Our Invited Speakers
Jelena Grujić
|
Sandro Meloni
|
Francisco Santos
|
Boyu Zhang
|
Their Abstracts
Jelena Grujić - Network reciprocity, a mechanism of cooperation or not - what do experiments say?
The emergence of cooperation in societies is one of the most important unanswered questions in science at the moment. Numerous theoretical explanations based in Game Theory have been offered and in the last 10 years a number of experimental works started to be show up to test these theories. Here, I will present the results of the series of experiments designed to test network reciprocity. At the end I will present a model, which originate from neuroscience and can explain the reaction times distribution observed in these experiments which gives us a new insight into what is deliberate and what intuitive behaviour.
Sandro Meloni - Adding realism to Evolutionary Game Theory: Individual activity patterns and heterogeneous resource allocation
For more than fifty years Evolutionary Game Theory (EGT) provided an unified theoretical framework to understand biological, social and economic interactions. However, recent experimental evidence highlighted the presence of complex scenarios that cannot be captured by the classical assumptions behind EGT. In this talk I will show how simple modifications to classical social dilemmas can give rise to a plethora of complex behaviours that better describe social and economic systems. In particular, for 2-person games, the presence of more realistic activity patterns can lead to new dynamical phenomena characterized by instability and chaos but, at the same time, preserves the overall evolution of cooperation. For group interactions, instead, I will present a refinement of classical Public Goods Games on networks that is able to reproduce typical traits of economic societies like the Pareto Law for wealth distributions. Finally, I will also demonstrate that such results imply a conceptual change on how we think about cooperation in collective dilemmas.
Francisco Santos - Linking individual and collective behavior in adaptive social networks
One of the greatest challenges of complex systems research and computational social science is to understand how societies behave as a collective, knowing how individuals interact with each other. Social networks offer compelling examples of this, commonly affecting in such a fundamental way the dynamics of the population they support that the global, population-wide behaviour that one observes often bears no relation to the individual processes it stems from. In this talk, I will discuss a way of establishing a link between individual (micro) behaviour and collective (macro) behaviour for coevolutionary processes in adaptive networks. I will resort to the paradigmatic example of cooperation dilemmas showing how adaptive social networks may act to change the dilemma that individuals engage on, revealing a very different behaviour at a global level. In particular, I will discuss how an adaptive network transforms a two-person social dilemma locally faced by individuals into a collective dynamics associated with an N-person coordination game, whose characterisation depends sensitively on the relative time scales between the entangled behavioural and network evolutions. I will also discuss the complex interplay between selection pressure and network topology in the effective dilemmas individuals face. This framework is general enough to be applicable to other processes occurring in static and evolving networks, notably disease spreading and the emergence of non-trivial patterns of peer-influence.
Zhang Boyu - Evolution of fairness on social networks
Experiments on the Ultimatum Game (UG) have shown time and again that people’s behavior is a far cry from what a rational person would do. Several theoretical works have proposed that these results may arise evolutionarily when subjects act in both roles and there is a structure in the population specifying who plays with whom. Here we introduce two networked UG experiments to test the above ideas. In experiment in which subjects play a single role, social networks promote the emergence of communities and diversity but do not affect the average fairness level. In experiment in which subjects play simultaneously both roles, acceptance levels of responders approach rationality, while proposers accommodate their offers to their environment. Furthermore, we find that proposers follow a best-response-like approach to choose their offers, but subjects keep their acceptance levels all the time.
The emergence of cooperation in societies is one of the most important unanswered questions in science at the moment. Numerous theoretical explanations based in Game Theory have been offered and in the last 10 years a number of experimental works started to be show up to test these theories. Here, I will present the results of the series of experiments designed to test network reciprocity. At the end I will present a model, which originate from neuroscience and can explain the reaction times distribution observed in these experiments which gives us a new insight into what is deliberate and what intuitive behaviour.
Sandro Meloni - Adding realism to Evolutionary Game Theory: Individual activity patterns and heterogeneous resource allocation
For more than fifty years Evolutionary Game Theory (EGT) provided an unified theoretical framework to understand biological, social and economic interactions. However, recent experimental evidence highlighted the presence of complex scenarios that cannot be captured by the classical assumptions behind EGT. In this talk I will show how simple modifications to classical social dilemmas can give rise to a plethora of complex behaviours that better describe social and economic systems. In particular, for 2-person games, the presence of more realistic activity patterns can lead to new dynamical phenomena characterized by instability and chaos but, at the same time, preserves the overall evolution of cooperation. For group interactions, instead, I will present a refinement of classical Public Goods Games on networks that is able to reproduce typical traits of economic societies like the Pareto Law for wealth distributions. Finally, I will also demonstrate that such results imply a conceptual change on how we think about cooperation in collective dilemmas.
Francisco Santos - Linking individual and collective behavior in adaptive social networks
One of the greatest challenges of complex systems research and computational social science is to understand how societies behave as a collective, knowing how individuals interact with each other. Social networks offer compelling examples of this, commonly affecting in such a fundamental way the dynamics of the population they support that the global, population-wide behaviour that one observes often bears no relation to the individual processes it stems from. In this talk, I will discuss a way of establishing a link between individual (micro) behaviour and collective (macro) behaviour for coevolutionary processes in adaptive networks. I will resort to the paradigmatic example of cooperation dilemmas showing how adaptive social networks may act to change the dilemma that individuals engage on, revealing a very different behaviour at a global level. In particular, I will discuss how an adaptive network transforms a two-person social dilemma locally faced by individuals into a collective dynamics associated with an N-person coordination game, whose characterisation depends sensitively on the relative time scales between the entangled behavioural and network evolutions. I will also discuss the complex interplay between selection pressure and network topology in the effective dilemmas individuals face. This framework is general enough to be applicable to other processes occurring in static and evolving networks, notably disease spreading and the emergence of non-trivial patterns of peer-influence.
Zhang Boyu - Evolution of fairness on social networks
Experiments on the Ultimatum Game (UG) have shown time and again that people’s behavior is a far cry from what a rational person would do. Several theoretical works have proposed that these results may arise evolutionarily when subjects act in both roles and there is a structure in the population specifying who plays with whom. Here we introduce two networked UG experiments to test the above ideas. In experiment in which subjects play a single role, social networks promote the emergence of communities and diversity but do not affect the average fairness level. In experiment in which subjects play simultaneously both roles, acceptance levels of responders approach rationality, while proposers accommodate their offers to their environment. Furthermore, we find that proposers follow a best-response-like approach to choose their offers, but subjects keep their acceptance levels all the time.
