To vaccinate or not to vaccinate? A coevolutionary dilemma
Held by Dr. Alessio cardillo (Ecole Polytechnique Federale de
24.11.2016, 14:15 Uhr
Vaccination is considered the most efficient way to prevent the spreading of an infectious disease; and, through the years, public health stakeholders have fostered the adoption of proactive vaccination attitudes. Despite such tremendous efforts, we are witnessing the emergence of growing anti-vaccination movements in most of the more economically developed countries. The voluntary decision of getting vaccinated can be thought as an act of cooperation since it implies the payment of some costs to provide benefits for the whole population. On the other hand, free-riders decide to avoid the payment and assume the risk of getting exposed to the infection. However, if all the members of the population except one vaccinate, we end up in a case in which the free-rider benefits of herd immunization effects; thus placing himself in a position of advantage with respect to the rest of the population. Such scenario gives rise to a social dilemma that can be studied within the framework of evolutionary game theory. Both the emergence of cooperation and the spreading of a disease have been studied for years separately in populations where interactions are encoded by complex networks. However, a coevolutionary approach where the two processes are intertwined constitutes a better description of the vaccination issue rather than the one based on the two processes independently. In this talk, I will present a coevolutionary vaccination model to mimic the adoption of vaccine in the case of an influenza-like disease. In particular, the agents decide whether getting vaccinated for the next season, or not, according to the experience they – and their neighbours -- have accumulated in the previous season. Such experience depends on the dynamical outcome of a spreading process based on the SEIR compartmental model. Assuming that the contexts through which the decision and the spreading take place are not the same, one may decide to represent the structure of the interactions by means of a multiplex network; where each layer is associated to a distinct dynamical process. More specifically, I consider a duplex network (i.e. a multiplex made of only two layers) where the spreading takes place on a Random Geometric Graph (RGG) to account for face-to-face contact interactions; the evolutionary game, instead, takes place on a Barabasi Albert Scale-Free (SF) network to mimic social interactions. Finally, to account for the existence of people strongly adverse to vaccination, whose decision cannot be altered even in presence of compelling evidence, I consider a fraction, Z, of zealots agents and study the response of the system to the presence of such individuals.