Complex Systems Summer School 2014

In this work we introduce the new approach of Dynamical Game Theory (DGT). In this approach, individuals perceive their payoffs differently as a function of their previous gain history and intrinsic characteristics. We present the dynamical donation game as a first case in this framework and analyze its corresponding dynamics through analytical and numerical approximations. We find that a high level of cooperation can be achieved and maintained when the dynamical donation game is played. The introduction of dynamical games opens new horizons in the explanation of the evolution of cooperation in social environments.LINK

A multiplex network is an ideal representation for modeling agents that interact in different ways. The present paper develops a dynamic agent-based study of revolutions under authoritarian regimes. We argue that formal and informal social interactions influence the political stance of the population in very different ways, particularly in contexts in which not all political opinions can be freely expressed. To account for this, society is modeled using a multiplex network with two layers: the formal network and the informal one. The formal network is a simple balanced tree that represents the set of professional interactions, with the dictator at the root. The informal network is a Barabasi-Albert network that represents friendship relationships. Every period some agents interact in one of the two networks. After interacting, depending on his position in the formal network and the information he receives, an agent decides whether to support or oppose the dictator. We study the fraction of agents who oppose the dictator as a function of two parameters: the fraction of agents that engage in social interaction p in each time step and the sensitivity of agents to their position in the formal network α. We find a sharp phase transition as p increases above zero, illustrating the importance of social interactions in facilitating revolutions.LINK

It is well known that social norms and peer pressure have a big influence on individual decisions and behavior. How do they change and evolve? How can one affect or modify them? And how can an individual break out of the role imposed by them? In this paper we develop two dynamical theoretical models to study these and other related questions. We model a collection of agents forming beliefs about the optimality of different possible choices or states, which will determine their identity. Our first model is based on excitable linear systems. Each agent updates its disposition for each of the different possible states, as a weighted average of the personal beliefs, its neighbors’ beliefs, and the influence of an external source of information. In the second model agents follow the same principle, but in this case they observe the actions taken by the whole population, and they modify their dispositions following a Bayesian update. In both models agents’ beliefs about the optimality of the different states will determine if they keep their current state or if they move to a new one, changing their identity. These models allow us to study how fast social norms emerge or change, as a function of the strength of the external information and the divergence between agents’ own preferences and the established social norms.LINK

The Bitcoin marketplace provides a unique opportunity for information and social scientists to explore familiar patterns in new light. Trade manias, also often referred to controversially as economic bubbles, have been widely discussed in political-economy. In this paper, we identify moments of transition from sharp increases to sharp drops in the price of Bitcoin and apply network and conversation analyses around them. We isolate our analysis to the four largest peaks in the history of Bitcoin. Our findings illustrate how computationally intensive techniques may uncover signals of emergence of such phenomena in complex social systems.LINK

Processes in which diseases spread and sustain inside of different types of populations were always interesting subjects for epidemiologists. In recent years, ecologists draw the attention to the fact that not only hosts but also the other species in the community can affect the process of disease spread and in that way influence the result of the infection. In nature, we find many parasites and infectious agents with complex life cycles and which can be transmitted in the ecological communities through contact interaction or through feeding interaction. Multiplex networks are layered networks that can represent different types of interactions between agents. In this networks the agents are part of all the layers, but the structure of their interaction is distinct in each layer. We recognise multiplex networks approach as a way in which we can question the importance of the different forms of transmission for the disease spread in the ecosystems. Our results show that for Trypanosoma cruzi, a parasite that can be transmitted through arthropod vectors or through feeding on infected prey, the infection is more widespread when considering both layers in the process. When considering the multiplex structure, species that are not connected through the food web can be affected because of the inclusion of the vectorial layer. The frequency of vectors in the community also influenced the infection spread, increasing the speed of infection in the hosts. We conclude that the multiplex approach is extremely powerful when dealing with different types of interactions and that non-trivial results can be found when the multilayered structure of the process is considered.LINK

We propose the existence of a fundamental tradeoff between division of labor (i.e. cooperation, lack of redundancy, or interconnectedness) and robustness in complex, adaptive systems. It is unlikely that a universal optimum for this tradeoff exists, and we suggest that context-dependent optima are likely tuned by the stability of the environment in which a system evolves. Here, we describe the mechanistic basis for this tradeoff in both natural and social systems, and show how these systems should self-organize under differing levels of environmental noise. We use agent-based modeling to explore the way simple, generalized systems optimize along this tradeoff curve under different perturbation regimes. We then investigate real-world data sets from qualitatively different environments, with differing levels of stability. We also look at more quantitative relationships between environmental disturbances and biological network structure, using microbial mesocosms under controlled laboratory conditions. Finally, we compare structural properties of natural systems to our model outputs. Overall, we find that complex networks maintain a high degree of connectivity under a large range of environmental variability, but the prevalence of high-degree nodes (agents/organisms with large numbers of connections) drops to zero above a noise threshold. Cole Mathis, Arizona State University (US), contributed substantially to this project.LINK

An approach based on Artificial Intelligence is proposed for the design of conscious behavior in artificial agents, with the objective that it can inform theories of consciousness in neuroscience. Points of comparison between neuroscientific theories of consciousness and the artificial intelligence based approach are presented and analyzed. The significance of an interdisciplinary approach to studying consciousness is emphasized by highlighting connections between anthropology, neuroscience, and artificial intelligence.LINK<

Pollination systems are composed of flowering plants and flower visitors, engaging into mutualistic interactions. However, the flower visitors include true pollinators, which pollinate the flower by visiting it through the legitimate way, and also by cheaters, which use the flower0 s resources (e.g. nectar and pollen) without pollinating it or been just marginally efficient on pollination. On the one hand, plants have different flower structures, as shallow and tubular flowers, which can provide some protection agains the cheaters effects or higher efficiency when visited by pollinators. Even though cheaters can damage flowers, there is evidence that cheaters can have a positive effect on the pollination service. In fact, the existence of cheaters decreases the amount of reward provided by plants in a given environment. Therefore, pollinators travel further in order to visit more flowers or even spend a longer time in each flower to collect enough resources. It increases the cross-pollination rate and the pollination success, especially to auto-incompatible plant species. The presence of cheaters in these systems represent a delicate trade-off when mutualistic interactions when cheaters effects are taken into account. In this work, we are interested to understand how pollination systems allow the persistent coexistence of the two types of visitors and plants. We developed a mean field analytical model relying on game theory, with a bipartite network of two kinds of plants (shallow and tubular flowers) and two kinds of visitors (pollinators and cheaters). Our analytical and numerical results confirm the presence of metastable states of persistent coexistence of the above-mentioned visitors and plants. In order to better describe additional real-world features of pollination systems (i.e. the spatial distribution of flowers, the depletion of resources, and the crossing pollination effect) we also implement an agent-based model. In this case, we observed coexistence of the two visitors and two plants when we included the space. We are still studying the agent-based mode approach to understand, for instance, how spatial structures (as the ones resulting from mankind actions) can affect pollination systems.LINK

Electrical communication between cardiomyocytes can be perturbed during arrhythmia, but these perturbations are not captured by conventional electrocardiographic metrics. We developed a theoretical framework to quantify electrical communication using information theory metrics in two-dimensional cell lattice models of cardiac excitation propagation. The time series generated by each cell was coarse-grained to 1 when excited or 0 when resting. The Shannon entropy for each cell was calculated from the time series during four clinically important heart rhythms: normal heartbeat, anatomical reentry, spiral reentry and multiple reentry. We also used mutual information to perform spatial profiling of communication during these cardiac arrhythmias. We found that information sharing between cells was spatially heterogeneous. In addition, cardiac arrhythmia significantly impacted information sharing within the heart. Entropy localized the path of the drifting core of spiral reentry, which could be an optimal target of therapeutic ablation. We conclude that information theory metrics can quantitatively assess electrical communication among cardiomyocytes. The traditional concept of the heart as a functional syncytium sharing electrical information cannot predict altered entropy and information sharing during complex arrhythmia. Information theory metrics may find clinical application in the identification of rhythm-specific treatments which are currently unmet by traditional electrocardiographic techniques.2014. Modelling the heart as a communication system. The Royal Society Interface. DOI: 10.1098/rsif.2014.1201

In this paper, we propose a strategy to control the self-organizing dynamics of the Bak-Tang-Wiesenfeld (BTW) sandpile model on complex networks by allowing some degree of failure tolerance for the nodes and introducing additional active dissipation while taking the risk of possible node damage. We show that the probability for large cascades significantly increases or decreases respectively when the risk for node damage outweighs the active dissipation and when the active dissipation outweighs the risk for node damage. By considering the potential additional risk from node damage, a non-trivial optimal active dissipation control strategy which minimizes the total cost in the system can be obtained. Under some conditions the introduced control strategy can decrease the total cost in the system compared to the uncontrolled model. Moreover, when the probability of damaging a node experiencing failure tolerance is greater than the critical value, then no matter how successful the active dissipation control is, the total cost of the system will have to increase. This critical damage probability can be used as an indicator of the robustness of a network or system.LINK

Network ecology is a rising field of quantitative biology representing ecosystems as complex networks. A suitable example is parasite spreading: several parasites may be transmitted among their hosts through different mechanisms, each one giving rise to a network of interactions. Modelling these networked, ecological interactions at the same time is still an open challenge. We present a novel spatially embedded multiplex network framework for modelling multi-host infection spreading through multiple routes of transmission. Our model is inspired by Trypanosoma cruzi, a parasite transmitted by trophic and vectorial mechanisms. Our ecological network model is represented by a multiplex in which nodes represent species populations interacting through a food web and a parasite contaminative layer at the same time. We modelled Susceptible-Infected dynamics in two different scenarios: a simple theoretical food web and an empirical one. Our simulations in both scenarios show that the infection is more widespread when both the trophic and the contaminative interactions are considered with equal rates. This indicates that trophic and contaminative transmission may have additive effects in real ecosystems. We also find that the ratio of vectors-to-host in the community (i) crucially influences the infection spread, (ii) regulates a percolating phase transition in the rate of parasite transmission and (iii) increases the infection rate in hosts. By immunizing the same fractions of predator and prey populations, we show that the multiplex topology is fundamental in outlining the role that each host species plays in parasite transmission in a given ecosystem. We also show that the multiplex models provide a richer phenomenology in terms of parasite spreading dynamics compared to more limited mono-layer models. Our work opens new challenges and provides new quantitative tools for modelling multi-channel spreading in networked systems.LINK