β = .131, p ⩽ .001; Switzerland: β = .21, p ⩽ .001; United Kingdom: β = .40, p ⩽ .001; United States: β = .28, p ⩽ .001). On the basis of this finding, H1d was accepted.

H2a and b addressed the relationship between trust in politics and spreading disinformation. Two measures were used to examine trust in politics, one for politicians of the government and the other for politicians of the opposition. The results showed a differentiated picture: negative relationships between trust in government politicians and disinformation for France (β = −.09, p ⩽ .05) and Switzerland (β = −.11, p ⩽ .05) but positive relationships for the United Kingdom (β = .18, p ⩽ .001) and the United States (β = .16, p ⩽ .001). Similarly, I found mixed results for trust in opposition politicians. The analysis revealed positive relationships for Belgium (β = .13, p ⩽ .001) and France (β = .13, p ⩽ .001) but no relationships for the remaining countries. H2a and H2b were partially accepted.

H3a and H3b postulated a negative relationship between spreading disinformation and democratic attitudes. For satisfaction with democracy, this relationship could again only be established for certain countries, namely Belgium (β = −.10, p ⩽ .01) and Germany (β = −.12, p ⩽ .01). For the United States, a positive relationship was found, contrary to the theoretical assumptions (β = .11, p ⩽ .001). Therefore, H3a was only partially accepted. For the importance that the respondents attached to democracy, an effect was found only for Switzerland (β = −.08, p ⩽ .05), so H3b was also only partially accepted.

In summary, news media trust was not related to the spread of disinformation, except for Belgium. However, highly significant effects of trust in news on social media on the willingness to share, like, or comment on disinformation were found in all countries studied. Trust in politicians who were members of the government at the time of this study was weakly negatively related to the outcome in France and Switzerland but strongly positively related to those in the United Kingdom and United States. By contrast, trust in the opposition during the study period was positively related to willingness to further disseminate disinformation in Belgium and France and negatively related to the same in the United Kingdom. Democratic attitudes such as satisfaction and importance showed negative correlations with the outcome in several countries, except in the United States, where a positive effect was found for satisfaction.

Conclusion

The spread of disinformation gained momentum during the COVID-19 pandemic, with partisan groups strongly attached to alternative theories explaining the pandemic via anti-elitist narratives and conspiracy theories (Newman et al. 2021). Disinformation operations strategically manipulate an environment that perfectly exploits the affordances that digital media environments offer for intentional deception (Howard and Bradshaw 2018). To expand knowledge about why disinformation spreads so rapidly, this study examined how democratic attitudes and trust in news media and politics increase citizens’ willingness to interact with disinformation. Given the prevalence of disinformation on social platforms (Brennen et al. 2020), where citizens function as drivers of disinformation campaigns (Buchanan and Benson 2019), intentions to like, share, or comment on false articles were examined from a cross-national perspective.

Although trust in the media is not associated with reactions to disinformation, citizens with high trust in news on social media were most willing to like, share, or comment on disinformation across all countries. News on social media, like other news, are diverse and include content from reputable actors (e.g., professional news media) and dubious sources (e.g., alternative media). Thus, trust in such news does not mean that individuals are only exposed to disinformation and blindly believe it. However, previous research has shown that trust in news on social media is associated with willingness to spread disinformation (Sterrett et al. 2019). Individuals who trust social media content appeared to be less critical and may be more likely to share content, whether true or false, thoughtlessly. This result suggests that competencies such as digital news literacy are fundamental to combat the massive spread of disinformation. Research on news literacy has shown that citizens who are more literate and knowledgeable about media structures see and share less content online and are more skeptical of information quality on social media (Vraga and Tully 2021). When messages on social media are unquestioningly trusted, falsehoods can spread more easily.

Researchers have argued that media trust is a complex concept that must be studied at different levels (Engelke, Hase, and Wintterlin 2019; Strömbäck et al. 2020). Hence, this study distinguished between trust in news media in general, the news media used by individuals themselves, and trust in journalists. Trust arguably lies at the heart of the “disinformation order” (Wardle and Derakhshan 2017) because it is ultimately about believing facts or “alternative truths” (Frischlich and Humprecht 2021). Disinformation campaigns work collaboratively, as their support for such messages might affect the willingness to share them with social contacts (Allington et al. 2021; Chadwick and Vaccari 2019). The goal of this study was to explore this relationship, but it is unclear what respondents mean by media trust or what role it plays in their evaluation of (dis)information in social media. More in-depth qualitative research is needed to gain deeper insights.

Another important finding of this study is that trust in the news media plays a less important role than trust in politicians and general attitudes toward democracy in the context of spreading disinformation. This finding can be interpreted as implying that political attitudes and ideologies are more important than factual knowledge in the context of disinformation dissemination. However, because news media use is about “identity, not information” (Kreiss 2018), media trust remains an important component in combating disinformation. When individuals lose their trust in legacy media and turn to other news sources such as partisan or alternative media, their exposure to disinformation is likely to increase (Frischlich and Humprecht 2021; Müller and Schulz 2021). If disinformation is consistent with their prior beliefs and increases distrust of other social groups (Rathje, Bavel, and Linden 2021), elites, and/or legacy news media (Zimmermann and Kohring 2020), users become more likely to spread it (Van Bavel et al. 2021).

The findings of this study regarding the role of political trust are in line with those of studies that showed that the willingness to spread disinformation is primarily due to ideology and partisan motivations (Shin and Thorson 2017; Van Bavel et al. 2021; Weeks 2015). The analysis revealed that the relationship between trust in politicians and willingness to share disinformation is complex. In the United Kingdom and United States, citizens who had high trust in the (conservative) government at the time of the study were more willing to disseminate disinformation, whereas those in France and Belgium who had high trust in opposition politicians were more likely to do so. It can be assumed that this finding is related to the topics of the social media posts used in this study. Previous research has shown that topics such as migration and climate change are particularly targeted by right-wing populist politicians (Blassnig, Rodi, and Tenenboim-Weinblatt 2019; Ernst et al. 2019). Moreover, conservative voters have been found to be more willing to share disinformation (Guess, Nagler, and Tucker 2019; Osmundsen et al. 2021). This is also likely the case in countries where conservative governments were in power at the time of the study, such as the United Kingdom and United States, where individuals who are willing to spread disinformation also have higher trust in the government. Thus, sharing disinformation can be considered as a form of political participation driven by partisan bias (Osmundsen et al. 2021). However, such a form of “dark” participation (Quandt 2018) can lead to the polarization of public opinion, the spread of conspiracy theories, and the erosion of social cohesion, which can harm democratic institutions and the broader social fabric.

The finding that low satisfaction with democracy is related to willingness to disseminate disinformation can also be interpreted in this context. Rejection of democratic values can lead to a lack of trust in traditional sources of information—a void that is potentially filled by alternative sources of information that are more susceptible to disinformation. This can create a vicious cycle that undermines democracy and citizens’ ability to make informed decisions.

The different political situations in the countries studied could also partly explain the results on media trust: some legacy news media in these countries convey the political messages of certain political camps. If the respective political actors disseminate misleading or even false information, the news media may (sometimes unintentionally) contribute to the spread of deceptive narratives. An example might be U.S. Fox News’s coverage, which both is critical of immigration and downplays human responsibility for climate change (Hoewe et al. 2020).

That people tend to process information in an ideologically motivated manner has already been established in selective exposure literature (Knobloch-Westerwick et al. 2021) but may be especially problematic in the context of disinformation spreading. Producers of disinformation exploit confirmation biases and purposefully target responsive audiences on social media (Weeks and Gil de Zúñiga 2021). Research has consistently shown that misperceptions are especially persistent when they align with prior beliefs (Hameleers and van der Meer 2020; Lewandowsky et al. 2012; Thorson 2016). Returning to the overarching research interest, individuals who have trust in certain political actors appear to be more inclined to spread disinformation that they assume is in line with the views and values of these political actors. Taken at the societal level, this finding underscores the role of influential political actors in different countries. By fomenting uncertainty or even disseminating misleading information, politicians can (un)intentionally contribute to the spread of disinformation. Especially in times of crisis such as during pandemics or military conflicts, these actors bear a particularly huge responsibility.

Several limitations compromise the results of this study. First, the estimates varied across countries, which indicates that the results are unlikely to be transferable to other nations. Information environments offer unique opportunity structures for the dissemination of disinformation (Humprecht, Esser, and Van Aelst 2020). This study was limited to six Western democracies, and more research is needed beyond the Global North on the mechanisms that lead to spreading disinformation. Second, this study focused on the relationships between trust in media, social media, and politicians and willingness to spread disinformation. Other factors that play roles in this context, such as the general willingness to share content on social media, were not included, as they have already been examined in previous studies (Humprecht et al. 2021; Yu et al. 2022). Third, the participants were shown disinformation articles on polarizing topics such as COVID-19, immigration, and climate change, which can trigger strong emotions in the respondents (Weeks and Gil de Zúñiga 2021). As emotional and cognitive responses are linked (Ryan and Krupnikov 2021), future research should replicate the study with less polarizing topics.

This study is in line with a previous work that showed the prevalence of disinformation among a group of citizens (Hopp, Ferrucci, and Vargo 2020). Although the results highlight the central role of individual differences, the country differences show how important political information environments are for vulnerability to deception that can lead to the collective disruption of society when systematically exploited by disinformation agents. Disinformation can erode trust in the democratic process, as people lose confidence in the fairness and transparency of elections, government policies, and public discourse. This can lead to political polarization, social unrest, and the breakdown of democratic norms and values. Digital journalism studies can help identify effective strategies for combating disinformation and restoring trust in news media and political institutions. This includes developing tools and techniques for fact-checking, increasing transparency in news production and distribution, promoting news literacy and critical thinking skills, and fostering a more diverse and inclusive media landscape that promotes an informed, engaged, and democratic society.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study is part of a project funded by the Swiss National Science Foundation (Grant No. 100017L_182253).