[From Dag Forssell (2006 0101 1910 PST)]
Dick, Bryan,
I registered at NY times so I could access the article, complete with
slide show. Amazing how this kind of dreck is received and published
as respectable output by "researchers". The trick must be to provide
fancy, colorful illustrations. I can't read this stuff.
For those who can, I am attaching a pdf file that holds the entire
article as well as the slide show.
Best, Dag
Cells that read minds1.pdf (1.21 MB)
From Dan Miller (2006.0110.1345 EST)
Dag and Others:
I agree that articles about scientific studies that appear in the popular
press are nearly always inaccurate and suffer from serious hyperbole. I've
read much of the original research on mirror neurons and I'm intrigued if
not convinced. I hesitate to assign these ideas to the "dreck" bin. Some
interesting neurophysiological process may be afoot. The ideas do seem to
fit with interpersonal processes identified by symbolic interactionists -
not that this means all that much.
Best,
Dan
Dan Miller
Department of Sociology
University of Dayton
Dayton, Ohio 45469-1442
937.229.2138
Dan.Miller@notes.udayton.edu
Dag,
Thanks for vetting it and composing it in a PDF for us! 
--Bry
-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@LISTSERV.UIUC.EDU]On Behalf Of Dag Forssell
Sent: Tuesday, January 10, 2006 11:08 AM
To: CSGNET@LISTSERV.UIUC.EDU
Subject: [CSGNET] NYTimes.com: Cells That Read Minds
[From Dag Forssell (2006 0101 1910 PST)]
Dick, Bryan,
I registered at NY times so I could access the article, complete with
slide show. Amazing how this kind of dreck is received and published
as respectable output by "researchers". The trick must be to provide
fancy, colorful illustrations. I can't read this stuff.
For those who can, I am attaching a pdf file that holds the entire
article as well as the slide show.
Best, Dag
Dan et al.,
Well the thing is that this article gives us a phenom to pick at. Brain
activity during observation that seems similar to the brain activity during
manipulation. How do WE describe it is the issue, not setting the baloney
detector to high gain and blasting the article to ashes...
I will look at that article in PDF with new eyes then.
--Bry
-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@LISTSERV.UIUC.EDU]On Behalf Of Dan Miller University of
Dayton
Sent: Tuesday, January 10, 2006 12:43 PM
To: CSGNET@LISTSERV.UIUC.EDU
Subject: Re: [CSGNET] NYTimes.com: Cells That Read Minds
From Dan Miller (2006.0110.1345 EST)
Dag and Others:
I agree that articles about scientific studies that appear in the popular
press are nearly always inaccurate and suffer from serious hyperbole. I've
read much of the original research on mirror neurons and I'm intrigued if
not convinced. I hesitate to assign these ideas to the "dreck" bin. Some
interesting neurophysiological process may be afoot. The ideas do seem to
fit with interpersonal processes identified by symbolic interactionists -
not that this means all that much.
Best,
Dan
Dan Miller
Department of Sociology
University of Dayton
Dayton, Ohio 45469-1442
937.229.2138
Dan.Miller@notes.udayton.edu
[From Dick Robertson,2006.01.10.1600CDT]
I wonder if a case couldn't be made for the cells he describes in the article as 7th order reference signal-generators?
Best,
Dick R
Dan Miller University of Dayton wrote:
From Dan Miller (2006.0110.1345 EST)
Dag and Others:
I agree that articles about scientific studies that appear in the popular
press are nearly always inaccurate and suffer from serious hyperbole. I've
read much of the original research on mirror neurons and I'm intrigued if
not convinced. I hesitate to assign these ideas to the "dreck" bin. Some
interesting neurophysiological process may be afoot. The ideas do seem to
fit with interpersonal processes identified by symbolic interactionists -
not that this means all that much.Best,
DanDan Miller
Department of Sociology
University of Dayton
Dayton, Ohio 45469-1442
937.229.2138
Dan.Miller@notes.udayton.edu
Re: NYTimes.com: Cells That Read
Minds
[Martin Taylor 2006.01.12.15.43]
Brian Thalhammer Tue, 10 Jan 2006
15:12:43 -0600
Dan et al.,
Well the thing is that this article gives us a phenom to pick at.
Brainactivity during observation that seems similar to the brain activity
duringmanipulation. How do WE describe it is the issue, not setting the
baloneydetector to high gain and blasting the article to ashes…
I will look at that article in PDF with
new eyes then.
I thought you might be interested in an article from Science last
April, with commentary. Google Scholar (http://scholar.google.com/scholar?q=link:http%3A%2F%2Fwww.sciencemag.org%2Fcgi%2Fcontent%2Fabstract%2F308%2F5722%2F662
already lists 7 subsequent publications that cite it. I imagine that
some of these are what the NYT article is dealing with, though I
haven’t read any of them (or the one I’m attaching, not in any
detail).
Brian says, the question isn’t that the interpretations aren’t PCT and
therefore it’s nonsense, but that something is happening and is it to
be expected within the framework of PCT? If not, does some new
hypothesis have to be added to PCT to accommodate the data?
Martin
From my cursory reading, there seems to be something to it. As
[From Bjorn Simonsen (2006.01.12,22:35 EUST)]
From Dan Miller (2006.0110.1345 EST), Brian Thalhammer Tue, 10 Jan 2006
15:12:43 -0600 , [Martin Taylor 2006.01.12.15.43]
I have read some of Rizzolatti's papers. In one of them "Action recognition
in the premotor cortex" (Gallese, Fadiga, Frogassi and Rizzolatti), he has
a paragraph named "Mirror systems in humans". Here he refers to Fadiga et
al. 1995. He writes: ..... In these experiments, subjects were stimulated
during observation of an experimenter grasping three-dimensional objects,
during observation of the same objects, and during the detection of the
dimming of a small spot of light. The results show a significant increase of
motor evoked potentials recorded from the hand-arm muscles during grasping
observation with respect to the other conditions. The Increase was present
only in those muscles that were recruited when the subjects made the
observed movements actively. ..................."
I have order the Fadiga article and should maybe have waited with this mail
till the article was studied, but I will come with a comment here. I have
marked the sentence: "The results show a significant increase of motor
evoked potentials ...." The concept _significance increase_ tells me that
Fadiga had some problems with his experiments. As Bill said in BCP,
""Significance" in experimental results had come to mean something other
than "importance". It now means a little triumph over nature's noise level".
I will come back when I have studied his article, but I don't trust these
mirror systems in human brain. And how much shall we trust the information
from a transcranial magnetic stimulation (look summary below ). Maybe
somebody could tell me that.
A Summary.
Motor facilitation during action observation: a magnetic stimulation study
L. Fadiga, L. Fogassi, G. Pavesi and G. Rizzolatti
Istituto di Fisiologia Umana, Universita di Parma, Italy.
1. We stimulated the motor cortex of normal subjects (transcranial magnetic
stimulation) while they 1) observed an experimenter grasping 3D-objects, 2)
looked at the same 3D-objects, 3) observed an experimenter tracing
geometrical figures in the air with his arm, and 4) detected the dimming of
a light. Motor evoked potentials (MEPs) were recorded from hand muscles. 2.
We found that MEPs significantly increased during the conditions in which
subjects observed movements. The MEP pattern reflected the pattern of muscle
activity recorded when the subjects executed the observed actions. 3. We
conclude that in humans there is a system matching action observation and
execution. This system resembles the one recently described in the monkey.
bjorn
[From Erling Jorgensen (2006.01.12 2100 EST)]
Dick, Bryan, Dag, Dan, et al. –
Somewhat belatedly, this is with reference to the New York Times
article, “Cells That Read Minds” by Sandra Blakeslee, published on
January 10, 2006. (Thank you, Dick, for pointing the article out.
And thank you, Dag, for making the article readable here.)
The article mentioned, “Every time the monkey grasped and moved
an object, some cells in that brain region would fire.”
When I read this, the first things that struck me – thinking from the
vantage point of Perceptual Control Theory – were the following.
To “grasp” is an Event perception. To “move” is a Transition perception.
An “object” is a Configuration perception.
The article continued, “when the student raised the cone to his lips, the
monitor sounded - brrrrrip, brrrrrip, brrrrrip - even though the monkey
had not moved but had simply observed the student grasping the cone
and moving it to his mouth.”
“Raised-the-cone-to-his-lips” sounds like a Sequence perception. “To”
is a Relationship perception. And so forth. It seems as though it might
be possible to deconstruct what the article is describing into corresponding
PCT concepts.
In other words, PCT is a kind of conceptual map, to be overlaid on top
of a descriptive territory, to see if any increment of understanding can
arise from comparing the two, (keeping in mind all the time that “the map
is not the territory.”) The most useful kind of understanding is what we
call “explanation,” where a complex territory of relationships can be
effectively mapped onto a much smaller set of relationships (called
tautologies) on the map, with minimal loss of richness.
So, I would start quite simplistically. Hierarchical Perceptual Control
Theory has a heuristic for a bare bones model, consisting of levels of
perception that would need to be brought to their respective reference
standards, in order to accomplish a coordinated set of goals. The
phenomenon being raised by this research (in the NYTimes) is the
apparent observation that similar neurons may be involved, whether
acting oneself or observing another doing something similar.
So then, one of the descriptions from the New York Times article
stated, “The same brain cells fired when the monkey watched humans
or other monkeys bring peanuts to their mouths as when the monkey
itself brought a peanut to its mouth.”
A first approximation at deconstructing some of this into an HPCT
formulation might be as follows:
Program perception (implied) of “Eating”
~ IF no peanut, Grasp
~ ELSE, next
~ IF not at mouth, Bring toward
~ ELSE, Eat
including >
Category referents of “Peanut” and “Hand” and “Mouth”
and enacted via >
Sequence perception of “Grasp, then Bring toward, then Eat”
Ref. = Event-A, Relationship, Event-B
enacted via >
Relationship perceptions of “Distance-from-peanut” and “Distance-
from-mouth”
Ref. = zero
alternating with >
Event perceptions of “Grasp” and “Eat”
Ref. = Start & Stop points
enacted via >
Transition perceptions of “Hand/arm movements” and “Mouth movements”
Ref. = coordinated opponent process muscles
enacted via >
Configuration perceptions of “Body positions” and “Joint angles”
enacted via >
Sensation perceptions of “Finger pressure” and “Tendon stretch”
etc.
Such a series of enacted goals and sub-goals involves a number of
different perceptions. Many of these perceptions could be held in
common, whether the monkey was observing another or doing the
acting itself. This is less likely (I believe), the lower in the hierarchy
the perceptions are occurring. But it is certainly conceivable that
similar Event perceptions, Relationship perceptions, Sequence
perceptions, or even Program perceptions could have been activated
in either the observation condition or the acting condition. So, one
candidate for the “mirror neurons” of the article would be cells
corresponding to the perception nodes of the HPCT model.
I initially thought that another candidate could be cells corresponding
to the reference nodes of the HPCT model. My reasoning was that
the proposed origin of reference signals within the HPCT model is
essentially remembered perceptions, replayed into the downward
channel of descending control loops. That’s one reason I tried to
specify some of the reference standards, in the simplified PCT
description above.
But having done that, it does not seem that such things could be
objects of perception of the monkey. We would have to propose -
(and maybe the article is doing exactly this) – that the monkey is
imagining its own enactment of the perceptual goals that would be
needed, were it to be doing what it is observing. That may be the
import of the statement in the article about the gross physiology being
studied; namely, “Thin wires had been implanted in the region of its
brain involved in planning and carrying out movements.” According
to an extended version of the HPCT proposal, “planning” would be
a function carried out by the so-called imagination connection.
At any rate, I know this is simply launching off of a popularized
treatment in the mainstream media of research that I have not read
in the original journals. And I know this doesn’t do justice to the
complexity of the neurophysiology involved. But what we are looking
for here is an initial mapping of the tautological relationships within a
simplified model (i.e., a hierarchical arrangement of PCT control loops),
onto the territory of some neurophysiological observations.
The PCT model (and its HPCT elaboration) has found it fruitful to
operate with a very limited number of conceptual nodes. The argument
has been that such a vast simplification shows something tremendously
important about the functional arrangement known as negative feedback
control. Perhaps the recording of “mirror neurons” is one way that the
fiction of perceptual nodes manifests itself in the brain.
I’ll have to leave it to others to determine whether searching in such
a direction, with such a map, leads to any increments of understanding
or not.
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.
<<<>>>
Erling, Dick, Dag, and Dan, et al…
Oh, yes, oh yes, that is exactly what I was thinking too, when I browsed the article Dick sent. But you did all the work. Wow.
Quite a number of times I get visual images when I read about such phenomena from outside PCT. And it just rang out, perceptual hierarchy here! That was the direction I had hoped folks would take at first.
So, thanks!
–Bry
-----Original Message-----
From: Control Systems Group Network (CSGnet) [mailto:CSGNET@LISTSERV.UIUC.EDU]On Behalf Of Erling Jorgensen
Sent: Thursday, January 12, 2006 8:27 PM
To:
CSGNET@LISTSERV.UIUC.EDU
Subject: Re: [CSGNET] NYTimes.com: Cells That Read Minds
[From Erling Jorgensen (2006.01.12 2100 EST)]
Dick, Bryan, Dag, Dan, et al. –
Somewhat belatedly, this is with reference to the New York Times
article, “Cells That Read Minds” by Sandra Blakeslee, published on
January 10, 2006. (Thank you, Dick, for pointing the article out.
And thank you, Dag, for making the article readable here.)
The article mentioned, “Every time the monkey grasped and moved
an object, some cells in that brain region would fire.”
When I read this, the first things that struck me – thinking from the
vantage point of Perceptual Control Theory – were the following.
To “grasp” is an Event perception. To “move” is a Transition perception.
An “object” is a Configuration perception.
The article continued, “when the student raised the cone to his lips, the
monitor sounded - brrrrrip, brrrrrip, brrrrrip - even though the monkey
had not moved but had simply observed the student grasping the cone
and moving it to his mouth.”
“Raised-the-cone-to-his-lips” sounds like a Sequence perception. “To”
is a Relationship perception. And so forth. It seems as though it might
be possible to deconstruct what the article is describing into corresponding
PCT concepts.
In other words, PCT is a kind of conceptual map, to be overlaid on top
of a descriptive territory, to see if any increment of understanding can
arise from comparing the two, (keeping in mind all the time that “the map
is not the territory.”) The most useful kind of understanding is what we
call “explanation,” where a complex territory of relationships can be
effectively mapped onto a much smaller set of relationships (called
tautologies) on the map, with minimal loss of richness.
So, I would start quite simplistically. Hierarchical Perceptual Control
Theory has a heuristic for a bare bones model, consisting of levels of
perception that would need to be brought to their respective reference
standards, in order to accomplish a coordinated set of goals. The
phenomenon being raised by this research (in the NYTimes) is the
apparent observation that similar neurons may be involved, whether
acting oneself or observing another doing something similar.
So then, one of the descriptions from the New York Times article
stated, “The same brain cells fired when the monkey watched humans
or other monkeys bring peanuts to their mouths as when the monkey
itself brought a peanut to its mouth.”
A first approximation at deconstructing some of this into an HPCT
formulation might be as follows:
Program perception (implied) of “Eating”
~ IF no peanut, Grasp
~ ELSE, next
~ IF not at mouth, Bring toward
~ ELSE, Eat
including >
Category referents of “Peanut” and “Hand” and “Mouth”
and enacted via >
Sequence perception of “Grasp, then Bring toward, then Eat”Ref. = Event-A, Relationship, Event-B
enacted via >
Relationship perceptions of “Distance-from-peanut” and “Distance-
from-mouth”Ref. = zero
alternating with >
Event perceptions of “Grasp” and “Eat”Ref. = Start & Stop points
enacted via >
Transition perceptions of “Hand/arm movements” and “Mouth movements”Ref. = coordinated opponent process muscles
enacted via >
Configuration perceptions of “Body positions” and “Joint angles”
enacted via >
Sensation perceptions of “Finger pressure” and “Tendon stretch”
etc.
Such a series of enacted goals and sub-goals involves a number of
different perceptions. Many of these perceptions could be held in
common, whether the monkey was observing another or doing the
acting itself. This is less likely (I believe), the lower in the hierarchy
the perceptions are occurring. But it is certainly conceivable that
similar Event perceptions, Relationship perceptions, Sequence
perceptions, or even Program perceptions could have been activated
in either the observation condition or the acting condition. So, one
candidate for the “mirror neurons” of the article would be cells
corresponding to the perception nodes of the HPCT model.
I initially thought that another candidate could be cells corresponding
to the reference nodes of the HPCT model. My reasoning was that
the proposed origin of reference signals within the HPCT model is
essentially remembered perceptions, replayed into the downward
channel of descending control loops. That’s one reason I tried to
specify some of the reference standards, in the simplified PCT
description above.
But having done that, it does not seem that such things could be
objects of perception of the monkey. We would have to propose -
(and maybe the article is doing exactly this) – that the monkey is
imagining its own enactment of the perceptual goals that would be
needed, were it to be doing what it is observing. That may be the
import of the statement in the article about the gross physiology being
studied; namely, “Thin wires had been implanted in the region of its
brain involved in planning and carrying out movements.” According
to an extended version of the HPCT proposal, “planning” would be
a function carried out by the so-called imagination connection…
At any rate, I know this is simply launching off of a popularized
treatment in the mainstream media of research that I have not read
in the original journals. And I know this doesn’t do justice to the
complexity of the neurophysiology involved. But what we are looking
for here is an initial mapping of the tautological relationships within a
simplified model (i.e., a hierarchical arrangement of PCT control loops),
onto the territory of some neurophysiological observations.
The PCT model (and its HPCT elaboration) has found it fruitful to
operate with a very limited number of conceptual nodes. The argument
has been that such a vast simplification shows something tremendously
important about the functional arrangement known as negative feedback
control. Perhaps the recording of “mirror neurons” is one way that the
fiction of perceptual nodes manifests itself in the brain.
I’ll have to leave it to others to determine whether searching in such
a direction, with such a map, leads to any increments of understanding
or not.
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.
<<<>>>
[From Bjorn Simonsen (2006.01.13,11:45 EUST)]
From Erling Jorgensen (2006.01.12 2100 EST)
So then, one of the descriptions from the New York Times article
stated, "The same brain cells fired when the monkey watched humans
or other monkeys bring peanuts to their mouths as when the monkey
itself brought a peanut to its mouth."
I think this is wrong. If we look in Bill's "Making sense of behavior", he
has convincing paragraph on page 22 and so on, where he describes how
different the perceptions the student perceive from the perceptions the
monkey perceive. He mentions a "student" grasping for a glass of water,
lifting the glass to the mouth and drinking the water. The "monkey" watching
this doesn't grasp, he doesn't feel the glass in his hand, he doesn't lift
the glass and he doesn't feel the swallowing of water. The monkey sees what
the "student" is doing, he sees the back of the student fingers and he sees
the glass tilting away from the monkey. There is an enormous difference
between what the "student" is perceiving and what the "monkey" perceives.
And when the New York Times article says "the same cells fired when the
monkey watched "The same brain cells fired when the monkey watched humans or
other monkeys bring peanuts to their mouths as when the monkey itself
brought a peanut to its mouth.", it is opportune to ask how they know it is
the same cells.
I think the "monkey" is able to imagine how it is to eat peanuts when he is
watching other monkeys eating peanuts, but, and there is a big but. Is there
a reference in the "monkey's" system signalling that he wishes to imagine
how it is to eat a peanut?
bjorn
[From Dick Robertson,2006.01.13.0952CST]
Erling Jorgensen wrote:
[From Erling Jorgensen (2006.01.12 2100 EST)]
I like it. Do you think it would be worthwhile for you to send your
analysis to those guys?
Best,
Dick R
Dick, Bryan, Dag, Dan, et al. –
Somewhat belatedly, this is with reference to the New York Times
article, “Cells That Read Minds” by Sandra Blakeslee, published on
January 10, 2006. (Thank you, Dick, for pointing the article out.
And thank you, Dag, for making the article readable here.)
The article mentioned, "Every time the monkey grasped and moved
an object, some cells in that brain region would fire."
When I read this, the first things that struck me – thinking
from thevantage point of Perceptual Control Theory – were the following.
To “grasp” is an Event perception. To “move” is a Transition
perception.An “object” is a Configuration perception.
The article continued, "when the student raised the cone to his
lips, themonitor sounded - brrrrrip, brrrrrip, brrrrrip - even though the monkey
had not moved but had simply observed the student grasping the cone
and moving it to his mouth."
“Raised-the-cone-to-his-lips” sounds like a Sequence
perception. “To”is a Relationship perception. And so forth. It seems as though it
mightbe possible to deconstruct what the article is describing into
correspondingPCT concepts.
In other words, PCT is a kind of conceptual map, to be overlaid
on topof a descriptive territory, to see if any increment of understanding
canarise from comparing the two, (keeping in mind all the time that "the
mapis not the territory.") The most useful kind of understanding is what
wecall “explanation,” where a complex territory of relationships can be
effectively mapped onto a much smaller set of relationships (called
tautologies) on the map, with minimal loss of richness.
So, I would start quite simplistically. Hierarchical Perceptual
ControlTheory has a heuristic for a bare bones model, consisting of levels of
perception that would need to be brought to their respective reference
standards, in order to accomplish a coordinated set of goals. The
phenomenon being raised by this research (in the NYTimes) is the
apparent observation that similar neurons may be involved, whether
acting oneself or observing another doing something similar.
So then, one of the descriptions from the New York Times article
stated, "The same brain cells fired when the monkey watched humans
or other monkeys bring peanuts to their mouths as when the monkey
itself brought a peanut to its mouth."
A first approximation at deconstructing some of this into an
HPCTformulation might be as follows:
Program perception (implied) of “Eating”
~ IF no peanut, Grasp
~ ELSE, next
~ IF not at mouth, Bring toward
~ ELSE, Eat
…including >
Category referents of “Peanut” and “Hand” and “Mouth”
…and enacted via >
Sequence perception of “Grasp, then Bring toward, then Eat”
Ref. = Event-A, Relationship, Event-B
…enacted via >
Relationship perceptions of “Distance-from-peanut” and "Distance-
from-mouth"
Ref. = zero
…alternating with >
Event perceptions of “Grasp” and “Eat”
Ref. = Start & Stop points
…enacted via >
Transition perceptions of “Hand/arm movements” and “Mouth movements”
Ref. = coordinated opponent process muscles
…enacted via >
Configuration perceptions of “Body positions” and “Joint angles”
…enacted via >
Sensation perceptions of “Finger pressure” and “Tendon stretch”
…etc.
Such a series of enacted goals and sub-goals involves a number
ofdifferent perceptions. Many of these perceptions could be held in
common, whether the monkey was observing another or doing the
acting itself. This is less likely (I believe), the lower in the
hierarchythe perceptions are occurring. But it is certainly conceivable that
similar Event perceptions, Relationship perceptions, Sequence
perceptions, or even Program perceptions could have been activated
in either the observation condition or the acting condition. So, one
candidate for the “mirror neurons” of the article would be cells
corresponding to the perception nodes of the HPCT model.
I initially thought that another candidate could be cells
correspondingto the reference nodes of the HPCT model. My reasoning was that
the proposed origin of reference signals within the HPCT model is
essentially remembered perceptions, replayed into the downward
channel of descending control loops. That’s one reason I tried to
specify some of the reference standards, in the simplified PCT
description above.
But having done that, it does not seem that such things could be
objects of perception of the monkey. We would have to propose -
(and maybe the article is doing exactly this) – that the monkey is
imagining its own enactment of the perceptual goals that would be
needed, were it to be doing what it is observing. That may be the
import of the statement in the article about the gross physiology being
studied; namely, "Thin wires had been implanted in the region of its
brain involved in planning and carrying out movements." According
to an extended version of the HPCT proposal, “planning” would be
a function carried out by the so-called imagination connection.
At any rate, I know this is simply launching off of a
popularizedtreatment in the mainstream media of research that I have not read
in the original journals. And I know this doesn’t do justice to the
complexity of the neurophysiology involved. But what we are looking
for here is an initial mapping of the tautological relationships within
asimplified model (i.e., a hierarchical arrangement of PCT control
loops),onto the territory of some neurophysiological observations.
The PCT model (and its HPCT elaboration) has found it fruitful
tooperate with a very limited number of conceptual nodes. The argument
has been that such a vast simplification shows something tremendously
important about the functional arrangement known as negative feedback
control. Perhaps the recording of “mirror neurons” is one way that the
fiction of perceptual nodes manifests itself in the brain.
I’ll have to leave it to others to determine whether searching
in sucha direction, with such a map, leads to any increments of understanding
or not.
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.
<<< >>>
[From Bryan Thalhammer (2006.01.13.1013 CST)]
Yes, "same cells" seems rather vague. I wonder how accurate the device is that measures which cells? We do not know (do we?) the level of discreteness of perceptual control just as cognitive theorists don't know where a schema or script lies (area of the brain, group of cells, a cell, a particular dendrite...).
But now: "There is an enormous difference between what the "student" is perceiving and what the "monkey" perceives." [Or what the monkey eating the peanut perceives and what the monkey only watching another monkey eating a peanut perceives.]
Really? I tend to disagree. Because, true, while the observer monkey sees a different pattern with the visual cortex than the student or the monkey performing that same action, the understanding of that action (category, sequence, relationship, event, transition, configuration, sensation) is the same, so long as the observer monkey sees enough of what is going on, to want to do it.
Humans (kids and adults) and monkeys do things because they want to control a perception. For example, they experience "not-peanut" and want remove that error. They learn because they experience error of not being able to control something. I have describe this in the past as a function of learning. Sensing a lack of control for a skill or understanding (chronic error) reorganization begins in terms of preparing a program for getting that skill (and we could observe results in the actor of programs anywhere from hopelessly inept to very successful). Learning is error reduction.
Mr. Monkey knows what the peanut is, and has a control system for what happens when he put a peanut in his mouth. MMMMMmmmm...! Here the perceptual control is error correction, too. But in the observer monkey, it never gets to happen.
"I think the "monkey" is able to imagine how it is to eat peanuts when he is
watching other monkeys eating peanuts, but, and there is a big but. Is there
a reference in the "monkey's" system signalling that he wishes to imagine
how it is to eat a peanut?"
The monkey is controlling for the action of eating a peanut, too. Powers describes imagination, which perceptual control without outputs getting into the actual environment, all right (no hands moving, or lip smacking?) but outputs may go down to the same areas the monkey activates in both situations (eating and wanting to eat that peanut). But the poor monkey can't act on the perceptual outputs from higher level control systems, either because the hands are held down or there is obviously no peanut he can grab and eat. Would that make the gain even higher on the outputs being sent down?
So, I would say that the issue is what "same cells" really implies.
--Bryan
-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@LISTSERV.UIUC.EDU]On Behalf Of Bj�rn Simonsen
Sent: Friday, January 13, 2006 4:41 AM
To: CSGNET@LISTSERV.UIUC.EDU
Subject: Re: [CSGNET] NYTimes.com: Cells That Read Minds
[From Bjorn Simonsen (2006.01.13,11:45 EUST)]
From Erling Jorgensen (2006.01.12 2100 EST)
So then, one of the descriptions from the New York Times article
stated, "The same brain cells fired when the monkey watched humans
or other monkeys bring peanuts to their mouths as when the monkey
itself brought a peanut to its mouth."
I think this is wrong. .... There is an enormous difference
between what the "student" is perceiving and what the "monkey" perceives.
And when the New York Times article says "the same cells fired when the
monkey watched "The same brain cells fired when the monkey watched humans or
other monkeys bring peanuts to their mouths as when the monkey itself
brought a peanut to its mouth.", it is opportune to ask how they know it is
the same cells.
I think the "monkey" is able to imagine how it is to eat peanuts when he is
watching other monkeys eating peanuts, but, and there is a big but. Is there
a reference in the "monkey's" system signalling that he wishes to imagine
how it is to eat a peanut?
bjorn
Is there less here than meets the eye? When a monkey grasps something, it presumably perceives itself grasping something. So when it perceives another monkey grasping something, there is in both cases a perception of grasping. It would appear to be compatible with pretty much any theory of how brains work, PCT or otherwise, for there to be neurons to be active in both cases, and to be inactive in situations in which there is no grasping. It may be interesting that there is such a localisation of specific (classes of) perception to specific neurons, but the popular extrapolations about how this explains empathy seem to me unwarranted. What am I missing?
--
Richard Kennaway, jrk@cmp.uea.ac.uk, http://www.cmp.uea.ac.uk/~jrk/
School of Computing Sciences,
University of East Anglia, Norwich NR4 7TJ, U.K.
[From Erling Jorgensen (2006.01.13 1200 EST)]
Bjorn Simonsen (2006.01.13,11:45 EUST)
Hi Bjorn,
This is the kind of layering of critique & greater sophistication
that I was hoping for, when I put forth my simplistic first
approximation.
The “monkey” watching this doesn’t grasp, he doesn’t feel the
glass in his hand, he doesn’t lift the glass and he doesn’t
feel the swallowing of water. The monkey sees what the “student”
is doing, he sees the back of the student fingers and he sees
the glass tilting away from the monkey. There is an enormous
difference between what the “student” is perceiving and what
the “monkey” perceives.
I agree. These are the crucial details that get lost behind a
broad-stroke formulation such as doing the “same” thing. All
doing needs to be deconstructed into the perceptions that would
need to be controlled for such a doing-Event to be controlled.
Then we need to ask, as you are doing above, whether the
observer could possibly perceive the same variables the same way.
This starts to rule out many of the possible perceptions I
raised in my schematic of the situation. Actually, my main
candidate for a perception that might be similar, whether
observing or acting, is the Relationship perception of
“toward-the-mouth.” The reference for that relationship
(i.e., “whose mouth?”) would differ between the two conditions.
But the “toward-ness” might be the same. Were we doing the
research, that would generate a hypothesis, to put to the
Test of a Controlled Variable. If such a “mirror neuron”
corresponds to a Relationship perception such as “toward-ness”
or “toward-(a)-mouth”, then one could test whether that
observation holds under different disturbances (e.g., toward-
something-else, or not-toward).
And when the New York Times article says "the same cells fired
when the monkey watched “The same brain cells fired when the
monkey watched humans or other monkeys bring peanuts to their
mouths as when the monkey itself brought a peanut to its mouth.”,
it is opportune to ask how they know it is the same cells.
Agreed. That is very important to check out, & something I
have not done by checking the original article(s). It would
also be important to check what the “same firing” means.
I think the “monkey” is able to imagine how it is to eat
peanuts when he is watching other monkeys eating peanuts,
but, and there is a big but. Is there a reference in the
“monkey’s” system signalling that he wishes to imagine how
it is to eat a peanut?
This elaborates on the question I was alluding to. It is not
clear to me whether we should look for correspondence –
between territory & map – between “mirror neuron” and a PCT
perception node or a PCT reference node.
If we invoke imagining as part of the way we overlay our map
on their territory, aren’t we necessarily talking about
reference nodes in some way. Isn’t that part of our model
for how we envision imagining taking place? Imagining proceeds
the way all behavior does, except it is part of a truncated
or short-circuited control process. So wouldn’t there have
to be requisite reference signals, for the imagining to even
get started? Maybe that’s exactly the point you’re making.
Sorry if I’m being dense.
Thanks for the exchange.
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.
<<<>>>
[Martin Taylor 2006.02.13.16.40]
I feel rather silly butting in when I haven't properly read the relevant papers, but to me the crucial thing I noted was that different cells fired depending on the _intent_ the monkey saw in the action. If I read it properly, "picking up" something is different if the "picking up" is seen as being for the purpose of eating the thing, rather than for the purpose of setting it aside. The Perspective article from Science by Nakahara and Miyashita that I posted as PDF even has "Understanding Intentions" as its main title.
If I misunderstand what I read in skimming, or if they misinterpreted the data, it becoems less interesting.
As a disclaimer, I should point out that I've long thought that perceiving intent is at the real heart of social interaction, and an ability to perceive intent should be expected in any social species, including monkeys and people (and dogs). My Layered Protocol Theory of human interaction is based around the idea that the purpose of dialogue is to perceive the other's intent, and to have the other perceive one's own intent -- at least when the dialogue is cooperative.
Martin
[From Rick Marken (2006.01.13.1630)]
Martin Taylor (2006.02.13.16.40) --
I feel rather silly butting in when I haven't properly read the relevant papers, but to me the crucial thing I noted was that different cells fired depending on the _intent_ the monkey saw in the action. If I read it properly, "picking up" something is different if the "picking up" is seen as being for the purpose of eating the thing, rather than for the purpose of setting it aside. The Perspective article from Science by Nakahara and Miyashita that I posted as PDF even has "Understanding Intentions" as its main title.
So these cells do the test for the controlled variable? Amazing, indeed!
I think the way to determine whether these cells actually detect intention is to present them with stimuli like those in my "Detection of Intention" demo (http://www.mindreadings.com/ControlDemo/FindMind.html). Just present one of the moving squares to the cell on each trial. A "mind readings" cell should only fire when the moving square that is presented is the one that is moving intentionally. The cell should not fire when presented with a passively moved square, even though this square exhibits the same movement pattern as the square that moves intentionally. My guess is that the cell would fire in the same way to the passively moved and intentionally moving square. I think the idea that cells differentially respond to intentional and unintentional movement reflects a deep lack of understanding of the nature of intentionality (purpose).
I think Richard Kennaway (Fri, 13 Jan 2006 16:18:09) provided the best analysis of this "Cells That Read Minds" report: There is less here than meets the eye!
Best
Rick
---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400
[From Bruce Nevin (2006.01.13.1959 EST)]
That's assuming that there is one and only one way to construct a
perception that for the organism represents the intention of the other.
Doing the test for controlled variables is the only correct way that we
know, the only basis for a scientific investigation of intention. A part
of that methodology, we are very careful (or should be) to avoid
projecting our expectations of what we would do if we were that other.
We must compensate for our very natural anthropomorphizing of animals
and, shall we say, "personomorphizing" of other people. We have to
resist these intrusions of inappropriate subjectivity into the objective
procedures of science.
But what is it that we are resisting? (And properly so.) Does it not
amount to constructing perceptions, however inaccurate and fallible they
may be, of the perceptions of others? It is this, I believe, that Martin
is talking about. It is this construction of perceptions -- perception
of other, perception of other's perception of the environment as we
perceive it, perception of other's perception of ourselves, perception
of other's intention -- that we share with some animals, with people
untutored in PCT, and also with people who are expert in PCT (obviously,
or there would be no need to resist them and compensate for them). These
constructed perceptions are unreliable, incomplete, blinkered by
limitations of knowledge, presupposition, expectation, and prejudice,
they are unscientific and unsuitable for building a science, but they
are absolutely appropriate subject matter for PCT.
/Bruce
-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@LISTSERV.UIUC.EDU] On Behalf Of Rick Marken
Sent: Friday, January 13, 2006 7:28 PM
To: CSGNET@LISTSERV.UIUC.EDU
Subject: Re: NYTimes.com: Cells That Read Minds
[From Rick Marken (2006.01.13.1630)]
Martin Taylor (2006.02.13.16.40) --
I feel rather silly butting in when I haven't properly read the
relevant papers, but to me the crucial thing I noted was that
different cells fired depending on the _intent_ the monkey saw in the
action. If I read it properly, "picking up" something is different if
the "picking up" is seen as being for the purpose of eating the thing,
rather than for the purpose of setting it aside. The Perspective
article from Science by Nakahara and Miyashita that I posted as PDF
even has "Understanding Intentions" as its main title.
So these cells do the test for the controlled variable? Amazing, indeed!
I think the way to determine whether these cells actually detect
intention is to present them with stimuli like those in my "Detection of
Intention" demo (http://www.mindreadings.com/ControlDemo/FindMind.html).
Just present one of the moving squares to the cell on each trial. A
"mind readings"
cell should only fire when the moving square that is presented is the
one that is moving intentionally. The cell should not fire when
presented with a passively moved square, even though this square
exhibits the same movement pattern as the square that moves
intentionally. My guess is that the cell would fire in the same way to
the passively moved and intentionally moving square. I think the idea
that cells differentially respond to intentional and unintentional
movement reflects a deep lack of understanding of the nature of
intentionality (purpose).
I think Richard Kennaway (Fri, 13 Jan 2006 16:18:09) provided the best
analysis of this "Cells That Read Minds" report: There is less here
than meets the eye!
Best
Rick
---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400
[From Rick Marken (2006.01.13.1730)]
Bruce Nevin (2006.01.13.1959 EST)
That's assuming that there is one and only one way to construct a
perception that for the organism represents the intention of the other.
I think you're replying to me. No, what I said does not assume that. It assumes that the researches were talking about cells that respond differently to intentional and unintentional behavior.
Doing the test for controlled variables is the only correct way that we
know, the only basis for a scientific investigation of intention. A part
of that methodology, we are very careful (or should be) to avoid
projecting our expectations of what we would do if we were that other.
We must compensate for our very natural anthropomorphizing of animals
and, shall we say, "personomorphizing" of other people. We have to
resist these intrusions of inappropriate subjectivity into the objective
procedures of science.But what is it that we are resisting? (And properly so.) Does it not
amount to constructing perceptions, however inaccurate and fallible they
may be, of the perceptions of others? It is this, I believe, that Martin
is talking about.
If so then he is talking about something different that what the researchers were talking about. What you are describing here is cells that perceive intentionality (a relationship level perception, at least), whether the behavior perceived was actually intentional or not. I would guess that there are such cells to be found because, as you note, we perceive intentionality in behavior whether or not the behavior was actually intentional. For example, I've sometimes perceived the actions of the wind as intentional when I was trying to read papers outside. But the researchers are saying, not that the cells are constructing perceptions of intention (whether or not the action perceived is _actually_ intentional). They say specifically that these cells "fire in response to chains of actions linked to intentions". In other words, these cells differentially respond to intentional vs unintentional actions in the same way that some cells respond differentially to long vs short wavelength light.
It is this construction of perceptions -- perception
of other, perception of other's perception of the environment as we
perceive it, perception of other's perception of ourselves, perception
of other's intention -- that we share with some animals, with people
untutored in PCT, and also with people who are expert in PCT (obviously,
or there would be no need to resist them and compensate for them).
I agree. And I believe that there must, then, be afferent neurons, at high levels in the nervous system, that carry signals that are perceptions of intentionality. But I don't believe (as the researchers do) that our perceptions of intentionality are "veridical".
These
constructed perceptions are unreliable, incomplete, blinkered by
limitations of knowledge, presupposition, expectation, and prejudice,
they are unscientific and unsuitable for building a science, but they
are absolutely appropriate subject matter for PCT.
Of course. But the conclusion of the researchers -- that they have found cells that detect intentionality (that they are cells that read minds) -- is almost certainly incorrect. And it's incorrect because they don't know what intentionality _is_.
Best
Rick
Phil Runkel replying to Martin Taylor [2006.02.13.16.40]:
"... I've long thought that perceiving intent is at the real heart of social interaction, and an ability to perceive intent should be expected in any social species, including monkeys and people (and dogs)."
Hear! Hear! Of course! Certainly! And we make our estimates, seems to me, just as Marken's models do.
--Phil R.
[From Rick Marken (2006.01.14.0840)]
Here's a post from yesterday that didn't make it through because I posted it from the wrong address:
-----
[From Rick Marken (2006.01.13.1100)]
Richard Kennaway
Is there less here than meets the eye? When a monkey grasps
something, it presumably perceives itself grasping something. So
when it perceives another monkey grasping something, there is in both
cases a perception of grasping. It would appear to be compatible
with pretty much any theory of how brains work, PCT or otherwise, for
there to be neurons to be active in both cases, and to be inactive in
situations in which there is no grasping. It may be interesting that
there is such a localisation of specific (classes of) perception to
specific neurons, but the popular extrapolations about how this
explains empathy seem to me unwarranted. What am I missing?
My guess is that you are (as usual) missing nothing.
We know that people (and probably animals) sometimes imagine themselves doing what they see others doing. PCT provides on explanation of how this might be done. However it's done, it's done in the brain. So there must be some physiology that underlies this phenomenon. The physiological result described in the article _might_ be related to this phenomenon, but -- as you so clearly point out -- it might not. I think the main point is that you can't look to the physiology for explanations of functional phenomena. That seems to me like looking at the hardware of a processor to understand how a software program works.
As you say, the same physiological result may be consistent with many different functional explanations of a phenomenon. I think physiology can show how a particular explanation might be implemented in "wetware", but in order to show this the physiological work must be based on the functional model. That is, once you have a model of the software that produces the behavior (in our case, PCT) you can see how that software might be implemented in hardware -- what kinds of hardware processes, like AND and NOR gates, are required to carry out the proposed functions of the software. If the physiologists did the research to test a particular functional model of imagination then the results might be interesting. Of course, if they did that, then they would surely have done the research quite differently and they would probably not have come to rather far-fetched conclusion that there are special "mirror cells" that detect intentions.
Best
Rick
---
Richard S. Marken, PhD
Psychology
Loyola Marymount University
Office: 310 338-1768
Cell: 310 729 - 1400