Biology Reference
In-Depth Information
FIGURE 27.1
Different scale structures used in epidemic modeling. Circles represent individuals and each color corresponds to a specific stage of
the disease. From left to right: homogeneous mixing, in which individuals are assumed to interact homogeneously with each other at random; social
structure, where people are classified according to demographic information (age, gender, etc.); contact network models, in which the detailed network of
social interactions between individuals provide the possible virus propagation paths; multi-scale models which consider subpopulation coupled by
movements of individuals, while homogeneous mixing is assumed on the lower scale; agent-based models which recreate the movements and interactions
of any single individual on a very detailed scale (a schematic representation of a city is shown).
based on the construction of highly detailed synthetic
societies. Within such a framework computer simulations
acquire a new value and allow on the one hand the creation
of in silico experiment hardly feasible in real systems, and
access to quantity and observables across different models
on the other. This computational thinking approach will be
also the guide to the understanding of typical non-linear
behavior and tipping points not accessible by analytical
means.
Although many basic conceptual questions remain
unresolved, the major challenge in the development of
models able to capture the behavior of large-scale techno-
social systems is their sensitivity and dependence on social
adaptive behavior. In the absence of a stress on the system,
a stationary state is reached in which the feedback between
the social behavior and the environment determines the
details of how the dynamical process of interest plays out.
Social behaviors react, adapt and define new way of
interacting as the dynamics of the system evolves. This
complicates the problem tremendously and clearly shows
the limits in our understanding of human behavior. The
view we have of human mobility, for example, is the daily
normal activity of individuals. In the case of a major event,
such as the spread of a novel pandemic, all the techno-
social systems we are part of can be pushed out of equi-
librium. Under stress individuals can act differently from
usual: they can decide to stay home, to avoid crowded
places and prevent children attending school. In general,
they can take any action to reduce their risk by self-initiated
behavioral changes. Contrary to what happens in physical
systems, the global evolution of the system and our
knowledge of it are part of the system dynamic.
Unfortunately, we are still not able to capture and
deeply understand social adaptation and quantify the
consequent changes on the dynamics of processes that
trigger it. While some of the above issues may find a partial
solution by improving the accuracy and reliability of
models, it is clear that social adaptation to predictions
presents us with new methodological and ethical problems.
Addressing these problems involves tackling three
major scientific challenges. The first is the gathering of
large-scale data on information spread and social reactions.
This is not currently out of reach, thanks to large-scale
mobile communication databases (mobile telephone,
Twitter logs, social web tools) operating at times of specific
disaster or crisis events. Second is the formulation of
formal models that make it possible to quantify the adap-
tation, changes and reactions of individuals as a function of
the dynamic processes occurring in the system. The third
challenge concerns the deployment of monitoring infra-
structures capable of informing computational models in
real time.
NETWORK THINKING
The study of contagion processes is a very active field of
research that crosses different disciplines. Epidemiologists,
computer scientists, and social scientists share a common
interest in studying spreading phenomena, and rely on very
similar models for the description of the diffusion of
viruses, knowledge, and innovation
[19
23]
. All these
processes define a contagion dynamic, which can be an
actual biological pathogen that spreads from host to host, or
a piece of information/knowledge that is transmitted during
social interactions. The connectivity patterns describing the
interactions of individuals as well as their mobility from
place to place are one of the crucial ingredients in the
understanding of contagion phenomena. The importance of
networks in epidemiology need not be stressed here. The
recent advances in the field, however, stem from the
increased ability to gather data on several large sets of
networked structures and populations. For a long time
approaches to human interactions and mobility have relied
mostly on census and survey data on social interactions,
e
