Session F01: Statistical Physics of Social Systems

Noise-mediated emergent phenomena in worms and plants

Throughout nature, interactions between individuals in a collective system give rise to emergent phenomena across scales in both space and time: motile systems like insect swarms and bird flocks often operate at much more rapid timescales than sessile systems like slowly growing plants. In this talk, I will discuss how fluctuations and symmetry breaking mediate self-organization and emergent collective behavior in models of (motile) entangled worm blobs and (sessile) mutually-shading sunflowers. Following experiments that reveal emergent locomotion in dense blobs of aquatic worms, I show how symmetry breaking and positive feedback in an active polymer model of worm behavior enable the blob to cohesively traverse a temperature gradient [1]. Next, I describe how information can propagate through an interacting system of plants, and demonstrate how self-organization emerges in the presence of random fluctuations [2]. By leveraging variability and disorder, collective living systems can adapt to perturbations and achieve emergent functionalities to enhance their survival.

[1] Chantal Nguyen, Yasemin Ozkan-Aydin, Harry Tuazon, Daniel I. Goldman, M. Saad Bhamla, and Orit Peleg. Front. Phys. 9:734499 (2021)

[2] Chantal Nguyen, Imri Dromi, Aharon Kempinski, Gabriella E. C. Gall, Orit Peleg, and Yasmine Meroz. arXiv:2206.05540 (2022)   

Sociophysics of Polarization

We proposed (M. Kaufman et al Entropy 2022, 24(9), 1262) a statistical mechanics model for exploring dynamics of political polarization in the US. We use it to explore the effect of leadership actions and of external events on polarization. We consider three US political groups: democrats and republicans (currently highly polarized), and independents. We assume that each individual (in each of the three groups) has a stance ranging between left-leaning and right-leaning. We quantify the noise in this three groups system as a social temperature T. Individuals interact with each other within their own group and with individuals of the other groups.  We generate attitude scenarios in time and explore whether concerted interventions, or unexpected shocks can alter them. We study the model with mean field (long-range) interactions and with short-range interactions via Monte Carlo simulations.

 

    

Using stochastic thermodynamics to analyze non-thermodynamic properties of networked dynamical systems

Social networks of political voters, gene regulatory networks, recurrent neural networks or groups of flocking birds, all are examples of out-of-equilibrium systems of interdependent and co-evolving units. Even though there is no thermodynamic interpretation, it is still useful to quantify the irreversibility of these systems in terms of stochastic thermodynamics, namely by calculating the entropy production of the whole system as well as for each subsystem and relate it with the topological properties of the underlying network. First, we show the total entropy production increases for the networks with high non-reciprocity of the nodes (i.e., the absolute difference between in-degree and out-degree). The subsystem entropy production also increases with the distance from the reciprocal case (i.e., the same number of in-links and out-links). Finally, we show the validity of the other results of stochastic thermodynamics as speed limit theorems and thermodynamic uncertainty relations.   

Double transitions in heterogeneous contagion processes on networks

In many real-world contagion phenomena, the number of contacts to spreading entities for adoption varies for different individuals. Therefore, we study a model of contagion dynamics with heterogeneous adoption thresholds. We derive mean-field equations for the fraction of adopted nodes and obtain phase diagrams in terms of the transmission probability and fraction of nodes requiring multiple contacts for adoption. Our contagion model exhibits a rich variety of phase transitions such as continuous, discontinuous, and hybrid phase transitions, criticality, tricriticality, and double transitions. In particular, we find a double phase transition showing a continuous transition and a following discontinuous transition in the density of adopted nodes with respect to the transmission probability. We show that the double transition occurs with an intermediate phase in which nodes following simple contagion become adopted but nodes with complex contagion remain susceptible.   

The Physics of Economic Tipping Points Via Robophysics

We have created an analog economic system running by a population of intelligent robots as workers and firm owners, and a central computer plays the role of a central bank. The robots move over a two-dimensional world in which multiple information fields are represented by chromatic lights coming from a large LED array. Each robot has to make financial decisions partially based on the local state in which they themselfes also influence. The communication network is via bluetooth, which allows trading, non-local information-exchange and global business. Having build an economy concretely from the ground up, we can now study how different macro-behaviors can emerge from micro-interactions, such as the tipping point in banking and the risk of economy collapse when there is a drastic change to the world.   

Active flow of pedestrian crowds: from large-scale measurements to variational modeling

Pedestrians walk and choose their direction based on individual objectives and instantaneous traffic. This yields high variability in crowd dynamics from diluted to dense regimes. Despite the unpredictability of single individuals, ensemble-level universal physical features emerge. These encompass frequent fluctuations and rare events within «solo» dynamics, mutual interactions, as well as routing choices. Reaching a quantitative understanding of these features is a major scientific challenge retaining great societal impact (e.g., in the design of civil infrastructures or crowd management measures), and sharing deep connections with the statistical physics of active matter.

 

To move towards a quantitative physics-based understanding of crowd dynamics, over the previous years, we established large-scale observational experiments held in real-life settings  (stations, festivals, museums), aimed at investigating statistical features of pedestrian motion. Via homemade high-fidelity tracking systems, we have collected datasets including millions of trajectories acquired with and without external influencing stimuli (i.e., crowd control measures, like signage or visual cues). In this talk, I will discuss the issue of modeling in (statistically) quantitative terms via Langevin-like equations and variational principles the observed dynamics, including PDFs of individual velocity, position, contact-avoidance, as well as path choice.   

Quantitative analysis on order splitting behavior of individual traders and the long memory of the order flow.

Revealing the origin of stylized facts in financial markets is a popular research topic in econophysics. Recently, high-quality data allowed us to understand stylized facts based on microscopic observations. One such successful research was provided by Lillo et al. 2005, proposing the Lillo-Mike-Farmer (LMF) model to describe the dynamics of order-splitting traders (STs). This model explained the origin of the long-range correlation (LRC) in the order flow from the viewpoint of the order-splitting hypothesis. While the plausibility of that scenario was qualitatively verified by Toth et al. 2015, there has been no solid support of the LMF prediction at a quantitative level. In this presentation, we investigate the quantitative prediction by the LMF model through microscopic data analysis. We analyzed a large quote dataset from the Tokyo stock exchange nine years long, including the virtual server account information. We classified traders into STs and random traders and found that the metaorder size distribution submitted by STs has a power-law tail. We finally analyzed the joint distribution of two power-law exponents for the metaorder-size distribution and autocorrelation function to confirm the validity of the LMF prediction.   

Planting memes: the shape of information flow in social networks

The movement of information through virtual social networks has proven to be a primary factor in people's knowledge of, perspective on, and engagement with local and global events; understanding the function of these evolving infrastructures is paramount. Although recent work has studied the spread of (mis)information through hashtags and metadata and theories of information diffusion have been developed, the behavior of networks depends sensitively on their topology, which is difficult to access and characterize. We study the informatic impulse of image-based internet memes through social networks by applying principles of analysis inspired by condensed matter physics. Internet memes - which contain an identifiable image that is unchanging with time and text containing evolving sentiments - are an excellent parcel of information that can be tracked to examine network topology. Memes are scraped from social media networks, and the underlying images and texts are categorized using support vector and text recognition algorithms, respectively. Sharing metadata is used to construct the branching path of information transfer. This way we treat each new meme species as an informatic impulse into a network and observe its propagation, decay, and robustness within the network topology.   

Predicting scientific output from interactions at conferences

Across scientific fields, it is common to have annual meetings, conferences, or symposia, where scientists can meet, present their work and hear about their peers' progress. These meetings often serve as incubators for new collaborations. But how good are they at generating successful collaborations that actually lead to scientific output? Here we analyze a unique dataset that enables us to draw connections between initial meetings of scientists and eventual peer-reviewed publications years after. We analyze data from thousands of pairs of scientists spanning multiple conferences in different fields. We propose a simple mathematical model to predict which pairs of authors will end up publishing, and show that it outperforms competing models. This provides new insight into the drivers of collaboration, but also enables us to make predictions as to whether a pair of scientists will be productive together or not — perhaps enabling us to design future conferences more effectively.   

Physics of social interaction at virtual and in-person conferences: a model for the formation of scientific collaborations

The COVID-19 pandemic has sparked a growing debate about the value of in-person compared to virtual interactions. Here, we show that properly designed virtual meetings generate novel collaborations. We present a nonlinear dynamical model for the origin of scientific collaborations at conferences, inspired by the physics of catalytic processes. We tested the model using data we constructed as part of a longitudinal dataset of “Scialog” conferences, including room-level participation data from four in-person and six virtual meetings, each with about 50 participants. We show that interaction in assigned groups was a better predictor of who ultimately formed teams at virtual compared to in-person scientific conferences. We attribute this to informal interaction playing a greater role in team formation during in-person rather than virtual meetings. This observation is supported by anaysis which shows that in-person conferences strengthen network connectedness more than virtual conferences. This suggests that virtual conferences are better at engineering team formation, while in-person conferences are superior for strengthening participants’ awareness of each other.

    

Idea engines: Unifying innovation & obsolescence from markets & genetic evolution to science

Innovation and obsolescence describe dynamics of ever-churning and adapting social and biological systems, concepts that encompass field-specific formulations. We formalize the connection with a toy model of the dynamics of the "space of the possible" (e.g. technologies, mutations, theories) to which agents (e.g. firms, organisms, scientists) couple as they grow, die, and replicate. We predict three regimes: the space is finite, ever growing, or a Schumpeterian dystopia in which obsolescence drives the system to collapse. We reveal a critical boundary at which the space of the possible fluctuates dramatically in size, displaying recurrent periods of minimal and of veritable diversity. When the space is finite, corresponding to physically realizable systems, we find surprising structure. This structure predicts a taxonomy for the density of agents near and away from the innovative frontier that we compare with distributions of firm productivity, covid diversity, and citation rates for scientific publications. Remarkably, our minimal model derived from first principles aligns with empirical examples, implying a follow-the-leader dynamic in firm cost efficiency and biological evolution, whereas scientific progress reflects consensus that waits on old ideas to go obsolete. Our theory introduces a fresh and empirically testable framework for unifying innovation and obsolescence across fields.   

Majority-vote model on continuous networks

The dynamics of opinion formation in societies is a complex phenomenon where collective herd behavior and personal ideas drive essential grouping mechanics. This work investigates the evolutionary dynamics of opinion formation on a continuous network of social interactions. We use the two-state majority-vote model with noise, where an individual adopts the opinion of the majority of its neighbors with probability 1 − q, and a different opinion with chance q, where q stands for the noise parameter. This model presents three collective social opinion states: consensus, polarization, and fragmentation. In the continuous network framework, the interacting population consists of N individuals randomly positioned in a continuous square area of side L = 1, with periodic boundary conditions. The position of every individual assumes real-valued coordinates constrained to the square area, and we relate the average connectivity of each individual with their social interaction radius. We employ Monte Carlo simulations and finite-size scaling analysis to estimate the critical noise parameter as a function of the average connectivity and obtain the phase diagram and its critical exponents β/ν, γ/ν and 1/ν. We observe that the critical noise is an increasing function of the interaction radius R and that a higher R-value favors consensus.   

Author not Attending

Land use change driven by rapid urbanization, climate change-induced migration, and renewable energy generation and distribution poses major challenges for humanity in the coming decades. However, a long history of past practices in land use management has produced globally pervasive systemic inequity and sustainability concerns. The advent of Multi-Objective Land Allocation (MOLA) approaches could open the possibility of increased objectivity and transparency in land use planning. Here, we use techniques from statistical physics to show that generic planning criteria in MOLA generate a series of "flashpoints" where minute changes in planning priorities produce macroscopic changes in land use outcomes. We find that flashpoints are generic features of MOLA models and signal inherent instabilities in land use planning regardless of whether planning is explicitly formulated quantitatively. These instabilities lead to ambiguities in planning outcomes that we term "grey areas". By directly mapping grey areas between planning priorities and outcomes, we reduce a combinatorially large space of land use patterns to a finite, characteristic set that can facilitate dialogue among planners and stakeholders.