Opinion dynamics with backfire effect and biased assimilation
Open accessThe DOI matches its Crossref record; 45 of the 45 references it cites were found in a registry or an index.
Abstract
The democratization of AI tools for content generation, combined with unrestricted access to mass media for all (e.g. through microblogging and social media), makes it increasingly hard for people to distinguish fact from fiction. This raises the question of how individual opinions evolve in such a networked environment without grounding in a known reality. The dominant approach to studying this problem uses simple models from the social sciences on how individuals change their opinions when exposed to their social neighborhood, and applies them on large social networks. We propose a novel model that incorporates two known social phenomena: (i) Biased Assimilation : the tendency of individuals to adopt other opinions if they are similar to their own; (ii) Backfire Effect : the fact that an opposite opinion may further entrench people in their stances, making their opinions more extreme instead of moderating them. To the best of our knowledge, this is the first DeGroot-type opinion formation model that captures the Backfire Effect. A thorough theoretical and empirical analysis of the proposed model reveals intuitive conditions for polarization and consensus to exist, as well as the properties of the resulting opinions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures



![Fig 4 For the Karate network. (a) the distribution of βP for 10,000 random opinion vectors (uniform on [−1, 1]); For one of the opinion vectors, (b) the variance of all converged y as β increases from 0 to 10; (c) consensus opinion values for β ∈ [0, 2.1]; (d) final converged opinions for each of the nodes.](https://figures.cochat.ai/paper-figures/625471df82b44e819209022855d83f0a/opinion-dynamics-with-backfire-effect-and-biased-assimilation-figure-4.jpg)



Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Collective dynamics of ‘small-world’ networks
Watts, D. J., & Strogatz, S. H. (1998) · Nature · doi:10.1038/30918 · Cited by 43,060 · cited 1×
“• Watts-Strogatz (WS) networks G ( n, K, 1) have the small world property of with K being the average degree, and we fix the rewiring probability to be 1 (i.e., random graph), thus only refer to K [ Watts and Strogatz, 1998 ] .”
BackgroundStatistical mechanics of complex networks
Albert, R., & Barabási, A.-L. (2002) · Reviews of Modern Physics · doi:10.1103/revmodphys.74.47 · Cited by 20,471 · cited 1×
“• Barab ´ asi-Albert (BA) networks G ( n, M 0 , M ) are scale-free, where M 0 is the number of initial nodes and M the number of nodes that a new node is connected to [ Albert and Barab´asi, 2002 ] .”
BackgroundBollobás, B. (1985) · Cited by 6,196
An Information Flow Model for Conflict and Fission in Small Groups
Zachary, W. W. (1977) · Journal of Anthropological Research · doi:10.1086/jar.33.4.3629752 · Cited by 4,704 · cited 2×
“The real datasets include Zachary’s Karate Club network [31] where we use synthetic opinion vectors, and six Twitter networks from [32, 33] that are gathered with real opinions (computed using sentiment analysis) for different events ranging from political elections to sports.”
BackgroundMethodLord, C. G., Ross, L., & Lepper, M. R. (1979) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.37.11.2098 · Cited by 4,583 · cited 2×
“Two observations from psychology and social sciences relating to our work are the biased assimilation and backfire effect [14, 15], which state that individuals are more inclined to accept opinions closer to their own [11], and that, when exposed to the opposite opinions, individuals entrench themselves in their own o…”
BackgroundStatistical physics of social dynamics
Castellano, C., Fortunato, S., & Loreto, V. (2009) · Reviews of Modern Physics · doi:10.1103/revmodphys.81.591 · Cited by 4,119 · cited 3×
“Much of this research leverages preexisting opinion formation models that have been studied for decades [4, 5].”
BackgroundInfluentialDegroot, M. H. (1974) · Journal of the American Statistical Association · doi:10.1080/01621459.1974.10480137 · Cited by 4,004 · cited 8×
“The DeGroot model is elegant and intuitive, and it guarantees that the opinions converge towards a consensus [4, 6].”
BackgroundMethodInfluentialJackson, M. O. (2010) · Princeton University Press eBooks · doi:10.2307/j.ctvcm4gh1 · Cited by 3,606 · cited 5×
“We study the common setting where opinions are formalized as real values, formed through social interactions (see [Jackson, 2008] and [Castellano et al.”
BackgroundInfluentialWhen Corrections Fail: The Persistence of Political Misperceptions
Nyhan, B., & Reifler, J. (2010) · Political Behavior · doi:10.1007/s11109-010-9112-2 · Cited by 3,173 · cited 1×
“…2009], which state that individuals are more inclined to accept opinions closer to their own [Lord et al. , 1979], and that, when exposed to the opposite opinion, individuals entrench themselves in their own opinion [Nyhan and Reifler, 2010; Chong and Druck-man, 2007; Herr, 1986], respectively.”
BackgroundJackson, M. O. (2008) · Princeton University Press eBooks · doi:10.1515/9781400833993 · Cited by 2,666
Cited by
Recent advances in opinion propagation dynamics: a 2020 survey
Noorazar, H. (2020) · The European Physical Journal Plus · doi:10.1140/epjp/s13360-020-00541-2 · Cited by 177
“Chen [97] adds a repulsive behavior to the model of Dandekar et al.”
MethodFalse Alarm: The Truth About Political Mistruths in the Trump Era
Porter, E., & Wood, T. J. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108688338 · Cited by 63
From classical to modern opinion dynamics
Noorazar, H., Vixie, K. R., Talebanpour, A., & Hu, Y. (2020) · International Journal of Modern Physics C · doi:10.1142/s0129183120501016 · Cited by 133
“[30].”
Opinion dynamics: Statistical physics and beyond
Starnini, M., Baumann, F., Galla, T., Garcia, D., Íñiguez, G., Karsai, M., Lorenz, J., & Sznajd-Weron, K. (2026) · Reviews of Modern Physics · doi:10.1103/j1zg-ddqv · Cited by 73
Learning Opinion Dynamics From Social Traces
Monti, C., De Francisci Morales, G., & Bonchi, F. (2020) · Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining · doi:10.1145/3394486.3403119 · Cited by 31
“The idea that interactions can have different effects is a key concept in several opinion dynamics models [1, 5, 11, 20, 29].”
BackgroundQuantum-Mechanical Modelling of Asymmetric Opinion Polarisation in Social Networks
Maksymov, I. S., & Pogrebna, G. (2024) · Information · doi:10.3390/info15030170 · Cited by 31
“…there exist classical models of confirmation bias [40,41], opinion formation and polarisation [44–56], human morality [57] and backfire effect [58,59], the core principles of quantum mechanics exploited in this paper have not been previously applied to the aforementioned psychological…”
BackgroundOnline Fake News Opinion Spread and Belief Change: A Systematic Review
Altoe, F., Moreira, C., Pinto, H. S., & Jorge, J. A. (2024) · Human Behavior and Emerging Technologies · doi:10.1155/2024/1069670 · Cited by 17
“The results obtained in [16] showed con fi rmation bias in combination with the back fi re e ff ect to be strong drivers of polarization.”
BackgroundInfluentialNon-consensus states in circular opinion model with repulsive interaction
Han, W., Gao, S. D., Huang, C., & Yang, J. (2021) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2021.126428 · Cited by 12
Zareer, M. N., & Šelmić, R. R. (2024) · Social Network Analysis and Mining · doi:10.1007/s13278-024-01402-x · Cited by 11
“Additionally, the integration of social psychology phenomena such as biased assimilation and the backfire effect within a DeGroot-based model is explored in Chen et al. (2021).”
Selective agreement, not sycophancy: investigating opinion dynamics in LLM interactions
Cau, E., Pansanella, V., Pedreschi, D., & Rossetti, G. (2025) · EPJ Data Science · doi:10.1140/epjds/s13688-025-00579-1 · Cited by 9
“Recent models also incorporate the backfire effect [9, 10], where individuals become more entrenched in their opinions when confronted with contradictory information [11].”
Shin, Y. A., Constantino, S. M., Beckage, B., & LaCasse, K. M. (2025) · Current Opinion in Behavioral Sciences · doi:10.1016/j.cobeha.2025.101528 · Cited by 13
Emergence mechanisms of group consensus in social networks
Wang, M., & Zhang, Z. (2023) · Frontiers of Engineering Management · doi:10.1007/s42524-023-0277-x · Cited by 8
“Chen et al. (2021) introduced bias coefficients for each agent to investigate how bias levels within the network structure and nodes affect the achievement and stability of the population’s final equilibrium.”
Low, N. K. Y., & Melatos, A. (2021) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2021.126722 · Cited by 8
Change My Mind: Data Driven Estimate of Open-Mindedness from Political Discussions
Pansanella, V., Morini, V., Squartini, T., & Rossetti, G. (2023) · Studies in computational intelligence · doi:10.1007/978-3-031-21127-0_8 · Cited by 9
“However, the effects of open-mindedness alone and in interplay with other factors have been extensively studied through opinion dynamics models, extending the baseline [8] with additional agent-level characteristics [15], considering multi-dimensional opinions [7] or adding repulsive behaviors [4].”
Towards a Personalized Online Fake News Taxonomy
Altoe, F., & Pinto, H. S. (2023) · ACM Conference on User Modeling, Adaptation and Personalization (UMAP) · doi:10.1145/3565472.3592963 · Cited by 7
“Backfire effect is the fact that an opposite opinion may make people’s opinions more extreme instead of moderating them [12].”
BackgroundA Multi-agent Model for Opinion Evolution in Social Networks Under Cognitive Biases
Alvim, M. S., da Silva, A. G., Knight, S., & Valencia, F. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-62645-6_1 · Cited by 5
Discerning media bias within a network of political allies and opponents: Disruption by partisans
Bu, Y., & Melatos, A. (2023) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2023.128958 · Cited by 5
“However, extensions of the DeGroot and Deffuant-Weibuch models certainly exist, that consider antagonistic interactions without partisans [8, 9, 37, 38].”
MethodModeling Polarization Caused by Empathetic and Repulsive Reaction in Online Social Network
Hirakura, N., Aida, M., & Kawashima, K. (2022) · IEICE Transactions on Communications · doi:10.1587/transcom.2021ebp3150 · Cited by 5
Modeling and simulation analysis of public opinion polarization in a dynamic network environment
Yuan, J., Shi, J., Che, J., Xu, C., & Wang, J. (2020) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.5771 · Cited by 5
Fairness and consensus in an asynchronous opinion model for social networks
Aranda, J., Betancourt, S., Díaz, J. F., Paz, J., & Valencia, F. (2026) · Journal of Logical and Algebraic Methods in Programming · doi:10.1016/j.jlamp.2026.101140 · Cited by 4
Doroszkiewicz, H., & Sierakowska, M. (2021) · Journal of Clinical Nursing · doi:10.1111/jocn.16138 · Cited by 3
Probabilistic analysis of agent-based opinion formation models
Devia, C. A., & Giordano, G. (2023) · Scientific Reports · doi:10.1038/s41598-023-46789-3 · Cited by 4
Graphical analysis of agent-based opinion formation models
Devia, C. A., & Giordano, G. (2024) · PLoS ONE · doi:10.1371/journal.pone.0303204 · Cited by 5
“…we apply the new graphical analysis technique to the classical Friedkin-Johnsen model [29, 30] and the recently proposed Classification-based model [63]; the Bounded Confidence model [23] and the recently proposed Backfire Effect and Biased Assimilation model [31] are also analysed, in the S1 File.”
MethodA Model of Polarization on Social Media Caused by Empathy and RepulsionPreprint
Hirakura, N., Aida, M., & Kawashima, K. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.08141 · Cited by 3
Modeling Competing Agents in Social Media Networks
Zareer, M. N., & Šelmić, R. R. (2022) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv57592.2022.10004247 · Cited by 2
“…agents [25]–[28], with homophily (where individuals with similar opinions tend to connect more then those with opposing opinions) [29], with bias [30], with opinion manipulation [28], [31]–[33], and with repulsion (where individuals with extreme opinions polarize further after interactions) [34].”
BackgroundA Multi-Agent Model for Opinion Evolution under Cognitive BiasesPreprint
Alvim, M. S., da Silva, A. G., Knight, S., & Valencia, F. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2402.17615 · Cited by 2
“[13] extends the model [4] to include the effect of backfire-effect as well.”
BackgroundMisinformation and polarisation as manifestations of social identity: a literature review
Zhi, Y., & Wang, J. (2026) · Journal of Information Communication and Ethics in Society · doi:10.1108/jices-07-2025-0195 · Cited by 2
Online opinion dynamics model considering derivative topics based on complex network
Yao, C., Lin, Y., He, Y. L., Wang, D., Yang, Y., & Fu, D. (2025) · Information Technology and People · doi:10.1108/itp-10-2024-1325 · Cited by 1
Content-Agnostic Moderation for Stance-Neutral Recommendations
Li, N., Kang, B., & Bie, T. D. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761111 · Cited by 3
Wyse, S. K. (2024) · cIRcle (University of British Columbia) · doi:10.14288/1.0443976 · Cited by 1
Evolution of Opinions with Estimation and Interference
Liu, Y., & Cheng, L. (2022) · Chinese Control Conference (CCC) · doi:10.23919/ccc55666.2022.9901556
“The bias phenomenon is studied in [9], where the entrenchment parameter is introduced to model the bias and backfire effect, and therefore this model allows the polarization.”
Bu, Y., Melatos, A., & Evans, R. J. (2025) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2025.130792
“Opponents in a network have been studied with multi-valued belief PDFs [29, 30, 31] as well as with deterministic models, which focus on when and why opinions diverge [48, 35, 49, 50].”
Coevolution of opinion and consumption behavior under a two-layer network framework
Yin, L., Chen, J., Gao, F., & Wu, X. (2025) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2025.131020
MADmax: Multi-agent Trust Dynamics and Influence Maximization
Sorensen, A., Sweitzer, M. D., Naugle, A., Doyle, C., & Krofcheck, D. (2025) · Springer proceedings in complexity · doi:10.1007/978-3-031-89692-7_5
Yang, F., Bai, J., Li, L., & Zeng, D. D. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228193 · Cited by 1
“Furthermore, in the genetically modified food (GMF) dataset, we observe the backfire effect, that is individuals tend to be more stubborn when confronted with opinions that contradict their own [12].”
H-NeiFi: Non-Invasive and Consensus-Efficient Multi-Agent Opinion Guidance
Guo, S., Xu, H., Yang, Y., Guan, Z., Zhao, W., Zhang, X., & Song, Y. (2026) · International World Wide Web Conference (WWW) · doi:10.1145/3774904.3792113
“…science foundation and may trigger psychological resistance mechanisms, such as the backfire effect [9]. d Effective strategies must, therefore, balance immediate efficiency with long-term stability, avoiding self-reinforcing echo chambers while steering consensus toward socially aligned outcomes.”
The Spiral of Silence in Multi-agent Models for Opinion Formation
Aranda, J., Díaz, J. F., Gaona, D., & Valencia, F. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-11176-0_24
Analysis of Opinion Evolution Based on Hegselmann–Krause Model with Historical Opinion
Zhou, Y., & Sun, J. (2026) · Entropy · doi:10.3390/e28050541
“…attention has also been paid to psychological and cognitive mechanisms, such as cognitive dissonance [21], backfire effects and biased assimilation [22], self-persuasion [23], attitude hiding [24], and the combined effects of cognitive dissonance and introversion on echo chamber formation [25].”
Consensus Under Cognitive Biases in an Asynchronous Multi-agent Model of Opinion Evolution
Aranda, J., Díaz, J., Dreher, J., Muñoz, M., Paz, J., & Valencia, F. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-39395-1_10
Maksymov, I. S. (2026) · Cambridge University Press eBooks · doi:10.1017/9781108939010.034
Pogrebna, G., & Hills, T. T. (2026) · Cambridge University Press eBooks · doi:10.1017/9781108939010.024
Opinion Polarisation in Social Networks
Maksymov, I. S. (2026) · Springer eBooks · doi:10.1007/978-3-032-25965-3_2
Learning About Polarization from Simulation: A Comparative Analysis of Four Models
Baillie, E., Smith, W., & Geard, N. L. (2026) · Springer proceedings in complexity · doi:10.1007/978-3-032-31712-4_1
Learning Social Influence with Unobserved AttentionPreprint
Thompson, W., Tian, M., Huynh, L., Wiedemann, A., Modisette, V., Chodrow, P. S., & Brooks, H. Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.12616
“Some DeGroot-type backfire constructions keep opinions in a compact interval by clipping, introducing nonsmoothness [55].”
Metadata
- Authors
- Chen, X., Tsaparas, P., Lijffijt, J., & De Bie, T.
- Affiliations
- Ghent University (BE)
- University of Ioannina (GR)
- Corresponding author
- Chen, Xi
- Venue
- PLOS ONE
- Publisher
- Public Library of Science (PLoS)
- Published
- 2021-09-01
- Volume
- 16
- Issue
- 9
- Pages
- e0256922
- ISSN
- 1932-6203 (ISSN-L)
- PMC
- PMC8409649
- OpenAlex
- W2924693993
- Cited by
- 39
- References
- 45
- Citations by year
- 2026: 7 · 2025: 9 · 2024: 5 · 2023: 6 · 2022: 3 · 2021: 3
- FWCI
- 1.66
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS ONE (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Ghent University Academic Bibliography (Ghent University) (submittedVersion, other-oa)
- Ghent University Academic Bibliography (Ghent University) (submittedVersion, cc-by)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Funders
- FP7 Ideas: European Research Council (615517)
- Fonds Wetenschappelijk Onderzoek (G0F9816N, G091017N, 3G042220, 665501)
- Vlaamse regering (AI Research Programme)
- Topics
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Opinion Dynamics and Social Influence
- Social Sciences › Communication › Social Media and Politics
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Complex Network Analysis Techniques
- Keywords
- opinion dynamics, backfire effect, biased assimilation, DeGroot model, opinion formation, polarization, consensus, social influence, misinformation, echo chambers, extreme opinions
- MeSH terms
- Online Social Networking, Attitude, Humans, Models, Psychological, Prejudice, Social Media
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.