Interconnectedness and (in)coherence as a signature of conspiracy worldviews

Abstract

Conspiracy theories may arise out of an overarching conspiracy worldview that identifies common elements of subterfuge across unrelated or even contradictory explanations, leading to networks of self-reinforcing beliefs. We test this conjecture by analyzing a large natural language database of conspiracy and nonconspiracy texts for the same events, thus linking theory-driven psychological research with data-driven computational approaches. We find that, relative to nonconspiracy texts, conspiracy texts are more interconnected, more topically heterogeneous, and more similar to one another, revealing lower cohesion within texts but higher cohesion between texts and providing strong empirical support for an overarching conspiracy worldview. Our results provide inroads for classification algorithms and further exploration into individual differences in belief structures.

Loose and Tight: Creative Formation but Rigid Use of Nominal Compounds in Conspiracist Texts

“…Language is a complex and multifaceted domain, and capturing all the nuances of genre differences using one or few predictors could be indeed challenging. Even though the study of conspiracist language is still in its infancy, it has already yielded promising outcomes (Fong et al, 2021;Klein et al, 2019;Meuer, Oeberst, & Imhoff, 2023;Miani et al, 2021Miani et al, , 2022Samory & Mitra, 2018;Tangherlini et al, 2020). Future endeavors may benefit further from leveraging data-driven machine learning techniques to assess the relative contributions of individual lexical markers (Douglas et al, 2023).…”

“…To be more credible and memorable, thus increasing persuasion, these narratives could rely on creativity. A recent large‐scale text analysis found that, compared to mainstream narratives, conspiracist narratives are internally less coherent but more interconnected in a dense and unstructured network of shared topics (Miani et al., 2022). These patterns suggest that semantic associative networks of conspiracist texts are different from those of mainstream texts, converging with existing research on creative individuals (Kenett, Anaki, & Faust, 2014).…”

“…In this process, believers, who often have low tolerance to ambiguity (Marchlewska, Cichocka, & Kossowska, 2018), may jump to conclusions quicker than nonbelievers (Kuhn, Lieb, Freeman, Andreou, & Zander-Schellenberg, 2021) and identify meaningful relationships among randomly co-occurring events (van der Wal et al, 2018), sometimes confusing aspects of reality, such as believing that prayers have the capacity to heal (Lobato, Mendoza, Sims, & Chin, 2014), and endorsing paranormal beliefs (Darwin, Neave, & Holmes, 2011). Such a tendency to connect disparate ideas may play a pivotal role in generating the unconventional narrative patterns observed in conspiracist narratives (Bruns, Harrington, & Hurcombe, 2020;Miani et al, 2022;Samory & Mitra, 2018;Tangherlini, Shahsavari, Shahbazi, Ebrahimzadeh, & Roychowdhury, 2020).…”

“…This makes them more memorable and more easily transmitted (Norenzayan, Atran, Faulkner, & Schaller, 2006). Conspiracist narratives are based on a persuasive rhetoric that mimics the formal features of academic discourse, using a multitude of arguments (Miani, Hills, & Bangerter, 2022; Oswald, 2016). Likewise, malicious rumors (vs. non‐malicious ones) feature high levels of linguistic sophistication (Vosoughi et al., 2017).…”

IRMA: the 335-million-word Italian coRpus for studying MisinformAtion

“…We also tested the degree to which credibility was linked to interconnectedness, that is how multiple ideas form a dense and highly interconnected network, a property of conspiracy narratives (Miani et al, 2022a). To this purpose, we created networks for each website from the cooccurrence of the top-fifty most frequent words extracted from the TF-IDF (see variable tfidf10 in Table 3).…”