Exemplar Democracy
When reporting about social problems, the media typically use exemplars i.e. the narrative of a concrete person to illustrate the broader issue. Exemplars can make otherwise complex political issues accessible and interesting to people. Yet, they can also provide an extreme perspective which triggers (1) factual misperceptions in the mass public and (2) biases people’s democratic responsibility attributions of whether the government or the individual is to blame for the problem.
The EXDEM project seeks to understand:
- Why are some types of media exemplars so powerful in shaping factual misperceptions and responsibility attributions in the mass public while others are uninfluential?
- How can the effects of powerful but unrepresentative exemplars be corrected to combat misinformation in the mass public?
About the project
About the project
- Why are some types of media exemplars so powerful in shaping factual misperceptions and responsibility attributions in the mass public while others are uninfluential?
- How can the effects of powerful but unrepresentative exemplars be corrected to combat misinformation in the mass public?
The EXDEM project is theoretically ambitious and will develop a new interdisciplinary theory integrating insights from psychology, communication, cognitive science, and political science to answer the research questions above.
In different lines of research the journalistic practice of using the narrative of a concrete person to illustrate the broader issue has been referred to as “exemplars”, “human interest frames”, “case reports” and “episodic frames”. So the project includes and contributes to these lines of research.
By answering the research questions the EXDEM project has the ambition to advance scientific knowledge of the consequences of media exposure as well as the sources of misinformation and biased democratic responsibility attributions in the mass public and develops new best practice guidelines for how to mitigate them. Relevant sources of theoretical inspiration include (but is certainly not limited to) research on cognitive biases and heuristics, emotions, interpersonal communication, and identity.
Methodologically, the project implements a unique cross-national research design combining different types of survey experiments with analyses of people’s real-world reactions to real-world media tweets. The project will develop and implement a series of parallel experiments in the United States, France, and Denmark to study the research questions. Additional countries may be included among the sites of study.
Work packages
Work packages
The ExDem project is organized into four work packages:
WP1. Exemplar strength on factual perceptions and accountability
WP1. Exemplar strength on factual perceptions and accountability
WP1 focuses on the first step of the communication flow where people are directly exposed to exemplars in the media. The work package provides a cross-national investigation of whether bias-matching exemplars in news stories have a stronger effect on factual perceptions and responsibility attribution than exemplars without a psychological match. To test this, we offer a novel and ambitious cross-national research design with online survey experiments collected in Denmark, France, and the United States.
WP2. Exemplar strength in interpersonal information transmission
WP2. Exemplar strength in interpersonal information transmission
WP2 focuses on the second step of the communication flow where people tell each other about exemplars from the news. The work package asks whether bias-matching exemplars are shared and recollected more and with greater impact in interpersonal communication than exemplars without a psychological match. To answer the research question WP2 includes analyses of real-world Twitter data on news articles featuring exemplars from leading news media Twitter profiles in Denmark, France and the United States, and the number of re-tweets they receive. Second, to maximize internal validity and test the effects of recollected exemplars on factual perceptions and democratic responsibility attribution, we will implement a unique experimental design from cognitive psychology – the chain transmission design in studies in all three countries.
WP3. Among whom are bias-matching exemplars strong?
WP3. Among whom are bias-matching exemplars strong?
WP3 investigates the segments of citizens who are most susceptible to the effects of bias-matching exemplars. The project will study a series of psychologically relevant individual differences. Methodologically, the project will rely on validated survey instruments collected in cross-national surveys as part of the data collection for WP1 and WP2.
WP4. How can the effects of unrepresentative bias-matching exemplars be corrected?
WP4. How can the effects of unrepresentative bias-matching exemplars be corrected?
WP4 asks how the effects of unrepresentative bias-matching exemplars can be corrected. WP4 implements cross-national survey experiments in Denmark, France, and the United States to test whether the impact of an unrepresentative bias-matching exemplar can be corrected by providing a bias-matching counter exemplar and through optimized reporting of statistical base-rate information. The ambition is to provide an evidence-based foundation for developing best-practice guidelines for reporting of statistical information and counter-exemplars.
News, publications and research dissemination
News, publications and research dissemination
New workshop
June 18th we host the interdisciplinary workshop "Advances in Research on Public Opinion in Digitalized Democracies" at Aarhus Institute of Advanced Studies. Program coming soon
AIAS Seminar
Lene Aarøe presented joint work with Miceal Canavan, Dept. of Political Science, Aarhus University, and Julian Christensen, VIVE, on how politicians and citizens use different evidence to update their policy preferences., May 27, 2024
Talk at the HBES conference 2024
We presented our resent research on the impact of exemplars and statistical information on citizens and politicians opinions competitive information environements at the conference of the Human Behavior and Evolution Society, May 23, 2024
Talk at the "Navigating the Complexities of Information Integrity in a Digital Age" conference
Lene Aarøe gave the talk "Do The Facts Matter? The Impact of Statistical Evidence and Single Exemplars on Policy Opinions Among Citizens and Politicians in Digital Democracies" at the Communication Horizons Conference with the theme is "Navigating the Complexities of Information Integrity in a Digital Age", UC Davis, May 3-5, 2024.
Research dissemination
The ExDem project was represented at the conference ”Neutralitet eller aktivisme? Journalister og forskere i dialog” https://www.dmjx.dk/neutralitet-eller-aktivisme-journalister-og-forskere-i-dialog where Lene Aarøe participated in the panel debate ”Repræsentation og præsentation: Hvordan præsenteres forskere og deres forskning i medierne?”.
Abstract accepted for the 2024 APSA conference
Abstract accepted for the 2024-APSA conference: "Persuasive Implications of Experts’ Gender and Visual Gender Expression" by Kjær, Kristine Juul, Dvinge, Mathilde, Laustsen, Lasse & Aarøe, Lene
Research Dissemination: How News Affect Us? Hearts & Minds Festival
The EXDEM project participated at the Hearts and Minds Festival organized by Folkeuniversitet. Saturday September 28, Lene Aarøe debated with associate professor Hans Henrik Knopp, Danish School of Education, Aarhus University, and journalist and podcast editor Peter Læhde Thomsen the moest recent research findings on how the news affect us.
Invited Talk at the Misinformation Seminar, Sciences Po
Lene Aarøe gave an invited talk "Do The Facts Matter? The Impact of Statistical Evidence and Single Exemplars on Policy Opinions Among Citizens and Politicians in Digital Democracies at the Misinformation Seminar, at Sciences Po, October 25, 2024.
Workshop: Future directions
The ExDem project organized a midway workshop looking ahead on future research directions, Monday November 11, at Aarhus Institute of Advanced Studies (AIAS). External discussants were professor Stuart Soroka, UCLA and professor MSO Morten Skovsgaard, University of southern Denmark. Preliminary findings from the project "Exemplars and Responsibility Attributions: A meta-analysis" were discussed. The project is co-authored by Lene Aarøe, and research assistants Erik H. Knudsen and Mateusz Psujek.
New publication
Aarøe, L., Andersen, K., Skovsgaard, M., Svith, F., & Schmøkel, R. (2024). The journalistic preference for extreme exemplars: educational socialization, psychological biases, or editorial policy? Journal of Communication, 74(1), 48-62.
Keynote at the Norface Governance Final Conference
Lene Aarøe gave the keynote "Do the facts matter? The persuasive power of statistical evidence and single exemplars among citizens and elites" at the Norface Governance final confererence Democratic Governance in a Turbulent Age?, University of Vienna, Feb. 14-16, 2024.
Research dissemination
Lene Aarøe participated in the panel debate "Ret og retfærdighed i et presset sundhedsvæsen" at the Festival of Research at Aarhus University, April 2024, and talked about the impact of exemplars in public debate on target group members' health and on public opinion. 
Conference presentation, Annual Meeting of the Danish Political Science Association
We presented the following project at the annual meeting of the Danish Political Science Assocation, Nyborg, November 29, 2023: Kristine, Juul Kjær, Mathilde Dvinge Pedersen, Lasse Laustsen & Lene Aarøe (2023). ”How experts’ Gender and gender typicality shape their persuasivness, perceived traits and what people learn from them”.
Public lecture
Lene Aarøe gave the public lecture ”Hjernen, medierne og vores politiske meningsdannelse” live streamed at the Hearts and Minds knowledge festival. https://fuau.dk/aarhus/program/livestream/hjernen-medierne-og-vores-politiske-meningsdannelse-livestream-med-lene-aaroee-2321-530
New preprint
We have the following pre-print ready: Kristensen-McLachlan, R. D., Canavan, M., Kardos, M., Jacobsen, M., & Aarøe, L. (2023). Chatbots Are Not Reliable Text Annotators. arXiv preprint arXiv:2311.05769.
Research dissemination
Lene Aarøe gave an interview to the national Danish news paper kristeligt Dagblad on the impact of exemplars in political communication https://www.kristeligt-dagblad.dk/danmark/statsministeren-fremhaever-ham-som-et-ideal-en-folkeskolelaerer-tidligere-elever-udfordrer
Conference presentation
We presented the following paper at the annual meeting of the Danish political science association: Aarøe, L, Canavan, M., & Christensen, J. (2023). "Do the facts matter? How Do Politicians and Citizens use Different Evidence to Update Their Policy Preferences." Nyborg, November 29, 2023.
Abstract accepted for the 2024 HBES conference
Aarøe, Lene, Canavan, Miceal & Christensen, Julian. (2024). Do the facts matter
Paper presented at the 2024 APSA Conference, Philadelphia
On September 7, we presented the paper "How experts' gender and gender stereotypicality shape their persuasiveness, perceived traits and what people learn from them" at the Annual Meeting of the American Political Science Association in Philadelphia, September 5-8. The paper is co-authored by Kristine Juul Kjær, Mathilde Dvinge Pedersen, Lasse Laustsen and Lene Aarøe.
Research Dissemination: Public debate
Lene Aarøe participated in the debate "Artificial intelligence. A new chapter in our formative history" together with professor Mads Rosendahl Thomsen, Professor of Comparative Literature, and Cathrine Hasse, Professor of cultural anthopology and learning and moderated by science journalist Line Friis Frederiksen. The debate was organized by the Royal Society of Science and the Letters and Folkeuniversitet, Tuesday, October 22, 2024 in Odense. Photo: Martin Dam Kristensen.
Invited talk with Professor Stuart Soroka, UCLA
November 10-15 the ExDem project was visited by professor Stuart Soroka from UCLA. November 12, Professor Soroka gave the talk "Negativity and Misinformation" on how negativity biases in news account for a substantial portion of longstanding inaccuracies (or “misinformation”) in coverage of a broad range of social, medical, environmental, political, and economic domains.
Contact
For questions regaring the Exemplar Democracy project (ExDem), please contact:
The EXDEM project is led by Professor of Political Science Lene Aarøe and funded by the Independent Research Fund Denmark.
