Re: cognitive != observable? (project three)

I don't see how you can talk about babies "learning to hear";
the startle reflex for loud noises is present at birth and
probably in the womb.

Similarly, almost certainly the ability to detect patterns in
what is heard must be inate, for without it there would be no
mechanism by which to learn it, just as the ability to produce
sounds in a repertory of patterns, which can then be reinforced,
is apparently inate.

The "learning" begins with matching those patterns to other
events in the real world, both on input and on output.

A baby that can say "da da da" is massively reinforced in
an environment where the parent wants the baby to say "dada",
and so it begins, but the ability to connect reinforcement
to pattern must pre-exist or no learning could take place.

xanthian.

[It is interesting that though a mockingbird learns the songs
of other birds throughout its life, the one I was raising
could not be taught the first few bars of "Listen to the
Mockingbird" for all my trying.  Apparently my whistling is
not within the set of patterns its inate recognition covers.]


-- 
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG