Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.

Dismiss

Re: Perception, Behaviour and Quantum Information

6 views

Skip to first unread message

Stuart Hameroff

unread,

May 10, 2005, 12:05:52 PM5/10/05

Hi everyone

Quoting Alex Green <drale...@YAHOO.CO.UK>:
> The big surprise [in Libet's results] is that both stimulus
> and response can be completed
> reflexly (non-consciously) before the 0.5 secs required to model them
> and hence have them in conscious experience. It is this that seems to
> have made Penrose shake his pen in disbelief.

Stuart
(I suspect Roger shook his head first.) In many processes we *seem* to
be acting consciously. So the first problem is that this implies conscious
control of many actions is an illusion. Consciousness is merely along for the
ride (though it may serve other functions).

The second question is whether there is sufficent time
for even unconscious processes. John McCone (no friend of the
quantum idea) says (1999): [for] tennis players facing a fast serve, even
if awareness were actually instant, it would still not be fast enough.....

Others have questioned whether hitting a fast pitched baseball or fast-bowled
cricket ball are physiologically possible (by current explanations).

> Alex
> What I do not understand is what aspect of Libet's results show
> actual intervention backwards in time. In a hierarchical processor
> like the brain it seems natural that a set of cortical processors
> might receive a stimulus and another set react to it and then a third
> set produce this as a bound conscious experience.

Stuart
Lets take the medial lemniscus studies. When he stimulated the medial lemniscus
(just below thalamus) in awake subjects with a train of pulses that persisted
for 500 msec AND caused cortical activity for those 500 msec AND caused an
evoked potential at around 20 msec, the subject reported conscious experience
at 20 msec (the precise time was determined indirectly, cleverly, to avoid
delay due to verbal report).

OK, fine. You could say that the 500 msec cortical activity was sequlae of a
stimulus which was above threshold, having caused conscious activity at 20
msec.

But, if the stimulation caused the evoked potential but then was stopped
prematurely, before the 500 msec, no conscious experience occurred.

The brain *seemed to know in advance* that the neuronal adequacy, as Libet
called it, would or would not be completed.

> Alex
> In novel situations the algorithmic brain produces the response. Put
> a person who has never driven before in a car. Let a person walk in front
> of them once they have got into third. Almost every learner will run down
> the pedestrian. Thank God for dual controls. When the learner driver and
> instructor have wiped away the sweat they will spend a few lessons on
> the skill of stopping a car.

Stuart
What about rapid conversation? Its almost continually novel. You are suggesting
that the brain responds unconsciously before the words to which it is
responding are comprehended. Even when contextual effects are considered this
explanation doesnt work. Some people go shooting their mouth off, but not
everyone.

Alex
> In the phi illusion the delay is sufficient to allow the flashing
> bars to be modelled
> as a motion.The phi illusion is evidence for a modelling delay. See:
> http://www.users.globalnet.co.uk/~lka/handle.exe for an example of the phi
> illusion showing that modelling occurs. Temporal binding also requires a
> modelling delay.

Stuart
But the problem is that when you try and fool the subject, the brain seems to
know in advance what is (or is not) coming. In color phi, after a series of red
dots on the left alternating with green on the right (in which the subject fills
in and consciously experiences the red turning to green halfway across)
the tricky experimenter shows red on the left followed by red on the right, the
subject does not see the red changing to green halfway across.

Similarly in the cutaneous rabbit anomaly (5 taps oon the wrist, followed by 5
taps on the forearm, followed by 5 taps at the elbow - the blindfolded subject
feels 15 taps hopping incrementally up the arm) if you try and fool the subject
and not tap at the forearm or elbow, the subject feels all 5 initial taps at
the wrist.

Alex
> I am not convinced that consciousness is necessary even for the veto. See:
> http://en.wikipedia.org/wiki/Halting_problem
> However, I think it probably does get used for vetoing because, like you, I
> think it selects/collapses a reality.

Stuart
Agreed, but lets distinguish between the Copenhagen interpretation in which
consicousness causes collapse, and our (Penrose and mine, and maybe yours,
though I'm not completely sure) position in which consciousness IS a (type of)
self-collapse. The former puts consciousness outside of science, the latter
puts it on the edge between the quantum and classical worlds.

Alex
> The real need for consciousness seems to be the need for something to
> implement Ellis McTaggart's 'A Series'. It is required for the act of
> 'becoming', for there to be a past and present. Such a requirement would only
> occur if the universe were truly QM such as modern multiple universe/
> many minds theories propose. Why else would a temporal series exist? See:
> http://www.ditext.com/mctaggart/time.html

Stuart
I agree. Outside consciousness, there is no flow of time. Consciousness creates
the flow of time. I have a paper on this from a conference/book on the nature
of time. See http://www.quantum-mind.org/Time.htm

But we dont necessarily need multiple universes or minds (per person). Penrose
OR avoids the need. The multiple possibilities are self-limited, and reach
threshold for (conscious) self-collapse before branching off to form a new
universe. Things are much tidier that way.

But if you dont see a flow of time outside consciousness, why are you bothered
by a timeless unconscious which manifests as backward flow in the consciously
observed classical world? It rescues us from being mere helpless spectators.

cheers

Stuart
www.consciousness.arizona.edu/hameroff

>
> Best Wishes
>
> Alex Green
>
>
>
> Stuart Hameroff <hame...@U.ARIZONA.EDU> wrote:Hi everyone
>
> Alex Green
>> Huxley's epiphenomenalism [we are 'merely helpless spectators']
>> is often rejected simply because of this bleakness.
>
> Stuart
> Unless one takes Libet's backward referral literally, data and prevalent
> theory point to conscious perception occurring after many actions deemed
> (seeming as though they are) conscious. Huxley's description of 'helpless
> spectators' may be overly bleak and dramatic, but according to conventional
> views consciousness remains epiphenomenal. We are merely happy, helpless
> spectators.
>
>> Alex
>> As far as I can see Libet showed that about 0.5 secs after
>> a stimulus a conscious experience containing the stimulus
>> occurs:
>>
>> "In other words, it is the content of the subjective experience, of the
>> neuronally delayed awareness, that is modified by the referral to
>> the earlier
>> timing signal." Libet, B. (2000). Time factors in conscious processes:
>> Reply to Gilberto Gomes. Consciousness and Cognition, 9, 1#8211;12.
>
> Stuart
>
> Libet is basically following Dennett's retrospective reconstruction (or
> Orwellian revisionism). We just remember things happening in real
> time, rather
> than a half second later. Dennett has mental time following its own
> meandering
> course, looping backwards in real time. But then one must define mental time.
> The problem is just compounded.
>
> In his 2000 book Libet say:
> 'the timing of a sensation is subjectively referred...not that the conscious
> sensation itself jumped backwards in time...the content of the subjective
> experience...is modified by the referral to the earlier timing signal.'
>
> But with all due respect to Ben Libet, that doesnt explain his results
> (unless the direct experience associated with neuronal adequacy is
> delayed yet
> another 500 msec in his comparisons of direct cortical stimulation, hand
> stimulation and medial lemniscus stimulation).
>
> If one suspends disbelief in actual backward time referral,
> the data show that the conscious experience occurs at the time of the evoked
> potential ('the timing signal'), 20 to 30 msec after stimulus and several
> hundred msec before the neuronal brain activity necessary (neuronal adequacy,
> as Libet put it) for conscious awareness.
>
> Certainly evolution would favor real time conscious actions rather
> than delayed
> nonconscious actions.
>
> Alex
>> This interpretation is consistent with what we know about
>> the speed of the processing required to construct the content
>> of conscious experience. For example, the various perceptual
>> rivalries operate on this time base, the auditory continuity and
>> similar illusions have the same timings. If conscious
>> experience is the output of cortical modelling then c.0.5 secs
>> would be expected because sensory and motor data from
>> all modalities must be integrated.
>
> Stuart
> Agreed. But the experience and apparently conscious actions occur far
> earlier.
>
>> Alex
>> As Penrose says:
>>
>> "But I have to confess to a disbelief in the possibility that there
>> can be no role for consciousness in such rapid activities
>> like ordinary conversation.." (Shadows of Mind p387).
>>
>> I have two principal objections to this.
>>
>> Firstly, I can find no evidence for processes in conscious
>> experience. Things just pop into mind. If I try to construct
>> things consciously any building block just pops into mind,
>> any connection between blocks just pops up. Conscious
>> experience does not create it's own content.
>
> Stuart
> Agreed. Unconscious/preconscious processes create the content. But
> they create
> numerous possible contents. In the quantum mind view, preconscious processes
> are quantum information - superpositions of multiple possibilities - one of
> which pops into mind at the moment of quantum state reduction. Quantum
> mechanics allow such quantum information to travel backward in
> classical time.
>
> Alex
>> Secondly, there is an alternative to conscious control of
>> behaviour: conscious training of behaviour. Instead of
>> processing individual behaviours conscious experience
>> provides intuitions of whether certain behaviours match the
>> world and/or the worldline of the individual. A matching
>> signal reinforces these behaviours. This is 'free will' after
>> the event.
>
>
> Stuart
> The training argument is cogent, but lets first clarify free will.
> There are two issues being conflated: one is whether we have conscious
> control in real time, or after the fact (as we discussed in the context of
> Libet).
>
> But regardless of *when* and regardless of whether our actions are chosen
> consciously or unconsciously, the other issue is whether our actions are
> completely deterministic/algorithmic. Only Penrose non-computability gets us
> out of algorithmic responses.
>
> As for the training argument, yes, up to a point. But how do we
> respond in novel
> situations? And what about the color phi effect, and Libet's results, and
> temporal binding? One must postulate unconscious actions masquerading as
> consciousness due to training, AND retrospective reconstruction.
>
> I know you say that consciousness helps the training, like Libet's veto, or
> Jeffrey Gray's idea of consciousness as a planner. But none of those
> functions
> need to be conscious. So (teleologically) why bother with
> consciousness at all?
>
> My own view is not that consciousness evolved from biology, but that biology
> adapted to ubiquitous protoconsciousness.
>
> Alex
>> So there is no need for direct connection to the past,
>> Penrose's disbelief can be explained by pointing out that
>> humans endlessly train themselves by rehearsing
>> conversation with inner speech.
>
> Stuart
> If by inner speech you mean unconscious processing, then it may be
> coming from
> the near future. (Backward referral means from the future to the present).
>
>> Alex
>> I hope I demonstrated above that conscious experience need not be
>> epiphenomenal even if it could not intervene in the past.
>
> Stuart
> The question is whether consciousness can intervene in the present.
> But you havent demonstrated that is not epiphenomenal, at least not to me.
>
> Alex
>> There is some proof that conscious experience is classical. Prior to
>> decoherence theory classical structures were known to be generated by the
>> possibility of a classical observation by a conscious observer.
>
>
> Stuart
> I basically agree that conscious experience is classical.
> Consciousness is the process of quantum information
> reducing/collapsing to classical states.
>
> But the conscious observer (the Copenhagen interpretation) is
> misleading, for it
> places consciousness outide science. In my view (and that of Roger Penrose)
> consciousness does not cause collapse, consciousness IS collapse (a
> particular
> type of collapse). And decoherence theory doesnt solve the problem
> for several
> reasons. It doesnt address the fate of isolated systems, and it doesnt really
> collapse the wave function, merely bury it in noise.
>
>
> Alex
>> My own take on this is that the mind offers a preferred basis for the
>> brain and this basis is expressed as the geometry of conscious experience.
>> The qualia, as you and Brian Flanagan suggest, are certain QM fields but are
>> particular values of those fields superselected by the geometry.
>
> Stuart
> I think I agree :)
>
> cheers, and thanks for your comments
>
> Stuart
> www.consciousness.arizona.edu/hameroff
>
>
>
> ---------------------------------
> Yahoo! Messenger - want a free & easy way to contact your friends online?

Alex Green

unread,

May 10, 2005, 4:15:43 PM5/10/05

Quoting Alex Green :
> The big surprise [in Libet's results] is that both stimulus
> and response can be completed
> reflexly (non-consciously) before the 0.5 secs required to model them
> and hence have them in conscious experience. It is this that seems to
> have made Penrose shake his pen in disbelief.

Others have questioned whether hitting a fast pitched baseball or fast-bowled
cricket ball are physiologically possible (by current explanations).

Alex
Having been a keen squash player I would put speedy responses
down to correct positioning in response to the opponent's
actions before they make the serve. At London University I
started playing regularly with a one armed opponent who was
strong and skillful. It took perhaps five matches before I got the
hang of responding to serves where the ball was skilfully thrown up in
the air with the racquet hand!

OK, fine. You could say that the 500 msec cortical activity was sequlae of a
stimulus which was above threshold, having caused conscious activity at 20
msec.

But, if the stimulation caused the evoked potential but then was stopped
prematurely, before the 500 msec, no conscious experience occurred.

The brain *seemed to know in advance* that the neuronal adequacy, as Libet
called it, would or would not be completed.

Alex
Libet's results are consistent with subliminal perception and backward
masking experiments. It is well known that a very short low intensity stimulus
will often result in reactions where the subject is unaware of the causative
stimulus. Similarly backward masking within the modelling delay can
suppress the stimulus.

Alex
I would propose that the words in rapid conversation are non-consciously
comprehended. Conscious experience operates a crude level of control,
intercepting problems. In normal speech we have no awareness of the
construction of the phonemes of words, words appear ready made, we do
not construct the grammar and syntax, it just happens. The non-conscious
speech generators are quite capable of taking care of 0.5 secs on their
own and they allow the speaker to formulate their thoughts at a non-verbal
level in response to the effect of the speech whilst speaking. This is
the advantage of a hierarchical architecture - you can walk, eat and
think all at once!

Alex
I have done the phi experiment with a reaction time test and the reaction to
the appearance of the blue bar always occurs about 150 msecs after the
the bar physically appears on the screen whether the bar is flashed on its
own or as part of a phi series of alternating flashes with a red bar. This
suggests that the apparent movement of the bar is a positioning error
in the construction of the mental model.

In the link given above:
http://www.users.globalnet.co.uk/~lka/handle.exe
the horizontal bars alternate with each other. However, the horizontal bars
grow and shrink as the main vertical bars move. This shows that the
vertical bars only move in our mental model after they have been placed on
the screen. The mental model has simply positioned them incorrectly and
then adjusts them back to their proper places, extending the horizontal
bars as it does so (focus on the point where the horizontal bars meet to
get the full pump handle effect). If you select 'test' and 'phi' then click the
'slower' option repeatedly until the delay between alternations is 0.6 secs
the phi illusion disappears, confirming a 0.5 secs modelling delay.

So, the reason the subject does not see red changing to green halfway
across when red is substituted for green is that the red is already
registered non-consciously. The modelling of the position of the bars
does not even begin until the bars are physically on the screen, as
the 'tricky' experimenter and the discussion above has shown - how
else could the subject fail to model green?

Stuart
Similarly in the cutaneous rabbit anomaly (5 taps on the wrist, followed by 5

taps on the forearm, followed by 5 taps at the elbow - the blindfolded subject
feels 15 taps hopping incrementally up the arm) if you try and fool the subject
and not tap at the forearm or elbow, the subject feels all 5 initial taps at
the wrist.

Alex
Again, the taps are already in the sensory pathways so the subject models
a tap on the elbow with an incorrect position towards the wrist. I would
predict that if the delay between taps is much more than about 0.6 secs the
illusion will disappear because a mental model with no taps can intervene.

Alex
Yes, I would agree that consciousness is associated with state vector
reduction.

Alex
The interval is only about 0.5 secs so we can adjust 'consciously' on quite
rapid time scales. However, accurate and skilful behaviour at intervals of less
than 0.5 secs will need training. That said, for a guitarist a suggestion derived
from non-conscious processing that is experienced as the precursor of an emotion
could lead to a non-conscious directive that in three bars a note will be played on
the guitar pianissimo. Three bars later the pianissimo will occur and 0.5 secs
later the guitarist may feel fulfilled. A hierachical processor can mix and match
its thoughts and actions.

So where does this leave us? In my model the conscious brain does not
actually intervene, it just selects. In your model it intervenes. In my model
there will be very little effect of conscious experience unless either a
many-minds model of the universe applies or conscious experience
is composed of something like an em field that has considerable QM
uncertainty even in a warm fluid bath.

Best Wishes

Alex Green

------------------------------------------------------------------------------------------
>
>
>
> Stuart Hameroff wrote:Hi everyone

cheers

Stuart
www.consciousness.arizona.edu/hameroff

>
> Best Wishes
>
> Alex Green
>
>
>
> Stuart Hameroff wrote:Hi everyone

---------------------------------
How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos. Get Yahoo! Photos

Stuart Hameroff

unread,

May 12, 2005, 12:17:35 PM5/12/05

Hello again

Alex Green

> Having been a keen squash player I would put speedy responses
> down to correct positioning in response to the opponent's
> actions before they make the serve. At London University I
> started playing regularly with a one armed opponent who was
> strong and skillful. It took perhaps five matches before I got the
> hang of responding to serves where the ball was skilfully thrown up in
> the air with the racquet hand!

Stuart
Of course. I played a lot of baseball, and learned, for
example, to anticipate whether the pitcher was throwing a fastball or a curve
ball by the position of his thumb as the ball was released. And in tennis I
guess where the serve is going (my forehand or backhand) by the toss of the ball
before the racquet is even brought in position to hit it, and the angle of the
racquet as it approaches the ball.

But none of these account for *consciously*
seeing the ball late enough in its trajectory to hit it.
Jeffrey Gray (in his 2004 book) said:
In top level tennis the speed of the ball after a serve
is so great, and the distance to travel so short, that the player
must strike it back before consciously seeing the ball leave the servers racket.
Conscious awareness comes too late.

This isnt as strong as McCones statement that there isnt enough time for even
unconscious processing. But the problem is that players apparently consciously
see the ball as they prepare to hit it. And there isnt enough time for that,
apparently.

Alex
> Libet's results are consistent with subliminal perception and backward
> masking experiments. It is well known that a very short low intensity
> stimulus
> will often result in reactions where the subject is unaware of the causative
> stimulus. Similarly backward masking within the modelling delay can
> suppress the stimulus.

Stuart
You are omitting a rather significant difference. Libet's subjects were
consciously aware of the stimulus. Take away consciousness, and we have no
problem.

Alex
> I would propose that the words in rapid conversation are non-consciously
> comprehended. Conscious experience operates a crude level of control,
> intercepting problems. In normal speech we have no awareness of the
> construction of the phonemes of words, words appear ready made, we do
> not construct the grammar and syntax, it just happens. The non-conscious
> speech generators are quite capable of taking care of 0.5 secs on their
> own and they allow the speaker to formulate their thoughts at a non-verbal
> level in response to the effect of the speech whilst speaking. This is
> the advantage of a hierarchical architecture - you can walk, eat and
> think all at once!

Stuart
I am not arguing against hierarchical architectures. Are you saying that
conversation is non-conscious? Obviously much of the processing in any modality
is nonconscious, and some choice of verbal response may be reflexive. But do you
really think all rapid conversation is completely unconscious?

In speech, 0.5 secs is too late. Evoked potentials indicating conscious word
recognition occur at about 400 msec after auditory input, however semantic
meaning is appreciated (and response initiated) after only 200 msec. As Velmans
points out, only two phonemes are heard by 200 msec, and an average of 87 words
share their first two phonemes. Even when contextual effects are considered,
semantic processing and initiation of response occur before conscious
recognition (Van Petten et al 1999).

Alex
>> In the phi illusion the delay is sufficient to allow the flashing
>> bars to be modelled

>> as a motion. The phi illusion is evidence for a modelling delay. See:

>> http://www.users.globalnet.co.uk/~lka/handle.exe for an example of the phi
>> illusion showing that modelling occurs. Temporal binding also requires a
>> modelling delay.

Stuart
Modelling delays are Dennetts retrospective reconstruction, or Orwellian
revisionism. Thus what we call conscious experience (according to Dennett and,
apparently, you) is an illusory, after-the-fact reconstruction. And our rapid
responses are unconscious. We are back to epiphenomenal helpless spectators.
OK, maybe we are. But you still have Libet's data unexplained.

Alex
> Yes, I would agree that consciousness is associated with state vector
> reduction.

Stuart
Good. But you still havent declared yourself: 'associated with' could pertain to
both Copenhagen (consciousness causes collapse) and Penrose, Stapp, myself and
others (consciousness IS collapse).

Alex

> So where does this leave us? In my model the conscious brain does not
> actually intervene, it just selects. In your model it intervenes.

Stuart
The first problem is what exactly the conscious brain is. I assume you are
saying that the conscious brain is the observer, collapsing the wave function
and thus selecting a particular conscious experience/reality.
To my taste that is a bit circular.

I would say that consciousness is the selection process itself.

Alex

> In my model
> there will be very little effect of conscious experience unless either a

> many-minds model of the universe applies....

Stuart
So you are saying that every superposition branches off either a) to form a new
universe containing one version of the observers mind, or b) just the mind (but
then what happened to the universe?). So the effect of consciousness then would
be....what?

Alex
> ...or conscious experience

> is composed of something like an em field that has considerable QM
> uncertainty even in a warm fluid bath.

Stuart
It sounds like quantum field theory in which the vacuum
(zero point energy) fluctuates, and you are putting consciousness as an
intrinsic part of it. But arguments can be made that a) the vacuum is a higher
level manifestation of underlying quantum gravity, and b)
uncertainty/randomness in QM occurs strictly in the measurement process or
decoherence. The underlying quantum world may be highly ordered and laden with
information.

I dont mean to be picky. I think your views are closer to mine than most others,
and therefore (in my humble opinion) closer to being correct.

0 new messages