Sage Journals: Discover world-class research

Abstract

Although a growing body of scholarship examines who believes conspiracy theories (CTs) and why, less is known about why people share CTs. We test the impact of three independent motives on people’s willingness to share CTs on social media: bolstering their or their group’s beliefs (motivated sharing), generating collective action against their political outgroup because of losing (sounding the alarm), and mobilizing others against the political system (need for chaos). Using an original survey of US adults (N = 3336), we test these three motives together and find strong evidence for motivated sharing and need for chaos, but no evidence for sounding the alarm. Our findings suggest that motivated sharing—when measured directly as belief in the CTs—is the strongest predictor of willingness to share CTs on social media. Need for chaos has less of an effect on sharing than belief but a consistently stronger effect on sharing than partisanship and ideology. Altogether, we demonstrate that sharing CTs on social media can serve both motivated and mobilizing functions, particularly for those who believe the CTs or seek to challenge the political system, rather than impugn their political rivals.

Keywords

Conspiracy theories motivated reasoning social media online sharing need for chaos misinformation mobilization

Introduction

A growing body of literature suggests that the social and political consequences from the spread of conspiracy theories (henceforth, CTs) are profound (e.g., Allcott and Gentzkow, 2017; Douglas, 2021; Farhart et al., 2022). While the study of who believes CTs and why burgeons across the social sciences (see Douglas et al., 2019 for a comprehensive review), evidence suggests that exposure to a CT is a necessary condition for belief: One cannot believe and potentially act on the belief that 9/11 was an inside job, for example, unless they encounter the theory (see e.g., Berinsky, 2017; Pennycook et al., 2018).

A substantial literature on selective exposure explores the conditions under which people are exposed (or not exposed) to attitude/identity consistent (or inconsistent) information like CTs (e.g., Hart et al. 2009). Although this research finds that ideological or identity-based echo chambers can affect who is exposed to attitude/identity consistent CTs, people must share a CT for it to take hold (e.g., Del Vicario et al., 2016). In turn, and because belief in the veracity of a CT need not be a necessary condition for sharing it, fully understanding the socio-political implications of CTs requires studying the antecedent motivations for sharing.

In psychology, early studies of rumors established that people use rumors to help justify and interpret the world, but doing so is “seldom, if ever, understood by the rumor spreader” (Allport and Postman, 1947: 38). Following the advent of the Internet and social media, recent work demonstrates why people selectively share negative content online (e.g., Shin and Thorson, 2017; Shmargad and Klar, 2019), how emotions elicit information seeking and sharing on social media (e.g., Hoewe and Parrot, 2019), and that negative content becomes more memetically powerful over time (e.g., Gerts et al., 2021). Although CTs are an extreme form of negative, and often political, content (Douglas et al., 2019), there exist other motivations to share negative political content (like CTs) on social media besides the desire to impugn the other side (Petersen et al., 2023). Altogether, this suggests that the motivations to share CTs can operate independently of belief—and may operate independently of partisanship altogether.

In this paper, we explore the effect of three motives on people’s willingness to share CTs on social media: bolstering their or their group's beliefs (motivated sharing), generating collective action against their political outgroup (sounding the alarm), and mobilizing others against the political system (need for chaos). We argue that, whereas motivated sharing is more internally focused and dependent on belief, sounding the alarm is more externally focused on defeating an outgroup and may operate independently of CT belief. We also echo Petersen et al. (2023) in arguing that chaotic motivations can supersede those rooted in partisanship; here, the motive for sharing a CT is to “burn it all down” regardless of which party is in power and belief in the CT (4).

We pit these motives against each other and find not only that belief is the strongest predictor of sharing CTs, but also that people are more likely to share partisan/ideologically consistent CTs than partisan/ideologically inconsistent CTs (motivated sharing). Our results also show that, whereas feeling one’s side is currently losing in politics more often than winning has a surprisingly negative effect on sharing CTs (sounding the alarm), the need for chaos is positively associated with willingness to share all CTs (but has a smaller effect on sharing than belief). Collectively, our findings add to evidence demonstrating that sharing CTs on social media can reinforce beliefs and serve a mobilizing purpose, but the motivation is to disrupt the entire system, rather than rally one’s partisan ingroup against the outgroup. Before elaborating on our research design and findings, we first discuss the theoretical rationale for each of the three sharing motives.¹

Motivated sharing

Research consistently finds that CT endorsement is, in part, a form of motivated reasoning (or “directional reasoning,” see Kunda, 1990) aimed at protecting/bolstering existing attitudes in the face of attitude-challenging information (e.g., Kahan et al., 2011; Lewandowsky et al., 2013; Lodge and Taber, 2013; Miller et al., 2016; Taber and Lodge, 2006). Given the powerful role motivated reasoning plays in CT belief and sharing attitude/identity consistent information online (e.g., Allcott and Gentzkow, 2017; Barberá et al., 2015), we posit that people are willing to share CTs on social media to engage in motivated sharing, that is, to bolster or protect their or their group’s beliefs. Specifically, we expect to find that people who believe a CT will express greater willingness to share it. We also anticipate people will express greater willingness to share partisan/ideologically consistent CTs more than partisan/ideologically inconsistent CTs.

Sounding the alarm

Whether some individuals are motivated to share CTs regardless of their or their group’s beliefs, however, may depend on political context. As Imhoff et al. (2022) described, “Political control deprivation can result from losing elections, so that one’s political values are not represented by governing parties. The experience of lack of control, in general, stimulates a desire to make sense of the social environment,” (2). Because the underlying assumption is that “conspiracy theories communicate information to generate collective action in the face of threat” (Douglas et al., 2019: 13), previous evidence demonstrated two necessary conditions for the loser perceptions argument: (1) People endorse CTs to impugn political outgroups and (2) doing so requires identifying whether CTs come from their group or the outgroup (Smallpage et al., 2017).

We argue there is a third necessary condition: For CTs to be an “early warning system” (Uscinski and Parent, 2014: 131), ingroup members must be aware of those CTs. Because sharing CTs is inherent to awareness, we hypothesize those who feel their side is currently losing more often than winning are more motivated to “sound the alarm”; in other words, they are more willing to share CTs for the purpose of generating collective action against their political outgroup. In this way, sounding the alarm is more externally focused and less dependent on belief than motivated sharing (especially, perhaps, when one feels their group is on the losing side).

Need for chaos

Rallying a partisan ingroup against a partisan outgroup is not the only type of mobilization that could motivate the sharing of CTs. As Petersen et al. (2023) argued, the private act of believing is different from the public act of sharing, and thus sharing “hostile political rumors” like CTs may require different motivations than believing (1). Having posited those high in the “need for chaos” (NFCh) would “indiscriminately share hostile political rumors as a way to disrupt the established democratic ‘cosmos’ and start anew,” they developed and validated a seven-item NFCh scale to measure individual differences in extreme discontent with—as well as the concomitant desire to mobilize against—the established democratic system (2; see also Arceneaux et al., 2021; Petersen et al., 2018). As expected, Petersen et al. (2023) found those with greater NFCh expressed greater willingness to share hostile political rumors online, with the effects of NFCh superseding those of partisanship.

While we also expect to find the effect of NFCh will supersede partisan motivations for sharing (either the motive to bolster one’s partisan/ideological worldview or the partisan motive to mobilize one’s ingroup against the outgroup), Petersen et al. (2023) tested the effect of partisanship on belief and sharing rather than pitting belief and partisanship against each other on willingness to share. We argue testing these mechanisms together is important given one’s motivation to share CTs can stem from the desire to preserve their and/or their group’s beliefs (motivated sharing), but belief is not a necessary condition for sharing in service of the group (sounding the alarm). As such, we extend Petersen et al.’s (2023) analysis to posit that those with higher NFCh will express greater willingness to share CTs on social media, independent of partisanship and belief—that is, independent of motivated sharing and sounding the alarm.

Hypotheses

We ask whether one’s motivation to: bolster their or their group’s beliefs (motivated sharing), generate collective action against their political outgroup because of losing (sounding the alarm), and/or mobilize others against the political system (need for chaos) affects their willingness to share CTs on social media. We hypothesize:

H1a: Individuals who believe a CT are more willing to share it on social media than those who do not believe it (motivated sharing).

H1b: Individuals are more willing to share partisan/ideologically consistent CTs than partisan/ideologically inconsistent CTs (motivated sharing).

H2: Individuals who feel that they are on the losing side of politics are more willing to share CTs on social media than those who feel that they are on the winning side of politics, independent of their or their group’s beliefs (sounding the alarm).

H3: Individuals with higher NFCh are more willing to share CTs on social media than those with lower NFCh, independent of their or their group's beliefs and loser perceptions (need for chaos).

Methods

We administered an original survey in December 2018 over Lucid,¹¹ a platform used in similar research (e.g., Guay and Johnston, 2021; Kreps and Kriner, 2022) and that uses quota sampling to recruit online survey respondents demographically in line with US Census margins (Coppock and McClellan, 2019). Of the 3336 survey respondents, 1772 identified as Democrats/leaned Democrat and 1564 identified as Republicans/leaned Republican.²

All variables included in our analysis are recoded to range from 0 to 1, and all table/figure values are unstandardized OLS regression coefficients.³ We include further information on Lucid, variable coding, complete question wording, and the descriptive statistics of our sample in the Supplementary Materials.

Independent variables

Motivated sharing

We operationalize motivated sharing in three ways. We first directly assess the degree to which respondents believe CTs: three we expect Democrats/liberals to be more likely to believe and three we expect Republicans/conservatives to be more likely to believe.⁴ In short, we expect to find those who believe a CT will express greater willingness to share it.

We also operationalize motivated sharing as partisanship (operationalized as a dummy variable for Republicans, with Democrats as reference) and self-reported ideology (on a seven-point scale ranging from extremely liberal to extremely conservative).⁵ Here, we posit that

Democrats/liberals [Republicans/conservatives] are more willing to share partisan/ideologically consistent CTs than partisan/ideologically inconsistent CTs.

Sounding the alarm

We asked respondents the extent to which, on issues that matter to them, their side has been winning more often than losing or losing more often than winning. We hypothesize people who perceive their side as losing more often than winning are more willing to share CTs than their counterparts.

Need for chaos

We employed an eight-item NFCh scale (Petersen et al., 2023; see also Arceneaux et al., 2021; Petersen et al., 2018).⁶ Item and factor analyses further validate these eight items as a unidimensional scale ( $α =$ 0.94) and confirm NFCh is substantively different from other variables in our model.⁷ We expect to find NFCh is positively associated with the indiscriminate willingness to share CTs--in short, regardless of their truth value or partisan/ideological nature.

Control variables

We also control for strength of partisan identity, authoritarianism, need for closure, trust, interest, religiosity, education, income, gender, age, ethnicity, and race.

Dependent variable

Given space constraints and the potential for priming, we did not ask every respondent about their willingness to share every CT. Instead, respondents were randomly assigned questions about their belief in and willingness to share one of the three Democratic CTs and one of the three Republican CTs.⁸

Results

It is instructive to first examine what predicts belief in each of the six CTs. Consistent with our expectations, results across all models (Figure 1) suggest Democrats [Republicans] are more likely than Republicans [Democrats] to believe a Democratic [Republican] CT. Ideology displays a slight asymmetry, as liberals were more likely than conservatives to believe all Democratic CTs, but conservatives were more likely than liberals to believe two of the three Republican CTs. Loser perceptions are positively associated with belief in two of three Democratic CTs and negatively associated with belief in all three Republican CTs. NFCh is distinct in that it is positively associated with belief in all six CTs.

Figure 1.

Predictors of belief in conspiracy theories.

H1a. Motivated Sharing: Turning to our first hypothesis, do people who believe a CT express a greater willingness to share it on social media? As expected, the first and most direct measure of belief in the CT is positively associated with expressed willingness to share each of the six CTs on social media (Figure 2).

Figure 2.

Predictors of willingness to share conspiracy theories.

H1b. Motivated sharing: We also predicted that people would be more willing to share partisan/ideologically consistent, rather than partisan/ideologically inconsistent, CTs. Contrary to expectations (Figure 2), partisanship is not significantly associated with willingness to share any Democratic or Republican CTs. Ideology is significantly associated with sharing two of the six CTs; here, more liberal individuals expressed greater willingness to share two Democratic CTs.

Altogether, belief is a strong predictor of willingness to share (H1a); however, partisanship and ideology (independent of belief) are not robust predictors of willingness to share CTs on social media (H1b).

H2. Sounding the alarm: Although the weak effects of partisanship and ideology could be interpreted as indicating that sharing does not serve a “sounding the alarm” motive, we expected that the motive to generate collective action against one’s partisan outgroup would be stronger amongst people who believe that they are on the losing, as opposed to the winning, side of politics. Contrary to H2, however, loser perceptions are negatively associated with willingness to share two of the Democratic and two of the Republican CTs (Figure 2). In other words, those who perceived their side as currently winning more often than losing (on issues that matter to them), expressed greater willingness to share CTs than their counterparts.⁹

H3. Need for chaos: Is NFCh positively associated with willingness to share CTs on social media? Consistent with H3, NFCh is statistically significant and positively associated with expressed willingness to share all six CTs, even after controlling for belief, partisanship, ideology, loser perceptions, and all other antecedents (Figure 2). The effect of NFCh is larger than the motive to sound the alarm (which, as measured by loser perceptions, is in the opposite direction than hypothesized) but smaller than that of belief. Together, these results not only add to evidence showing NFCh supersedes partisanship (Petersen et al., 2023), but also indicate that NFCh and belief are strong and consistent predictors of willingness to share CTs on social media.

Robustness check

Given the collinearity between belief and partisanship (see Figure 1) and modeled after Petersen et al.’s (2023) analysis, we ran all CT models again without the direct measure of belief. Unsurprisingly, and consistent with H1b (see Figure 3), partisanship is a stronger predictor of sharing: Specifically, when belief is not included in the model, results indicate both Democrats and Republicans expressed greater willingness to share two of “their” CTs. The effects of ideology, loser perceptions, and NFCh remain largely unchanged.

Figure 3.

Predictors of willingness to share conspiracy theories (without belief).

Discussion and conclusion

We tested the impact of three motives on people’s willingness to share CTs on social media. Our results provide support for the motivated sharing hypothesis (sharing to bolster/protect one’s beliefs), especially when measured directly as belief. Our findings also provided no support for the sounding the alarm hypothesis (sharing to generate collective action against one’s partisan outgroup, resulting from feelings of losing) and strong support for the need for chaos hypothesis (sharing to mobilize others against the entire political system).

In pitting these three motives against each other, we ultimately find that while NFCh had smaller effects on sharing compared to belief, it consistently had greater effects on sharing compared to partisanship and ideology—indicating that the motivation to challenge the entire political system, rather than partisan/ideological rivals, is a strong driver of willingness to share CTs on social media.

Before concluding, a few caveats are in order. First, although the Lucid sample tracks with US Census demographic margins, it is not nationally representative. Second, the current study is observational. Although it would be unethical to experimentally induce chaotic motives, we did attempt to experimentally induce loser perceptions and to “turn off” partisan motivated reasoning (by giving some respondents an opportunity to affirm closely held values, see e.g., Lyons et al., 2021); both efforts were ultimately unsuccessful. Finally, the external validity of our findings is limited in that our measured outcome relies on self-reported willingness to share instead of actual measured behavior. Still, like Petersen et al. (2023), we point to evidence that suggests self-reported intentions to share are indicative of actual sharing behaviors (Mosleh et al., 2020) and look forward to research that can more successfully manipulate these motives and capture actual, individual-level sharing behavior.

Altogether, our findings reveal that prior notions of partisanship and chaos as drivers of sharing hostile political rumors (including CTs) are perhaps more nuanced than extant literature suggests. While our results are consistent with Petersen et al. (2023) in that the effect of partisanship on sharing pales in comparison to NFCh, we examine the effects of NFCh and belief together to reveal both factors are strong and consistent predictors of willingness to share CTs on social media. Our findings also account for loser perceptions and ideology—and are robust to a range of other antecedents—providing strong evidence to suggest individuals are willing to share CTs on social media not only to reinforce existing beliefs, but also to mobilize others against the entire political system. In turn, we advise scholars to pay particular attention to differences between these measures when drawing conclusions about the motives for sharing CTs.¹⁰

Supplemental Material

Supplemental Material - By any memes necessary: Belief- and chaos-driven motives for sharing conspiracy theories on social media

Supplemental Material for By any memes necessary: Belief- and chaos-driven motives for sharing conspiracy theories on social media by Christina E. Farhart, Erin B. Fitz, Joanne M. Miller, and Kyle L. Saunders in Research & Politics

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Kyle L. Saunders

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Allcott

Gentzkow

(2017) Social media and fake news in the 2016 election. The Journal of Economic Perspectives 31(2): 211–236. DOI: 10.1257/jep.31.2.211

Allport

Postman

(1947) The Psychology of Rumor. New York: Henry Holt and Company.

Arceneaux

Gravelle Osmundsen , et al. (2021) Some people just want to watch the world burn: the prevalence, psychology and politics of the ‘Need for Chaos’. Philosophical Transactions of the Royal Society B: Biological Sciences 376(1822): 20200147. DOI: 10.1098/rstb.2020.0147

Barberá

Jost Nagler , et al. (2015) Tweeting from left to right Psychological Science 26(10): pp. 1531–1542. DOI: 10.1177/0956797615594620

Berinsky

(2017) Rumors and health care reform: experiments in political misinformation. British Journal of Political Science 47(2): 241–262. DOI: 10.1017/S0007123415000186

Coppock

McClellan

(2019) Validating the demographic, political, psychological, and experimental results obtained from a new source of online survey respondents. Research and Politics 6(1): 1–14. DOI: 10.1177/2053168018822

Del Vicario

Bessi

Zollo

, et al. (2016) The spreading of misinformation online. Proceedings of the National Academy of Sciences 113(3): 554–559. DOI: 10.1073/pnas.1517441113

Douglas

(2021) COVID-19 conspiracy theories. Group Processes and Intergroup Relations 24(2): 270–275. DOI: 10.1177/1368430220982068

Douglas

Uscinski

Sutton

, et al. (2019) Understanding conspiracy theories. Political Psychology 40(S1): 3–35. DOI: 10.1111/pops.12568

10.

Farhart

Douglas-Durham Lunz Trujillo , et al. (2022) Vax attacks: how conspiracy theory belief undermines vaccine support. Progress in Molecular Biology and Translational Science 188(1): pp. 135–169. DOI: 10.1016/bs.pmbts.2021.11.001

11.

Gerts

Shelley

Parikh

, et al. (2021) “Thought I’d share first” and other conspiracy theory tweets from the COVID-19 infodemic: exploratory study. JMIR Public Health and Surveillance 7(4): e26527. DOI: 10.2196/26527

12.

Guay

Johnston

(2021) Ideological asymmetries and the determinants of politically motivated reasoning. American Journal of Political Science 66(2): 285–301. DOI: 10.1111/ajps.12624

13.

Hart

Albarracín

Eagly

, et al. (2009) Feeling validated versus being correct: a meta-analysis of selective exposure to information. Psychological Bulletin 135(4): 555–588. DOI: 10.1037/a0015701

14.

Hoewe

Parrott

(2019) The power of anger: how emotions predict information seeking and sharing after a presidential election. Atlantic Journal of Communication 27: 272–283. DOI: 10.1080/15456870.2019.1614925

15.

Imhoff

, et al. (2022) Conspiracy mentality and political orientation across 26 countries. Nature Human Behavior 6: 392–403. DOI: 10.1038/s41562-021-01258-7

16.

Kahan

Jenkins-Smith

Braman

(2011) Cultural cognition of scientific consensus. Journal of Risk Research 14(2): 147–174. DOI: 10.1080/13669877.2010.511246

17.

Kreps

Kriner

(2022) The COVID-19 infodemic and the efficacy of interventions intended to reduce misinformation. Public Opinion Quarterly 86(1): 162–175. DOI: 10.1093/poq/nfab075

18.

Kunda

(1990) The case for motivated reasoning. Psychological Bulletin 108(3): 480–498. DOI: 10.1037/0033-2909.108.3.480

19.

Lewandowsky

Oberauer

Gignac

(2013) NASA faked the moon landing-therefore, (climate) science is a hoax. Psychological Science 24(5): 622–633. DOI: 10.1177/0956797612457686

20.

Lodge

Taber

(2013) The Rationalizing Voter. Cambridge: Cambridge University Press.

21.

Lyons

Farhart

Hall

, et al. (2021) Self-affirmation and identity-driven political behavior. Journal of Experimental Political Science 9(2): 225–240. DOI: 10.1017/XPS.2020.46

22.

Miller

Saunders

Farhart

(2016) Conspiracy endorsement as motivated reasoning: the moderating roles of political knowledge and trust. American Journal of Political Science 60(4): 824–844. DOI: 10.1111/ajps.12234

23.

Mosleh

Pennycook

Rand

(2020) Self-reported willingness to share political news articles in online surveys correlates with actual sharing on Twitter. PLoS One 15(2): e0228882. DOI: 10.1371/journal.pone.0228882

24.

Nisbett

Wilson

(1977) Telling more than we can know: verbal reports on mental processes. Psychological Review 84(3): 231–259. DOI: 10.1037/0033-295X.84.3.231

25.

Pennycook

Cannon

Rand

(2018) Prior exposure increases perceived accuracy of fake news. Journal of Experimental Psychology: General 147(12): 1865–1880. DOI: 10.1037/xge0000465

26.

Petersen

, et al. (2018) A ‘need for chaos’ and the sharing of hostile political rumors in advanced democracies. Paper presented at the annual meeting of the American Political Science Association, Boston, MA, 30 August-2 September.

27.

Petersen

Osmundsen

Arceneaux

(2023) The “need for chaos” and motivations to share hostile political rumors. American Political Science Review 1–20. DOI: 10.1017/S0003055422001447

28.

Shin

Thorson

(2017) Partisan selective sharing: the biased diffusion of fact-checking messages on social media. Journal of Communication 67: 233–255. DOI: 10.1111/jcom.12284

29.

Shmargad

Klar

(2019) How partisan online environments shape communication with political outgroups. International Journal of Communication 13: 2287–2313.

30.

Smallpage

Enders

Uscinski

(2017) The partisan contours of conspiracy theory beliefs. Research and Politics 4(4): 1–7. DOI: 10.1177/2053168017746554

31.

Sunstein

Vermeule

(2009) Conspiracy theories: causes and cures. The Journal of Political Philosophy 17(2): 202–227. DOI: 10.1111/j.1467-9760.2008.00325.x

32.

Taber

Lodge

(2006) Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science 50(3): 755–769. DOI: 10.1111/j.1540-5907.2006.00214.x

33.

Uscinski

Parent

(2014) American Conspiracy Theories. Oxford: Oxford University Press

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

4.01 MB