/sci/ - Science & Math

>implying I am not studying computer science to make this happen.

Consciousness is a mystery. It is scientifically unexplainable and undetectable. The only being in the universe you can say with certainty is concious is YOU. Everyone else could just be a soulless biological robot doing what natural selection has programmed them to do.

http://en.m.wikipedia.org/wiki/Philosophical_zombies#section_3

Besides, even if you could prove these "minds" were indeed true concious minds, supernaturalists would just argue that their supernatural being of choice gave these machines souls. Supernatural beliefs are most definitely irrational, but they can never be disproven.

>>5266931
>Consciousness is a mystery. It is scientifically unexplainable and undetectable.
Yeah, it really is an unfalsifiable proposition rendered untouchable by special pleading. Science shouldn't even bother with it and leave it to the new agers and religious nuts.

>>5266888
Here's some updates and annotations to that pic.


>>5266931
> It is scientifically undetectable.
Brain scanning can detect patterns that are typical for consciousness, so technically it's detectable.
Or maybe you'd like to say our sun is undetectable, we can only see the light coming from it.

>>5266943
>Brain scanning can detect patterns that are typical for consciousness, so technically it's detectable.
Brain mechanisms are detectable and "mappable", but the subjective experience that constitutes for consciousness is not, which is why it's not really science. Look up "qualia", if you want to waste your time.

There is no guarantee that there is anything it is like to be a brain uploaded into a computer. Scientifically, conscious experience does not even exist. There is no scientific basis to believe that a program running in a computer has a there there.

>>5266949
>Look up "qualia"
Why should I bother with philosophical bullshit when this is a science board?

Brain mechanisms are detectable and mappable, and they in turn correlate with subjective experiences. And stimulation of brain structures can induce subjective experiences. Everything points at subjective experiences being a secondary effect of neurons doing their job, just because we don't have an elegant explanation with obvious and intuitive punching power yet doesn't make it a magic mystery that will never be explained. Though of course don't tell that to philosophers as they'll throw a hissy fit.

>>5266962
They correlate, but they are not identical to them.

>>5266939
There is something it is like to be me. In principle it should be possible to determine why using science. Real science. New agers and religious nuts do not do science and won't every have answers.

>>5266962
I think you are mistaken what philosophers actually do.

>>5266957
>wat

>>5266931
Oh, the damned philosophy. "I think, therefore, I am." is a phrase the most people think is a solid proof of your personal existence. I forget which philosopher it was, but I do know somebody came up with a philsophical idea that proved that proof wasn't reliable.

That's just philosophy. No good answers, just brilliant questions.

As far as the religions go, I don't care and neither should the scientific community (which they don't). Although there are many religious people who do agree with science and understand there are fallacies in religion logic, the one's who don't are not worth the effort.

However, scientifically there are quesitons related to the after life and life in general that have not been solved yet. Although they are not the same, some physicist believe there is a sort of god or some central force in the universe, and experiments with artificial life could help uncover some of the mystery.

prepare for the massive storm of transhumanists, who never took a physics or computer science course in their life, but know with absolute certainity that "the singularity" will happen within the next 5 years.

>>5266969
>They correlate, but they are not identical to them.
Because we at best measure the equivalent of a dot of ink of a massive moving picture, we need more data to distinguish the pattern, we need even more to distinguish the pattern of movement and even further to understand how the picture is made.

>>5266980
>I forget which philosopher it was

René Descartes.

>>5266985
That isn't what I meant. You are not going to find consciousness no matter how well you describe the system.

>>5266986
>René Descartes
I know he as the one who stated, "I think, therefore, I am." But was he also the one who disproved the reliability of that claim?

In any case, philosophy is great for creating perspective, but it is not great for doing anything useful in itself.

>>5266980
I assume you have something it is like to be you. However, you have no basis to believe that if you created a computer program that somehow emulated you perfectly that that program would have anything it was like to be a computer program.

You cannot just assume that a computer program can be conscious just because it behaves and thinks identically to you.

For some this doesn't matter. If you cannot prove it scientifically then it doesn't matter. And on this matter we just don't know.

>>5266988
>You are not going to find consciousness
So when I can put you in a brain scanner and hear your internal monologue through the speakers, and see what you visualize on a computer screen, including behavioural predicions for the coming ten minutes, I've still not found your conscioussness, but just some crude representation?

You might as well claim that we haven't found the sun yet, and never will, just some light with correlations but not the real sun.

>>5266996
...and then somebody goes and provides a perfect example of incredibly muddy thinking.

>>5266993
It provides more than perspective. Philosophy also puts serious constraints on our thinking so that we don't have muddled ideas. It forces us to be very precise about what we mean and even if it cannot disprove something it certainly puts constraints on what it would take for things to be true.

>>5267000
>Ad hominem
You philosophers always contribute so much to scientific debate.

>>5267001
What part of that was ad hominem?

>>5266996
>consciousness

Not this word again. Unscientific dualism nonsense belongs on >>>/x/

>>5267007
I know you. I've hated you before. Don't you have other threads you can ruin?

>>5266994
I guess the real topic of this thread is consciousness and what it means to be conscious.

I feel as though there may not be a direct proof of conscious, but perhaps some sort of indirect proof that can only resolve to an actual consciousness.

Computers are unable to feel, but if this brain simulation is able to feel, all on its own, then could we assume it has a conscious? I mean, if you left the computer-brain alone, and it became bored, lonely and it started coming up with new ideas that originated out of no where, then could we assume the brain is unlike anything else on earth except conscious life? And if the only thing it resembles is conscious life, then we would know it is either conscious life or some new form of life. Either way ground breaking, and either way enough to say that our ideas of life would need to be re-worked.

>>5267012
It would only be that way because it is programmed to be that way. It doesn't prove "consciousness" in the computer anymore than it does in yourself.

>>5267010
Ignore the troll bud

>>5267010
>ad hominem
Go back to >>>/b/

>>5267012
In your first sentence you are saying that the topic of this thread belongs on /x/. Then why are you still here? Untestable spiritual claims have no place on the science board.

>>5266983

>implying Transhumanism == Singularitarianism

Oh man. Some of us actually work to make these things happen ;_;

>Especially with how close we are to successfully emulation a human brain.

It has already happened.

http://wiki.transhumani.com/index.php?title=Whole_Brain_Emulation#Large-Scale_Emulations_So_Far

See the blue line.

>>5267024
>why are you still here
Because the thread is still here
Go away wannabe

>>5267022
>promotes dualism nonsense on a science board
>calls others "trolls"

I'll give you a 1/10

>>5267024
What part of that was ad hominem?

>>5267030
>implying dualism

>>5267012
>Computers are unable to feel
Neurons are unable to feel too. But they can generate a pattern that feels.
The reason why we think computers are soulless stems from the fact that they are too precisely manufactured machines. They do precisely as we tell them to, every single computer anywhere will slavishly follow a ultra deterministic pattern.

If your computer could not be loaded to run arbitrary programs at will. If it consisted of a massive bank of write-once-memory and an amalgamation of various specialized CPUs that crystallized into a software speciality that requires some quirks to operate well, then everyone would have a lot easier time to belive they could be conscious.

>>5267036
>not implying neutral monism
>2012

>>5267017
>It doesn't prove "consciousness" in the computer anymore than it does in yourself.

Than what else do you want?

I mean, there are many ideas that aren't necessarily proven, but everything else fits the idea in a way that it is the only logical answer. Look at the Higgs Boson particle, it wasn't even viewed. But the before and after was, and this allowed for scientist to come to the conclusion that the particle did exist.

If our conscious can't be proven, then who would expect the computer's conscious to be proven? We would simply have to correlate similar traits between the two and come to a logical conclusion of a present "conscious" or not.


I am loving this thread.

>>5267031
Please take the topic to >>>/x/ where it belongs. We do not want untestable claims without evidence here.

>>5267036
Exactly. That poster was implying dualism. His proposed soul / consciousness entity has no evidence and no testable effects.

>>5267037
You cannot even prove that neurons form "patterns that feel", so how do you expect to show that a computer does?

>>5267039
Honestly I do not know what I want. Some are content to simply write it off as another non-existent thing that people hypothesized. Like the word "sunrise" they see the words we use to describe our subjective selves as fictions from when we didn't know better.

I'm willing to admit I don't know and anybody here would be honest if they said the same.

>>5267042
>You cannot even prove that neurons form "patterns that feel"
I have a brain. It's made of neurons. I feel. QED.
If you're going to focus on shit tier philosophical aspects you'll continue to get shit tier philopsophical answers.

>>5267040
>That poster was implying dualism.

>referring to yourself in the third person
What the fuck

>>5267037
Then I guess computers and human's are just the same things, but made of different parts. But you could say the same thing for everything in the universe. From a human to a computer to a rock, we all do something, and in the big picture, we all do the same thing, and that is exist.

I guess the computer is no different from the human. With or without what believe is a conscious. We both do things, it's just that a computer does different things. But even if we did make a computer do exactly what human's do, than we really haven't proved anything.

I don't even know, my brain is full of fuck right now.

>>5267055
It is "shit-tier" now to say plainly the current state of knowledge? Maybe I should have said "prove scientifically", because I myself see that my brain is producing a "pattern that feels", but since it is private and unavailable then it does not really constitute proof of anything.

>>5267054
I did not reply to myself. I replied to another poster.

>>5267055
Non-sequiturs and baseless claims without evidence. Why are you here, if you totally ignore science and rationality?

>>5267066
>Then I guess computers and human's are just the same things, but made of different parts
This is correct. Humans are biological machines. Our cells are made of molecules and molecules obey the laws of physics. Reactions to physical input are entirely deterministic.

>>5267052
I'd hope nobody here claimed to know what they we're talking about. But still, you don't need a degree in computers or physics or philosphy to ask the right questions and come up with the right answers.

But I really don't know what the questions we are tying to answer are.

>>5267066
My main assertion is this. If you believe in qualia, that there is something it is like to be you, then you should be dubious about uploading your brain into a machine. It may all be quarks, but it has a different arrangement than your biological brain, and since you have no idea what makes qualia happen you'd have every reason to believe such a transfer would snuff out your subjective experience.

If you don't, then it doesn't matter. It would just be an upgrade.

>>5267076
And the laws of physics, which we do not fully understand, support the emergence of a mind.

>>5267088
What do you mean by mind and emergence? My immediate thought is that you are being sloppy, but then I figured I'd give you a chance.

>>5267086
>If you believe in qualia
Back to >>>/x/ please.

>>5267088
No, they do not. The laws of physics do not imply the existence of untestable non-physical and non-interacting entities without evidence. This is complete and utter bullshit. Take that spiritualism nonsense to >>>/x/

>>5267093
Why are you getting pissy because I posed a hypothetical question? If you X then Y, if you not X then Y'

I do not even think you read my post.

>>5267096
He has no idea what he's talking about. He's an /x/tard trying to justify magic.

>>5267101
>metaphysical naturalism
>magic

>>5267066
>I guess the computer is no different from the human.
Well, we both consist mostly of protons and neutrons in various configurations. The point here being that there's an infinite amount of patterns, not just static patterns but reactive patterns.

Your conscioussness is an emergent phenomenon, multiple highly refined functions come together to form it. The same is true for the display picture you see now, it's not generated by any single transistor, but several hundred million transistors and other fancy tech working together, it's emergent.

Now I could remove your occipital cortex, and you'd stop seeing, yet you'd still be conscious, and I could disable the screen rendering output of a computer, yet it could still track a virtual desktop inside itself.

And I could degrade your brain functions, your short term memory would last 0.5 seconds, you'd not remember what the last sentenced you read was about, and you'd still be conscious. I could rollback the computer, cut down on RAM, diskspace, clock frequency, and it'd still be a computer.

Imagine if moores law went backwards. And your consciousness degraded at a similar rate. When would you stop considering the computer a computer and yourself conscious?

The border is extremely ill defined,and yet in every good damned discussion about conscioussness we're trying to pin down a precise and laser cut definition, and it fails every god damn time because it's a spectrum, it's emergent, it's made out of pieces where every piece contributes a bit, with none defining of conscioussness.

In the same way we could probably improve of conscious experience with neuroprostethics, perfect recall, intuitive understanding of other persons due to sharing of feelings, sharing of visualisations, sharing of knowledge, sharing of innate conceptual understanding, sharing of senses and so on.

pic unrelated

>>5267107
>unobservable non-interacting entities without evidence
>magic

Choose two and go back to >>>/x/

>>5267110
> mfw I went to /x/ and we figured all this shit out
> mfw I have no face.

>>5267025
>http://wiki.transhumani.com/index.php?title=Whole_Brain_Emulation#Large-Scale_Emulations_So_Far

2005.
Never heard of it before.
Why.
This is a pretty fucking big deal.
WHY.

>>5267110
Did you reject science or something?

>>5267115
There actually isn't. There are something like 10^80 configurations of the space you are currently occupying.

>>5267115
>remove your occipital cortex, and you'd stop seeing, yet you'd still be conscious
I wouldn't be too sure of that.

>>5267115
Kind of like a taking a grain of sand from a pile of sand.

>>5267123
Well, that is for all practical purposes enough availible configurations to be percived as infinite by the human brain and the ordinary scales of numbers and variations we percive. Though I guess that number is further reduced by availibility of elements and energy and stability and whatnot but the concept is still true: the number is abundantly large.

>>5267142
To be fair it is an amazingly large number.

>>5267139
Well, I could damage it in a precise and controlled manner to ensure you end up blind without lopping it off with a chef knife and risking massive further trauma that kills you. People with occipital damage due to various trauma have lost their vision this way without turning into vegetables.

>>5267115
>Your conscioussness is an emergent phenomenon
That's a funny way to say that you believe in magic. Using the word "emergence" makes only sense when there is an observable phenomenon that can be considered to emerge. Claiming that an unobservable metaphysical phenomenon without evidence is "emergent" is pure bullshittery. "Hurr, the non-interacting invisible demon in my closet is an emergent phenomenon."

>>5267126
I am a scientist and I embrace rationality. Rationality tells us not to believe in claims that are untestable and have no evidence.

>>5267153
Do you believe in the sun, yes or no?

>>5267153
Rationality tells us to believe in claims that are possibly testable and therefore would have evidence. How will we know for sure if we don't attempt to devise an experiment?

>>5267153
So conscioussness is not emergent according to you, what is it then?
How could it not be emergent? Do you belive in a conscioussness particle while in the same post claiming you embrace rationality?

>>5267166
In case you have not figured it out, he simply does not believe in it AT ALL.

And "believe" is the correct word, because there is no definition of it that can be scientifically tested at this time.

>>5267162
>hurr durr let's all believe in ghosts

>>5267165
I can see the sun.

>>5267166
>what is it then?
How about you prove that your magical soul entity exists in the first place before asking for an explanation? Care to show me any evidence? No? Then fuck off to >>>/x/

>>5267172
You know you can see, ergo you're conscious.

>>5267172
>I can see the sun.
That wasn't the question. I assume that means yes.
Back to >>>/x/ please.
The sun is not observable, its just a construct to explain where a lot of photons on Earth come from. /sci/ is not for unscientific untestable irrational bullshit.

>>5267172
>How about you prove
No, the burden of proof is on you this time. Though of course you'll shift it right back to me because you're an incompetent shitposter that can only make baseless claims and shit up this board with replies so pointless even the most edgy teenage presudophilosopher would be embarassed to associate with you.
in b4 >ad hominem

>>5267176
Bullshit. The ability to perceive visual input is a physical feature. It does not require magic.

>>5267178
Troll harder.

>>5267185
The burden of proof is on the one who makes a claim of existence. You claim a non-interacting entity residing in your head exists, so it's up to you to show the evidence. It is only rational not to believe in claims without evidence.

>>5266888
>>5266943
As a guy with a degree in neuroscience and who does neuroscience research, these pictures are mindboggingly retarded.

You can't describe the brain's processing power in computer science terms. It works in a completely different manner from a computer. There's no point in computing power where the computer suddenly becomes sentient, nor is there a point where you can go "yup, we can upload a brain to it now". It's all about the design. If you were to try to upload a brain to a computer, the computer would need to be a brain emulator, which would, I imagine, entail being able to perfectly simulate not only every neuron but also every biochemical process involved.

Even assuming we had the computing power to do this, there are two massive issues completely unrelated to computing power:
1) You need an interface to "read" the brain. We do not have anything like this yet. The best we have are structural brain imagers (CT, MRI), but the resolution of those technologies is way too low to image every cell in your brain. Even if we did have that kind of resolution, those technologies still couldn't tell you anything about the molecular biology of each cell (the properties of neurons are rather fluid and much of cognition is based upon how your neurons change themselves in response to input).

2) We don't know nearly enough about the brain to simulate it. This post is already too long so I won't go into details here, except to say that our current understanding of neuroscience is like knowing what the tip of an iceberg looks like but still being ignorant of the 90% that's under the water. And then every year as new discoveries are made neuroscientists realize that the iceberg is actually 10x bigger than we previously thought.

>>5267194
>Bullshit. The ability to perceive visual input is a physical feature. It does not require magic.
Prove you can see or go back to >>>/x/.

<span class="quote deadlink">>>5267203[/spoiler]
Stop it. Trolling isn't allowed outside of /b/.

>>5267198
This.

>>5267198
Computer scientist here. If people like you are all neuroscience has, then I bet it's gonna take a long time...

> It works in a completely different manner from a computer.
I lol'd.

>>5267198
> It's all about the design.
learn2churchturingthesis, fagoot

>>5267217
Do you know anything about neuroscience? I'm guessing not.

Saying that a brain works the same as a computer is like saying that a bird works the same way as an airplane. It's retarded.

>>5267237
>Saying that a brain works the same as a computer
Supercomputers doesn't resemble nuclear weapons at all, yet we're still doing simulations of them. And given that the brain is doing computations, we really just need to replicate the computational aspect of it, no need to simulate phospholipid bilayers and other wetware.

The point is not to create an identical simulation of a human person with all drawbacks of a body, but to replicate the intelligence and the cognitive aspects. If the replication gives superior results to the original person, even better.

>>5267237
see
>>5267233

It's ridiculous how you have no idea what you are talking about. I mean, obviously, the presented charts are blatant speculation and no one knows exactly how things will turn out. But you act like you could proof that uploading a brain to a computer with today's design cannot work. If you actually could, I'd suggest you waste no time talking to us but do the following:
1. Publish proof.
2. Disprove Church-Turing-thesis.
3. Acquire Turing Award, Fields medal, w/e
4. ???
5. PROFIT

If you are still answering to this instead of becoming famous: Yeah, thought so. Now please stop talking about computers, complexit, simulation or anything related ever again.

>>5267244
>If the replication gives superior results to the original person
Though of course we don't necessarily need to have an original person, it could learn and grow up as a purely digital person.

Or we could scan the brain of a deceased person and see if we can revive him in silico.

>>5267237
Another computer scientist here. Ignore that guy, apparently he is a fool.

>>5267248
Church-Turing thesis is not a scientific statement, it is a mathematical one.

>>5267258
Did I imply anything about it? The important fact is that to this day, no one has found a definite counter-example to it. If our neuroscience guy had one, it would cause a revolution in several parts of the structural sciences.

Rationality tells us consciousness exists, but is agnostic about how it exists.

Monoists,dualists/anything other than agnostics on the issue are not being rational.

>>5267260
You seemed to imply that it showed that computers and brains work the same way.

>>5267248
Did you even read my first post? There are major issues completely unrelated to the computational aspect of it. It's not enough to have a computer that is powerful enough to emulate a brain.

>>5267244
Except that the properties of receptors, channels, transporters, etc. have a big impact on how excitable a cell is. If you're going to emulate a brain you have to be able to recognize when every single one of those things changes. Even if your emulator is just going to say "okay, this neuron now has an excitability coefficient of .95" you still need to recognize when a change in the physiology of the cell has made it more or less excitable, which requires knowing every process. There is so much processing going on at every level of the brain, you just can't model the high level systems and call it a day. A LOT of processing goes on at the cellular and molecular levels as well, and you just can't ignore it. A single flaw in a single receptor subunit can be the difference between a genius and retard (eg: subunit replacement in NMDA receptors influences how permeable the cell is to Ca++, which in turns determines how easily the cell's synapses undergo LTP and LTD).

You can't compare a nuclear explosion (a very simple thing to model like all physical phenomena) to a biological process (extraordinarily complex). There's a reason why biologists still do all their research in living organisms instead of just renting time at a supercomputer to run a simulation.

>>5267262
It does, but only to a certain extent. Let me explain.

The Church-Turing thesis states that the model of computation given by, for example, the Turing machine is universal. That means that every process that can be accurately described can be simulated by a Turing machine, and thus, probably by a computer someday.

The neurscience guy claimed he knew a process that could be subject to science but COULD NOT be modeled by a Turing machine, which may exist, but discovering it is a big thing, really.

On a further note, the human brain is turing-complete. Thus, if the Church-Turing thesis holds, brains and turing machines are comparably powerful, which in fact may cause them to share some properties. However, obviously, I'd never imply they work in a similar way internally.

>>5267282
>On a further note, the human brain is turing-complete
note that if something is turing-complete, this means it can simulate a turing-machine. This does not necessarily mean that a turing machine can simulate it.

>>5267282
>The neurscience guy claimed he knew a process that could be subject to science but COULD NOT be modeled by a Turing machine
No I didn't you fucking retard. I never said it wasn't possible, I said it isn't going to happen anytime soon because:

1) We don't know enough about the brain to create a working model.
2) We don't have a way to read information from the brain like you would read information from a disk.

The stuff about computation power was explaining that it's dumb to say that computing power is sufficient to model a brain, as if once a computer gets this fast it's as powerful as a brain. You cannot describe a brain's processing power in terms of FLOPS or hertz, you cannot describe it's memory in terms of bytes. It does not work that way. You can run a simulation of it, sure, but that will require a hell of a feat of science that comes from complete understanding of the brain. To say "oh, once computers can do this many FLOPS we'll be able to simulate the brain" is non-nonsensical. That's like someone saying "once we discover how to combust fuel we'll be able to travel to other galaxies". Yes, you do need to know how to utilize fuel to gain energy in order to get into space, but that's hardly sufficient for space travel.

>>5267281
>It's not enough to have a computer that is powerful enough to emulate a brain.
You need a brain to emulate and an emulation of course too, then it's enoguh.

>>5267282
>the Turing machine is universal
where universal was defined by Turing to be everything that a Turing machine can do. The Church-Turing thesis had little to say about physical systems or science as a whole.

>>5267290
Obviously. Mathematically speaking, let us write A < B for B can be reduced to A (i.e. A can simulate B).

Then, with the brain obviously being Turing-complete, it holds that... (as you clearified)
Turing < Brain

So (axiom A) IF the brain is understandable by science (i.e. can be described by rules based on logic) and (axiom B) IF the Church-Turing thesis holds, then a Turing machine must be able to describe it, too, because according to axiom B it can describe any ruleset based on logic. Thus:
Brain < Turing

Thus, what I wanted to say was that if we grant A and B, we can deduce that Turing = Brain, where = is an equivalance relation over the computational power of a given model.

>>5267281
>A single flaw in a single receptor subunit

There won't be any subunits in the simulation.

>>5267290
Obviously. Mathematically speaking, let us write A < B for A can be reduced to B (i.e. B can simulate A).

Then, with the brain obviously being Turing-complete, it holds that... (as you clearified)
Turing < Brain

So (axiom A) IF the brain is understandable by science (i.e. can be described by rules based on logic) and (axiom B) IF the Church-Turing thesis holds, then a Turing machine must be able to describe it, too, because according to axiom B it can describe any ruleset based on logic. Thus:
Brain < Turing

Thus, what I wanted to say was that if we grant A and B, we can deduce that Turing = Brain, where = is an equivalance relation over the computational power of a given model.

(deleted the post, fucked up my reduction relation...)

>>5267310
As I explained IN THIS POST, the Turing machine was taken for the definiton of "universality" because up to this day, it is able to execute any computation we could ever think of.
This didn't affect other sciences, because no one ever had a problem with a Turing machine being able to execute his theories.

Seriously, we're not doing history of science here...

>>5267313
are you making claims about imaginary things that don't exist, or do you have some reason to make this claim?

>>5267313
But you still need to take them into account. Otherwise you can't correctly model learning. Just assuming that "this stimulation = LTP/LTD" isn't good enough, because a person may have a mutation in one of those receptor subunits that allows in more Ca++ which in turn makes their cells more prone to LTP/LTD. It could, in fact, be a major contributor to why they have an IQ of 130 instead of 100.

>>5267332
Not the guy but I think the general idea is that the perceived complexity in biological systems like the brain isn't inherent to the problem they solve but only due to a lack of abstraction in the models of current biology and related fields.

Especially when it comes to computational power, all models of computation we have so far (and note that these are all able to execute every precisely describeably algorithm we could think of so far) are extremely simple for what they can do.

Thus, the... let's say most hopeful... assumption is that modelling a certain abstraction of a human brain is sufficient to grasp the parts we are interested in. Under this assumption, the neuroscientist's claim sounds kinda like "You cannot model the solar system unless you understand the weather on earth". And the transhumanist people are just hoping that "Mars won't give a fuck".

Either way, I guess it WILL be very helpful to get to a comparable computational power (however that was measured in the first place), because it will tell us that all we need to do is try to play around with the model now, even if we don't gain any significant insight from neuroscience. So we could have luck, but objectively speaking it does look kinda hopeless atm.

All these threads turn into crazy useless philosophical nonsense.

Why not argue about how we'll have to modify our laws to account for persons with multiple, simultaneous physical instantiations? How do you arrest them, et cetera?

>>5267332
Are you entirely ignorant or do you just like to be really obnoxious with your arguments?

Do you think we have to simulate protons and neutrons too or the simulation will fail? No, well, how about the intermolecular bonds between amino acids and protein folding? No? Why then should we arbitrarily draw the line at subunits that determine Ca flux? Why not just assign a virtual hand of god to drive the Ca flux, that way we can change a number in the code to change the flux as opposed to going through a protein folding simulation to get novel subunits every fucking time we want to change a value.

Are you entirely ignorant of programming and computer science?

>>5267351
That isn't science. Philosophy is.

>>5267331
It is not able to compute every computation we can think of unless you define computation to be "what a turing machine can do"
What about the busy beaver sequence? The halting problem? Perfectly well-defined problems with solutions that no program can produce.

>>5267352
>Why not just assign a virtual hand of god to drive the Ca flux
Not the guy you're responding to (I'm the other guy), but that's what I've been saying this whole time. You don't have to model the chemical elements like protons and folding, but you do need to model the molecular elements, like the qualities of the receptors. To make that "hand of god" you're talking about, you still have to know the structure of the receptor and what the implications of that structure are in order to know how much to change the relevant variable in the code. So even by your own admission you're fucking wrong.

>>5267359
well-defined problem =/= there is a well-defined algorithm to solve it

>>5267352
I'm sorry if I seem(ed) obnoxious.
It seems to me like neuroscience guy is saying
>we don't have a nice model of how the brain works, so to 'upload' a brain, it seems like we'll need to simulate every atom, and then we're at the mercy of cascading errors
and your response was
>this system won't have errors, or if it does, they'll be too small to matter
which is a pretty wild claim.

Maybe I'm just not paying enough attention (I'm alternating grading and browsing 4chan), but that's how I read the conversation.

>>5267374
so you are defining computation to mean "algorithm" or "something a computer can do"
your reasoning is circular.

>>5267261
Believing in "consciousness" means dualism. This is irrational. Stop it or take it to >>>/x/

>>5267373
>that's what I've been saying this whole time.
>So even by your own admission you're fucking wrong.

You're saying you're wrong?


>you still have to know the structure of the receptor and what the implications of that structure are in order to know how much to change the relevant variable in the code.
Of course you need to know the structure, function and dynamics of what you're simulating even if you simplify it.
What's the fascination with nitpicking here? This fucking discussion is regressing to more convoluted and retarded naysayer arguments faster than I can type, I'm out.

>>5267395
>a computer can do computation
No shit, sherlock.

An algorithm is a precisely defined series of actions. For more, you can google yourself. As of now, everything that could somehow be called algorithm could be executed by a Turing machine, which is why it was used as the basis of modern computability theory. The Church Turing thesis says that this will apply to any sufficiently precise action description yet to come.

Thus, if you are not to disprove the Church Turing thesis, you can only claim that...
1. The behavior aof the brain cannot be precisely understood.
2. The behavior of the brain can be modeled by a turing machine.
Or you could always prove Church and Turing wrong, but as I said, this is a bit too big to be done in a /sci/ thread.

The neuroscientist and I seem to agree now, anyway. I'm just very allergic whenever I think I see a claim regarding the impossibility of something, because it's a very big statement to make. As long as we're mainly talking about if it will take another 10 years or another 100 years, I don't have much of an argument for either case, except I will TRY to make it short.

>>5267318
> brain obviously being Turing-complete

please tell me more about how this is obvious without going full retard by assuming the brain is deterministic.

>>5267490
With enough ink, paper and manual labour you can do the same thing as any turing machine can.

This is a learned task of course, but to be able to learn this emulation task it would need to be turing complete. Or something like that, I'm not the guy who made the initial claim.

>>5267440
look, according to wiki:
>Church–Turing (...) "thesis" [is] about the nature of functions whose values are effectively calculable; or, in more modern terms, functions whose values are algorithmically computable. In simple terms, the Church–Turing thesis states that a function is algorithmically computable if and only if it is computable by a Turing machine.

This has nothing to do with brains or chemistry or physics or anything physical at all. This only has anything at all to do with math (which modern computers simulate (imperfectly))
I don't know why you're bringing it up at all.

I wish I could talk to you in person because I think we're talking at cross purposes.

You seem to be arguing that someday we might be able to simulate a human brain in a computer. Fine.
My argument is that this may also not happen. This may not even be possible.
I see no compelling reason why I should believe it is possible.

As you are "allergic" when you see people claiming things are impossible, I am "allergic" when I see people claiming things must be.

>>5267397
>believing in consciousness means believing in dualism
You funny guy.

>>5267490
a system is turing complete if it can simulate a turing machine.
given enough time and perhaps paper, I can simulate a turing machine, so I'm turing complete.
This means I'm at least as complex as a turing machine. This does not necessarily mean a computer can simulate me however because I may be more complex than a turning machine.

>>5267541
It's true.

http://en.wikipedia.org/wiki/Dualism_%28philosophy_of_mind%29

>>5267593
It's not.
www.google.com/?q=metaphysical+naturalism

>>5267541
>>>/x/
don't come back

>>5267608
>metaphysical

Stopped reading there. Back to >>>/x/ please. Science deals with testable and observable things.

>>5267653
Why do you not like philosophy?

>>5267608
You are implying the existence of a non-interacting and unobservable entity. That is dualism

>>5267657
So? What is wrong with dualism?

<span class="quote deadlink">>>5267655[/spoiler]
I do like philosophy. It doesn't belong here though. This is a science board. What you posted isn't even philosophy btw. It's just edgy teenager redditor babble.

<span class="quote deadlink">>>5267659[/spoiler]
It is not science.

>>5267490
I am the original guy but I think you got the answer you wanted: Stating the brain is turing-complete works the other way aorund. A human brain can simulate a Turing machine, in principal. That said, Turing machines have indefinite memory and human brains probably don't, we can still say that for practical purposes, a brain can do at least what a Turing machine can do.

>>5267539
>I don't know why you're bringing it up at all.
The Church-Turing thesis describes computation, and has thus to do with EVERYTHING. And it states that brains or chemistry or physics cannot provide a more powerful computational model than the Turing machine.

So what I'm stating is that it (most probably) cannot happen that chemistry or physics or neuroscience give us a logical, mathematical model of how the brain functions but a Turing machine cannot compute it. Basically I wanted to state that we have only two scenarios:

>scenario 1
The brain cannot be fully understood by logic-based sciences. Then a Turing machine cannot simulate it, either.

>scenario 2
The brain can be fully understood by logic-based sciences. Then a Turing machine can simulate it, too.

What I'm argueing against is the in-between case where people think science will be able to generate a functioning model of brain behavior, but we will never be able to apply it to a computer.

Also,
>I see no compelling reason why I should believe it is possible.
There isn't atm. Because today's sciences don't really understand the brain, there always is the worst case scenario where they simply don't get any better. And I'm not saying anything is going to happen for sure or anything is certainly possible. My whole point is that IF we can completely understand the brain using today's scientifc method, it is possible to simulate it using today's computer architecture.

>>5267025
Thanks a lot for the link, anon.
>Observed Alpha and Gamma rhythms.

That's pretty fucking impressive.

>>5267980
I didn't have a chance to reply to all this earlier, but you have a fundamental confusion here. The Turing machine is a mathematical concept and what can be said about it applies only to math.

There are a great many things that are not computable and although we may not be able to build a physical COMPUTER that is better than a Turing machine, that does not in any way imply that every aspect of the physical universe is COMPUTABLE.

That indeed means that the behavior of a brain may not be a computable function that can be described and computed by a computer.

Turing machines are actually incredibly limited, and even as a computer scientist I believe that most of the interesting math is beyond its reach.

I almost consider anything that is computable to be boring.