RE: [xmca] LSV-Sakharov Procedure: The role of the experimenter

["Paula M Towsey" <paulat@johnwtowsey.co.za>]

Dear Mike and Likeminded Mavens

Question One:
Your apparent laziness, Mike, has nothing to do with it: I should have made
it more explicit in the text that what was happening with the
three-year-olds, after the H&K/Sakharov script had been duly followed, was a
deviation from the script - and this is why...

Not one of the three-year-olds responded to the request to find blocks they
thought could be the same as the mur exemplar: in fact, they apparently
ignored it almost completely and went about doing their own thing (so, the
instructions and the words cev, bik, mur, and lag did not serve to organise
their activity with the blocks).  We did get wonderful things like the Big
Bad Wolf and trucks and drivers and coloured people and mummy, daddy, and
baby blocks (and rainbows and the green moon (green semi-circle)) from some
of them, and very little verbal response from others.  Because there was no
take-up with the artificial concepts, I wondered if I could find out what
they really thought about the things their preschool programme seemed to be
(disconcertingly) preoccupied with teaching them: colours and shapes.  So, I
asked these three-year-olds to find colours and shapes for me, and, as I
mentioned to David recently, they didn't mix up the names for colours and
shapes (much), like saying "Lello" for blue, or "triangle" for square (I
remember battling with these names when I was little).  But they did give us
wonderful examples of pseudoconceptual reasoning (lack of hierarchy,
instability, not abstracting the quality "circle" from the other
characteristics - blue - in which it is embedded), along with their
syncretic connections - it is so (because I say so); "Let's do them to
roll!": and all of this despite the programme's insistence on colour and
shape...

Otherwise, the script was as per Hanfmann and Kasanin 1942 (because the two
versions of T&S that I consulted at the time - 1986 and 1962 - both cited
their interpretation of the procedure as the apparent authority) coupled
with what I gleaned from Sakharov's 1928/30 paper.  I avoided the rather
behaviourist connotation of providing a gift as enticement to join in the
activity (from Sakharov), but carefully presented two options to
participants upfront.  They could put together all the potential mur
candidates and then begin to check this group (Sakharov), or, if they wanted
to, could sort all of the blocks into four groups from the start (H&K).
Other than that, I followed H&K to the letter.  (Details are in an appendix
to my research report, and also in H&K's 1942 monograph (and in their
appendix of that work too).)

Perhaps I should put together some notes on my role as the researcher so we
can compare it to yours from the 80s.  What I did try very hard to stay with
was the avoidance of leading questions, and "When all else fails, plead the
fifth amendment" - iow, stick to the script.  Some of Sakharov's one-by-one
reason at a time became a bit pedantic with the older participants, so, as
long as we could get participants to "think aloud" (This is a talking and
thinking activity), we would watch for the next few moves, and then ask
again.  I think though, that this element of the study would probably be a
paper in itself (for interested mavens at any rate...).

Question Two:
If there is scepticism needed in this part of the paper involving my
interpretation and how I inserted the Van der Veer & Valsiner quotation,
then this is possibly because of my treatment of it - or is this more a
question you have about abstraction and thinking patterns of adults anyway?
In either case, my take on this element is along these lines:

I find many examples of pseudoconceptual thinking in my own day-to-day
activities, for example, but most especially when I am faced with learning
something new (see my earlier discussion with David Kellogg on "flying great
circles"). Learning new things (Alex Kozulin writes lots about
pseudoconceptual thinking in learning a new language) can reveal our
pseudoconceptual approach in that we are not certain which of the
characteristics before us are the essential ones, and which are the
functional ones.  These essential characteristics are the ones which we
assign hierarchy to; but we have to be able to abstract them from the
situation they're embedded in - and we need to be able to do this
consistently.  (As I put it in Monday last week's posting: "when they
(adults) encounter something beyond their own field of expertise - nuclear
physics or rocket science - and when they use short-hand to create
generalised representations of things (everyday concepts;
concepts-for-ourselves-and-not-for-experts)... the signifying function of
language would possibly also be augmented because adults (and adolescents)
will tend use one system (like maths or ethics) to understand another."

An example of using one system to understand another from the study comes
from an adult who, for obvious reasons, doesn't feature in this paper on
pseudoconcepts...

"At 13 minutes into the session, this subject (SX09M) said "I've got an
idea.  The common thing between these shapes is that the triangles come in
four different sizes" (this despite there being five triangles).
 
This subject, SX09M, said that "How I deduced this categorisation is that
the common thing in the triangles is that they are the ones which seem to
differentiate on the height and the size.  And so there's this in the
circular ones as well - they also have that characteristic."  He had then
extrapolated this principle to the other blocks and solved the problem -
statistically and mathematically - by analysing the characteristics of
groups of blocks to establish where the areas of commonality lay, which
would form the basis for sorting the blocks."

So, if you compare this level of abstraction and generalisation within the
system of stats/maths (it's what he does for a living, no surprise there!),
with the reasoning of many of the examples that we've given in the paper
(and in colour, thanks to you!), you may see that it is different in the
following three main areas:

"What each of these instances was found to share, to varying degrees of
complexity, was a tendency to link concrete, factual, and functional
attributes rather than logical, abstract(ed), essential characteristics or
principles; an insensitivity to inconsistencies
and contradictions; and the lack of a system to compare or juxtapose one's
actions against." (Wolves, p. 240)

I think establishing differences between goats and gophers and wolves and
shaggy coats really is as difficult as Vygotsky said it would be - I'd been
over the "There are only two semi-circles" recording over and over again
before the penny dropped for me (never mind her!), and can think of many
other likeminded imposters...  

Till next batch of questions?
Take care.
Paula  

-----Original Message-----
From: xmca-bounces@weber.ucsd.edu [mailto:xmca-bounces@weber.ucsd.edu] On
Behalf Of Mike Cole
Sent: 09 August 2009 07:08 PM
To: eXtended Mind, Culture,Activity
Cc: Elena Yudina
Subject: [xmca] LSV-Sakharov Procedure: The role of the experimenter

Hi Paula, Carol, and other mavens of the LSV-Sakharov procedure.

I have been re-reading the wolves in sheep's clothing article and ruminating
about the
procedures you were using in relation to those used by LSV/Sakh and
Hanfmann/Kasanin (which
i have not gone back and studied carefully, so maybe the answer is easily
attributed to my laziness).

My major question has to do with the questions asked by the experimenter.
For example, did LSV/Sakh
ask questions such as that for your Figure 4 -- find blocks that are a
different color? When we tried
replicating this work back in the 1980's, we came a cropper of uncertainty
about what questions the
expermenter should be asking and focused as closely as we could on our
interpretation of the procedures
we had read about which seemed to start with one block and ask for others
which were in the same category,
then turning over a counter example. This got more and more complex the
further through the set we got.
And A LOT depended upon what the experimenter thought the kid was thinking.
That is, to introduce a phrase
from an adjacent conversation, there was recipient design in the adult
question, a guess at what conceptual
schema the kid was using at the time.  And these questions did not involve
word like "color" but word like "cev."
Is color an auxiliary stimulus in the experiment the same way "cev" is" ?

This aspect of the procedure is not clear to me.

Secondarily, don't we have to be sceptical about such as the van der Veer
and Valsiner quote on p. 236 in your
sentence on p. 236:  What Vygotsky is implying is that real cnocepts depend
on particular kinds of characteristics
or features that are abstract, abstracted, and system-related. This would
result in meanings that children attach to
words, and those of adults, being at different levels: "concrete features
vs. abstract definitions.".

But we can be pretty sure that adults, (the contributors to this discussion,
for example) routinely, even in our writing,
are operating at the level of pseudo concepts and, I believe, often at the
level of chaining..... at least our discussions
over time certainly have that characteristic.

in short, this old goat is having difficulty telling not only sheep from
wolves in sheeps clothing, but goats from gophers!!

mike
_______________________________________________
xmca mailing list
xmca@weber.ucsd.edu
http://dss.ucsd.edu/mailman/listinfo/xmca

_______________________________________________
xmca mailing list
xmca@weber.ucsd.edu
http://dss.ucsd.edu/mailman/listinfo/xmca