The Black Swan Thinking Project - Session 2: How The Leopard Didn’t Get Its Spots

The Black Swan Thinking Project - Session 2: How The Leopard Didn’t Get Its Spots

Ruben R. Puentedura, Ph.D.

— Nov 24, 2020

ConverSketch by Karina Branson

My thanks to everyone who came to the second Black Swan Thinking Session. The recording of the session, and Karina Branson's superb graphic summary are linked above.

If you’re just joining the Project, and would like to catch up, the recording and files for the first session are also available on the blog.

Let’s take a moment here to recap some key points of what’s been covered thus far:

  • The System Dynamics topics covered in the first session outline a world in which feedback loops, couplings between systems, and delays in responses can lead to intuition-defying results.
  • At the same time, studying and playing with a set of central System Dynamics Archetypes can help develop new intuitive insights, which can in turn be used to construct new frameworks for action that are less failure-prone in a broader range of scenarios.
  • The topics introduced in the second session add nuance by bringing in the world of the agents that make up systems, with their corresponding introduction of new model behaviors. These new behaviors range from small fluctuations to dramatic deviations from the predictions of simpler models.
  • A key point covered in the discussion portion of the second session is that agent models also allow for a better understanding of real-world scenarios, by providing a basis for differential model comparisons across time and societal contexts. The setup and summary of these discussions can be found in the video above, starting at 43:20.
  • Looking forward, the third session will look at how to collect information and use it to discover connections between agents, and how the resulting networks fill in the final piece of the puzzle needed to circumscribe the unexpected, and allow for the creation of tactical and strategic plans that are robust or antifragile, even when expectations break down.

Here are the links to the toolset and sample models discussed in the second session:

Finally, here are some additional resources on agent-based modeling:

Projects

Ruben R. Puentedura, Ph.D.