[dao-forum] schema theory

[Kenneth Fields <ken@xxxxxxxxxxx>]

> This is understandable... I wouldnt move my communication to dedicated
> off-ground field also... DAO should find a way to link and cross reference
> such threads on lists without interfiering with dynamics.

This is interesting... I was thinking of pasting some listserv excerpts into criticalware's liki - http://www.criticalartware.net/ . See how its processed. Regarding copyright, maybe you could call this, DJing, or discourse jockey. Using theory loops.

Why go through all that trouble you may ask... we could just take every word we see (re digital arts), any media books index (The Media Reader) and start hierarchicalizing. Well,  I'm not sure, maybe I'm just driven to see the emergent, rather then a  quick hack. We'll see how it goes.


> Unfortunatly there is clear mix of different approaches - same taking more
> context driven labeling system, some more technology, few more
> semiotic/aesthetic/economic(even)... but rather than deciding on best (who
> would ever manage to inplement this decision anyway)... it would be sane
> to list them and see where are +/- of all.

Wordnet uses the psycholexicological (nice word) approach. The basis is synonym sets or more fundamentally schema or prototype theory. It's intuitive and not a controversial theory at this point in time. Its a good read if you have the time.

Speaking of which, I just found this old paper of mine. I'll quote a section on schema theory - just for the record. There's some references at the end which might be interesting.

Regularity and Randomness

Generally, a CAS gathers information about itself and its own surroundings. Perceived regularities are identified in this experience and stored (compressed) in schemas; randomness is filtered though not discarded (implicit memory). There are three important terms in the equation of schema: pattern, randomness, and effective complexity. A continuum exists from completely predictable pattern (minimal complexity), to complete randomness (also minimal complexity). Effective complexity occurs between these two terms; it differs from crude complexity in that effective complexity is only the identification of abstracted regularities not the complete compressed description.

Douglas Hofstadter with the Fluid Analogies Research Group (FARG), has done work in the area of sequence abstraction or extrapolation. He describes the essence of intelligence as pattern-finding, and points to other work in this field: Simon & Lotovsky, Pivar & Finkelstein, and Persson. This research, an AI approach, works on strategies for controlling a search, reduction of a sequence to simpler sequences, general and expert knowledge, and esthetics-driven perception (he describes mathematical discoveries as being driven by a search for elegance or consistency in concept integration). One individual may be more sensitive to pattern recognition then the next, expertise aside. The issue is how to generalize or conceive rules for deriving schema or representation.

Schema

Bartlett’s theory, that memory is constructive, showed by the recalling of stories that we instantiate a schema that represents novel information in terms of an existing conceptual organization. The result is that all sensation is assimilated through schema. Through Piaget and Bartlett, this became an element of cognitive theory. A problem with early formulations of schema theory was the lack of consideration of the adaptive/dynamic nature between representation and world.

Piaget was first to investigate experimentally how schema are formed and changed. He proposed the term accommodation to account for uneasily assimilated new experience, restructuring schemata to better account for experience. Rumelhart went in the direction of an connectionist, object-oriented organization, where the schema have default values, input variables, and can be recursive -- historically embedded (usurped schema remain with some activation and may have a nonlinear effect in producing useful variation later). In addition, he introduced the concept of generativity (or restructuring), that is, simple modification of existing schema (Eysenck).

Direct experience is problematic regarding issues of realism vs constructionism. Therefore Ben Martin discusses the issue of internalization as “encoding regularities in the world in terms of the internal states of a system (e.g., the mind).” This is an important point for situating the experience of a CAS in terms of open systems vs closed systems (interesting discussion in Maturana and Varela). Current thinking in this regard is Kantian (the origination of the word schema) in that experience is direct “precisely because [time, space, and causation] are not observable in the world but are rather a property of apperception itself” (Martin).
 

 “Schemata and the organization among them determines their representational
 character and capacity. This is a critical feature of schemata: they exist in a system   
  constrained to certain possible states and so can only represent the world in a way
 that is consistent with the organizational principles of that system” (Martin).
 
Going one physiological step further, Patricia Smith Churchland, predicts that the brain does not cause consciousness, the brain is consciousness: “Certain patterns of brain activity are the reality behind the experience.” This is simply a deduction however, pointing out the fact that “electricity is not caused by moving electrons, it is moving electrons; genes are not caused by base pairs in DNA, they are base pairs; temperature is not caused by mean molecular kinetic energy, it is mean molecular kinetic energy.” Thus, there may be a perception of the brain’s state which is the schema.

Variation of Schemata

Schank introduced a dynamic memory theory (scientific theories are themselves a perfect example of adaptive schema), breaking schema into Memory Organization Packets (MOPS), and higher level Thematic Organizational Points (TOPS), elemental units which can be combined and recombined. Dynamic memory theory was an attempt to introduce more flexibility into his earlier script theory, but did not “fulfill the intuitive flexibility of the schematic approach” (Eysenck).

Schemas guide action which effect the environment which feeds back into the organism giving rise to competing schema. In this process mutation or adaptation occurs, referred to as tuning by Rumelhart. He developed a connectionist approach which could physically explain how this dynamic behavior might happen in the brain. This is envisioned as a large data structure with simple processing units (modules, nodes) which “interact to produce a global representational state” (Martin). A connectionist network can represent environmental information by altering weights so as to change the stable state of the system; sub-networks (sub-schema) within the global network can be utilized individually or be combined with other modules. A further modification in the connectionist model can be made by referring to active/weighted modules as “basins of attraction” (Martin).

It is proposed that an evolutionary selective process among schema is suited to the continued existence of the organismic host. Conversely, there are occasions when non-adaptive or maladapted schemas confer advantage. A resistant, nonadaptive tendency is akin to the physical property of maintenance or stability of state. It may also be that misperception can also act as an informational filter that benefits attention.


Reference.

Bak, Per. Self-Organized Criticality, in Complexity: Metaphors, Models, and Reality;
 (Cowan, Pines, and Meltzer, Editors). Santa Fe Institute. Addison-Wesley Pub. 1994.

Churchland, P.S. Can Neurobiology Teach Us Anything About consciousness, in The Mind,
 the Brain, and Complex Adaptive Systems; (Morowitz and Singer, Editors).
 Santa Fe Institute Studies in Complexity Proceedings. Addison-Wesley Pub. 1995.

Eysenck and Keane. Cognitive Psychology. Psychology Press. 1995.

Gell-Mann, Murray. Quark and the Jaguar. Freeman. 1994.

Gell-Mann, Murray.  The Mind, the Brain, and Complex Adaptive Systems    (above).

Hofstadter, Douglas. Fluid Concepts and Creative Analogies. Basic Books. 1995.

Holland, John. Echoing Emergence, in Complexity: Metaphors, Models, and Reality (above).

Kihlstrom, J. F. The Rediscovery of the Unconscious, in The Mind, the Brain, and CAS
 (above).

Martin, Ben. The Schema, in The Mind, The Brain, and Complex Adaptive Systems (above).

Maturana, Humberto. Autopoiesis and Cognition. Reidel Pub. 1972.

Stevens, Charles. Conplexity of Brain Circuits, in The Mind, the Brain, and Complex Adaptive   Systems (above).

Varela, Francisco. Principles of Biological Autonomy. North Holland. 1979.  

 _________________________________________________________________
Message Archives: http://dao.cim3.net/forum/dao-forum/
Subscribe/Unsubscribe/Config: http://dao.cim3.net/mailman/listinfo/dao-forum/
Shared Files: http://dao.cim3.net/file/
Community Wiki: http://dao.cim3.net/wiki/
Community Portal: http://dao.cim3.net/
To Post: mailto:dao-forum@dao.cim3.net    (01)