Wonderful, do you have a diagram?
~PM
(thinking in pictures...)
Date: Sun, 4 May 2014 22:30:26 -0700
Subject: Re: [agi] Situation Induction
From: [email protected]
To: [email protected]
Here is something I'm working on relevant to the discussion of
situations. It is not yet at point where it can be demonstrated,
but here's an abstract description that may present a unique and
unexplored perspective on situations. What is proposed is a
bicameral architecture of situated and embodied semiotic
control. Embodied domain state control is constrained by
situational control. At the embodied control level, effects are
adjustments of state, led by sentiments (positive and negative
preference) considered as Peirce's thirdness. The same is true at
the situational level (which constrains the embodied sentic
level) effects in the space of situations are led by purposes.
Embodiment
1. Control domain
2. Effect
3. Feel w/rt Effect
Situation
4. Situational domain
5. Effect upon situational domain (decision)
6. Purpose w/rt Effect
Peirce's modes are greatly illuminated when cast in terms of
Perceptual Control Theory, which I think can be generalized as a
Semiotic Control Theory. This may be a novel finding, but it
seems more likely that I just didn't realize what Peirce meant
until I thought of his work in terms of PCT. The main insight
that I think PCT brings to the Peircean modes is that qualitative
categories can be derived from purely enactive hierarchical
control. In other words, for example "feeling" is a thirdness
phenomenon strictly because it is an instance of directing the
selection of effect (2ndness-level) toward a reference condition,
established by the organism as a control system. Similarity an
"agreement" belongs to the situational domain of 4thness because
it it establishes parameters constraining 3rdness-level activity.
Again, this seems so obvious now that I am working with the
ideas, that it is likely an understanding shared by others as
well.
An interesting fact about this is that it produces an
intriguingly natural definition of purpose in the context of
control systems. First off, purpose is about controlling effects
in a situational domain. This leads to the more interesting
observation that purpose is about those signs that designate
collectively or individually the direction of changes in some
situation domain that in terms of the control system are
favorable. I suppose "goal" and "purpose" are fairly
interchangeable here, but "purpose" seems to more cleanly
accentuate the easily intersubjective nature of signs acting as
purposive operations.
AGI | Archives
| Modify
Your Subscription
-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com
