Rosennean ideas and better modelling...

[Judith Rosen <***>]

Hi Folks,

I'm formatting this email in plain text rather than HTML so that Athel can
read it without having to go through strange contortions. (By the way, my
reason for preferring "rich text", which is html, is that it gives me far
more options in my writing, such that my writing is more expressive than the
words themselves can be. In plain text all I can do it capitalize to create
emphasis. In rich text, I have options like bold, italics, underlining,
enlarging some or all of the text or making it smaller, changing the font on
some or all of the text, etc.... or some variation of two or more of those
options. In short, I can put my "voice" into my writing. Remember, I'm a
writier-- that's what I DO, most of my professional time. So I'm conditioned
to prefer formats that give me more choices as a writer whether I'm
expressing ideas, thoughts, reactions, imagination, etc.).

Anyway, I have been working on a paper with John K. about Anticipatory
Systems and Time, among other things, and an insight came to me yesterday
that impacts directly the discussion we've been having on the list about
building better models. I decided to share it with the list rather than
hoard this insight to myself. So, here goes:

If it is true that an organism evolves with the aspects of its environment
built in to its own organization (which has tobe the case;... Fish extract
oxygen from water and mammals extract it from air: this reflects certain
conditions of the environment in which these organisms evolved. Thus, their
entire physiology organized around the behaviors of their evolutionary
environments)..... and..... If all organisms have an 'internal predictive
model' as my father put it, which directs the behavior of each organism
according to a predicted future condition in its environment (such as
seasonal changes in weather), THEN the definition of an Anticipatory System
is one in which it's not just the future acting as a causal agent on the
organism in the present but that the past is acting as a causal agent in
many different ways, including interacting with the future in a certain
sense. Here's why I say that: In all organisms, it's true that the past is a
causal agent in the ordinary sense. But, in anticipatory behavior there is
evidence that the past also acts as a causal agent in the present and future
in extraordinary ways-- in which the past cycles of the environment THROUGH
TIME are encoded somehow into the organization of the living system
(organism). I think that this is what constitutes the 'internal predictive
model' my father was referring to! It is how a single celled organism or a
tree "predicts the future". It has the past cycles organized into it in some
way and it isn't just the environmental conditions that an organism evolves
to fit into, but the environmental conditions through time.

Furthermore, it is not just the range of the changes in the environment the
system evolved in, but also the SEQUENCE THROUGH TIME of the changes that
are encoded into the organization of each species of organism. A maple tree
can be killed by the very same cold temperatures that it easily survives in
winter, if it happens out of sequence according to its "model". The model is
not just a predictive model, it is also an evolutionary model. The
evolutionary past predicts the future. This is why the organism can be
"fooled" by human intervention. It's model doesn't adapt. It can only
evolve. Organisms can evolve but they can also sometimes adapt as well. This
is what humans do when we want to live outside our environment (outer space,
under water, etc): We adapt by changing the immediate environment to suit
our model.

If we want to create better models of biological systems, we need to include
this aspect of their behavior (the anticipatory aspect). In order to do
that, there are certain aspects that need to be included in the model/s we
make of any given biological system: 1) We need to include that system's
evolutionary context. 2 ) We need to include how that context behaves over
time. And 3 ) We need to keep the sequences correct. The sequence is every
bit as crucial as the range of variation in conditions caused by external
changes (like rotation of the planet, orbit around the sun, etc). All of
these things play a role in how biological systems organize themselves

The more I manipulate these ideas, the more seems to make itself apparent.
Fascinating!

Judith