Research
Language Dynamics
Our research work in Language Dynamics is supported by the Estonian Research Council, through the Grant PRG 1059, "Learning Processes in Language Dynamics".
Language dynamics employs complex systems models and tools to describe competition, evolution, and spreading of languages. We study mesoscopic ecological-like models of language competitions, as well as microscopic (possibly heterogeneous) many-agent models to study the spreading and evolution of languages.
On the top of this, we have also an interest and an active research line in language/dialect comparison and classification using statistical tools from complex data analysis.
We also studied the statistical properties of written text in relation to the appearance of power-laws, e.g. the Zipf's law.
Chek the publications below for further details.
PUBLICATIONS
A three-state language competition model including language learning and attrition,
Stefano Scialla, Jens-Kristjan Liivand, Marco Patriarca, Els Heinsalu,
Front. Complex Syst. 1-1266733 (2023),
https://doi.org/10.3389/fcpxs.2023.1266733The Physics of Languages
Marco Patriarca, Els Heinsalu, and David Sánchez
Physics World, 13 June 2023
https://physicsworld.com/a/the-physics-of-languages/The role of bilinguals in the Bayesian naming game,
Gionni Marchetti, Marco Patriarca, Els Heinsalu.
Physica D 428, 133062 (2021) doi:10.1016/j.physd.2021.133062 arXiv:2106.00069A Bayesian Approach to the Naming Game Model,
Gionni Marchetti, Marco Patriarca, and Els Heinsalu,
Frontiers in Physics 8, 10 (2020) doi:10.3389/fphy.2020.00010 arXiv:1911.13012"Languages in Space and Time: Models and Methods from Complex Systems Theory",
Patriarca, Marco; Heinsalu, Els; Leonard, Jean Leó.
Cambridge University Press (2020).A Bird’s-Eye View of Naming Game Dynamics: From Trait Competition to Bayesian Inference,
Gionni Marchetti, Marco Patriarca, and Els Heinsalu,
Chaos 30, 063119 (2020) doi:10.1063/5.0009569 arXiv:2004.01994Patterns of linguistic diffusion in space and time: The case of Mazatec,
J.L. Léonard, M. Patriarca, E. Heinsalu, K. Sharma, A Chakraborti.
in: "Complexity Applications in Language and Communication Sciences",
À, Massip-Bonet, G. Bel-Enguix, A. Bastardas-Boada, Editors,
p. 139-170 Springer (2019) doi:10.1007/978-3-030-04598-2_9Applicazioni alla linguistica dei metodi e modelli della teoria dei sistemi complessi,
Marco Patriarca, Els Heinsalu, Jean Léo Léonard,
p. 103-143, Il Calamo (2018).
in: "Mutamento Linguistico e Biodiversità"
(Atti del XLI Convegno della Società Italiana di Glottologia, Perugia, 1-3 dicembre 2016)
L. Costamagna, E. Di Domenico, A. Marcaccio, S. Scaglione, and B. Turchetta, editors,
II Calamo, Roma, 2018. ISBN: 9788898640317
Resonances in cell-networks
The function and properties of biological cell networks depends on their global dynamical properties. Indeed, the very reason for the existence of cell networks is that they have special features that single cells don't have. We investigate various types of cell networks and the resonances appearing in their collective behaviors.
In our last paper "Diversity-induced decoherence" we study a network of coupled oscillators with a strong time scale separation uncovering a new nontrivial effect that we name diversity-induced decoherence (DIDC), in which heterogeneity modulates the mechanism of self-induced stochastic resonance to inhibit the coherence of oscillations.
We have studied also mathematical models β-cell networks, showing that the emergence of pacemakers (or hubs) in the system is a natural consequence of the oscillator population diversity. We have also studied the interplay of noise and disorder in cell networks, showing that it presents a wide range of effects, which can be synergistic or independent of each other.
PUBLICATIONS
Diversity-induced decoherence
Marius E. Yamakou, Els Heinsalu, Marco Patriarca, and Stefano Scialla,
Phys. Rev. E 106, L032401 (2022)
doi: 10.1103/PhysRevE.106.L032401 https://arxiv.org/abs/2206.08064The interplay between diversity and noise in an excitable cell network model,
Stefano Scialla, Marco Patriarca, Els Heinsalu.
Europhysics Letters 137, 51001 (2022) doi: 10.1209/0295-5075/ac5cdb arXiv:2203.12506Hubs, diversity, and synchronization in FitzHugh-Nagumo oscillator networks: Resonance effects and biophysical implications,
Stefano Scialla, Alessandro Loppini, Marco Patriarca, and Els Heinsalu.
Phys. Rev. E 103, 052211 (2021) doi:10.1103/PhysRevE.103.052211 arXiv:2105.05652
Heterogeneity and diffusion in Ecological Competition
Competition models are a fundamental paradigm in complex systems for the study of natural selection in different processes, from ecological competition to cultural diffusion and epidemic spreading.
The interplay between diffusion and competition is known to lead to relevant phenomena such as the appearance of wave fronts.
In our research we take into account two other crucial features of ecological systems.
First, we consider the finite-range character of interactions,. In this case the system can develop patterns (in the form of clumped distributions).
Furthermore, we include heterogenity in the diffusion properties of the individuals. We find that the natural selection process can proceed in a very different way and, in principle, any individual, either slow or fast, can win the competition process, depending on the system parameters.
PUBLICATIONS
Influence of invasion on natural selection in dispersal-structured populations,
David Navidad Maeso, Marco Patriarca, Els Heinsalu,
Physica A 547, 124427 (2022),
doi: 0.1016/j.physa.2022.127389 arxiv.org/abs/2204.11899The dynamics of natural selection in dispersal-structured populations,
Heinsalu, Els; Navidad Maeso, David; Patriarca, Marco,
Physica A 547 (2020) 124427 doi:0.1016/j.physa.2020.124427 arXiv:2004.14689The role of dispersal in competition success and in the emerging diversity,
Els Heinsalu, David Navidad Maeso, and Marco Patriarca,
Eur. Phys. J. B 91, 255 (2018) doi:10.1140/epjb/e2018-90372-5 arXiv:2004.06088
Power-Law distributions in Economics and Text Analysis
Power-law distributions are ubiquitous in many systems, from the Zipf's law in text analysis to the Pareto law found in wealth distributions.
We have studied microscopic models that leads to the spontaneous emergence of power-law distribution, as a consequence of the diversity of the constituent units. Thus, a possible interpretation of the appearance of power-laws is that it is dynamically originated from the heterogeneity of the system.
PUBLICATIONS
In: "Econophysics and Sociophysics: Recent Progress and Future Directions", Frédéric Abergel & al. Editors, Springer (2017) ISSN 2039-411X:
Kinetic Exchange Models as D Dimensional Systems: A Comparison of Different Approaches (p.147), Marco Patriarca, Els Heinsalu, Amrita Singh, and Anirban Chakraborti.
The Microscopic Origin of the Pareto Law and Other Power-Law Distributions (p. 159), Marco Patriarca, Els Heinsalu, Anirban Chakraborti, and Kimmo Kaski.