two words:

prisoners dilemma

I'll be honest, Im new to this stuff. Like, this post is the first step. I had to re-read things a couple times to decode them for myself, to get a handle on what was being talked about.

In discussion here, I think I will quite often resort to lay-person vocab, so I apologize if that slows things down.

To clarify, Im trying to understand - are you are making the connection that the 'available act-types, potential pathways through that dynamical self-organization' Juarrero refers to be 'the concept of self-organization ... abstracted from modeled or observed instances'?

If you are, then you are, on one hand, making assumptions about the evolving process of Neuro-activity. And then working to apply those assumptions to systems outside neuro-activity?

Im still trying to comprehend the question at hand.

I think that 'available act-types, potential pathways through that dynamical self-organization' applies very wonderfully to examples of plant growth through a biological community. ie: non-neural growth based on genetics, resource availability, and physical limitation.

is that a base comment? am I stating the obvious?

I've been studying these matters on my own account for some time now, and have been able to recognize this apparently premature tendency to impose the notions of systems theory to several situations which, after some fashion, resembled structures/behaviors/developments commonly seen/described in systems theory.

There are two sides to the question you posed:
1) To what extent is it warranted that the models that proved useful in systems theory will apply to the situation/context at hand?

2) How can I demonstrate if such a partial (and useful) mapping exists between the theory and practical situation?

Both seem to be alike, but they are radically different.

The first has to do with what perceptions/cognitive mechanisms or categories led me to believe that such and such notion from systems theory could be brought to a particular problem.

The second has to do much more with how clearly defined/recognizable are the criteria for determining/recognizing if the notions brought/overlayed proved useful in the situation.

The first question is purely intuitive, and it has a lot to do with the particular way in which you acquired and practiced the notions of systems theory while you were learning them. The situation that seems fit for the application of such general principles will most likely appear to be so because of your perceptual capabilities/limitations, in a very general sense. If I learned that a chaotic systems, even if unpredictable with a comfortable amount of detail, eventually founds itself moving within the bounds of a certain region in a vector space, this will apparently lead me to believe that chaotic systems are not that chaotic at all, even though I cannot predict their behavior accurately.

Would applying this metaphor to any seemingly unpredictable system be irresponsible/unwarranted? Perhaps the person who learned them now considers useful to see his/her family as such a chaotic system, and will notice that, not surprisingly enough, her family, even though it presents certain 'random' behavioral fluctuations, never actually traspasses/goes beyond them. So the matter now would be finding those 'attractor' regions, what are their characteristics, how are they constituted, what provides them with stability and under what conditions would this region would morph or disappear, etc.

The analogy is unwarranted, and the very few categories that systems/complexity theory and the family seem to share, are the ones that the person recognized and used as anchors and points of reference for sustaining/justifying the imposed identity. Not much more can be said besides this as to what connects one thing with another. This leads me to the second question.

How can I determine if the model, through its apparent similitude with the current situation, will serve as a tool for prediction and further understanding?
One can move along the ontological dimensions that this problem poses, or one can choose to be pragmatic about it. Of course between this two extreme positions there are countless shades of gray.

If the person, by seeing his/her family as a chaotic system, is able to transport notions/tools from complexity theory, implement them (to the extent such a thing is possible) and monitor/evaluate the results, what does this say about the warranty of identity between the systems? That, if such an identity is not entirely demonstrable/warranted, you can still use the model and derive some benefit from it.

What is the likelihood that the model will eventually be found wanting as more and more situations do not conform/comply with its predictions? It is high. This is of no importance, because such is the destiny of a model, any model: to be eternally insufficient, to be in constant reevaluation and redefinition, to constantly lack the variables/correlations that convey precision in measurement and prediction in the future.

Premises about the world are built, premises tumble down. What remains constant is our ability to generate such models, and fine-tune them to better reflect reality, which, in pretty simple terms means to encompass those recent situations which the model didn't contemplate.

I apologize if my reply was too long; it's just that you posed an interesting question and, for some time, I had an episode similar to the one that you recognized in the philosopher.

It would be great to hear your thoughts on the matter.