GESIS Leibniz Institute for the Social Sciences: Go to homepage

CPIN: Critical Phenomena In Networks

A NetSci 2025 Satellite Symposium

This is the homepage for the NetSci 2025 satellite symposium Critical Phenomena In Networks, to be held in Maastricht, Netherlands, on June 2, 2025.

Critical Phenomena In Networks

Critical phenomena subsume phenomena related to critical points in complex systems. Critical points are specific values of so-called control parameters that govern the macro-level behavior of systems. When control parameters are at their critical values, the macro-level system states reside at phase transitions between order and disorder. In physics, the state at the phase transition is described by the divergence of macro-level parameters, scaling relationships among system variables, fractal behavior, non-ergodicity, and universality.

Critical phenomena are well-established for physical systems as well as random graphs. However, in network science, we are concerned with networks as representations of complex adaptive systems which are “alive” and reconfigure in response to endogenous or exogenous triggers (e.g., biological, brain, or social networks). In those systems, critical phenomena are harder to demonstrate, and the idea itself is contested. Nevertheless, evidence is amassing which suggests that critical phenomena do occur in those networks (e.g., fractal structures and dynamics, scaling relationships such as network densification, critical transitions, avalanches of activity).

Symposium

This symposium aims at progressing our understanding of, and knowledge about, critical phenomena in networks. How do classical percolation and renormalization theories apply to complex networks? How do the core findings (e.g., scale-free and small-world networks, community structure) of network science relate to critical phenomena? These days, much network data is temporal in nature (link streams). Is there a benefit in leveraging dynamic phase transition theory? Are complex networks attracted by critical points as in the theory of self-organized criticality? Other than physical systems, adaptive systems are functional and bear meaning, entailing theoretical questions. What can the meaning of criticality be in non-physical systems? What can the equivalent of temperature as a control parameter be?

Translating Philip W. Anderson, social networks are governed by the rules of biological networks but not just by those; biological networks are governed by the rules of chemical networks but not just by those; etc. According to this hierarchy, we can expect to find signals and meanings of criticality in the most complex systems. The symposium welcomes empirical and modeling contributions regarding all kinds of complex networks. Interdisciplinary exchange is stressed. Accordingly, contributors are asked to present their research in a way that listeners from different disciplines can follow. Sessions give room to discussions. We try to make this a symposium series to make sustained progress.

Sponsors

CPIN is sponsored by GESIS – Leibniz Institute for the Social Sciences and the University of Exeter.

Important Dates

  • Call for contributions public: January 17, 2025
  • Deadline for contributions: February 15, 2025
  • Notification for contributions: February 18, 2025
  • Deadline for early bird registration: February 25, 2025
  • Symposium date: June 2, 2025
  • Conference dates: June 2 to 6, 2025

Call for Contributions

We invite contributed talks. Please submit an extended abstract until February 15, 2025, by email to cpin(at)gesis(dot)org. The symposium is non-archival, which means, there is no publication arrangement. Your work can be new or previously published. We look forward to receiving proposals from many disciplines. Though the subject originates in the natural sciences, we also encourage submissions from the social sciences and humanities where the concept of criticality may hide behind other theoretical ideas (e.g., order and chaos). A non-exhaustive list of topics of interest includes:

  • Critical phenomena in physical, chemical, biological, brain, ecological, social, cultural, linguistic, financial, and technological networks.
  • Critical phenomena in static, weighted, adaptive, temporal, multilayer, and higher-order networks.
  • Phase transitions and tipping points in networked systems.
  • Evidence and mechanisms of self-organized criticality in adaptive systems.
  • Critical transitions and avalanches of activity in networks.
  • Fractal structures and dynamics in networked systems.
  • Scaling laws and their relation to critical phenomena.
  • Feedback loops and reconfiguration in response to endogenous or exogenous triggers.
  • Functional implications of criticality in non-physical networks.
  • The meaning of non-human or human behavior between order and chaos.
  • Adaptive systems and the role of control parameters in non-physical networks.
  • Dynamic modeling and simulation of critical phenomena in complex systems.
  • Novel computational tools for detecting and characterizing criticality.
  • Applications of critical phenomena in epidemics, diffusion, and resilience.
  • Cross-disciplinary approaches to understanding criticality across physical, biological, and social domains.
  • Philosophical and conceptual frameworks for criticality in adaptive and functional systems.

The Submission requirements are as for the main conference:

  • 1-page abstract with text, one figure with caption, and selected references.
  • PDF format.
  • Created using the LaTeX template here (PDF example here).
  • Submission deadline: February 15, 2025.

Submissions will be considered for contributed talks based on the adherence to the symposium theme, originality, novelty, and scientific quality. We will send out notifications by February 18, 2025. Once an abstract has been accepted, at least one author is required to register and attend the satellite to present the work.

Send your contribution

Keynotes

Dr. Stefan Bornholdt

Professor of theoretical physics at the University of Bremen, Germany. He leads the Complex Systems Lab within the Institute for Theoretical Physics, focusing on understanding the principles of complex systems in nature through statistical physics. His research spans interdisciplinary topics such as complex networks, theoretical systems biology, self-organized criticality, and the application of physics to economics and social behavior (socio- and econophysics).

Self-organized criticality in neural networks and beyond

Self-organized criticality (SOC) provides a framework to explain how dynamical systems with many degrees of freedom can tune themselves to the critical point of a phase transition of the system without requiring external fine-tuning. In this talk, I will review mechanisms for self-organized criticality in adaptive networks. Such networks are usually characterized by a fast dynamics of variables on the nodes, and a much slower adaptive dynamics of the links or even the topology itself. This separation of dynamical time scales allows for robust implementations of SOC, where the self-organization takes place on the slow, not the fast, variables. Such SOC models are robust against noise and thus have a good chance to be at work in real world systems, much more than the first generation of SOC models where criticality is stored in the fast dynamical variables of the system as, for example, in the sandpile model. As one prominent example I will discuss the role self-organized critical adaptive networks play as toy models for criticality in the brain. The specific properties of neural networks in the brain motivate an interesting new perspective on the phase transition in the classical spin glass model of random neural networks. This opens for possibilities for an alternative SOC mechanism and poses questions about the interplay between topology and self-organization. Last, not least, I will discuss the possible relevance of such mechanisms to fields beyond neural networks, ranging from the genome to social and technological networks.

Dr. Marija Mitrovic Dankulov

Research professor at Scientific Computing Laboratory and Head of Innovation Center at the Institute of Physics Belgrade, University of Belgrade. She has extensive knowledge and experience in theoretical and computational physics. Her primary research interest is statistical physics of socio-economic systems and complex networks theory.

Mechanisms of self-organized criticality in social dynamics

The concept of self-organized criticality (SOC) has been widely studied in various fields, including physics, biology, and computer science. SOC provides a mechanism for the diversity of spatiotemporal scales that optimizes the system’s response to omnipresent driving forces. In the past two decades, signatures of SOC have been found in many collective social behaviors, including online social dynamics. In online social dynamics, this concept can shed light on the emergent patterns and collective behaviors observed in social media platforms and online communities. In this talk, we highlight prominent features of human activity patterns, conditioned by circadian cycles and content-related interactions, that can affect the course of the dynamics from the elemental to the global scale. We show that SOC states can be detected in collective emotions and knowledge-building dynamics through avalanche and fractal time series analysis. Precisely, we determine avalanches of emotional comments exhibiting multifractal scaling, scale-invariant inter-avalanching behaviors, and temporal correlations, which are the leading indicators of the existence of SOC states. Through analysis of social network topology, we show that the existence of community structure is inextricably linked with the emergence of SOC.

Program

Time Slot Speaker and title
08:30 – 08:40 Opening remarks The organizers
08:40 – 09:20 Invited 1 Stefan Bornholdt: “Self-organized criticality in neural networks and beyond”
09:20 – 09:40 Contributed 1 Rashid Bekhbudov: “Which complex networks are fractal?”
09:40 – 10:00 Contributed 2 Minsuk Kim, Hanlin Sun, Robert Jankowski, Lorenzo Cirigliano, Anna Poggialini, & Filippo Radicchi: “Shortest-path percolation on complex networks”
10:00 – 10:20 Contributed 3 Eva Rifà, Julian Vicens, & Emanuele Cozzo: “Early warning signals in rumor models: Distinguishing organic growth from astroturfing”
10:20 – 10:30 Discussion
10:30 – 11:00 Coffee break 
11:00 – 11:40 Invited 2 Marija Mitrovic Dankulov: “Mechanisms of self-organized criticality in social dynamics”
11:40 – 12:00 Contributed 4 Ren Manfredi & Rossana Mastrandrea: “Enhancing cooperation in public good game by introducing adverse event in structured populations”
12:00 – 12:20 Contributed 5 Şiir Çınar Uysal: “Topological signatures of critical transitions in complex systems’ networks”
12:20 – 12:30 Discussion

Organizers

Dr. Haiko Lietz: Senior researcher at GESIS – Leibniz Institute for the Social Sciences in Cologne, Germany. He holds a Ph.D. in sociology from the University of Duisburg-Essen. He is interested in everything related to fractals and applying percolation theory and complexity theory to social systems. At GESIS, he coordinates consulting on computational social science methods and digital behavioral data.

Dr. Marcos Oliveira: Lecturer in Computer Science at the University of Exeter. He is interested in data- and theory-driven explicit models to unveil social mechanisms behind phenomena ranging from social inequality to urban crime.

Dr. Jun Sun: Postdoctoral researcher at GESIS – Leibniz Institute for the Social Sciences in Cologne, Germany. He holds a PhD in computer science from the University of Stuttgart. He is interested in analysing and modelling complex emergent phenomena in social systems and digital behavioral data through interdisciplinary methodologies.

Dr. Mathieu Génois: Associate professor at Aix Marseille Université in Marseille, France.

The organizers review the proposals for contributed talks.

Mailing List

We have started a mailing list to foster exchange on critical phenomena in networks. We will also use it for announcements regarding this symposium and possible future iterations.

To subscribe, visit the list’s website.