Random Variations to Natural Information Stores

Random Variations to Natural Information Stores
Variations to natural information stores occur randomly. Random genetic
variation mechanisms are well known. Mutation and sexual recombination
result in random variations and without these mechanisms, no natural
alterations to a genetic code would occur. Barring deliberate human action,
there is no other mechanism available. Similarly, and as indicated earlier,
barring knowledge held in long-term memory indicating which moves to
make when faced with a problem, moves can only be generated randomly as
indicated on the left side of the elements combinations continuum of the
matrix of continua. Until the knowledge base can be brought into play
allowing movement to the right side of the elements combinations
continuum, move generation is necessarily random, just as mutation and
genetic recombination are random. Material deliberately intended to have
an educative function provides the only exception to these mechanisms.
Education techniques can reduce or eliminate the random generation of
problem-solving moves (see later), just as the deliberate alteration of a
genetic code substitutes for the random variations due to mutation and
genetic recombination.
Both the historical reasons for and the consequences of the concept
of random variations to natural information stores need to be carefully
noted. Random variation was required to explain the evolution of
species through natural selection without a guiding intelligence and provides
one of the major functions of the theory of evolution. In other words,
evolution by natural selection does not have a ‘‘central executive’’ to guide
the process. Indeed, in the many theologically motivated debates concerning
the theory of evolution, there appears to have been a tacit consensus that no
238 John Swellernatural, as opposed to supernatural, candidate for an intelligence guiding
the evolution of species was available. All of the ‘‘intelligence’’ of the system
resides in genes. A requirement for a second intelligence to guide the manner
in which genes evolve would require a third to guide the second and so on,
resulting in an infinite regress. Random mutation and natural selection act
as substitutes for an additional intelligence.
One purpose of this chapter is to suggest that human cognitive
architecture similarly has no natural intelligence in the form of a central
executive guiding the generation of novel procedures. There is a natural
intelligence in the form of schemas held in long-term memory that
guide previously learned procedures that have been established as eVective.
Those schemas govern the vast bulk of human behavior, including
determining what new material should and should not be learned. As
indicated previously and as is the case for evolution by natural selection,
that stored information incorporates intelligence. An additional intelligence
(or central executive) would require an infinite regress to function. When
schematic knowledge held in long-term memory is not available or when
guidance from other humans who hold such knowledge is not available,
only random selection of mental actions is possible. Of course, knowledge
gained from those randomly selected mental actions can be retained in
long-term memory, which ensures that subsequent actions are intelligent
rather than random. Analogously, genetic codes provide a natural intelli-
gence to guide the continuation of a successful species. When suitable codes
are not available, random mutations determine which codes will be oVered
to the environment for testing as part of the processes of natural selection