Description of the Symbol Level
Cognitive architectures require some sort
of internal representation in order to achieve Turing completeness; that
is, the ability to act as a universal machine. In other words, for
knowledge to be used explicitly, it must be represented in some way.
Such a representation is necessarily symbolic. Symbol systems (which
include cognitive architectures) have the following properties related to
the symbols they manipulate:
- Memory: a way to store symbols
- Symbols: patterns of (distal) representation
- Operations: creation, modification, reproduction and destruction of
symbols
- Interpretation: transformations of symbols via operators
- Capacities: sufficiency and (universal) completeness of the system
Such a system produces through time a slowly changing collection of
symbolic structures. The symbolic structures available at any one time
are necessarily only a subset of the possible symbolic representations and
the world such symbols represent.
One of the fundamental tenets of
artificial intelligence is that a physical
symbol system has the necessary and sufficient means for general
intelligence. This is known as the physical symbol system hypothesis.
Symbols represent knowledge. Thus, above at levels above the symbol
(or architecture) level, knowledge may mediate behavior. This level is
known as the knowledge level.
Newell characterizes the symbol level in
humans as the
cognitive band.
Press
UP to go to the list of theories.
Press
HOME to go to the Table of Contents.