Modeling Resistance and Stubbornness in Opinion Dynamics

You are reading version 1 of this direction, frozen 2026-10-04. The latest version is version 2.

Version 1 of 2127 papers218 conceptsfrozen 2026-10-04

Traditional models of social influence, such as the DeGroot model, often assume that individuals are perfectly plastic—that they update their beliefs by simply averaging the opinions of their neighbors. In these frameworks, a network will almost always reach a global consensus given enough time. However, real-world social systems rarely reach total agreement; instead, they are characterized by persistent disagreement, polarization, and echo chambers.

A growing body of research suggests that the missing ingredient in these models is stubbornness: the inherent resistance an individual has toward changing their mind. By treating susceptibility to influence as a dynamic, state-dependent variable rather than a constant, researchers can better explain why opinions stabilize into clusters rather than merging into a single consensus.

The Mechanism of State-Dependent Susceptibility

In a standard linear influence model, your openness to persuasion is fixed. However, Angelia Nedic, Ji Liu, and Mengbin Ye have demonstrated that susceptibility is often state-dependent. Specifically, individuals become less susceptible to change as their opinions approach the boundaries of the opinion spectrum—a phenomenon known as "extremity-biased resistance" [Discrete-Time Polar Opinion Dynamics with Susceptibility].

This mechanism captures a psychological reality: the more entrenched or radical an individual's view becomes, the more they "tune out" conflicting information. This creates a non-linear feedback loop where moving toward an extreme position simultaneously increases the resistance to moving back toward the center.

Zealots as Network Anchors

While some agents are partially stubborn, others are "zealots"—individuals who refuse to change their opinions regardless of social pressure. Antoine Vendeville and Benjamin Guedj show that these inflexible agents act as permanent sinks for influence, preventing the network from ever reaching a unified consensus [Voter model with zealots for social opinion evolution].

Mathematically, this shifts the study of opinion dynamics from "initial value problems" (where the final state depends on the average of starting views) to boundary value problems. As Amir Leshem and Vikram Krishnamurthy argue, these stubborn "boundary nodes"—whether they are advertisers, firm believers, or malicious actors—effectively dictate the final equilibrium of the entire network, regardless of what the majority initially believed [Boundary value problems in consensus networks].

The Cognitive Origins of Resistance

Why does stubbornness exist in the first place? One perspective, rooted in Rational Inattention, suggests that resistance is a byproduct of cognitive effort. Research on social learning indicates that the degree of influence a network exerts on an individual is inversely proportional to the effort that individual invested in forming their initial belief. If you spent significant mental energy researching a topic, you are mathematically more likely to resist social signals that contradict your findings.

Furthermore, this resistance is not just a personality trait but a dynamic variable that evolves through exposure. Xinyu Li and Jinyang Huang have developed models where "attitude persistence" increases as a user encounters more information aligned with their current views [Information Dissemination Model Based on User Attitude and Public Opinion Environment]. This creates a computational explanation for echo chambers: repeated exposure to similar views reinforces a user's resistance, making them increasingly immune to outside influence over time.

Impact on Convergence and Stability

Stubbornness fundamentally changes the temporal dynamics of a network. Ge Chen and Wenjun Mei recently provided a complete proof that while heterogeneous models (where agents have different levels of open-mindedness) still eventually converge to a stable state, the process can be arbitrarily slow [Convergence of the Heterogeneous Deffuant-Weisbuch Model].

In practical terms, this means that a highly stubborn population might appear to be in a state of permanent "dissensus" or disagreement, even if a mathematical equilibrium exists in the infinite future. This "stalling" of consensus is a direct result of low weighting factors in opinion updates, where agents prioritize their internal anchors over social feedback [Convergence, Consensus and Dissensus in the Weighted-Median Opinion Dynamics].

Strategic Implications

Understanding resistance allows for more sophisticated interventions. Jon Kleinberg and Rediet Abebe have shown that shifting the equilibrium of a network is often more effective when targeting an individual's susceptibility rather than their specific opinion [Opinion Dynamics with Varying Susceptibility to Persuasion]. By making a few key influencers slightly more or less stubborn, one can steer the entire collective opinion more efficiently than by trying to change the beliefs of the majority directly.

Papers in this version

Showing 60 of 127 papers.

History & citationVersion 2 · 2026-10-04

Current · Permanent link

  1. Version 2 (latest) · 2026-10-04 · 99 papers
  2. Version 1 · 2026-10-04 · 127 papers
Cite this direction

Latest version (a living page):

@misc{lacuna_stubborn_agent_influence_2e24a9b0,
  title = {stubborn agent influence},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/stubborn-agent-influence/rdr_2e24a9b0568043839558c9948573b539},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_modeling_resistance_and_stubbornness_in__2e24a9b0_v1,
  title = {Modeling Resistance and Stubbornness in Opinion Dynamics},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/modeling-resistance-and-stubbornness-in-opinion-dynamics/rdr_2e24a9b0568043839558c9948573b539/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}