Julian Kauk
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Julian Kauk has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Stefan Robert Schweinberger
Department for General Psychology and Cognitive Neuroscience, Institute of Psychology, Friedrich Schiller University of Jena, Jena, Germany
7 shared publications - Helene Kreysa
ORCID http://orcid.org
3 shared publications - Edda Humprecht
Norwegian University of Science and Technology
2 shared publications Helene Kreysa
2 shared publications- Tobias Rothmund
Institute of Communication Science Friedrich Schiller University Jena Jena Germany
2 shared publications André Scherag
Institute of Medical Statistics, Computer and Data Sciences, Jena University Hospital, Jena, Germany
1 shared publication
Latest Publications
- Open Access
- 25 August 2026
- Published by Center for Open Science in PsyArXiv
LLMs reveal how ideology and time shape intuitive vs. deliberative rhetoric across 31 European parliaments
Democratic deliberation depends on how political actors balance intuitive, affective appeals with analytically structured argumentation. Yet how ideology and time shape this rhetorical balance across diverse national contexts at large scale remains poorly understood. Informed by dual-process theories, we leverage large language models (LLMs) to analyze 3.3 million speeches from 31 European parliaments (2004–2023), quantifying two fundamental rhetorical styles: intuitive rhetoric, characterized by affective and heuristic expression, and deliberative rhetoric, characterized by analytical and evidence-oriented language. Across parliaments, ideological position is systematically associated with rhetorical style through curvilinear relationships. Intuitive rhetoric increases toward both ideological extremes, whereas deliberative rhetoric peaks near the political center and declines toward the fringes, forming a robust ideological horseshoe across institutional contexts. Temporal dynamics further document a shift toward greater reliance on intuitive rhetoric, accompanied by a modest decline in deliberative rhetoric, while the curvilinear ideological structuring of rhetorical styles persists throughout. These trends are consistent with a gradual adaptation to attention-driven communication environments, which structurally favor intuitive and affective expression over analytical depth. Together, our findings map the ideological and temporal contours of deliberative erosion in European democracies, with implications for institutional design and the resilience of democratic discourse.
- 25 August 2026
- Published by Center for Open Science in PsyArXiv
Open AccessLLMs reveal how ideology and time shape intuitive vs. deliberative rhetoric across 31 European parliaments
Democratic deliberation depends on how political actors balance intuitive, affective appeals with analytically structured argumentation. Yet how ideology and time shape this rhetorical balance across diverse national contexts at large scale remains poorly understood. Informed by dual-process theories, we leverage large language models (LLMs) to analyze 3.3 million speeches from 31 European parliaments (2004–2023), quantifying two fundamental rhetorical styles: intuitive rhetoric, characterized by affective and heuristic expression, and deliberative rhetoric, characterized by analytical and evidence-oriented language. Across parliaments, ideological position is systematically associated with rhetorical style through curvilinear relationships. Intuitive rhetoric increases toward both ideological extremes, whereas deliberative rhetoric peaks near the political center and declines toward the fringes, forming a robust ideological horseshoe across institutional contexts. Temporal dynamics further document a shift toward greater reliance on intuitive rhetoric, accompanied by a modest decline in deliberative rhetoric, while the curvilinear ideological structuring of rhetorical styles persists throughout. These trends are consistent with a gradual adaptation to attention-driven communication environments, which structurally favor intuitive and affective expression over analytical depth. Together, our findings map the ideological and temporal contours of deliberative erosion in European democracies, with implications for institutional design and the resilience of democratic discourse.
- Open Access
- 31 July 2025
- Published by Center for Open Science in SocArXiv
Affective dynamics of online climate change denial
Climate change represents one of the most pressing challenges of our time, yet efforts to address it are still persistently obstructed by climate change denial. While emotions are recognized as crucial drivers of such denial, a comprehensive understanding of their dynamics and underlying affective dimensions is lacking. Here, we analyze over one million tweet cascades from a 10-year period (2013-2023) to compare the affective landscapes of climate denial and climate action posts. Our findings reveal that climate denial tweets consistently exhibit stronger negative affective signals, particularly anger and disgust, which have intensified over time and are strongly associated with higher user engagement. In contrast, climate action tweets show lower, more stable levels of negativity and are more frequently characterized by positive emotions like optimism and trust. A meta-regression analysis, framed by the Valence-Arousal-Dominance (VAD) model of affect, demonstrates that valence is the primary dimension explaining these affective differences, their evolution over time, and their impact on engagement. Results suggest that the sustained spread of climate denial is increasingly driven by a strategic mobilization of negative emotional valence. These findings offer a unified, theory-driven perspective on the affective mechanisms that shape the online climate change discourse and the propagation of climate misinformation.
- 31 July 2025
- Published by Center for Open Science in SocArXiv
Open AccessAffective dynamics of online climate change denial
Climate change represents one of the most pressing challenges of our time, yet efforts to address it are still persistently obstructed by climate change denial. While emotions are recognized as crucial drivers of such denial, a comprehensive understanding of their dynamics and underlying affective dimensions is lacking. Here, we analyze over one million tweet cascades from a 10-year period (2013-2023) to compare the affective landscapes of climate denial and climate action posts. Our findings reveal that climate denial tweets consistently exhibit stronger negative affective signals, particularly anger and disgust, which have intensified over time and are strongly associated with higher user engagement. In contrast, climate action tweets show lower, more stable levels of negativity and are more frequently characterized by positive emotions like optimism and trust. A meta-regression analysis, framed by the Valence-Arousal-Dominance (VAD) model of affect, demonstrates that valence is the primary dimension explaining these affective differences, their evolution over time, and their impact on engagement. Results suggest that the sustained spread of climate denial is increasingly driven by a strategic mobilization of negative emotional valence. These findings offer a unified, theory-driven perspective on the affective mechanisms that shape the online climate change discourse and the propagation of climate misinformation.
- Open Access
- 16 June 2025
- Published by Digital Games Research Association in DiGRA Digital Library
Gaming the Political Arena: An Analysis of Political Messaging on Twitch in Germany
- 16 June 2025
- Published by Digital Games Research Association in DiGRA Digital Library
Open AccessGaming the Political Arena: An Analysis of Political Messaging on Twitch in Germany
- Open Access
- 1 February 2025
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 4 (2) , pgaf028
Large-scale analysis of fact-checked stories on Twitter reveals graded effects of ambiguity and falsehood on information reappearance
Misinformation disrupts our information ecosystem, adversely affecting individuals and straining social cohesion and democracy. Understanding what causes online (mis)information to (re)appear is crucial for fortifying our information ecosystem. We analyzed a large-scale Twitter (now “X”) dataset of about 2 million tweets across 123 fact-checked stories. Previous research suggested a falsehood effect (false information reappears more frequently) and an ambiguity effect (ambiguous information reappears more frequently). However, robust indicators for their existence remain elusive. Using polynomial statistical modeling, we compared a falsehood model, an ambiguity model, and a dual effect model. The data supported the dual effect model (13.76 times as likely as a null model), indicating both ambiguity and falsehood promote information reappearance. However, evidence for ambiguity was stronger: the ambiguity model was 6.6 times as likely as the falsehood model. Various control checks affirmed the ambiguity effect, while the falsehood effect was less stable. Nonetheless, the best-fitting model explained <7% of the variance, indicating that (i) the dynamics of online (mis)information are complex and (ii) falsehood effects may play a smaller role than previous research has suggested. These findings underscore the importance of understanding the dynamics of online (mis)information, though our focus on fact-checked stories may limit the generalizability to the full spectrum of information shared online. Even so, our results can inform policymakers, journalists, social media platforms, and the public in building a more resilient information environment, while also opening new avenues for research, including source credibility, cross-platform applicability, and psychological factors.
- 1 February 2025
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 4 (2) , pgaf028
Open AccessLarge-scale analysis of fact-checked stories on Twitter reveals graded effects of ambiguity and falsehood on information reappearance
Misinformation disrupts our information ecosystem, adversely affecting individuals and straining social cohesion and democracy. Understanding what causes online (mis)information to (re)appear is crucial for fortifying our information ecosystem. We analyzed a large-scale Twitter (now “X”) dataset of about 2 million tweets across 123 fact-checked stories. Previous research suggested a falsehood effect (false information reappears more frequently) and an ambiguity effect (ambiguous information reappears more frequently). However, robust indicators for their existence remain elusive. Using polynomial statistical modeling, we compared a falsehood model, an ambiguity model, and a dual effect model. The data supported the dual effect model (13.76 times as likely as a null model), indicating both ambiguity and falsehood promote information reappearance. However, evidence for ambiguity was stronger: the ambiguity model was 6.6 times as likely as the falsehood model. Various control checks affirmed the ambiguity effect, while the falsehood effect was less stable. Nonetheless, the best-fitting model explained <7% of the variance, indicating that (i) the dynamics of online (mis)information are complex and (ii) falsehood effects may play a smaller role than previous research has suggested. These findings underscore the importance of understanding the dynamics of online (mis)information, though our focus on fact-checked stories may limit the generalizability to the full spectrum of information shared online. Even so, our results can inform policymakers, journalists, social media platforms, and the public in building a more resilient information environment, while also opening new avenues for research, including source credibility, cross-platform applicability, and psychological factors.
- Open Access
- 25 December 2024
- Published by Elsevier in Computers in Human Behavior
- Vol. 165
Large-scale analysis of online social data on the long-term sentiment and content dynamics of online (mis)information
- 25 December 2024
- Published by Elsevier in Computers in Human Behavior
- Vol. 165
Open AccessLarge-scale analysis of online social data on the long-term sentiment and content dynamics of online (mis)information
- Open Access
- 24 February 2024
- Published by Springer Nature in Journal of Big Data
- Vol. 11 (1) , 1-32
The adaptive community-response (ACR) method for collecting misinformation on social media
Social media can be a major accelerator of the spread of misinformation, thereby potentially compromising both individual well-being and social cohesion. Despite significant recent advances, the study of online misinformation is a relatively young field facing several (methodological) challenges. In this regard, the detection of online misinformation has proven difficult, as online large-scale data streams require (semi-)automated, highly specific and therefore sophisticated methods to separate posts containing misinformation from irrelevant posts. In the present paper, we introduce the adaptive community-response (ACR) method, an unsupervised technique for the large-scale collection of misinformation on Twitter (now known as ’X’). The ACR method is based on previous findings showing that Twitter users occasionally reply to misinformation with fact-checking by referring to specific fact-checking sites (crowdsourced fact-checking). In a first step, we captured such misinforming but fact-checked tweets. These tweets were used in a second step to extract specific linguistic features (keywords), enabling us to collect also those misinforming tweets that were not fact-checked at all as a third step. We initially present a mathematical framework of our method, followed by an explicit algorithmic implementation. We then evaluate ACR on the basis of a comprehensive dataset consisting of $$>25$$ million tweets, belonging to $$>300$$ misinforming stories. Our evaluation shows that ACR is a useful extension to the methods pool of the field, enabling researchers to collect online misinformation more comprehensively. Text similarity measures clearly indicated correspondence between the claims of false stories and the ACR tweets, even though ACR performance was heterogeneously distributed across the stories. A baseline comparison to the fact-checked tweets showed that the ACR method can detect story-related tweets to a comparable degree, while being sensitive to different types of tweets: Fact-checked tweets tend to be driven by high outreach (as indicated by a high number of retweets), whereas the sensitivity of the ACR method extends to tweets exhibiting lower outreach. Taken together, ACR’s capacity as a valuable methodological contribution to the field is based on (i) the adoption of prior, pioneering research in the field, (ii) a well-formalized mathematical framework and (iii) an empirical foundation via a comprehensive set of indicators.- 24 February 2024
- Published by Springer Nature in Journal of Big Data
- Vol. 11 (1) , 1-32
Open AccessThe adaptive community-response (ACR) method for collecting misinformation on social media
Social media can be a major accelerator of the spread of misinformation, thereby potentially compromising both individual well-being and social cohesion. Despite significant recent advances, the study of online misinformation is a relatively young field facing several (methodological) challenges. In this regard, the detection of online misinformation has proven difficult, as online large-scale data streams require (semi-)automated, highly specific and therefore sophisticated methods to separate posts containing misinformation from irrelevant posts. In the present paper, we introduce the adaptive community-response (ACR) method, an unsupervised technique for the large-scale collection of misinformation on Twitter (now known as ’X’). The ACR method is based on previous findings showing that Twitter users occasionally reply to misinformation with fact-checking by referring to specific fact-checking sites (crowdsourced fact-checking). In a first step, we captured such misinforming but fact-checked tweets. These tweets were used in a second step to extract specific linguistic features (keywords), enabling us to collect also those misinforming tweets that were not fact-checked at all as a third step. We initially present a mathematical framework of our method, followed by an explicit algorithmic implementation. We then evaluate ACR on the basis of a comprehensive dataset consisting of $$>25$$ million tweets, belonging to $$>300$$ misinforming stories. Our evaluation shows that ACR is a useful extension to the methods pool of the field, enabling researchers to collect online misinformation more comprehensively. Text similarity measures clearly indicated correspondence between the claims of false stories and the ACR tweets, even though ACR performance was heterogeneously distributed across the stories. A baseline comparison to the fact-checked tweets showed that the ACR method can detect story-related tweets to a comparable degree, while being sensitive to different types of tweets: Fact-checked tweets tend to be driven by high outreach (as indicated by a high number of retweets), whereas the sensitivity of the ACR method extends to tweets exhibiting lower outreach. Taken together, ACR’s capacity as a valuable methodological contribution to the field is based on (i) the adoption of prior, pioneering research in the field, (ii) a well-formalized mathematical framework and (iii) an empirical foundation via a comprehensive set of indicators. - Open Access
- 11 October 2023
- Published by Springer Nature in Behavior Research Methods
- Vol. 56 (5) , 5103-5115
The Jena Audiovisual Stimuli of Morphed Emotional Pseudospeech (JAVMEPS): A database for emotional auditory-only, visual-only, and congruent and incongruent audiovisual voice and dynamic face stimuli with varying voice intensities
We describe JAVMEPS, an audiovisual (AV) database for emotional voice and dynamic face stimuli, with voices varying in emotional intensity. JAVMEPS includes 2256 stimulus files comprising (A) recordings of 12 speakers, speaking four bisyllabic pseudowords with six naturalistic induced basic emotions plus neutral, in auditory-only, visual-only, and congruent AV conditions. It furthermore comprises (B) caricatures (140%), original voices (100%), and anti-caricatures (60%) for happy, fearful, angry, sad, disgusted, and surprised voices for eight speakers and two pseudowords. Crucially, JAVMEPS contains (C) precisely time-synchronized congruent and incongruent AV (and corresponding auditory-only) stimuli with two emotions (anger, surprise), (C1) with original intensity (ten speakers, four pseudowords), (C2) and with graded AV congruence (implemented via five voice morph levels, from caricatures to anti-caricatures; eight speakers, two pseudowords). We collected classification data for Stimulus Set A from 22 normal-hearing listeners and four cochlear implant users, for two pseudowords, in auditory-only, visual-only, and AV conditions. Normal-hearing individuals showed good classification performance (McorrAV = .59 to .92), with classification rates in the auditory-only condition ≥ .38 correct (surprise: .67, anger: .51). Despite compromised vocal emotion perception, CI users performed above chance levels of .14 for auditory-only stimuli, with best rates for surprise (.31) and anger (.30). We anticipate JAVMEPS to become a useful open resource for researchers into auditory emotion perception, especially when adaptive testing or calibration of task difficulty is desirable. With its time-synchronized congruent and incongruent stimuli, JAVMEPS can also contribute to filling a gap in research regarding dynamic audiovisual integration of emotion perception via behavioral or neurophysiological recordings.
- 11 October 2023
- Published by Springer Nature in Behavior Research Methods
- Vol. 56 (5) , 5103-5115
Open AccessThe Jena Audiovisual Stimuli of Morphed Emotional Pseudospeech (JAVMEPS): A database for emotional auditory-only, visual-only, and congruent and incongruent audiovisual voice and dynamic face stimuli with varying voice intensities
We describe JAVMEPS, an audiovisual (AV) database for emotional voice and dynamic face stimuli, with voices varying in emotional intensity. JAVMEPS includes 2256 stimulus files comprising (A) recordings of 12 speakers, speaking four bisyllabic pseudowords with six naturalistic induced basic emotions plus neutral, in auditory-only, visual-only, and congruent AV conditions. It furthermore comprises (B) caricatures (140%), original voices (100%), and anti-caricatures (60%) for happy, fearful, angry, sad, disgusted, and surprised voices for eight speakers and two pseudowords. Crucially, JAVMEPS contains (C) precisely time-synchronized congruent and incongruent AV (and corresponding auditory-only) stimuli with two emotions (anger, surprise), (C1) with original intensity (ten speakers, four pseudowords), (C2) and with graded AV congruence (implemented via five voice morph levels, from caricatures to anti-caricatures; eight speakers, two pseudowords). We collected classification data for Stimulus Set A from 22 normal-hearing listeners and four cochlear implant users, for two pseudowords, in auditory-only, visual-only, and AV conditions. Normal-hearing individuals showed good classification performance (McorrAV = .59 to .92), with classification rates in the auditory-only condition ≥ .38 correct (surprise: .67, anger: .51). Despite compromised vocal emotion perception, CI users performed above chance levels of .14 for auditory-only stimuli, with best rates for surprise (.31) and anger (.30). We anticipate JAVMEPS to become a useful open resource for researchers into auditory emotion perception, especially when adaptive testing or calibration of task difficulty is desirable. With its time-synchronized congruent and incongruent stimuli, JAVMEPS can also contribute to filling a gap in research regarding dynamic audiovisual integration of emotion perception via behavioral or neurophysiological recordings.
- Open Access
- 4 September 2023
- Published by Springer Nature in Research Square
The adaptive community-response (ACR) method for collecting misinformation on social media
Social media can be a major accelerator of the spread of misinformation, thereby potentially compromising both individual well-being and social cohesion. Despite significant recent advances, the study of online misinformation is a relatively young field facing several (methodological) challenges. In this regard, the detection of online misinformation has proven difficult, as online large-scale data streams require (semi-)automated, highly specific and therefore sophisticated methods to separate posts containing misinformation from irrelevant posts. In the present paper, we introduce the adaptive community-response (ACR) method, an unsupervised technique for the large-scale collection of misinformation on Twitter (now known as 'X'). The ACR method is based on previous findings showing that Twitter users occasionally reply to misinformation with fact-checking by referring to specific fact-checking sites (crowdsourced fact-checking). In a first step, we captured such misinforming but fact-checked tweets. These tweets were used in a second step to extract specific linguistic features (keywords), enabling us to collect also those misinforming tweets that were not fact-checked at all as a third step. We initially present a mathematical framework of our method, followed by an explicat algorithmic implementation. We then evaluate ACR on the basis of a comprehensive dataset consisting of > 25 million tweets, belonging to > 300 misinforming stories. Our evaluation shows that ACR is a valid and useful tool for large-scale collection of misinformation online: Text similarity measures clearly indicate correspondence between claims of false stories and collected tweets. Taken together, ACR's efficacy is based upon three pillars: (i) the adoption of prior, pioneering research in the field, (ii) a well-formalized mathematical framework and (iii) a robust empirical proof. This triad leads us to the conclusion that ACR has the potential to make a valuable contribution to further resolve methodological challenges of the field.
- 4 September 2023
- Published by Springer Nature in Research Square
Open AccessThe adaptive community-response (ACR) method for collecting misinformation on social media
Social media can be a major accelerator of the spread of misinformation, thereby potentially compromising both individual well-being and social cohesion. Despite significant recent advances, the study of online misinformation is a relatively young field facing several (methodological) challenges. In this regard, the detection of online misinformation has proven difficult, as online large-scale data streams require (semi-)automated, highly specific and therefore sophisticated methods to separate posts containing misinformation from irrelevant posts. In the present paper, we introduce the adaptive community-response (ACR) method, an unsupervised technique for the large-scale collection of misinformation on Twitter (now known as 'X'). The ACR method is based on previous findings showing that Twitter users occasionally reply to misinformation with fact-checking by referring to specific fact-checking sites (crowdsourced fact-checking). In a first step, we captured such misinforming but fact-checked tweets. These tweets were used in a second step to extract specific linguistic features (keywords), enabling us to collect also those misinforming tweets that were not fact-checked at all as a third step. We initially present a mathematical framework of our method, followed by an explicat algorithmic implementation. We then evaluate ACR on the basis of a comprehensive dataset consisting of > 25 million tweets, belonging to > 300 misinforming stories. Our evaluation shows that ACR is a valid and useful tool for large-scale collection of misinformation online: Text similarity measures clearly indicate correspondence between claims of false stories and collected tweets. Taken together, ACR's efficacy is based upon three pillars: (i) the adoption of prior, pioneering research in the field, (ii) a well-formalized mathematical framework and (iii) a robust empirical proof. This triad leads us to the conclusion that ACR has the potential to make a valuable contribution to further resolve methodological challenges of the field.
- Open Access
- 12 August 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (8) , e0256179
Understanding and countering the spread of conspiracy theories in social networks: Evidence from epidemiological models of Twitter data
Conspiracy theories in social networks are considered to have adverse effects on individuals’ compliance with public health measures in the context of a pandemic situation. A deeper understanding of how conspiracy theories propagate through social networks is critical for the development of countermeasures. The present work focuses on a novel approach to characterize the propagation of conspiracy theories through social networks by applying epidemiological models to Twitter data. A Twitter dataset was searched for tweets containing hashtags indicating belief in the “5GCoronavirus” conspiracy theory, which states that the COVID-19 pandemic is a result of, or enhanced by, the enrollment of the 5G mobile network. Despite the absence of any scientific evidence, the “5GCoronavirus” conspiracy theory propagated rapidly through Twitter, beginning at the end of January, followed by a peak at the beginning of April, and ceasing/disappearing approximately at the end of June 2020. An epidemic SIR (Susceptible-Infected-Removed) model was fitted to this time series with acceptable model fit, indicating parallels between the propagation of conspiracy theories in social networks and infectious diseases. Extended SIR models were used to simulate the effects that two specific countermeasures, fact-checking and tweet-deletion, could have had on the propagation of the conspiracy theory. Our simulations indicate that fact-checking is an effective mechanism in an early stage of conspiracy theory diffusion, while tweet-deletion shows only moderate efficacy but is less time-sensitive. More generally, an early response is critical to gain control over the spread of conspiracy theories through social networks. We conclude that an early response combined with strong fact-checking and a moderate level of deletion of problematic posts is a promising strategy to fight conspiracy theories in social networks. Results are discussed with respect to their theoretical validity and generalizability.
- 12 August 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (8) , e0256179
Open AccessUnderstanding and countering the spread of conspiracy theories in social networks: Evidence from epidemiological models of Twitter data
Conspiracy theories in social networks are considered to have adverse effects on individuals’ compliance with public health measures in the context of a pandemic situation. A deeper understanding of how conspiracy theories propagate through social networks is critical for the development of countermeasures. The present work focuses on a novel approach to characterize the propagation of conspiracy theories through social networks by applying epidemiological models to Twitter data. A Twitter dataset was searched for tweets containing hashtags indicating belief in the “5GCoronavirus” conspiracy theory, which states that the COVID-19 pandemic is a result of, or enhanced by, the enrollment of the 5G mobile network. Despite the absence of any scientific evidence, the “5GCoronavirus” conspiracy theory propagated rapidly through Twitter, beginning at the end of January, followed by a peak at the beginning of April, and ceasing/disappearing approximately at the end of June 2020. An epidemic SIR (Susceptible-Infected-Removed) model was fitted to this time series with acceptable model fit, indicating parallels between the propagation of conspiracy theories in social networks and infectious diseases. Extended SIR models were used to simulate the effects that two specific countermeasures, fact-checking and tweet-deletion, could have had on the propagation of the conspiracy theory. Our simulations indicate that fact-checking is an effective mechanism in an early stage of conspiracy theory diffusion, while tweet-deletion shows only moderate efficacy but is less time-sensitive. More generally, an early response is critical to gain control over the spread of conspiracy theories through social networks. We conclude that an early response combined with strong fact-checking and a moderate level of deletion of problematic posts is a promising strategy to fight conspiracy theories in social networks. Results are discussed with respect to their theoretical validity and generalizability.