My 2010 research paper for Cosmos & History is just published. It describes how scientists can guide their models of natural systems to change along with the evolutionary changes in the systems being modeled, “Models Learning Change” (PDF).
The classic case of systems that people in them will need to change their models for, is that of growth economies. Their growth is a process of rapidly evolving organizational development, with many natural limits as an environmental system.
Ways to anticipate when the ‘=’ signs of a model should be again considered as ‘?’ marks are discussed. A change the model is called for, for example, at the point when a growth economy’s environment becomes increasingly unresponsive to increasing investment.
If a model does not duplicate that important reversal in the sign of the environment’s responsiveness, switching from a prior consistently positive to negative response, it may guide policy makers to accelerate investment and so drive the whole system toward more rapidly diminishing and then vanishing returns. If the model changes with the new direction in the environmental response, the optimal path would reorient the system toward stabilizing investment at a peak level of returns and stable rates of change, rather than increasing instability.
I have a 2009 essay in Cosmos & History too, also on the organizational development processes of natural systems, Life’s Hidden Resources for Learning (PDF), discussing environmental systems that exhibit active learning processes in the organizational development of the whole, guided by the active learning of their living parts as they explore their own environments and relationships.
The journal Cosmos & History publishes research and essays on the philosophy of nature, edited by Arran Gare. His view is well exhibited in his recent Tribute to Brian Goodwin and essay on the history of modern philosophy, Philosophical Anthropology, Ethics and Political Philosophy in an Age of Impending Catastrophe
See also PFH publication list