Santa Fe Institute Collaboration Platform

COMPLEX TIME: Adaptation, Aging, & Arrow of Time

Get Involved!
Contact: Amy P. Chen, Program Manager,

Dynamic Multi-System Resilience in Human Aging/Ecology for doctors: system dynamics models as a tool to understand observed behavior

From Complex Time

November 13, 2018
9:40 am - 10:20 am


Ingrid vd Leemput (Wageningen Univ.)


There is common ground in analysing the health state of human beings and the state of ecosystems, especially in the need to identify conditions that dispose a system to be knocked from seeming stability into another, undesired state. In ecology, relatively simple system dynamics models have proven to be valuable to understand such dynamics. Examples include lakes that unexpectedly shift from a clear to a turbid state, or coral reefs that are suddenly overgrown by macroalgae. In this talk, I will introduce you to the world of dynamical models, and provide examples how they have proven their value in ecology. Based on that, I will discuss assumptions, benefits and limitations. Then, I will explain how the analysis of bifurcations has led to the development of Dynamic Indicators of Resilience (DIORs). Finally, I will make a bridge from ecology to health, and point to some open questions in relation to DIORs, networks, and oscillating dynamics.

Presentation file(s)
Download Presentation (Delete)
Related files

Post-meeting Reflection

Ingrid vd Leemput (Wageningen Univ.) Link to the source page

The workshop was a good mix of theoretical and practical insights on resilience in human aging. I see a lot of parallels with the work we do on ecological resilience, and I think we could learn a lot from each other. One of the things I noticed is the different definitions of resilience used (and other semantics), which may become confusing. It would be good to make an overview of the different definitions, and to not invent new words for the same concept.

To really get a proper understanding, and to develop well-grounded indicators of the system dynamics and resilience, a combination of the different presented methods would make a lot of sense. This ideal path in my opinion would be 1) for each 'sub-system', to develop a solid idea of the large-scale feedbacks and dynamics, and develop a mechanistic model based on that, taking into account temporal and spatial scale. This should lead to some idea on the stability, and dynamics of the particular sub-system: is the sub-system expected to have alternative stable states/ tipping points/ oscillating dynamics/ chaos/ flickering/ spiraling? 2) This basic understanding should lead to hypotheses on what type of indicators of resilience could be useful and realistic (e.g. perform stress tests, measure DIORs, potential analysis etc) 3) These hypotheses should be tested both in the field, and in more realistic, fully parameterized models, as presented by Alfons Hoekstra.

I enjoyed the talk by Porter a lot. On one hand, it touched me personally, to see how resilience these communities can be after so much suffering, but also it reminded to think about what makes a system more resilient? We talked about feedbacks a lot, but not so much about other relevant factors, such as functional redundancy, response diversity, and connectivity. A vary obvious example of functional redundancy is, I think, the two kidneys a human being has (you can survive without one).

In my opinion, the mouse models, while a mouse is not a human being, can be extremely useful to get a grip on the coupling between subsystems, and the way we could approach the resilience questions in human beings.

The last discussion was interesting, because Sanne Gijzel pointed to an example case, in which several sub-systems failed in a row. I think we can learn a lot from these type of examples (also the example of Heather Whitson) about the coupling between subsystems.

Reference Material