SpongeMan's Sensorimotor Manifesto [hide]
Search [hide]
Last forum posts [hide]
Last wiki comments [hide]
Last visitors [hide]
1)   JessMartone at Sep 03, 2010 [15:33]
2)   wangchao45 at Aug 29, 2010 [01:30]
3)   chris251984 at Aug 28, 2010 [02:31]
4)   dollievaruna at Aug 25, 2010 [03:26]
5)   FrankHill at Aug 22, 2010 [00:17]
6)   judy6kline at Aug 18, 2010 [02:04]
7)   jmjm123 at Aug 17, 2010 [05:56]
8)   allensmith at Aug 16, 2010 [10:51]
9)   resumeplus at Aug 13, 2010 [11:05]
10)   pattypetrillo at Aug 10, 2010 [17:57]
Last changes [hide]

Forum: Discussion on the sensorimotor approach

Forums->Discussion on the sensorimotor approach->"Skills," "attention," "access"

PaulReeve
"Skills," "attention," "access"


davidphilipona wrote:
> > it seems to me that the sony aibo is able to exercise a few sensorimotor skills, like walking, but I doubt very much that it "has feel", and I would suggest this is because it has not cognitive system. As far as I'm concerned, I would therefore on the opposite stress the notion of cognitive access, whatever this can mean.

kevin wrote:
> i agree witht his.

my reply:
Why do cognitive systems have feels and not non-cognitive systems, just incidentally?

kevinoregan wrote:
> concerning whether "skill" already implies cognition:
> first what is cognition? Your immune system could be considered to do
> categorisation of foreign cells... a soap bubble might be said to have cognitive
> capacities since it calculates its shape by making a minimum of surface area.

my reply:
This soap-bubble notion truly seems like a terrible idea. But in fact it seems to touch on a global problem with the skill/cognitive access/knowledge approach as you describe it. See Daniel Hutto's paper from the latest issue of Phenomenology and the Cognitive Sciences on the relationship between the laws of physics and sensorimotor skills and laws and "knowledge." (NB I don't endorse everything in that paper, but there's enough thoughtful material, including arguments already raised here, to warrant careful scrutiny.)

kevinoregan wrote:
> > Maybe david can find us a definition of cognitive capacities which is independent of the point of view in which you place yourself, the way in which you code the behaviors that you measure.

davidphilipona wrote:
> I just gave one suggestion. It's only a suggestion, maybe this one is not good, but the point is: I don't see why suggestions of that kind would have anything to do with the way you code behaviors.


my reply: The "coding" (theoretical or internal to the system? I don't actually get why this issue is raised here, but anyway) of behaviours does seem to be relevant here. Let me try and describe how it's relevant to my problems with the skill analogy.

Describing ordinary skills seems to involve describing the behaviours involved, in terms that relate to the goals achieved by the behaviours; in fact they are defined as skills exactly by being the ability to achieve particular behavioral goals in context-relevant ways. Cricket batting as Ed describes it seems to involve being sensitive to a range of factors in a situation - categorizing it, as Aline puts it, in a very fine-grained way - but then that categorization (as described) is actually defined by its contribution to shaping particular actions in the service of particular goals, and so the "coding" of the behaviours themselves seems to be fundamental to saying what the skill is. Sensory perception seems to encompass the sensitivities that are fundamental to the enabling of all and any physical behaviour; if describing behaviour is involved in describing some particular sensory-perceptual skill, the relevant subset of actions will be defined not by any particular set of goal-constrained actions, but by the range of information (find me another word here, Ed) about the situation the perceiver is in that that by virtue of the "skill" can have a shaping influence on any behaviour whatsoever. (So only those with a colour-seeing "skill" can go if the light is green and stop if it's red - but they can also run red lights to show off to friends, etc.) This may seem okay - you could try and describe the skill in terms of the way that sensory-perceptual sensitivities in given organism-environment situations can constrain whole space of physically possible actions, seeing the set of actions enabled by particular sensitivities as the behavioural side of the skill. But one worry for sensorimotorism that seems pretty obviously to loom up here: when we go to ask how we understand the ways that these so-called skills can "code" such generalized constraints on the space of possible behaviours, it seems hard to imagine that people won't just want to say that people have access to a "representation" of the situation in terms of relevant sensory-perceptual information that goes into shaping their behaviour. The role of action then becomes one of just abstractly fixing the "content" of perception, and we're back something very close to the view that we wanted to get rid of, right? And the problems that come along with it. Saying we have the "skill" of being able, say, to take any action whatsoever in response to the presence or absence of a certain colour doesn't seem to be a very convincing way of undermining the argument that there's something surprising about our having an experience linked into whatever it is that gives us that ability. And if the skill is just detecting the colour (this does seem to be the idea), once again: what is the value-added of the word "skill"?

Exactly the trouble I was trying to make clear in my talk in Antwerp in 2004.

kevin wrote: > Maybe david can find us a definition of cognitive capacities which is independent of the point of view in which you place yourself, the way in which you code the behaviors that you measure. But while we wait for him to find the solution, if there is one, I prefer to say that yes, all these things may be forms of cognition (with skills having the additional characteristic that they involve sensorimotor interactions).
>
> But when we have a feel, it is not sufficient for sensorimotor processes to be going on which can be considered "cognitive" or skillful. There has to be a kind of meta- level of cognition which could loosely be called "paying attention to": The organism, considered as an *individual*, acting in an *environment*, must be poised to make use of the skill in its rational behavior. Otherwise the aibo has feel.

my reply: "Rational"?!?!

This description of attention seems far less radical than the Merleau-Pontyan version of attention that Ed endorses. On this view perceivers are "always already" oriented in the world, and their perceiving is shaped by that orientedness, which combines their spatial focus, their motivations, their memories, etc. "Attention" is a kind of a crude way of describing some limited part of what the whole deployment of the perceiver's sensory skills entails; it has the merit of acknowledging that the skills are actually deployed, with consequences that include spatial focus, for example, but it seems on this view to be a confusion to suggest that this is an extra "meta" factor; it's just what it is to have and use the perceptual skills we actually have.

Then again I may be getting this all very wrong, Ed can tell you what he thinks better than I can.

davidphilipona wrote:
> > I must say that this notion of "focusing attention" is to me the most fuzzy concept of the whole theory (ce qui n'est pas une mince performance)
>
kevinoregan wrote:
> well i mean byfocussing attention the sameas what i mean by having cog access, I think.

my reply:
Again the confusion: is attention/access focused on the particular qualities picked up through "skilful" perceiving, or on the "skill" itself?!

aline wrote:
> > > and so on... are
> > >already included in access-consciousness so why not using
> > > it?
> >
> >
davidp wrote:
> > If I don't mistake, Kevin prefers to speak of cognitive access rather than A-consciousness because the later seems to him to be a stronger concept, involving the notion of self. (right, Kevin?)
>
kevin wrote:
> exactly. Cognitive access + notion of self having this access = conscious access.

my reply:
Whose idea is it that "access consciousness" is defined without regard to an "accessor"? Can you cite any reference that shows this concept to be different from Block's?

kevin wrote:
> everybody seems to dislike it when I say "having cog access to *the fact that* one is exercising a sm skill".
>
> The reason I say "the fact that" is that when you have a feel of red, you have no
> access to any of the details of the skills involved: you dont know what the rules
> are that the piece of paper is obeying as you move it around under the lights. If I
> had not said *the fact that*, then it sounds like you have access to the skill as a
> whole, all its details and interior workings. Maybe I should find another way of
> expressing the idea however, which is that while you dont have access to the
> interior workings, you CAN nevertheless distinguish the different laws, and you
> can even compare and contrast them: thus you DO have access to the fact that
> the current "red" rules are DIFFERENT from the rules of green, and that they are
> somewhat more similar to the rules of pink. Doing many athletic or motor tasks
> seems to me quite a good analogy: an athlete has no idea exactly what muscles to
> put into action in order to accomplish his feat, yet he knows how to do his feat in
> different ways. Dance teachers use all sorts of bizarre analogies to coach their
> students to move their bodies in certain ways (e.g. todo a "battement de jambe",
> you have to move out your foot as though it is a brush, brushing the floor on its
> way out). It's no use explaining to a dancer what muscles to use.

my reply:
See above, also Hutto paper, also earlier stages of the email discussion that preceded this thread. You're really not answering the challenging questions about the physical skill analogy at all.

You seem to be saying here that a) colours are distinguished by sensorimotor laws and b) we are able to distinguish colours. We very obviously don't know that we are ongoingly distinguishing them on this basis - and indeed the scientific evidence for this proposition actually seems for the moment to be weak at best. So we in fact don't at all have "access to the fact that the current "red" rules are DIFFERENT from the rules of green," we can just distinguish red and green! Again (again again again): what is the "access" here supposed to be access to?

Let me remake the same, clearly still 100% unanswered, point: motor skills clearly amount to the ability to physically do something; though we obviously don't know just what muscles we contract when we carry the actions out, they are identifiable as physical actions! What is the "skill" of seeing the ability to do?

I expect the answer goes something like this: the "seeing" or "colour seeing" skill is the ability to carry out actions as a function of the presence in our environment of stuff that is distinguished from other stuff by being in particular sensorimotor relationships to us. But see above. I also strongly recommend that you carefully read the Hutto paper in this context.

(Conceivably you could make this analogy run more smoothly by making it with abstract skills, like say mental calculations. ?)

aline wrote:
> > > it seems to me that part of the confusion comes from the
> > > mixing of A- and P-consciousness,
> > > then, when Kevin says that "reactivity and heaviness of the
> > > steering ... are aspects of the phenomenology of driving
> > > a car", he means that this level is phenomenally relevant
> > > (in opposition with lower levels, who have no phenomenal
> > > content)... though it could be that this level is just
> > > AC-relevant (a level of description!), because the
> > > community has words/concepts for them...
> > > so, one solution is to propose that phenomenal content is
> > > something that is added to AC when exercising a
> > > sensorimotor skill,
> > > then, only what is accessible to AC can have a phenomenal
> > > content,
> > > but, since not everything that is accessible to AC
> > > possesses a phenomenal content, the two notions remain
> > > distinct...
>
kevin wrote:
> yes, I think if I understand you correctly, that this is what I say: PC = AC + exercising a skill
>
> (although actually in order to let animals have phenomenality I say:
>
> "feel" = Cog Access of exercsiing a skill
>
> (this definition requires no self)
>
> Then, for humans who have selves:
>
> phenomenal consciousness = self knowing it has a feel

my reply:
I find all this totally incomprehensible.

aline wrote:
> > > "It suffices that the perceiver have cognitive access to
> > > the fact that particular sensorimotor laws are currently
> > > potentially applicable."
> > > I would refer to prediction here,
> > > what would it be to be "exercising a skill while currently
> > > not doing anything" if a prediction?
>
kevin wrote:
> I dont like the notion of preditions which carries the danger of an internal replica of what is being predicted

my reply:
Why don't SMCs run this danger?

aline wrote:
> > > about the "presence" of sensation and the fact that they
> > > are both under and not under our control,
> > > why do you mention this (rather undefined) word "presence"?
>
kevin wrote:
> I think the word "presence is used in the phenomenologist's tradition (Husserl?) to refer to the special aspect of sensations as compared to other mental events. Erik?
 
on: May 31, 2006 [15:30] score: 0.00 reads: 2923

Posted messages

Top
author message
wedding
wedding
on: Jun 24, 2010 [06:39] score: 0.00
Her company released the first line of imprintable wedding invitation and cheap wedding invitations , do-it-yourself bride who wanted a hand in making her own wedding invitation wording without sacrificing a professional quality dress up games for girls and dress up games for girls. Anna Griffin elegant traditional patterns and all the details that make fashion games, well, fine.

author message
wedding
Re: "Skills," "attention," "access"
on: Jun 24, 2010 [06:38] score: 0.00
In the years since, the name Anna Griffin has become synonymous with high quality paper crafting designs. Her company released the first line of imprintable wedding invitation and cheap wedding invitations for the home crafter, do-it-yourself bride who wanted a hand in making her own wedding invitation wording without sacrificing a professional quality dress up games for girls and dress up games for girls. Anna Griffin invokes vellum make fashion games, well, fine.

author message
KevinORegan
Re: "Skills," "attention," "access"
on: May 31, 2006 [19:34] score: 0.00
> davidphilipona wrote:
> > > it seems to me that the sony aibo is able to exercise a few sensorimotor skills, like walking, but I doubt very much that it "has feel", and I would suggest this is because it has not cognitive system. As far as I'm concerned, I would therefore on the opposite stress the notion of cognitive access, whatever this can mean.
>
> kevin wrote:
> > i agree witht his.
>
> my reply:
> Why do cognitive systems have feels and not non-cognitive systems, just incidentally?

they dont: cognitive systems have feels when they focus their attention on the exercise of sm skills. A system like a missile guidance system that has sensorimotor skills has no feel because it has no cogntive system to make use of the sm skills in its rational behavior.

>
> kevinoregan wrote:
> > concerning whether "skill" already implies cognition:
> > first what is cognition? Your immune system could be considered to do
> > categorisation of foreign cells... a soap bubble might be said to have cognitive
> > capacities since it calculates its shape by making a minimum of surface area.
>
> my reply:
> This soap-bubble notion truly seems like a terrible idea. But in fact it seems to touch on a global problem with the skill/cognitive access/knowledge approach as you describe it. See Daniel Hutto's paper from the latest issue of Phenomenology and the Cognitive Sciences on the relationship between the laws of physics and sensorimotor skills and laws and "knowledge." (NB I don't endorse everything in that paper, but there's enough thoughtful material, including arguments already raised here, to warrant careful scrutiny.)

very interesting article. Thx for pointing it out. I looked briefly at the section you mention. Globally my conclusion is the same as Hutto, as far as I can gather quickly: I considered the problem of what is meant by a sm law in the sm law section of this website. I came to the conclusion that the notion of a purpose was necessary, and depended on the "point of view" one takes in analysing a system...
>
> kevinoregan wrote:
> > > Maybe david can find us a definition of cognitive capacities which is independent of the point of view in which you place yourself, the way in which you code the behaviors that you measure.
>
> davidphilipona wrote:
> > I just gave one suggestion. It's only a suggestion, maybe this one is not good, but the point is: I don't see why suggestions of that kind would have anything to do with the way you code behaviors.
>
>
> my reply: The "coding" (theoretical or internal to the system? I don't actually get why this issue is raised here, but anyway) of behaviours does seem to be relevant here. Let me try and describe how it's relevant to my problems with the skill analogy.
>
> Describing ordinary skills seems to involve describing the behaviours involved, in terms that relate to the goals achieved by the behaviours; in fact they are defined as skills exactly by being the ability to achieve particular behavioral goals in context-relevant ways. Cricket batting as Ed describes it seems to involve being sensitive to a range of factors in a situation - categorizing it, as Aline puts it, in a very fine-grained way - but then that categorization (as described) is actually defined by its contribution to shaping particular actions in the service of particular goals, and so the "coding" of the behaviours themselves seems to be fundamental to saying what the skill is.

I agree with that!

>Sensory perception seems to encompass the sensitivities that are fundamental to the enabling of all and any physical behaviour; if describing behaviour is involved in describing some particular sensory-perceptual skill, the relevant subset of actions will be defined not by any particular set of goal-constrained actions, but by the range of information (find me another word here, Ed) about the situation the perceiver is in that that by virtue of the "skill" can have a shaping influence on any behaviour whatsoever. (So only those with a colour-seeing "skill" can go if the light is green and stop if it's red - but they can also run red lights to show off to friends, etc.) This may seem okay - you could try and describe the skill in terms of the way that sensory-perceptual sensitivities in given organism-environment situations can constrain whole space of physically possible actions, seeing the set of actions enabled by particular sensitivities as the behavioural side of the skill. But one worry for sensorimotorism that seems pretty obviously to loom up here: when we go to ask how we understand the ways that these so-called skills can "code" such generalized constraints on the space of possible behaviours, it seems hard to imagine that people won't just want to say that people have access to a "representation" of the situation in terms of relevant sensory-perceptual information that goes into shaping their behaviour. The role of action then becomes one of just abstractly fixing the "content" of perception, and we're back something very close to the view that we wanted to get rid of, right? And the problems that come along with it. Saying we have the "skill" of being able, say, to take any action whatsoever in response to the presence or absence of a certain colour doesn't seem to be a very convincing way of undermining the argument that there's something surprising about our having an experience linked into whatever it is that gives us that ability. And if the skill is just detecting the colour (this does seem to be the idea), once again: what is the value-added of the word "skill"?
>
> Exactly the trouble I was trying to make clear in my talk in Antwerp in 2004.
>

pff, I cant seem to get my head around what you wrote above. I'll try again later.


> kevin wrote: > Maybe david can find us a definition of cognitive capacities which is independent of the point of view in which you place yourself, the way in which you code the behaviors that you measure. But while we wait for him to find the solution, if there is one, I prefer to say that yes, all these things may be forms of cognition (with skills having the additional characteristic that they involve sensorimotor interactions).
> >
> > But when we have a feel, it is not sufficient for sensorimotor processes to be going on which can be considered "cognitive" or skillful. There has to be a kind of meta- level of cognition which could loosely be called "paying attention to": The organism, considered as an *individual*, acting in an *environment*, must be poised to make use of the skill in its rational behavior. Otherwise the aibo has feel.
>
> my reply: "Rational"?!?!

yes, that's what I claim. An animal has feel to the extent that it can make use of that feel in its purposive, rational, choiceful behavior. If an animal is just an automaton, it has no feel.

(in a way this is just a matter of definition. I think I'm just trying to encapsulate what everybody agrees distinguishes feel from no feel in animals of different complexity).

>
> This description of attention seems far less radical than the Merleau-Pontyan version of attention that Ed endorses. On this view perceivers are "always already" oriented in the world, and their perceiving is shaped by that orientedness, which combines their spatial focus, their motivations, their memories, etc. "Attention" is a kind of a crude way of describing some limited part of what the whole deployment of the perceiver's sensory skills entails; it has the merit of acknowledging that the skills are actually deployed, with consequences that include spatial focus, for example, but it seems on this view to be a confusion to suggest that this is an extra "meta" factor; it's just what it is to have and use the perceptual skills we actually have.
>
> Then again I may be getting this all very wrong, Ed can tell you what he thinks better than I can.
>
> davidphilipona wrote:
> > > I must say that this notion of "focusing attention" is to me the most fuzzy concept of the whole theory (ce qui n'est pas une mince performance)
> >
> kevinoregan wrote:
> > well i mean byfocussing attention the sameas what i mean by having cog access, I think.
>
> my reply:
> Again the confusion: is attention/access focused on the particular qualities picked up through "skilful" perceiving, or on the "skill" itself?!

neither... It's difficult to explain. The attention is focussed on the category (e.g. "red") that has been assigned to the skill. But the interior workings of the skill are not accessible.

>
> aline wrote:
> > > > and so on... are
> > > >already included in access-consciousness so why not using
> > > > it?
> > >
> > >
> davidp wrote:
> > > If I don't mistake, Kevin prefers to speak of cognitive access rather than A-consciousness because the later seems to him to be a stronger concept, involving the notion of self. (right, Kevin?)
> >
> kevin wrote:
> > exactly. Cognitive access + notion of self having this access = conscious access.
>
> my reply:
> Whose idea is it that "access consciousness" is defined without regard to an "accessor"? Can you cite any reference that shows this concept to be different from Block's?

I distinguish *cognitive* access from *conscious* access. The former need have no accessor!

Block uses the latter term. I think adding this distinction reveals the hidden notion of "self" in the notion of conscious access.


> kevin wrote:
> > everybody seems to dislike it when I say "having cog access to *the fact that* one is exercising a sm skill".
> >
> > The reason I say "the fact that" is that when you have a feel of red, you have no
> > access to any of the details of the skills involved: you dont know what the rules
> > are that the piece of paper is obeying as you move it around under the lights. If I
> > had not said *the fact that*, then it sounds like you have access to the skill as a
> > whole, all its details and interior workings. Maybe I should find another way of
> > expressing the idea however, which is that while you dont have access to the
> > interior workings, you CAN nevertheless distinguish the different laws, and you
> > can even compare and contrast them: thus you DO have access to the fact that
> > the current "red" rules are DIFFERENT from the rules of green, and that they are
> > somewhat more similar to the rules of pink. Doing many athletic or motor tasks
> > seems to me quite a good analogy: an athlete has no idea exactly what muscles to
> > put into action in order to accomplish his feat, yet he knows how to do his feat in
> > different ways. Dance teachers use all sorts of bizarre analogies to coach their
> > students to move their bodies in certain ways (e.g. todo a "battement de jambe",
> > you have to move out your foot as though it is a brush, brushing the floor on its
> > way out). It's no use explaining to a dancer what muscles to use.
>
> my reply:
> See above, also Hutto paper, also earlier stages of the email discussion that preceded this thread. You're really not answering the challenging questions about the physical skill analogy at all.
>
> You seem to be saying here that a) colours are distinguished by sensorimotor laws and b) we are able to distinguish colours. We very obviously don't know that we are ongoingly distinguishing them on this basis - and indeed the scientific evidence for this proposition actually seems for the moment to be weak at best. So we in fact don't at all have "access to the fact that the current "red" rules are DIFFERENT from the rules of green," we can just distinguish red and green!

right!

>Again (again again again): what is the "access" here supposed to be access to?

to the categories "red" and "green" that we associate with those skills, and which we can make use of in our car driving and other behavior, and which may be associated to conceptual categories as well.


>
> Let me remake the same, clearly still 100% unanswered, point: motor skills clearly amount to the ability to physically do something; though we obviously don't know just what muscles we contract when we carry the actions out, they are identifiable as physical actions! What is the "skill" of seeing the ability to do?

It's the ability to recognize that when I blink, move my eyes etc., sm relations are obeying the seeing-type laws. Here the situation is similar to the physical skill in the sense that we dont know the muscles we move, but further complicated by the fact that we dont know either the sensory input changes that are caused. nevertheless we can recognize that the laws are not, say, the hearing type laws, but the seeing-type laws.

.... more later.

author message
PaulReeve
Re: Re: "Skills," "attention," "access"
on: Jun 02, 2006 [12:30] score: 0.00
> > Why do cognitive systems have feels and not non-cognitive systems, just incidentally?
>
> they dont: cognitive systems have feels when they focus their attention on the exercise of sm skills. A system like a missile guidance system that has sensorimotor skills has no feel because it has no cogntive system to make use of the sm skills in its rational behavior.

??? What does "they" refer to here? Sorry, but I don't really understand what your answer has to do with my question. Which is:
"Why do cognitive systems have feels and not non-cognitive systems, just incidentally?"

> very interesting article. Thx for pointing it out. I looked briefly at the section you mention. Globally my conclusion is the same as Hutto, as far as I can gather quickly: I considered the problem of what is meant by a sm law in the sm law section of this website. I came to the conclusion that the notion of a purpose was necessary, and depended on the "point of view" one takes in analysing a system...

That's it?! You do realize that the entire article is devoted to explaining that in Hutto's view you're getting things completely wrong, right?

> > my reply: The "coding" (theoretical or internal to the system? I don't actually get why this issue is raised here, but anyway) of behaviours does seem to be relevant here. Let me try and describe how it's relevant to my problems with the skill analogy.
> >
> > Describing ordinary skills seems to involve describing the behaviours involved, in terms that relate to the goals achieved by the behaviours; in fact they are defined as skills exactly by being the ability to achieve particular behavioral goals in context-relevant ways. Cricket batting as Ed describes it seems to involve being sensitive to a range of factors in a situation - categorizing it, as Aline puts it, in a very fine-grained way - but then that categorization (as described) is actually defined by its contribution to shaping particular actions in the service of particular goals, and so the "coding" of the behaviours themselves seems to be fundamental to saying what the skill is.
>
> I agree with that!

That's fine, but that passage is a description of what a skill is in terms that do NOT apply to perceiving as you describe it! It's part of an argument that perceiving described in sensorimotor terms FAILS to qualify as a skill according to a common-sense understanding of what a skill is. You don't make clear an alternative notion of the meaning of "skill," you just help yourself to consequences of taking perceiving to be a skill in this common-sense way despite the fact that as far as I can tell it just isn't one!

> >Sensory perception seems to encompass the sensitivities that are fundamental to the enabling of all and any physical behaviour; if describing behaviour is involved in describing some particular sensory-perceptual skill, the relevant subset of actions will be defined not by any particular set of goal-constrained actions, but by the range of information (find me another word here, Ed) about the situation the perceiver is in that that by virtue of the "skill" can have a shaping influence on any behaviour whatsoever. (So only those with a colour-seeing "skill" can go if the light is green and stop if it's red - but they can also run red lights to show off to friends, etc.) This may seem okay - you could try and describe the skill in terms of the way that sensory-perceptual sensitivities in given organism-environment situations can constrain whole space of physically possible actions, seeing the set of actions enabled by particular sensitivities as the behavioural side of the skill. But one worry for sensorimotorism that seems pretty obviously to loom up here: when we go to ask how we understand the ways that these so-called skills can "code" such generalized constraints on the space of possible behaviours, it seems hard to imagine that people won't just want to say that people have access to a "representation" of the situation in terms of relevant sensory-perceptual information that goes into shaping their behaviour. The role of action then becomes one of just abstractly fixing the "content" of perception, and we're back something very close to the view that we wanted to get rid of, right? And the problems that come along with it. Saying we have the "skill" of being able, say, to take any action whatsoever in response to the presence or absence of a certain colour doesn't seem to be a very convincing way of undermining the argument that there's something surprising about our having an experience linked into whatever it is that gives us that ability. And if the skill is just detecting the colour (this does seem to be the idea), once again: what is the value-added of the word "skill"?
> >
> > Exactly the trouble I was trying to make clear in my talk in Antwerp in 2004.
> >
>
> pff, I cant seem to get my head around what you wrote above. I'll try again later.

Ok, I'll be waiting for that.

> > kevin wrote: > Maybe david can find us a definition of cognitive capacities which is independent of the point of view in which you place yourself, the way in which you code the behaviors that you measure. But while we wait for him to find the solution, if there is one, I prefer to say that yes, all these things may be forms of cognition (with skills having the additional characteristic that they involve sensorimotor interactions).
> > >
> > > But when we have a feel, it is not sufficient for sensorimotor processes to be going on which can be considered "cognitive" or skillful. There has to be a kind of meta- level of cognition which could loosely be called "paying attention to": The organism, considered as an *individual*, acting in an *environment*, must be poised to make use of the skill in its rational behavior. Otherwise the aibo has feel.
> >
> > my reply: "Rational"?!?!
>
> yes, that's what I claim. An animal has feel to the extent that it can make use of that feel in its purposive, rational, choiceful behavior. If an animal is just an automaton, it has no feel.
>
> (in a way this is just a matter of definition. I think I'm just trying to encapsulate what everybody agrees distinguishes feel from no feel in animals of different complexity).

Ok, I'll grant you this, that by "rationality" you actually seem to mean "purposiveness." Interesting though, that, everybody just agreeing on something in that way. How did everybody find out about these conditions on consciousness?

> > davidphilipona wrote:
> > > > I must say that this notion of "focusing attention" is to me the most fuzzy concept of the whole theory (ce qui n'est pas une mince performance)
> > >
> > kevinoregan wrote:
> > > well i mean byfocussing attention the sameas what i mean by having cog access, I think.
> >
> > my reply:
> > Again the confusion: is attention/access focused on the particular qualities picked up through "skilful" perceiving, or on the "skill" itself?!
>
> neither... It's difficult to explain. The attention is focussed on the category (e.g. "red") that has been assigned to the skill. But the interior workings of the skill are not accessible.

Now the attention is focused on the CATEGORY? Before I make any further comment on that: are you sure that's what you mean?

> > my reply:
> > Whose idea is it that "access consciousness" is defined without regard to an "accessor"? Can you cite any reference that shows this concept to be different from Block's?
>
> I distinguish *cognitive* access from *conscious* access. The former need have no accessor!

"Access" as a verb seems to be inherently transitive. Presumably its role in a functionalist view is that it's what happens when information is passed from some functional subpart of the system to some other part. No?

> Block uses the latter term. I think adding this distinction reveals the hidden notion of "self" in the notion of conscious access.

It would be great if the hidden notion of "cognitive access" was revealed too!

> > kevin wrote:
> > > everybody seems to dislike it when I say "having cog access to *the fact that* one is exercising a sm skill".
> > >
> > > The reason I say "the fact that" is that when you have a feel of red, you have no
> > > access to any of the details of the skills involved: you dont know what the rules
> > > are that the piece of paper is obeying as you move it around under the lights. If I
> > > had not said *the fact that*, then it sounds like you have access to the skill as a
> > > whole, all its details and interior workings. Maybe I should find another way of
> > > expressing the idea however, which is that while you dont have access to the
> > > interior workings, you CAN nevertheless distinguish the different laws, and you
> > > can even compare and contrast them: thus you DO have access to the fact that
> > > the current "red" rules are DIFFERENT from the rules of green, and that they are
> > > somewhat more similar to the rules of pink. Doing many athletic or motor tasks
> > > seems to me quite a good analogy: an athlete has no idea exactly what muscles to
> > > put into action in order to accomplish his feat, yet he knows how to do his feat in
> > > different ways. Dance teachers use all sorts of bizarre analogies to coach their
> > > students to move their bodies in certain ways (e.g. todo a "battement de jambe",
> > > you have to move out your foot as though it is a brush, brushing the floor on its
> > > way out). It's no use explaining to a dancer what muscles to use.
> >
> > my reply:
> > See above, also Hutto paper, also earlier stages of the email discussion that preceded this thread. You're really not answering the challenging questions about the physical skill analogy at all.
> >
> > You seem to be saying here that a) colours are distinguished by sensorimotor laws and b) we are able to distinguish colours. We very obviously don't know that we are ongoingly distinguishing them on this basis - and indeed the scientific evidence for this proposition actually seems for the moment to be weak at best. So we in fact don't at all have "access to the fact that the current "red" rules are DIFFERENT from the rules of green," we can just distinguish red and green!
>
> right!

"You're wrong! What you're saying is incoherent!"

"That's completely true! Next point."

Let me try and pick out the most revealing parts of that little sub-exchange.

kevin wrote:
"you DO have access to the fact that the current "red" rules are DIFFERENT from the rules of green, and that they are somewhat more similar to the rules of pink"

i wrote:
"we in fact don't at all have "access to the fact that the current "red" rules are DIFFERENT from the rules of green," we can just distinguish red and green! "

kevin wrote:
"right!"

????????


i wrote:

> >Again (again again again): what is the "access" here supposed to be access to?
>
> to the categories "red" and "green" that we associate with those skills, and which we can make use of in our car driving and other behavior, and which may be associated to conceptual categories as well.

I have no problem with the notion that we have "access" to colour differences through consciousness. I have a problem with any and all claims being made on the basis of an analogy between the ability to perceive in terms of these distinctions and physical skills. If the "skill" of perceiving consists in being able to make the distinctions that we actually can make in perception, then clearly no one could ever disagree that we have such a skill, and also no one could have any reason to change their mind about anything at all, because the claim would be vacuous.

> > Let me remake the same, clearly still 100% unanswered, point: motor skills clearly amount to the ability to physically do something; though we obviously don't know just what muscles we contract when we carry the actions out, they are identifiable as physical actions! What is the "skill" of seeing the ability to do?
>
> It's the ability to recognize that when I blink, move my eyes etc., sm relations are obeying the seeing-type laws. Here the situation is similar to the physical skill in the sense that we dont know the muscles we move, but further complicated by the fact that we dont know either the sensory input changes that are caused. nevertheless we can recognize that the laws are not, say, the hearing type laws, but the seeing-type laws.

The skill of seeing isn't the ability to see anything in particular, it's just the ability to recognize that seeing-type laws are applying in general in a situation? That doesn't seem right.

As far as I can tell this response has no bearing on the challenge I'm raising here. When you say that skill amounts to "recognizing" something, the relation between the idea and that of a physical skill becomes instantly problematic, for reasons discussed above that I'm not ready to try and formulate yet again for the moment. Again, I'll just wait on your response to all that.


Page: 1/1

Show posts:
 
Free Hit Counter
eBay