Re: sensitive period in language acquisition?

From: Marc Verhaegen
Message: 901
Date: 2000-01-13

>>>> Marc said: Some believe the fetus in utero already becomes accustomed
to its mother's sounds. Since about 3 months of age there are "dialogs"
between mother & child. But the first "words" appear at ca.1 year. This may
be called language already. What do you mean by template? Of course we have
brains that easily learn language (or better: speech -- chimps can perhaps
learn some sort of ASL but not to speak).

>>> By template I mean a predesign or a map that assists the infant in
learning his/her native tongue. I don't know how to scientifically prove
this notion, but I do know that a native born speaker learns his native
language with facility while a child age 1 or so would experience some
difficulty in learning the same language as a second language. My question
is why does this occur. Gerry

IMO we have what you could call templates for language in general, but not
for a specific language. IOW Russians have the same templates as Bushmen,
though perhaps there are small differences. If a second language is more
difficult, it's because it's learnt at a later age I think.

>> Hi Gerry: It's been 25 years since I got a BA degree in Pysch, or even
done any reading in the field, but I believe the basics haven't changed that
much in the physiology of learning theorey. One of the distinctions between
humans and other animals is the almost total lack of behavorial templates in
humans. Behaviorial templates in animals are called instincts, and beyond a
fear of heights, and quick sounds we are without them. The human
brain is very much like a new computer, just waiting to be programmed. But
the big difference between humans and computers is that are memories don't
get erased. What a memory is, is not fully understood. But we are pretty
certain that it involves the development of neural junctions called
synapses. This development involves the transmission of neural conductors
that actually change the nature of the synapse - in a very real sense the
brain, by thinking, experiencing etc -is altered. The more a neural pathway
is used (the same synapses is used) the more conducive that pathway becomes
(the faster it will transmit). In a sense we literally become predisposed to
thinking in certain ways. (There were a few other theories floating around
about field theories and gestalt netways some 25 years ago - but none of
them suggested anything like a inbuilt-template.) When we just start out,
almost nothing is programmed - beyond certain capabilities to recognise
certain visual stimuli (movement, edges, possibly circles etc) that have to
do with how receptors and inhibitor neurons interact with the synapses.
There are no inherent neural templates related to learning specific
languages. An infant learns fast because there is nothing in the way. Us
seniors can have a harder time at learning such things as language - because
our behavior as become so ingrained and most of us don't go to the trouble
to challenge it (burn in new neural pathways). Think about it. You have
been here in the USA for a long time. Ever hear of anyone saying that babies
born of English descended ancestry have a much easier time of learning
English than someone who is born of Greek, or French, etc. (Children born
to recent immigrants can have somewhat greater difficulty in learning
English, but that is invariably tied into use of a non-English language at
home and that is usually offset by the childs mastery of another language).
Infants/children simply learn easier, because of the developmental stage
they are in. I have simplified Learning theory a little bit here -
left out gobs of information about how the brain processes change with age
etc and the field is actually quite interesting, it was just starting to
merge with AI when I stopped trying to keep up. You may find more
information and better information about your question if you look there.
Best Regards Brent

> Hi Brent, I agree in general with all you have said. The only thing I'd
like to comment is the statement that people have no instincts but "a fear
of heights, and quick sounds". In my opinion we have inherited almost all
the instincts which mammals have. How do men behave in the presence of a
pretty girl (no matter what are their age, marriage status and probability
to establish romantic relations)? What does a man feel when suddenly know
that this child is his (even if absolutely no additional care is needed)? Is
there a big difference between hierarchy in a group of baboons and of
prisoners (I mean Russian jails)? The number of examples can be multiplied.
However you are absolutely right saying that the role of such "inborn
templates" in human is much less than in animals. They are usually
suppressed by mighty complex of conditioned reflexes and consciousness. I
think that as we us the term "instincts" to describe the aggregate of
behavioral unconditioned reflexes we may us term "culture" to describe the
aggregate of behavioral conditioned reflexes in human societies. Language is
an element of it. Alexander

I generally agree with you. We have a lot of instincts, not less than other
mammals. What we tend to learn is the result of what we're confronted with
(our family, school, society...) & of what our brain can learn. IOW we can't
learn anything. There are "empty places" in our brain that can be filled in,
but not everything can placed there.
There are grades of learning. Imprinting is very limited in what can be
filled in, eg, in young animals their mother's face or voice, the nestplace,
their father's & neighbours' song (in birds), the smell of their siblings,
etc., or later in life the features of the partner (in monogamous species)
or of the children. In humans the association areas in the neocortex are
enormously expanded, so that humans can learn a lot more than most if not
all nonhuman animals.

Marc