Phantoms in the Brain

CSGnet Archive CSG1998
1

[From Bruce Gregory (980926.0807 EDT)]

"Every time a "command" is sent from the supplementary motor area to the
motor cortex, it goes to the muscles and they move. At the same time,
identical copies of the same signal are sent to two other major "processing"
areas--the cerebellum and the parietal lobes--informing them of the intended
action.

Once these command signals are sent to the muscles, a feedback loop is set
in motion. Having received a command to move, the muscles execute the
movement. In turn, signals from the spinal muscles and joints are sent back
up to the brain, via the spinal cord, informing the cerebellum and the
parietal that "yes, the command is being performed correctly." These two
structures help you compare your intention with your actual performance,
behaving like a thermostat in a servo-loop, and modifying the motor commands
as needed (applying brakes if they are too fast and increasing the motor
outflow if it is too slow). Thus intentions are transformed into smoothly
coordinated movements."

_Phantoms in the Brain_ by V.S. Ramachandran and Sandra Blakeslee, William
Morrow, 1998.

2

[From Rick Marken (980926.0840)]

Bruce Gregory (980926.0807 EDT)--

"Every time a "command" is sent from the supplementary motor
area to the motor cortex, it goes to the muscles and they move.
At the same time, identical copies of the same signal are sent
to two other major "processing" areas--the cerebellum and the
parietal lobes--informing them of the intended action.

This is called the "efference copy" model of control. It's a
cause-effect, output generation model so it can't produce intended
_results_ in a disturbance prone environment. This approach to
control is actually used in the satellite control business; when
an operator sends a command (to do a delta-V say) he is also given
information (via a "copy" of the command sent) about what command
was actually sent. This is useful information but what the operator
really needs to see is the telemetry (perceptual) data that tells
him the actual _result_ of the command, which, even in space, is
not always exactly what was expected.

Having received a command to move, the muscles execute the
movement. In turn, signals from the spinal muscles and joints
are sent back up to the brain, via the spinal cord, informing
the cerebellum and the parietal that "yes, the command is being
performed correctly." These two structures help you compare your
intention with your actual performance

This does sound like a perceptual control loop. In this case, the
perception controlled is a perception of muscle tension or joint
angle.

Thus intentions are transformed into smoothly coordinated
movements."

"Thus", indeed. If only they could see that what is controlled is
not a "movement" but a "perception" ("signals from the spinal
muscles and joints"). Then they could see that their " phantom"
is a control loop that varies its actions (movements in this case)
as necessary to keep perceptions at their referecnes.

So near and yet so far.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

3

[From Bruce Gregory (980926.1235 EDT)]

Rick Marken (980926.0840)]

"Thus", indeed. If only they could see that what is controlled is
not a "movement" but a "perception" ("signals from the spinal
muscles and joints"). Then they could see that their " phantom"
is a control loop that varies its actions (movements in this case)
as necessary to keep perceptions at their references.

Actually, the "phantoms" are phantom limbs that appear after amputations.
Ramachandran uses mirrors to provide visual feedback on the "motions" of
these phantom limbs with very interesting results.

Bruce Gregory

4

[From Bruce Abbott (980926.1200 EST)]

Rick Marken (980926.0840) --

Thus intentions are transformed into smoothly coordinated
movements."

"Thus", indeed. If only they could see that what is controlled is
not a "movement" but a "perception" ("signals from the spinal
muscles and joints"). Then they could see that their " phantom"
is a control loop that varies its actions (movements in this case)
as necessary to keep perceptions at their referecnes.

So near and yet so far.

It sounds to me that Ramachandran and Blakeslee (1998) understand that they
have a control loop here. When they speak of controlling a movement, I
think there is little doubt they mean (in this example) limb positioning,
with "signals from the spinal muscles and joints" providing the readout of
limb position. The muscle tensions vary as necessary to bring about the
intended (reference) limb position. The arm moves to the intended position,
and as the reference changes, the limb position changes with it, more or
less as intended (their "controlled movement"), as reported by the sensor
signals.

I hate to burst your bubble [(;-> ], but if my interpretation of their words
is correct, they are not only near, they are right on.

Regards,

Bruce

5

[From Rick Marken (980927.1720)]

Bruce Abbott (980926.1200 EST)--

It sounds to me that Ramachandran and Blakeslee (1998)
understand that they have a control loop here.

That may be true. I was questioning whether they understood
that this loop controls a _perceptual variable_. Would it have
been so hard for Ramachandran and Blakeslee (1998) to say that
this control loop controls a perceptual representation of limb
position if, as you say, this is what they meant? Would it
be so hard for operant researchers to say that the rat controls
a perceptual variable that is disturbed by shock (or food) if,
as you have said, that is what they mean? Why doesn't any
psychologist _ever_ use the term _controlled variable_? Why won't
students of behavior simply say what they are doing (according
to you): studying how organisms control their own perceptions?
Why are students of behavior making it so hard for us to see that
they have been studying behavior as the control of perception for
the last 100 years?

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

6

[From Bruce Abbott (980927.0845 EST)]

Rick Marken (980927.1720) --

Bruce Abbott (980926.1200 EST)

It sounds to me that Ramachandran and Blakeslee (1998)
understand that they have a control loop here.

That may be true. I was questioning whether they understood
that this loop controls a _perceptual variable_. Would it have
been so hard for Ramachandran and Blakeslee (1998) to say that
this control loop controls a perceptual representation of limb
position if, as you say, this is what they meant?

One can properly _describe_ the system as controlling limb position _as
sensed_ or as controlling the sense of limb position. There's nothing
"hard" about doing it your way, it's just that it's not the only way, and
they chose to emphasize the observable consequences (controlled movements)
rather than the private ones (controlled perceptions). Perhaps if they had
only known that Rick Marken would make a big fuss about it, they would have
expressed themselves differently.

Regards,

Bruce

7

[From Rick Marken (980927.0840)]

Bruce Abbott (980927.0845 EST)

One can properly _describe_ the system as controlling limb
position _as sensed_ or as controlling the sense of limb
position.

But Ramachandran and Blakeslee (1998) didn't describe it in
either way. I believe they described it as control of movement.
I think you were the one who managed to see their comments as a
description of control of perception.

There's nothing "hard" about doing it your way, it's just that
it's not the only way, and they chose to emphasize the observable
consequences (controlled movements) rather than the private ones
(controlled perceptions).

My question was about why psychologists have consistently chosen
to emphasize controlled outputs rather than controlled perceptions?
Indeed, psychologists _always_ choose to describe behavior in terms
of observable outputs. As far as I can tell they _never_ describe
behavior in terms of controlled inputs. Why do you suppose this is?
Is it just historical bias? Why won't they _ever_ use the term
now that they (some of them) know it exists?

Perhaps if they had only known that Rick Marken would make a big
fuss about it, they would have expressed themselves differently.

I doubt it.

But apparently you _do_ think this is just an arbitrary choice.
The fact that psychologists _never_ talk about controlled variables
is also, I presume, a result of the fact that psychologists didn't
know that I would make a big deal out of it? Is that true too?
Are you saying that psychologists know all about controlled
variables but they just don't want to use the words "controlled
variable" to describe these variables? If that's true, could you
tell me what words they _are_ using to describe controlled
variables?

Best

Rick

···

--

Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

8

[From Bruce Abbott (980927.1200 EST)]

Rick Marken (980927.0840) --

Bruce Abbott (980927.0845 EST)

One can properly _describe_ the system as controlling limb
position _as sensed_ or as controlling the sense of limb
position.

But Ramachandran and Blakeslee (1998) didn't describe it in
either way. I believe they described it as control of movement.

Yes, indeed they did describe it as control of movement. And I explained
how this description is quite accurate. If you've forgotten already (or
overlooked it), please reread the post -- and think carefully about it --
before replying again.

There's nothing "hard" about doing it your way, it's just that
it's not the only way, and they chose to emphasize the observable
consequences (controlled movements) rather than the private ones
(controlled perceptions).

My question was about why psychologists have consistently chosen
to emphasize controlled outputs rather than controlled perceptions?

My assertion concerned only Ramachandran and Blakeslee, who in my judgment
were quite correct in their description. As I suggested in my post, the
emphasis on controlled outputs probably has something to do with the fact
that these are what are observed, whereas a person's perceptions can only be
inferred.

But apparently you _do_ think this is just an arbitrary choice.

It is in this case. And I can prove it.

The fact that psychologists _never_ talk about controlled variables
is also, I presume, a result of the fact that psychologists didn't
know that I would make a big deal out of it? Is that true too?
Are you saying that psychologists know all about controlled
variables but they just don't want to use the words "controlled
variable" to describe these variables? If that's true, could you
tell me what words they _are_ using to describe controlled
variables?

Really, Rick, what has this got to do with Ramachandran and Blakeslee?
That's who I'm talking about. I think they have it right. Do you have any
evidence that they don't? (I've already shown that use of the phrase
"control of movement" does not constitute such evidence, so you'll have to
look for something else.)

Regards,

Bruce

9

[From Rick Marken (980927.1710)]

Bruce Abbott (980927.1200 EST)--

Yes, indeed they did describe it as control of movement. And
I explained how this description is quite accurate.

I was just questioning whether there is any evidence that
Ramachandran and Blakeslee (1998) understand that behavior
is the control of perceptual variables. Nothing you have said
has provided me with any evidence that they do understand this.

My assertion concerned only Ramachandran and Blakeslee, who in my
judgment were quite correct in their description.

Their desciption of behavior as "control of movement" was not
"correct"; it was _ambiguous_. The word "movement" could refer
to a controlled perceptual variable or it could refer (as I
suspect it did) to a caused output variable. Did you present
evidence that Ramachandran and Blakeslee were using the word
"movement" to refer to a controlled perceptual variable? If so,
I would like to see that evidence.

By the way, is this your evidence? You said:

When they speak of controlling a movement, I think there is
little doubt they mean (in this example) limb positioning,
with "signals from the spinal muscles and joints" providing
the readout of limb position.

But I think it's a long way from knowing that "signals from the
spinal muscles and joints provide a readout of limb position"
to knowing that these signals are under _control_. So I am not
convinced by this evidence. Did Ramachandran and Blakeslee do
any testing to determine whether the observable correlates of
these signals were under control? Now _that_ would be evidence.

As I suggested in my post, the emphasis on controlled outputs
probably has something to do with the fact that these are what
are observed, whereas a person's perceptions can only be
inferred.

But I question whether the "controlled outputs" to which you
refer are understood to be _possible_ controlled variables. Limb
movements are only _possibly_ controlled variables because one
obviously has to test to determine that this variable is actually
under control. Did Ramachandran and Blakeslee explain how they
knew that limb position was under control?

Really, Rick, what has this got to do with Ramachandran and
Blakeslee? That's who I'm talking about. I think they have it
right. Do you have any evidence that they don't?

Actually, I have no idea. From what Bruce Gregory posted
originally it seemed pretty clear that Ramachandran and
Blakeslee have no idea what a controlled variable is. They
didn't seem to know that controlled variables might exist,
that these variables are _perceptual representations_ of
quantities in the world outside the nervous system or how to
test to determine whether a variable is under control. Maybe
they do. But I haven't seen any evidence yet that they do.

Actually, I've just seen another little piece of evidence that
suggests that Ramachandran and Blakeslee _don't_ know that
behavior is the control of perceptual variables. Bruce Gregory
(980927.1220 EDT) just posted this quote from Ramachandran:

It was as though Philip had some temporary inhibition or block
of the neural circuits that would ordinarily move the phantom
and the visual feedback had overcome this block."

I find it difficult to believe that someone who understands
that "visual feedback" is a controlled perceptual variable would be
able to say that visual feedback "overcame" anything. This suggests
that visual feedback _causes_ something; not that it is controlled.
Ramachandran may simply have an odd way of expressing the idea that
behavior is the control of perception; maybe he just meant to say:
"The mirror image of the actual arm provided a perception that
could now be virtually controlled by the existing lower order
muscle tension control systems". Maybe Ramachandran understands
that behavior is the control of perception but he just has a poor
way of expressing it. I'd appreciate seeing more evidence that
this is the case; it looks like Ramachandran has made some pretty
interesting robservations. I'm sure PCT could help hin; but maybe,
as you say, he already knows PCT, in which case there would really
be no need to get in touch with him.

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/