Geoscience Reference
In-Depth Information
a complete manner from the point of view of attributes and statistical relationships,
whereas in the second case, the agents are in general much more stylized. However,
in this latter case, the mechanisms of change are explicitly approached following a
generative principle [EPS 07]. That is to say, as it has been pointed out above, the
model makes explicit the rules which allow “generating” the structure to which the
system results. In order to characterize the structure of the system at the final step,
the concept of
resultant
would, thus, be more appropriate in the first case and that of
emergence
in the second case.
Finally, the relationship with the empirical reality also differs from one
application to the other using MAS. Therefore, the literature confronts the models
known as
Keep It Simple, Stupid (KISS)
, simple and stylized (Schelling's model is a
typical example), to the models known as
Keep It Descriptive, Stupid (KIDS)
favoring a more inductive approach mobilizing all the factors involved in the
evolution of a phenomenon (the example of the Pueblo in the Mesa Verde presented
earlier would be an example) [AXE 97, EDM 05, BAN 13]. Here again, we prefer to
emphasize the reciprocity of the two perspectives, one that allows us to bring
forward the generic mechanisms in the change, and the other to give their place to
the complex interactions operating in the empirical world.
