I was looking at the forest fire model here: https://complexityexplained.github.io/ and thinking that it was truly mesmerizing. This reminded me as well of the experience of watching network graphs self organize.
I think when a complex system is in motion, the brain can't figure out a mechanism of compressing the information, learning the patterns of motion, and hence the brain is fully occupied just tracking the unpredictable.
-=-=-
General feeling so far is the first 3 paragraphs (interactions, emergence, dynamics, self-organization, adaptation, interdisciplinary, methods) are well defined, after that seems to get lost, note how emergence and dynamics share keywords, and self-organization is hard to differentiate from emergence).
Also seems to be about how complex system manifest, not what are the underlying mechanisms (physics) of complex systems.
=-=-
Adaptation is when the higher level system communicates back to the lower level. Without that communication the lower level would never adapt. The challenge is still how to keep the level loosely coupled such that the feedback from the higher level does not dominate the lower level.
=-=-