Full publication list

Books

• Anna Zafeiris and Tamás Vicsek, Why we live in hierarchies: a quantitative treatise, SpringerBriefs in Complexity, 2018. (pdf)
  Summary: This book provides a quantitative framework for understanding hierarchical organization in biological and social systems. It combines empirical observations, mathematical modeling, and simulation results to argue that hierarchy often emerges as an efficient solution for decision-making and resource allocation.
  
  

Book chapters

• Anna Zafeiris and Tamás Vicsek, Advantages of hierarchical organization: from pigeon flocks to optimal network structures, in "Research in the Decision Sciences for Global Business: Best Papers from the 2013 Annual Conference" Gyula Vastag (Ed), FT Press Operations Management, USA, 2015. (pdf)
  Summary: This paper explores the prevalence and structure of hierarchical organization in both human and animal groups. Moving beyond the common binary leader-follower models, it presents biological observations and computational results that reveal inherently multi-level hierarchies. The study also introduces a quantitative measure for assessing hierarchical depth in complex networks, offering a unified framework for comparing organizational structures across systems.
  
  
• Anna Lázár, Karl Pauwels, Marc Van Hulle, Tamás Roska, Scene analysis of unstable video flows using multiple retina channels and attentional methods, in "Integrated Circuits, Photodiodes and Organic Field Effect Transistors", R. McIntire and P. Donnell (Eds.), NovaScience (NY), USA, 2009. (pdf)
  Summary: This book chapter presents a biologically inspired visual processing framework developed for the “Bionic Eyeglass Project,” aimed at assisting blind and visually impaired users. It introduces a novel image stabilization technique and real-time algorithms for detecting LED indicators, traffic signs, and indoor lighting from low-resolution, unstable mobile footage. Leveraging mammalian retina channel decomposition and saliency-based processing, the system achieves high accuracy under unconstrained real-world conditions, demonstrating the practical potential of neuromorphic vision models.
  
  

Articles

• Zsuzsanna Siklósi, Péter Csippán, Norbert Faragó Zsuzsa Hegedűs, Eszter Solnay, Anna, Szécsényi-Nagy, Márton Szilágyi and Anna Zafeiris "Individuals and communities, social networks and innovations in the Copper Age of the Carpathian Basin", In: Stavila, A; Bogdan, C; Cirt, R (Eds.) Interdisciplinarity in Archaeology UISPP 2023: Book of Abstracts, West University of Timisoara, 2023
  
  
• Máté Nagy, Anna Zafeiris, Attila Horicsányi, Enikő Kubinyi, Gábor Vásárhelyi and Tamás Vicsek, "Csoportos keresés labirintusban: a társaktól származó információ és a patkánycsoport változatos összetételének előnyei és hátrányai" (in Hungarian).
  En: "Group search in a maze: the advantages and disadvantages of social information and diversity in rat groups".
  Fizikai Szemle 73: pp. 305-310, 2023 (pdf)
  Summary: This study investigates how rats benefit — or are hindered — by group composition and peer-derived information during spatial search tasks in a maze. The results reveal that while social cues can accelerate exploration, diversity within the group affects efficiency in nuanced ways. The findings offer insights into the adaptive value of collective behavior and its limits in problem-solving contexts.
  
  
• Anna Zafeiris "Opinion polarization in human communities can emerge as a natural consequence of beliefs being interrelated", Entropy, 24(9), 1320, https://doi.org/10.3390/e24091320, 2022. (pdf)
  Summary: This paper introduces a computational model where individual belief systems are structured as networks of coherently connected concepts. It shows how polarization can emerge spontaneously if belief consistency is enforced locally — embodying the human tendency to avoid cognitive dissonance — thus offering a new explanation for widespread ideological divergence.
  
  
• Evelin Berekméri and Anna Zafeiris "Optimal collective decision making: consensus, accuracy and the effects of limited access to information", Scientific Reports, 10:16997, doi.org/10.1038/s41598-020-73853-z, 2020. (pdf)
  Summary: This paper explores the optimal structure of groups which are embedded into an external, observable environment for (i) reaching consensus (ii) having well-informed members, and (iii) for those cases when both aspects are equally important. The groups are characterised by their communication networks and individual properties. We find that the group structures promoting one or the other feature fundamentally differ from each other since having well-informed members requires highly specialised individuals embedded into a structured communication network, while consensus is promoted by non-hierarchical networks in which individuals participate equally. We also find that — contrary to intuition — high access to information calls forth hierarchy, and that suggestibility promotes accuracy, not consensus.
  
  
• Evelin Berekméri, Imre Derényi and Anna Zafeiris Optimal structure of groups under exposure to fake news, Applied Network Science 4:101, 2019. (pdf)
  Summary: Using network simulations, this work identifies how the topology of communication within groups affects their susceptibility to fake news. It finds that specific group structures can buffer better against misinformation than sertain other types, offering insights for communication design and media resilience.
  
  
• Maryam Zamani, Fereshteh Rabbani, Attila Horicsányi, Anna Zafeiris and Tamás Vicsek Differences in structure and dynamics of networks retrieved from dark and public web forums, Physica A, 525, pp. 326-336, 2019. (pdf)
  Summary: By comparing communication networks from dark web and public forums, we uncover significant structural and dynamical differences. The study highlights how anonymity and purpose influence online behavior and network centralization.
  
  
• Anna Zafeiris, Zsombor Komán, Enys Mones and Tamás Vicsek, Phenomenological theory of collective decision-making, Physica A, 479, pp. 287-298, 2017. (pdf)
  Summary: This paper introduces a quantitative framework for optimizing group composition when solving complex, multidimensional problems. It models collective decision-making as a negotiation process over independent sub-problems and shows that the most effective groups combine domain-specific specialists with generalists who possess partial insight into other fields. The counterintuitive result is that optimal performance requires not just expertise, but cross-domain awareness. The model's predictions are supported by empirical citation data, and the approach is adaptable to real-world scenarios involving weighted sub-tasks or social relations.
  
  
• Anna Zafeiris and Tamás Vicsek, Group performance is maximized by hierarchical competence distribution, Nature Communications, 4, Article number: 2484, doi:10.1038/ncomms3484, 2013. (pdf)
  Summary: This paper explores how the interplay between competence and influence shapes group performance in collective decision-making. It shows that optimal group outcomes arise when highly competent individuals act independently, largely ignoring others, while less competent members follow them—a form of structured herd behavior. The findings reveal that a hierarchical alignment of competence and influence is not only natural but also essential for maximizing group effectiveness.
  
  
• Tamás Vicsek and Anna Zafeiris, Collective motion, Physics Reports, 517(3-4), pp. 71-140, 2012. (pdf)
  Summary: This comprehensive review synthesizes decades of research on collective motion in biological and artificial systems—from bird flocks and fish schools to robotic swarms. It covers empirical findings, mathematical models, and simulation frameworks, emphasizing the role of local interaction rules in producing large-scale coordinated behavior. The paper serves as a foundational reference for understanding how simple individual dynamics can give rise to complex group movement.
  
  
• Anna Lázár, Dániel Ábel, Tamás Vicsek, Modularity measure of networks with overlapping communities, EPL (Europhysics Letters), 90, pp. 18001, 2010. (pdf)
  (2nd prize of the European Conference on Complex Systems 2010 "Best Papers Award" - pdf version of the conference paper)
  Summary: In this paper a non-fuzzy measure is being introduced which has been designed to rank the partitions of a network's nodes into overlapping communities. Such a measure can be useful for both quantifying clusters detected by various methods and during finding the overlapping community structure by optimization methods. The theoretical problem referring to the separation of overlapping modules is discussed, and an example for possible applications is given as well.
  
  
• Anna Lázár, Zoltán Vidnyánszky, Tamás Roska, Modeling stimulus-driven attentional selection in dynamic natural scenes, International Journal of Circuit Theory and Applications, 37(1), pp. 3-30, 2009. (pdf)
  Summary: This paper introduces a neuromorphic model of bottom-up visual attention based on a biologically inspired multi-channel retina system. By generating saliency maps for each channel and integrating them into a master map, the model predicts fixation points in dynamic natural scenes with high accuracy. Human eye movement data were used to optimize parameters, revealing that constant channel weighting outperforms adaptive strategies. The system achieves a ~74% prediction success rate - far above chance - demonstrating its potential for real-world vision applications.
  
  
• Kristóf Karacs, Anna Lázár, Róbert Wágner, B. Balint, Tamás Roska, Mihály Szuhaj Bionic Eyeglass: The first prototype A personal navigation device for visually impared. A review,1st International Symposium on Applied Sciences in Biomedical and Communication Technologies, ISABEL 2008, art. no. 4712625
  
  
• Anna Lázár, Tamás Roska, Human Tested Saliency Map Generation in the Bionic Eyeglass Project, Proceedings of The 10th IEEE International Workshop on Cellular Neural Networks and their Applications, pp.91-95, 2006, Istanbul, Turkey. (pdf)
  Summary: A biologically inspired saliency framework designed for assistive vision. This bottom-up attention model leverages mammalian retinal channels and receptive field dynamics to highlight relevant features in visual input. Its parameters were tuned using human testing, demonstrating strong alignment with perceptual salience and practical applicability in the Bionic Eyeglass Project.
  
  
• Tamás Roska, Dávid Bálya, Anna Lázár, Kristóf Karacs, Róbert Wágner, Mihály Szuhaj, System aspects of a bionic eyeglass, Proc. of International Symposium on Circuits and Systems ISCAS, pp. 161-164, 2006, Kos, Greece. (pdf)
  Summary: This work outlines the architecture of a wearable assistive system designed to support blind individuals in different daily environments (home, office, street). It combines spatial-temporal event detection and a biologically inspired retina model to interpret visual input in real-time. The approach integrates analogic and neural-inspired computing methods to recognize context-specific indoor and outdoor events.
  
  
• Kristóf Karacs, Anna Lázár, Róbert Wágner, Dávid Bálya, Tamás Roska, Bionic Eyeglass: an Audio Guide for Visually Impaired, Proceedings of the 1st Biomedical Circuits and Systems Conference, pp. 190-193, 2006, London, UK. (pdf)
  Summary: This study presents an assistive wearable system designed to translate visual input into auditory cues for blind users. It simulates human attention mechanisms by filtering relevant scene elements using a mammalian retina-inspired model and spatial-temporal computing. The system categorizes indoor and outdoor environments and tasks based on user-defined needs, with the goal of enabling intuitive real-time navigation and situational awareness.
  
  
• Dávid Bálya, Anna Lázár, Retinal processing, XI. MITT Kongresszus, 2005, Pecs.
  
  
• Z. Vidnyanszky, G. Kovacs, A. Lazar, Active vision , XI. MITT Kongresszus, 2005, Pecs.
  
  
• Anna Lázár, András Kocsor, An application of ranking methods: retrieving the importance order of decision factors, IEEE International Workshop on Soft Computing Applications SOFA 2005, Szeged, Hungary - Arad, Romania. (pdf)
  Summary: This paper introduces an inverse approach to the Analytic Hierarchy Process (AHP), designed to infer the relative importance of decision factors from observed rankings of human decision makers. Instead of prescribing weights to criteria, the method reconstructs those weights based on preference orderings, making it especially useful in studying human decision behavior. Tested on both synthetic data and a large-scale survey where participants ranked 100 sports, the algorithm achieved high accuracy—correctly identifying 90% of artificial importance orders, demonstrating its potential for applications in psychology, consumer studies, and decision support systems.
  
  
• Anna Lázár, Róbert Wágner, Dávid Bálya, Tamás Roska, Functional representations of retina channels via the refineC retina simulator, Cellular Neural Networks and their Applications. Proceedings of the 8th IEEE international workshop, pp. 333-338, 2004, Budapest. (pdf)
  Summary: Following the recent discovery of the multichannel retinal processing as well as the multilayer CNN models of it, the present paper reports results about some functional characteristics of three channels. In particular, qualitatively different maps of a video clip in the 'Local Edge Detector' channel, the 'Sluggish' channel and the 'On transient' channel have been found. The impacts on making retinal prostheses chips are also discussed.