[Bulk] Re: NYTimes.com: Cells That Read Minds

CSGnet Archive CSG2006
1

[Martin Taylor 2006.01.14.00.07]

[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 (Detection of Intention). 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.

Again, I know it's silly of me to enter this discussion without having properly read the papers, but my meager understanding suggests that this couldn't work, because the situation doesn't look to the observer like "that's what I would be intending if I were performing those actions". The observer can't see the computer's arm movements making the dots chase around the screen, so there's nothing to go on in inferring the reason it is making them go the way they do. The observer can see, and more importantly could do, the actions performed by the actor, from which the intent, if perceived, must be inferred.

[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.

I never read (or wrote) anything like that, but if you got that from a careful reading of the various papers, I'm not going to argue.

was it a monkey) saw another pick up something apparently intending to eat it, the cells that would have fired had the observer been intending to eat became active, but they didn't when the actor picked up something apparently intending to set it aside.

I can see I'm going to have to make time to read the relevant papers I can get my hands on, rather than waffling away like this.

Sorry for wasting your time.

Martin

···

From my cursory reading, I got the impression that when someone (or

2

[From Rick Marken (2006.01.14.0915)]

Martin Taylor (2006.01.14.00.07)

Rick Marken (2006.01.13.1630)

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 (Detection of Intention). 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.

my meager understanding suggests that this couldn't work, because the situation doesn't look to the observer like "that's what I would be intending if I were performing those actions".

It would certainly "work" as a method of determining whether these cells are actually detecting intention. You would have to adjust the methodology a bit to test the even more bizarre claim that these cells are detecting that "that's what I would be intending if I were performing those actions".

The observer can't see the computer's arm movements making the dots chase around the screen, so there's nothing to go on in inferring the reason it is making them go the way they do. The observer can see, and more importantly could do, the actions performed by the actor, from which the intent, if perceived, must be inferred.

As I read over the article I see that "intention" really has nothing to do with this. All the researchers are seeing is cells in an observer that make the same response to the same perception -- like the perception of a thrown ball -- whether that perception was produced by someone else or by the observer themselves. This is hardly a surprising result. It's like discovering a cell in an observer that responds to a circle whether the circle was drawn by the experimenter or by the observer himself. Again, I think Richard Kennaway put his finger on the essence of this research when he observed that there is less here than meets the eye. I would say far less.

From my cursory reading, I got the impression that when someone (or was it a monkey) saw another pick up something apparently intending to eat it, the cells that would have fired had the observer been intending to eat became active, but they didn't when the actor picked up something apparently intending to set it aside.

If the "picking up" with the intention to eat looked exactly like the "picking up" with the intention to put down and differed only in the underlying intent then you've, indeed, got some pretty amazing cells there: cells that detect intention. If, however, as seems more likely, the two examples of "picking up" are as different as two different wavelengths of light, then what you have are two different physical events and it is no surprise that a cell might respond differentially to these two events.

I can see I'm going to have to make time to read the relevant papers I can get my hands on, rather than waffling away like this.

Sorry for wasting your time.

Not a waste at all.

Best regards

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

3

[Martin Taylor 2006.01.16.16.40]

[From Erling Jorgensen (2006.01.16 1415 EST)]

On one run I did control for a zero-perception of the
cursor postion, by holding an envelope across the screen.
...
So, technically, that did help to define one limit
condition of your elegant simulation, that as it currently
stood, it could not differentiate passive observation
from controlling for a zero value of the perception.

You didn't control "for zero value of the perception"; you controlled for not having a defined value for the perception, which is a very different thing. I don't imagine that at the moment you perceive any particular value for the colour (right now) of the tie of the person living at 31 Planetree Drive, Upper Hertsworth, do you? If the person is wearing a tie, its colour isn't zero. For you, the colour is indeterminate, just as is the _position_ of the cursor when you obscure it from your vision. It's a perception that you don't have (other than in your imagination) at the moment.

A "zero value" for the perception of the position of the cursor is a very precise position perception. It says that the cursor is at whatever location you have specified to be represented by zero.

When you cover the screen with an envelope, the cursor's perceived position doesn't abruptly go from (say) "3 cm left" to dead centre (i.e. "zero"). But its precision goes from (say) +- 2mm at the moment you cover the screen to (say) 10 cm after several seconds have passed. How quickly you would lose precision would depend on the bandwidth of the cursor motion, but you would indeed lose it. A perception with no precision is not a perception that can be controlled.

Martin

4

[From Erling Jorgensen (2006.01.16 1800 EST)]

Martin Taylor 2006.01.16.16.40

A "zero value" for the perception of the position of
the cursor is a very precise position perception. It
says that the cursor is at whatever location you have
specified to be represented by zero.

So it appears that Rick's elegant simulation, as it stands,
can indeed model the condition of a zero-value of the
position-perception. In fact, that is the customary way
of doing the demo, using the center-target as an
arbitrary zero. I stand corrected, (well, sit, really).

For you, the colour is indeterminate, just as is the
_position_ of the cursor when you obscure it from your
vision. It's a perception that you don't have (other
than in your imagination) at the moment.

This would further indicate that my suggestion of breaking
the perceptual link to make it open loop (e.g., blanking
out a one inch band beneath the target) would be enough
to make the model match my unusual experimental condition,
with the (trivial) result of insuring essentially no
correlation (inverse or otherwise) between mouse movements
and the (unseen) disturbance function.

It seems, Rick's model could easily survive both these
falsifiability conditions.

You didn't control "for zero value of the perception";
you controlled for not having a defined value for the
perception, which is a very different thing.

Aren't there some situations where we control for an
absolute zero reference, rather than just an arbitrarily
defined zero value? My brain doesn't seem to be working
real well just now, but what about wearing an eye-mask
while trying to sleep during the daytime, to control for
zero amount of perceived sunlight?

In your example of the person's tie in Upper Hertsworth,
I wasn't controlling for not seeing it, I was not in a
postion to see it -- (I don't get around to Upper
Hertsworth much.) The color, for me, was indeterminate,
but so was my reference. It was not a situation that
mattered to me.

I am wondering if there a difference between actively or
deliberately ignoring a given perception -- a situation
of intent & thus control -- versus not having noticed (or
learned to notice) that perception? Would we model these
two situations differently?

I think it has implications for modeling learning. To me,
it seems that an agreed learning environment is one where
one is taught references for how to bring perceptions into
being that may not have been there before. A talented
teacher often conveys a sense of their own passion, that
these things matter and are worth caring about. And I
believe it may include, not so much learning answers, as
learning to ask relevant questions and care about the
answers.

These exchanges are helpful. Thank you to you & Rick.

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.
<<<<RCMH>>>>

5

[Martin Taylor 2006.01.17.13.41]

[From Rick Marken (2006.01.17.0800)]

... As far as zero value of the perception, there is presumably zero value of the cursor position perception when you are not looking at the cursor (or covering it up with something)

Having no value is not the same as having a value that happens for the moment to be zero.

If the value is zero, you can act on the environment to change it. If there is no value, nothing you do in the environment can alter the value so as to bring it closer to its reference (although you could alter the environmental feedback path so as to make the perception have a value, or you could imagine or remember a value, after which you would be able to act to change it).

Martin

6

[From Rick Marken (2006.01.17.1350)]

Martin Taylor 2006.01.17.13.41]

Rick Marken (2006.01.17.0800)--

... As far as zero value of the perception, there is presumably zero
value of the cursor position perception when you are not looking at
the cursor (or covering it up with something)

Having no value is not the same as having a value that happens for
the moment to be zero.

I don't see that. I was thinking of the perceptual signal, which is presumably represented
as rate of neural firing. So when the rate of neural firing is virtually zero impulses
per second (there is always _some_ firing) there is zero perception.

If the value is zero, you can act on the environment to change it. If
there is no value, nothing you do in the environment can alter the
value so as to bring it closer to its reference (although you could
alter the environmental feedback path so as to make the perception
have a value, or you could imagine or remember a value, after which
you would be able to act to change it).

I don't understand this. How does zero perception (in terms of neural perceptual
signals) differ from no value?

Best

Rick

Richard S. Marken, PhD
Psychology
Loyola Marymount University
Office: 310 338-1768
Cell: 310 729 - 1400

7

[From Rick Marken (2006.01.17.2200)]

I tried to post this earlier from work but it hasn't shown up yet. I may eventually but just for the sake of keeping up the conversation I'm sending it off again. I apologize for the likely repetition.

···

----
[From Rick Marken (2006.01.17.1350)]

>Martin Taylor 2006.01.17.13.41]
>
>> Rick Marken (2006.01.17.0800)--
>>
>>... As far as zero value of the perception, there is presumably zero
>>value of the cursor position perception when you are not looking at
>>the cursor (or covering it up with something)
>
>Having no value is not the same as having a value that happens for
>the moment to be zero.

I don't see that. I was thinking of the perceptual signal, which is presumably represented
as rate of neural firing. So when the rate of neural firing is virtually zero impulses
per second (there is always _some_ firing) there is zero perception.

>If the value is zero, you can act on the environment to change it. If
>there is no value, nothing you do in the environment can alter the
>value so as to bring it closer to its reference (although you could
>alter the environmental feedback path so as to make the perception
>have a value, or you could imagine or remember a value, after which
>you would be able to act to change it).

I don't understand this. How does zero perception (in terms of neural perceptual
signals) differ from no value?

Best

Rick

Richard S. Marken, PhD
Psychology
Loyola Marymount University
Office: 310 338-1768
Cell: 310 729 - 1400

Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

8

[Martin Taylor 2006.01.19.08.19]

[From Rick Marken (2006.01.18.1930)]

Rick Marken (2006.01.18.0830)

Martin Taylor (2006.01.18.09.51) --

I'm pretty busy today so I probably won't have time to give a complete reply until later this evening. I'll just quickly reply to one of your points.

Rick Marken (2006.01.17.0800)--

I don't see that. I was thinking of the perceptual signal, which is presumably
represented as rate of neural firing. So when the rate of neural firing is
virtually zero impulses per second (there is always _some_ firing) there is
zero perception.

Don't think of intensity perception, think of position, and think of
your demo.

I was. Remember, in PCT the value of a perceptual _variable_, be it intensity or principle, is represented by the rate of afferent neural firing.

Well, now that I have time I see that I don't really have much to add to what I said when I was rushed. So it's your turn;-)

Should I simply repeat my proposal for an addition to your demo? I think both the concept of it and the results of experiments with it would make the point pretty clearly. Or should I simply repeat the question raised in the first part of the message top which you claim to be responding? I think I'll do the latter, redundant though it may seem to be:

If I want to control for the cursor being at position zero, and it is in the middle of the screen, what do I do? If I want to control for the cursor being at the middle of the screen and I can't see it, what do I do?

Those two conditions are different, aren't they?

Another way of asking the question is: "Do I perceive the same thing when I can see that X has a value of precisely zero as I perceive when I can't see X at all?"

I guess a simple yes or no would do, but thinking about the implications of either answer might be better.

The rate of firing of a perceptual input function really isn't the issue at all.

I think that your difficulty with this may well be very closely linked to the problem you had with conceiving that a perceptual signal might have a value partly due to imagination. When you thoguht about that for a while, you seemed to understand it. Maybe thinking about the implications of a "yes" or a "no" answer to the quoted question might have a similar effect.

Martin

9

[From Rick Marken (2006.01.19.0830)]

Martin Taylor (2006.01.19.08.19)

Should I simply repeat my proposal for an addition to your demo?

Sure. I really didn't understand it at all. In particular, I didn't understand what it would be testing.

I think both the concept of it and the results of experiments with it would make the point pretty clearly.

To you perhaps. It made no sense at all to me. Maybe you could give it another try.

Or should I simply repeat the question raised in the first part of the message top which you claim to be responding? I think I'll do the latter, redundant though it may seem to be:

If I want to control for the cursor being at position zero, and it is in the middle of the screen, what do I do? If I want to control for the cursor being at the middle of the screen and I can't see it, what do I do?

Those two conditions are different, aren't they?

Another way of asking the question is: "Do I perceive the same thing when I can see that X has a value of precisely zero as I perceive when I can't see X at all?"

What is X that it can have a value of precisely zero? If X is the physical position of the cursor on the screen then any numbers we use to represent that variable are arbitrary, are they not. The same is true if X is the position of the image of the cursor on the retina.

I guess a simple yes or no would do, but thinking about the implications of either answer might be better.

I can't answer it yes\/no because I don't understand the question. I assume that X is the position of the cursor in the physical world (on the screen or on the retina). In that case, I have no idea what you mean when you ask what I see when X has a value of precisely zero. When does X have a value of precisely zero?

The rate of firing of a perceptual input function really isn't the issue at all.

It's the firing of the _output_ of the perceptual input function that I was talking about. And there is a physically meaningful zero point for that variable; it's when the output is not firing at all; zero (or nearly zero) impulses per second. For X out in the physical world, any physical position could be represented by zero, with positions to the left and right sides being positive and negative numbers, respectively.

I think that your difficulty with this may well be very closely linked to the problem you had with conceiving that a perceptual signal might have a value partly due to imagination. When you thoguht about that for a while, you seemed to understand it.

Actually, it wasn't "thinking about it for awhile" that helped; it was seeing it demonstrated in my open loop control demo (integral control case). I'm still not convinced that imagination is ever a particularly significant component of our everyday controlling however, since, as is shown in the demo, when you control a perception with a significant imagination component you are not really controlling at all.

Maybe thinking about the implications of a "yes" or a "no" answer to the quoted question might have a similar effect.

Perhaps. But you have a better chance if you can point me to a concrete demo that will help me understand what you mean.

Best

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

10

[From Rick Marken (2006.01.23.2150)]

>Rick Marken (2006.01.19.0830)--

>> Martin Taylor (2006.01.19.08.19)

>> Should I simply repeat my proposal for an addition to your demo?

>Sure. I really didn't understand it at all. In particular, I didn't
> understand what it would be testing.

Are you ever going to explain that proposal to me, Martin?

Sorry about the Canadian election, by the way. But a Canadian conservative can't possibly be as awful as an American one, can it? What did those Liberals do? Blow job again? :wink:

Best

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

11

[From Rick Marken (2006.01.24.0915)]

Martin Taylor (2006.01.24.01.20)

Rick Marken (2006.01.23.2150)

Are you ever going to explain that proposal to me, Martin?

When I think of some other way of suggesting that seeing that something is exactly zero is different from not seeing what its value is.

When you talk about neural firings going to zero, you really aren't talking in the same domain. If the model I conceived were to say that seeing something to be precisely zero yields the same perception as not seeing what the value is, then I would begin to think that my model might have some minor flaw.

I still don't understand what you mean by "seeing something to be precisely zero". We were talking about a specific example: the tracking task that is the first of my demos ("The Nature of Control"). What is being perceived is the position of a cursor and target. What is the state of that display that is seen as "precisely zero"?

You believe so strongly in your model that you don't think there should be a difference in the two perceptual situations.

What two perceptual situations? If you mean the difference between seeing the display and not seeing it then I think you should know that I consider these to be two very different perceptual situations.

In your mind, the model trumps the observations, and I don't know how to go about explaining it differently than I have heretofore.

I don't think so. I'm trying to think of how the model would differentially represent the two different situations that I experience: 1) the situation where I am controlling the distance between cursor and target (a perceptual variable that is presumably represented by the rate of firing in an afferent neuron) and 2) the situation where I can't see the screen at all (which I would represent in the model as zero firing in the afferent neuron that represents the perceptual variable that had been controlled).

Maybe if you were to make the modifications to your demo that I suggested, you could see what I'm trying to get across.

But the problem is that I don't understand the modifications you are suggesting.

Here's what you suggested:

Instead of having the mouse control the screen display directly, have
it control through a hidden (2-D) variable I'll call {P,Q}, and let
the screen {x,y} be {P+a,Q+b}, where a and b are variables. Give the
experimenter/subject two sliders for each of a and b, one of which
determines the standard deviation of the magnitude of a and b, the
other of which controls the bandwidth of the variation (which must be
slower than variation in P and Q. The subject's task now is to
control {P,Q}, not the screen position. Let a and b be zero for some
time at the start of a run before adding in the variation.

First of all, what's the 2-D variable? What affects the values P and Q? Why are we doing this in two dimensions (x,y) when the tracking tasks takes place in only one? What are the slides that affect the values of a and b? Are the sliders the mouse positions in x, y desk coordinates? If the mouse does affect a and b then are P and Q constants? If so, why have them? As I read on I see that the sliders actually affect the SD and magnitude of a and b so I guess a and b are varying over time on their own, the SD and magnitude of this variation being determined by the current settings of a and b. Is that right? So a and b are disturbances, right? So now I think P and Q may be the mouse positions. Is that right? Finally, you say that the subject is asked to control P Q. Since P and Q are presumably just numbers in the computer how does the subject even know what you're talking about? Since the subject can't perceive P Q he can't control P Q. So what am I missing here? And how does this suggestion pertain to the question of how we represent the situation where we no longer perceive a variable (such as the cursor-target relationship) because we've covered it up?

Sorry about the Canadian election, by the way. But a Canadian conservative can't possibly be as awful as an American one, can it? What did those Liberals do? Blow job again? :wink:

It may not be so bad.

Thanks for the analysis. I find your political analyses far more comprehensible (and compatible with my own understandings) than some of your scientific analyses;-)

Best

Rick

···

---
Richard S. Marken Consulting
marken@mindreadings.com
Home 310 474-0313
Cell 310 729-1400

12

[From Bryan Thalhammer (2006.01.24.1151 CST)]

That last message was not meant to be sent to the net. However, based on that event, I will now footnote.

Mooney, Chris (2005). The Republican war on science. New York: Basic Books. ISBN: 0-465-04675-4.

The text is a good read, and for scientists, it IS a good read, but terrifying when you think of the way research is being misused.

I do use alternate sources for news and information, since I do not trust the presently co-opted MSM (mainstream media). Therefore several good starting points might be:

Air America Radio
http://www.airamericaradio.com/
(in Canada, if you have a decent connection, you can hear it online).

Daily Kos
http://www.dailykos.com/
(Science and its misuse are also an important topic of discussion there.)

Truly, Conservative has many meanings. I prefer the meaning that is often used now, that conservative means saving something worthwhile, so I can be a Constitutional "Conservative", meaning that we have to begin saving our Constitution before it gets kicked into a corner.

Sorry for my candor, but I guess the Devil made me do it... Please throw stones in my general direction. :wink:

--Bryan

13

[From Rick Marken (2006.01.24.1350)]

Martin Taylor (2006.01.24.14.30]

Rick Marken (2006.01.24.0915) --

I still don't understand what you mean by "seeing something to be
precisely zero".

In the absence of some special functional organization not specified
in any model I know, this arrangement would mean that zero firing
rate corresponds to a perception of the most leftwards imaginable
position. However, you take this firing condition to mean that you
don't perceive the position to be anywhere at all. To me, that's a
contradiction.

I don't believe that cursor position is the variable controlled. And I don't think the left right position of the cursor is represented in terms of firing rate, with 0 being farthest one way and max firing being the farthest the other way. I think zero firing rate has to be reserved for zero of whatever perception is computed by the perceptual function; that is, zero firing has to mean no perception. I would guess that what is controlled in the tracking task is the distance of the cursor from the target and that this distance -- left or right of the cursor -- is represented by rate of firing, with zero distance being some intermediate firing rate, left distance being rates higher and right distance being rates lower than that intermediate value.

I can't see any way out of the problem that you are using the same
firing rate to result in two quite different perceptions at the same
time: that the position is maximally leftward, and that the position
is unknown.

Right. We can't let that happen. I makes most sense to me to have zero firing rate represent none of the perception and other firing rates represent other values of the perception. So if the perception is of zero deviation from the cursor the IS a perception -- there not NO perception -- so that zero deviation has to be represented by a positive rate of neural firing.

I do see a way out if there are two separate dimensions of the
perception: the most probable value of the thing perceived
simultaneously with a separate perception of the uncertainty of that
perception.

I like my way out better;-)

Maybe if you were to make the modifications to your demo that I
suggested, you could see what I'm trying to get across.

But the problem is that I don't understand the modifications you are
suggesting.

I'll try again, without reference to what I wrote before, so as to
make it more likely that one wording will illuminate the other.

You have a bunch of objects that move around the screen with random
relative motion. ..

So, what I now suggest is that you allow the subject to control
normally for the first part of the run, but at some moment you reduce
the contrast between the cursors and the background, either by
changing the background or by chaging the cursors. Changing the
background might be best, because it allows you to generalize to a
range of different experiments based on the spatial statistical
character of the background masking (changing brightness overall, or
putting random dots or lines or structured patterns into the
background, for example).

What you should be able to do is generate a graph of the time course
of loss of control as a function of the degree of contrast reduction
or masking, using the same statistics you use to "mind-read" which
cursor is being controlled.

I agree. This would be a fun augmentation of the demo. I'll consider it.

Best

Rick

Richard S. Marken, PhD
Psychology
Loyola Marymount University
Office: 310 338-1768
Cell: 310 729 - 1400