I challenge you: your best alternate to materialism

[MRC_Hans]

You don't seem to realise that in a scientific experiment, especially one examining consciousness all language and psychological bias must be excluded and the problem be reduced to its physical parameters.

In principle, yes. However, our definitions of 'conscious', 'self-aware', or even 'alive' are far too fuzzy and subjective to cater for a strict scientific protocol.

So if it is necessary to have a personal experience of being to produce self-conscious behaviour(absent meat), it is also necessary for a lyrebird to have a chainsaw in its throat.

Nonsense. You know better than this. A lyrebird, or a tape-recorder for that matter, will only reproduce sound. Obviously, there is a bit more to a chainsaw than the sound.

Hans

 

[mijopaalmc]

Which is flatly wrong. Aside from the error pointed out in mijopaalmc's post above, you can predict any computation's result if you have sufficient knowledge of the system. And that is, in fact, Koch's point; he is a neuroscientist and information scientist, and determining whether or not consciousness exists by examining the system is exactly what he does.

"Wet" is also a description off our experience with water, not an objective measurement of the physical properties of water.

 

[MRC_Hans]

Except a mimic is not what it appears to be, or sounds to be, in the case of the lyrebird.

Obviously.

Yes I agree, however an amoeba, or an ant has an awareness of some sort of its environment and behaves in response to this. The ant is clearly not self-conscious, but does appear to have some kind of consciousness, I.e. it is a being, with a personal experience, or presence as an entity.

So, now, what is it you are arguing? I'm pretty sure modern autonomous robots have behavour patterns a good deal more complicated than ants (and certainly than amoeba). If you want to argue such simple brains to have consciousness, you have clearly defeated the claim that life is needed for it.

Yes I agree that the group of life forms may have fuzzy edges. It is a distinct phenomena, resulting in distinct causal events. Especially in the case of the more complex lifeforms.

However, all this is really just a narrow slice of the problem of consciousness. How about a non-living entity, e.g. a super-large computer? How do we recognize consciousness in that?

Hans

 

[dlorde]

As a side note, in this interview, Christof Koch suggests he does not think that one can establish whether an entity is consciousness or not by a set of behaviors . . . he would need to look at the hardware to determine how the behavior was generated, and his skepticism grows as the entity is different than human physiology.

http://www.technologyreview.com/news/531146/what-it-will-take-for-computers-to-be-conscious/

Yes; and he does seem to have a major problem with computability and Church-Turing (although that may be an artefact of the interview)... He says that you could get consciousness with a particular kind of hardware (a neuromorphic model) that is closer to the neural behaviour and connectivity of a brain than transistor-based hardware. However, if it is computation that is key - and a neuromorphic computer is a computing device - then you can make a computationally equivalent neuromorphic computing device out of conventional transistors and software (which is what many neuromorphic chips & modules do).

I wonder whether he thinks that how an artificial neuron is implemented is important, and how he feels about a neuromorphic emulation in software running on conventional hardware...

He surely must have thought these things through - I need to do some reading.

 

[LarryS]

"Wet" is also a description off our experience with water, not an objective measurement of the physical properties of water.

If you watch the video, Koch is using 'wetness' in an objective manner. When one has sufficient amount of H2O one gets the behavior of wetness. However he claims that consciousness may be different, one still requires a sufficient amount of complex matter, but consciousness may be an intrinsic property of matter.

 

[Donn]

However he claims that consciousness may be different, one still requires a sufficient amount of complex matter, but consciousness may be an intrinsic property of matter.

Consciousness is an intrinsic property of matter just as music is. Pluck a guitar string sometime.

Get enough matter into complex shapes and all kinds of novel things start to happen.

Is your shadow intrinsic to you?

 

[mijopaalmc]

If you watch the video, Koch is using 'wetness' in an objective manner. When one has sufficient amount of H2O one gets the behavior of wetness. However he claims that consciousness may be different, one still requires a sufficient amount of complex matter, but consciousness may be an intrinsic property of matter.

The he's flat out wrong about "wetness" being an objective property of water: http://scitation.aip.org/content/aip/journal/jap/117/3/10.1063/1.4905616

 

[LarryS]

The he's flat out wrong about "wetness" being an objective property of water: http://scitation.aip.org/content/aip/journal/jap/117/3/10.1063/1.4905616

lighten up Francis, he's using 'wetness' as an analogy

 

[mijopaalmc]

lighten up Francis, he's using 'wetness' as an analogy

If the statement about the concrete referent of the analogy (here the assertion that wetness is an intrinsic property of "enough" water) is not true, the abstract property (here the emergence of phenomenal behavior from the constituents of a mechanistic system) is in fact not illustrated and the analogy fails.

 

[dlorde]

Even Koch suggests that being conscious is different than the neurons and etc from which it 'emerges' watch the top ~2 min video on this URL

http://nautil.us/issue/19/illusions/ingenious-christof-koch

Ouch... he comes across as really confused and conflicted in that video - first he offers an argument from incredulity that seems to misconstrue what emergence is (his incredulity about the difference between neurons and their function; i.e. neural networks, and the conscious sensations he has doesn't weigh against emergence, that kind of startling difference is a feature of emergence). In the absence of a model he finds acceptable, he decides that it just 'is' a property (but not an emergent property?) of certain organisational structures that allow integrated information (neural networks anyone?)...

 

[LarryS]

Ouch... he comes across as really confused and conflicted in that video - first he offers an argument from incredulity that seems to misconstrue what emergence is (his incredulity about the difference between neurons and their function; i.e. neural networks, and the conscious sensations he has doesn't weigh against emergence, that kind of startling difference is a feature of emergence). In the absence of a model he finds acceptable, he decides that it just 'is' a property (but not an emergent property?) of certain organisational structures that allow integrated information (neural networks anyone?)...

what I got from this (and other Koch videos) is he's suggesting that if there is a sufficient mass of the right type(s) of complexity of matter, that matter doesn't have to do anything (on a macro level) to be conscious, as it doesn't have to do anything to have mass.

 

[dlorde]

what I got from this (and other Koch videos) is he's suggesting that if there is a sufficient mass of the right type(s) of complexity of matter, that matter doesn't have to do anything (on a macro level) to be conscious, as it doesn't have to do anything to have mass.

No; as I understand it, there needs to be a sufficient degree of the kind of complexity that can support integrated information. Integrated information requires activity from the matter involved.

I just don't follow his argument for the appearance of consciousness in such arrangements not being emergent.

 

[dlorde]

I just watched Koch's video answer to whether IBM's Watson was conscious; he said no - but I would be interested to see a calculation of Watson's integrated information rating.

What troubled me about that video was his use of the simulation fallacy - saying that if you simulate weather on a computer nothing gets wet, which overlooks the fact that weather is a physical process involving air, water, etc., but if consciousness involves integrated information, it is a computational process, which any computer can replicate like for like - in the same way that you can emulate weather in a large enclosed volume with heaters, fans, water, etc. - in which case you will see weather events where things do get wet.

His underlying point seems to be that you won't get consciousness unless you "replicate the actual cause-effect repertoire of the brain". It's not clear from the video precisely what that means if not the computational processes involved.

 

[LarryS]

I just watched Koch's video answer to whether IBM's Watson was conscious; he said no - but I would be interested to see a calculation of Watson's integrated information rating.

What troubled me about that video was his use of the simulation fallacy - saying that if you simulate weather on a computer nothing gets wet, which overlooks the fact that weather is a physical process involving air, water, etc., but if consciousness involves integrated information, it is a computational process, which any computer can replicate like for like - in the same way that you can emulate weather in a large enclosed volume with heaters, fans, water, etc. - in which case you will see weather events where things do get wet.

His underlying point seems to be that you won't get consciousness unless you "replicate the actual cause-effect repertoire of the brain". It's not clear from the video precisely what that means if not the computational processes involved.

I did not write my post #1871 clearly - but trying again, I think he's suggesting that consciousness can be found present with a sufficient amount of the right complexity of matter, but c-ness does not emerge because consciousness is a fundamental property of that sufficient amount of right complexity of matter. In evolutionary terms, consciousness did not emerge from matter, consciousness evolved in complexity to harness an already existing potential property of matter. Today he might ask for a mulligan.

 

[Manopolus]

Hmm... best alternative to materialism....

Dadaism.


Put that one in your pipe and smoke it.

 

[Nonpareil]

In evolutionary terms, consciousness did not emerge from matter, consciousness evolved in complexity to harness an already existing potential property of matter. Today he might ask for a mulligan.

 

[punshhh]

In principle, yes. However, our definitions of 'conscious', 'self-aware', or even 'alive' are far too fuzzy and subjective to cater for a strict scientific protocol.

Perhaps so. This is why I am using a human being as a control, (the definition). If the experiment produces an object which appears to be as conscious as the control, then we move onto the next stage of the experiment, how to make it, or establish that it is, either conscious, or self-conscious(given my definitions).

Nonsense. You know better than this. A lyrebird, or a tape-recorder for that matter, will only reproduce sound. Obviously, there is a bit more to a chainsaw than the sound.

It is an imperfect analogy. It was to demonstrate that mimics can in principle exist. The important distinction between the mimic and the person or chainsaw it is mimicking is that the mimic of the person is not actually self-conscious, but performs parrot fashion all the behaviours of the person. Likewise the lyrebird is not actually a chainsaw but performs parrot fashion all the vibrations in the environment that a chainsaw would produce.

 

[doronshadmi]

Since x is non-particular (please see http://www.internationalskeptics.com/forums/showpost.php?p=10470356&postcount=1825) according to Neutral monism, existence itself is neutral w.r.t contradiction or tautology, as explained in http://www.internationalskeptics.com/forums/showpost.php?p=10437382&postcount=1405.

So instead of the traditional way of existence ("there exists x such that...") the neutral way is as follows:

"x (where x is neutral WFF), such that NOx|SOMEx|EVERYx|YESx (where these four states are x minimal properties)".

Direct awareness of x is neutral (it is not restricted to NOx|SOMEx|EVERYx|YESx) and enables that is directly aware of x to be developed into an harmoniums factor among x minimal restrictions NOx|SOMEx|EVERYx|YESx (as explained in http://www.internationalskeptics.com/forums/showpost.php?p=10440785&postcount=1481 and http://www.sciencechatforum.com/viewtopic.php?f=51&t=27823#p269301).

 

[punshhh]

No. As usual, you have it precisely backwards. Reality is not dictated by language. Language is dictated by reality. This is what the words need to mean for them to be at all functional.

language is an abstraction, what words mean is derived from intellect/mind. While consciousness is not fully understood, it remains a phenomena in the external world and should not be confused with the peculiarities and meanings of words used to refer to it or describe it.

Which is irrelevant, since we don't care if it's biological or not. Just if it's conscious.

It is the control, our control.

"All language must be excluded"? What are you on about?

Bias and confusion of reality with notions of reality.

Yes.

And yet it transpires that this computation includes all the chemistry in the brain. So either we have two black boxes with identical contents or the second box contains hardware carrying out a simulation of the missing contents of the control box?

You really need to define your terms.

intelligence in the usual definition. Any way in what you appear to be suggesting we now have intelligence supported by a simulation of the biological basis from which it emerges.

Again, this is a hypothetical situation in which we have a concrete definition. In this case, screening out imperfect mimics is trivial. You simply look at them and see that they do not meet the definition.

Except the ones we can't distinguish.

So have I.

The hardware is irrelevant so long as it can run the same software.

and simulation?

You do realize that death is the point where this would necessarily no longer work, yes? The brain stops working at brain death. That's why it's called brain death. What you are demanding is not proof of the computational basis of consciousness, but an extremely convoluted means of proving that we can revive the dead given technology of an appropriate level.

Death makes no difference surely from your perspective. Take a brain a millisecond after death, when the cause of death was not in the brain. You just flick the switch and it is self-conscious. What's wrong with that?

We could, but that's entirely beside the point. We still know that the computational functions of the brain are what result in consciousness.

self-consciousness, yes. Not the life of the being, or the basic consciousness of the body.

No. I have never talked about perfect replicas of humans. I have made this very clear over multiple posts. Pay attention.

Which given your definition does include a perfect simulation of the molecular state.

What I have talked about is perfect replicas of self-consciousness.

including the behaviour of the molecules found in the brain?

...Yes, it would.

Perhaps you feel it would be necessary for the simulation to be self-conscious, it really is not a requirement for computation, even simulation. I suppose we could have a simulation of self-consciousness make the object self-consciousness.

Yes. Those biochemical reactions are what is being computed.

What is your point?

Well apart from what I have stated above, there is the issue of simulating substance which is not fully understood. Oh the joys of speculation!

 

[Nonpareil]

Perhaps so. This is why I am using a human being as a control, (the definition).

That is not what "control" means.

Or what "definition" means.

It is an imperfect analogy. It was to demonstrate that mimics can in principle exist. The important distinction between the mimic and the person or chainsaw it is mimicking is that the mimic of the person is not actually self-conscious, but performs parrot fashion all the behaviours of the person. Likewise the lyrebird is not actually a chainsaw but performs parrot fashion all the vibrations in the environment that a chainsaw would produce.

Which is not at all what you actually illustrated. What your argument actually showed was that the same sound could be produced through two different means - that the means of production does not have to be identical to create the same product.

Which has been my point all along. You somehow try to twist this into "the lyrebird makes the same sound as a chainsaw and therefore must have a chainsaw in its throat", which is so entirely, staggeringly wrong that I honestly have no response to it, because it obviously isn't an actual response to what has been said.

And yet it transpires that this computation includes all the chemistry in the brain. So either we have two black boxes with identical contents or the second box contains hardware carrying out a simulation of the missing contents of the control box?

You do understand that two different computers can compute the same thing via different hardware, yes? Your desktop can do anything that ENIAC could do.

Again, the Church-Turing thesis is that any Turing-equivalent computer can compute anything that any other Turing-equivalent computer can. It might do so less efficiently, through different means, more slowly, or any number of other things, but it can still be done.

Except the ones we can't distinguish.

If it's impossible to distinguish, it's not an imperfect mimic.

If it's not an imperfect mimic, it is the thing it's mimicking.

and simulation?

Is not a perfect model.

Death makes no difference surely from your perspective. Take a brain a millisecond after death, when the cause of death was not in the brain. You just flick the switch and it is self-conscious. What's wrong with that?

If the cause of death was not the brain, then the brain is still active. The brain does not stop working so long as it is alive; the moment it does, you are dead.

And it only stops if it has been damaged in some way. What you are asking is not a matter of simply turning it back on, but restructuring what has been destroyed. Which is technically possible, but is a matter of exceedingly delicate neurosurgery.

So yes, it can be done, but the fact that we can't do it now isn't an indication that it's impossible. It's just damn hard.

Which given your definition does include a perfect simulation of the molecular state.

I have never talked about perfect replicas of humans. I have talked about perfect replicas of consciousness. This does not require duplication of humans.

Pay attention.

Perhaps you feel it would be necessary for the simulation to be self-conscious

No. It would, by definition, be self-conscious.

Whether or not I "feel it would be necessary" is irrelevant.

Well apart from what I have stated above, there is the issue of simulating substance which is not fully understood. Oh the joys of speculation!

You don't understand it.

Others don't have that problem.