Live Arm Demo: Pointing in the Dark

CSGnet Archive CSG1997
1

[from Gary Cziko 971202.0445 GMT]

Bill Powers (971130.1135 MST) wrote to Martin Taylor:

That makes two of us who are coming to a new view of
evolution, with others possibly travelling the same way. Gary Cziko, how
are you doing with all this?

I'm keeping up as best I can. Be assured that I am reading the posts on
evolution even though I simply cannot respond to all of them on CSGnet as
well as get my book finished. I'm afraid that responses (other than what
gets in my book) may have to wait until the CSG meeting this summer where I
look forward to some interesting discussions about evolution and PCT.

In the meantime, I have a new demo of perceptual control (or rather lack of
it) to report. I discovered it last night in my home office. I have a
power strip for my computer and peripherals which has an illuminated on/off
switch. I usually shut this off after shutting down my computer but before
shutting off the room lights. For some reason last night I shut off the
room lights first and then went to shut off the power strip in the darkened
room. I was quite surprised (although I shouldn't have been) when I
discovered that although the switch was plainly visible glowing in the
dark, I simply could not put my finger on it to shut it off!

The problem, of course, was that I could see the target but I could not see
my own finger. I then replicated this demo tonight in a room in my
basement where my daughter had affixed small glow-in-the-dark decals
(crescents like moons and circles like planets and stars) to the ceiling.
Since the ceiling is only about 7 feet high, I can easily stand on the
floor and touch any decal I wish. But when the lights are turned off is
very difficult for me to put my index finger over any targeted glowing
decal. It is a very strange feeling, after having missed the decal, to
have my finger touching the ceiling while seeing the decal nearby and
having absolutely no idea where the decal is in relation to my finger. I
had thought that proprioceptive feedback would be better than this, but
it's not really very good in this situation.

I found that I did get better if I attempted repeatedly to touch the same
target from the same floor position using the same posture. But if I moved
my feet or changed by pointing posture (such as pointing to my side instead
of in front of me) performance deteriorated again until I got used to the
new position and/or posture.

I can't imagine that no one else has discovered this phenomenon--perhaps it
has been mentioned on CSGnet and I missed it. I wonder how the Bizzis of
motor control can continue with their non-feedback reverse
kinematics/reverse dynamics approach to limb movements when this model
seems too easy to refute by just pointing in the dark.

--Gary

2

[Avery Andrews971202]
(Gary Cziko 971202.0445 GMT)

>I can't imagine that no one else has discovered this phenomenon--perhaps it

has been mentioned on CSGnet and I missed it. I wonder how the Bizzis of
motor control can continue with their non-feedback reverse
kinematics/reverse dynamics approach to limb movements when this model
seems too easy to refute by just pointing in the dark.

Well, that's easy! You just define the stuff you can't handle as
a `non-topic' (the exact word I've heard used in this kind of situation,
by an MIT faculty member, no less) and procede on course....

Avery

3

[From Bill Powers (971202.0714 MST)]

Gary Cziko 971202.0445 GMT --

In the meantime, I have a new demo of perceptual control (or rather lack of
it) to report. ... I was quite surprised (although I shouldn't have been)
when I discovered that although the switch was plainly visible glowing in
the dark, I simply could not put my finger on it to shut it off!

The problem, of course, was that I could see the target but I could not see
my own finger.

Lovely. Terrific.

It is a very strange feeling, after having missed the decal, to
have my finger touching the ceiling while seeing the decal nearby and
having absolutely no idea where the decal is in relation to my finger. I
had thought that proprioceptive feedback would be better than this, but
it's not really very good in this situation.

I'll try this as soon as possible. Obviously, it has also happened to me
but I just wasn't paying attention. You are a great attention-payer.

The first example probably happened because you had just turned off the
main room light and were't dark-adapted yet. I'm reminded of the tape from
Bruce Gregory's group. One of the demonstrations had to do with seeing in
the dark. The kids were convinced that all you had to do was wait long
enough and you would start seeing again, even in a completely blacked-out
room. They were quite surprised when, in a truly dark room, they couldn't
even see their own hands. One of them said "Well, we just didn't wait long
enough." These were science students, by the way -- but nobody had ever
asked them about their own theories of vision.

The point is that we aren't too likely to encounter your phenomenon in
ordinary dark rooms, because they aren't dark. Even a faint night-light
will provide enough illumination for you to see your own hand, if you wait
a while. However, right after the light is turned off, you can still see
the glow of the target but your eyes are not sensitive enough yet to see
the illumination of the hand, and bingo: the Cziko Effect. I can see how
this phenomenon could have been missed, even in pointing-in-the-dark
experiments. If you sit in a darkened room long enough, the light from the
target will be enough to let you see where your hand is.

For systematic testing, I suggest using a black glove.

I found that I did get better if I attempted repeatedly to touch the same
target from the same floor position using the same posture. But if I moved
my feet or changed by pointing posture (such as pointing to my side instead
of in front of me) performance deteriorated again until I got used to the
new position and/or posture.

I can't imagine that no one else has discovered this phenomenon--perhaps it
has been mentioned on CSGnet and I missed it. I wonder how the Bizzis of
motor control can continue with their non-feedback reverse
kinematics/reverse dynamics approach to limb movements when this model
seems too easy to refute by just pointing in the dark.

It's easy. First you become convinced that your explanation is right, and
then you select what you observe so as to avoid creating errors. It's done
all the time.

Best,

Bill P.