If one has to search the memory then one has to answer several
questions, for instance (i) Where to start? (ii) How far to go? (iii) Looking
for what? For the first question, where to start, it is helpful to assume that
the agent after its start will automatically activate those nodes in
his memory which have a similarity to his actual consciousness. As activation parameter it is proposed to use the parameter . A
possible definition could be to say:
representing no activation and
representing activation.
These nodes primarily activated by the actual consciousness can activate n-many more nodes.
Clearly there should be only a finite amount of associated activated nodes. All
those activated nodes together represent the activation area in the
memory. Thus, whatever an agent has to search will automatically start within
this activation area. And it can be assumed that the 'usual' search time is
always limited within a time window
inducing a sub space within
memory which induces boundaries within which usually the possible 'content'
will be found for 'remembered content'.
to be continued ....
Gerd Doeben-Henisch 2012-03-31