[From Erling Jorgensen (2011.11.11 11:00 EST)]
This thread used to be called “twilight of the idols,” although I forget
how it got that original title. I want to pick up on the question
Martin Taylor raised in various posts, about whether perceptions (or
other related aspects of control loops, such as reference outputs to
the next power level) can be considered or controlled as vectors,
or whether there is a theoretical or experimental necessity that they
be considered scalar signals.
I’ve been puzzling over that question. And while I don’t think I
necessarily have theoretical or experimental evidence, I think there
may be a phenomenological basis for considering that control becomes
possible when a perceptual input function creates a scalar signal.
What got me thinking in this vein was watching my daughter’s indoor
soccer game recently. She is part of a team that had sought to be in
a “recreational” league for her age group, but there were not enough teams
for that, so they got lumped together with the girls’ teams who were
definitely (and previously) in a more “competitive” grouping.
During their last game, they were decidedly overmatched. The girls on
the other team, from what I heard, had largely played together in other
leagues & venues. They knew how to set up & proceed down the field in
triangular patterns, so that they had an incredible passing game. They
regularly utilized outlet passing back to their defensive backs, which
seemed to give them a whole extra degree of freedom that our kids never
utilized. They even were well practiced with “give and go” attack
formations, that kept our kids stymied.
Our kids had a lot of good individual performances. But it was like
there was a whole different dimension of the game that they weren’t
keying in on. They weren’t even perceiving it. Call it “strategy,” or
“playing as a team,” or having well-rehearsed “programs or tactics” to
guide the unfolding play. Those were the kinds of perceptions that
members of the opposing team seemed to be creating & controlling, to
great effect (in terms of both score & effective time of possession, etc.)
Getting back to the vector versus scalar question. This may just be a
metaphorical use of the notion of vector on my part, which may not hold
up if it were implemented or simulated in some way. But I had the
impression that the aspects of play the other team utilized & controlled
so effectively were just being seen by our kids as collections of
unexpected perceptions, which kept cropping up, but which never really
cohered into a whole that they could then control for.
When the opposing team kept flowing into these triangular patterns, which
always allowed for an effective pass to take place, our kids would just
play the ball, but never knew how to get in the way of the passing lanes.
When their players would give & go, by making a pass & then racing ahead
to an open space for a quick return pass, our kids had no idea of how to
coordinate with teammates to stick with the one racing ahead while someone
else checked the one receiving the initial pass.
It is as if [there’s the metaphor] the perceptions for our kids were a
bunch of changing vectors, having no clear rhyme or reason to them (other
than the resulting score…), while the other team had coalesced the
lower level constituting perceptions into scalar(?) variables at a higher
level, rendering them control-able.
It makes me wonder whether the vector-vs.-scalar distinction might apply
to that interface where a different logical level of perception is being
constructed by a new set of perceptual input functions. To say it perhaps
too boldly, perceptions get controlled (or capable of being controlled)
when they become “scalar”. Prior to that, various lower level perceptions
are taking place, some of which are being controlled, but collectively
they still have a “vector-like” quality to them. Relative to the as yet
unformed higher level perceptual input function, there is an amalgam
character to them, which renders them uncontrolled as a whole. So then,
the term “vector” might be an appropriate one to capture the components
of what willl become a more singular perception. But until that “scalar”
function creation happens – bringing into being a new type of ‘invariance’
capable of being perceived – the components do not get controlled in
concert with one another.
To say this in another way. I am trying to ‘take the viewpoint’ of a
perceptual input function as it comes into being. The “vectors” refer to
the various dimensions of the lower level perceptions. A vector has not
just magnitude, but magnitude & direction. Yes, each lower level
perception already has a scalar form. But relative to a higher level
perceptual input function which may be controlled, it also represents an
implementing degree of freedom. That’s the piece that represents its
directionality.
In the example that Martin raised of the glass on the table, there is a
coherent Configuration grouping of perceptions, which remains invariant to
rotation of the glass (changing the oriented line perceptions), & invariant
to alterations in lighting (changing the shading perceptions), & invariant
to variations in position (changing relational delimiters such as what it
is “on”.) Those are all degrees of freedom which are free to vary,
probably within certain limits – for instance, if the lighting is dimmed
too much, the glass will not be monitored by sight, but perhaps by touch.
Nonetheless, those perceptions which constitute the Configuration
perception now cohere & seem to covary with one another. Each one has
its own current scalar value, but those have become degrees of freedom
relative to the glass configuration, once it has been constructed.
This makes me think of going into a highly cluttered gift store, with all
sorts of knick-knacks crammed on the shelves. I initially have the
experience of not really knowing where to look. The individual items
haven’t quite resolved for me. I just kind of see shelving, with a bunch
of lower level perceptions like color, kind of in a jumble. Relative to
seeing a particular “knick-knack”, or the even higher level perception of
a “plausible gift”, the perceptions just seem to exist as vectors to me.
And as such, they don’t yet seem control-able to me as singular objects.
Not sure if this argument is holding together or not. It’s at least
another take on how to view this matter of vectors.
All the best,
Erling
NOTICE: This e-mail communication (including any attachments) is CONFIDENTIAL and the materials contained herein are PRIVILEGED and intended only for disclosure to or use by the person(s) listed above. If you are neither the intended recipient(s), nor a person responsible for the delivery of this communication to the intended recipient(s), you are hereby notified that any retention, dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify me immediately by using the "reply" feature or by calling me at the number listed above, and then immediately delete this message and all attachments from your computer. Thank you.