if you could transfer

[NanoStuff]

Originally posted by: lyssword
if you could transfer brain functions/thought into a computer powered system, do you think that the said person would be able to think faster? Because all of the thought processes in brain are chemical, they are supposed to be a little slower than electrical as in computer/whatever system.

Thought processes are not chemical, they are informational, not limited to any specific transport. But the 'transport' is in fact electrical, chemicals only serve to control these electrical responses.

Originally posted by: lyssword
Now, I realize that there is no way that this will be even possible, since the human brain is so extremely complicated

That's not at all true considering technology grows exponentially in complexity. Replicated the brain is not a moving target for science. There's a number of interfaces and neuronal recording technologies at present, just enough to prove the concept.

But to answer your question, yes, of course, with a sufficiently capable computer, you would be able to think faster.

Originally posted by: lGannon
The brain is a special kind of quantum computer,

It's a concept that's been considered by a few, but the theory doesn't hold up so it's mostly disregarded. Existing brain models accurately represent natural responses with no consideration for quantum phenomena. So no, it's not a quantum computer.

Originally posted by: lGannon
Yeah and how much DO we really know?

There's plenty more to learn, but presuming we hardly know anything is an insult to scientists everywhere. We do in fact know a lot, but it takes a bit of exploration to understand the fact. Only the people that don't know a lot claim their ignorance as universal... no offense, by ignorance I simply mean lack of understanding

Originally posted by: lGannon
Given how complicated one cell would be to do accurately, doing a brains worth....well, that's an awful lot of computing power. You might be able to cut corners here and there, but I don't think you can do so without distorting what you're simulating - ie, while you're uploaded brain may be conscious in some manner, possibly, it won't be like a twin brain for the uploaded person. Even if the initial differences were small, given the way all the parts of the brain interact, a small difference would very rapidly butterfly into huge differences.

Good insight. It's true a simulation cannot perfectly replicate the system being simulated, it's a matter of the level of precision you want to achieve. The brain however would not be able to distinguish a subtle enough deviation, for example we can only distinguish between two different colors, tones or amplitudes that are a certain range apart. Realistically, the simulation only has to be 'accurate' enough, and preventing compounding rounding errors is something we can easily handle. It's also important to keep in mind we don't necessarily have to simulate the brain's biological state, as there are inherit, evolved limitations to our native brains. We may and probably will want to take a more ideal approach by reduction of algorithmic 'bloat' for what will likely prove to be detrimental processes. For example, you could waste a whole lot of computing time simulating neurotransmitter distribution through the brain to represent a biological scenario. However you could achieve the same end result by truncating the simulation to produce the end result without simulating the process of arriving at the result, thus not only achieving your desired outcome much faster, but freeing it from physical limitations that the simulation induces.

 

[dorion]

Originally posted by: NanoStuff We may and probably will want to take a more ideal approach by reduction of algorithmic 'bloat' for what will likely prove to be detrimental processes. For example, you could waste a whole lot of computing time simulating neurotransmitter distribution through the brain to represent a biological scenario. However you could achieve the same end result by truncating the simulation to produce the end result without simulating the process of arriving at the result, thus not only achieving your desired outcome much faster, but freeing it from physical limitations that the simulation induces.

By bloat and biological scenarios do you mean things like the mass of brain matter to handle physical movements, or even vision? As a simple brain in a box you would need only whats needed for thinking, step up to a robot body and there needs to be circuitry in place but it would be a whole different beast than the human body. What parts of the brain would you need to continue having to see with robot eyes? (this is a rhetorical question)

And why all this talk about simulating, I think a better approach is to circuitry more like the brain. What is going to be easier, a brain run on a super fast serial processor or a massively scalable parallel processor. If only Von Neumann hadn't died we might have the answer. Thats why I like the idea from nortexoid more than the download idea. A way to approach the complexity of the brain instead of our standard brute force that we apply to so many problems.

Lastly I find the idea of there being more "mes" that aren't "me" disturbing.

 

[Azulsky]

I would imagine the OP is expounding on the idea of transferring logic patterns

In my opinion what accounts for human intellect is more than simply logic.
Empathy is something that couldnt be accounted for by binary, it defies logic, just as logic defies it.
It isnt really known whether the brain is the seat of the "soul" or "spirit" or just a interface, because such cannot be measured.
We must also consider creative capacity diminishing, as stimuli are cutoff and experience terminated.

Interesting, an Anime Ghost in the Shell addresses an issue that one of the previous posters mentioned. If we transferred the brain into a computer system, then it would cease to function because it requires feedback from the entire human body. The Anime series speculates that cyborg bodies and computer systems have to trick the brain into thinking that is it still coexisting with the rest of the body.

Also to note that if the circuitry of the brain could be emulated by a computer, it still wouldn't accomplish anything. A humans capacity for logic is based on experience, aka memory, and bases decisions on past experiences. This also is addressed in another Anime series(sorta) NGE in the form of a supercomputer trio known as the Magi, which were like judges of a court making decisions based on preloaded statutes of its designer. The three most important aspects of the designers life(Woman, Mother, Scientist) were separated and each placed into a single computer. Decisions were based on a majority vote, allowing each computer to consider a different view of a contingency. We dont truly understand the brains capacity for information storage, or i havent seen any neurological studies dealing with things such as Savants and Photo Memory regarding this issue.

Just thought that aside was a little interesting. Overall i believe it is interesting that this topic is being discussed, as I believe the terminal result of software and hardware development is creating genuine intelligence.

 

[NanoStuff]

Originally posted by: dorion

Originally posted by: NanoStuff We may and probably will want to take a more ideal approach by reduction of algorithmic 'bloat' for what will likely prove to be detrimental processes. For example, you could waste a whole lot of computing time simulating neurotransmitter distribution through the brain to represent a biological scenario. However you could achieve the same end result by truncating the simulation to produce the end result without simulating the process of arriving at the result, thus not only achieving your desired outcome much faster, but freeing it from physical limitations that the simulation induces.

By bloat and biological scenarios do you mean things like the mass of brain matter to handle physical movements, or even vision? As a simple brain in a box you would need only whats needed for thinking, step up to a robot body and there needs to be circuitry in place but it would be a whole different beast than the human body. What parts of the brain would you need to continue having to see with robot eyes? (this is a rhetorical question)

And why all this talk about simulating, I think a better approach is to circuitry more like the brain. What is going to be easier, a brain run on a super fast serial processor or a massively scalable parallel processor. If only Von Neumann hadn't died we might have the answer. Thats why I like the idea from nortexoid more than the download idea. A way to approach the complexity of the brain instead of our standard brute force that we apply to so many problems.

Lastly I find the idea of there being more "mes" that aren't "me" disturbing.

I mean the requirement only to simulate the electrical properties of brain cells. Just get the bayesian information processes to work. You obviously would not want to simulate the path each electron takes, much like you wouldn't want to simulate the path each neurotransmitter takes, but arbitrate to simply configure the end result of such a process.

IBM intends to create such an algorithmic reduction in their model to have the processing capacity to simulate an entire neocortex. It's still on the table what such simplifications will do to the model, but I'm willing to bet it will improve communication. This may not be what they want, not at the time anyways. The idea is to get the most accurate representation of the brain, not necessarily the most efficient. But once it becomes possible to record the entire system in vivo, such efficiency and effectively improvement to the natural model will probably be desireable to those that choose to drop their biology. The patterning information for the brain and it's processes is probably extremely simple on the low level. It appears to be likely that in can be reduced to at most a KB-sized bayesian network that has a fractal-like expansion with very deterministic structural arrangement based on external input. The problem for living humans is much more complicated, being necessary to traverse every neuron to get the necessary information. But modelling a brain from scratch on a computer will likely prove to be a very simple model, simpler than that of the instruction set of a desktop processor. The limiting factor today doesn't seem to be what we know, but a lack of processing power allowing us to run a biological model of the brain, which is necessary before the entire system can be reduce to one relatively short magical algorithm. It seems to me the decade after the next is when the issue will begin to wrap up. Gives us time to consider the ethical implications of having slave robots with humanoid brains running around.

 

[NanoStuff]

Originally posted by: Azulsky
Empathy is something that couldnt be accounted for by binary, it defies logic, just as logic defies it.

Everything that can be accounted for can be accounted for in binary. Binary is the fundamental representation of truths and non truths, it's the core element to an entire turing model. As such, empathy can be accounted for in binary once the information processes responsible for it are known, even though not necessarily understood.

Originally posted by: Azulsky
It isnt really known whether the brain is the seat of the "soul" or "spirit" or just a interface, because such cannot be measured.

Such cannot be measured because such does not exist. If you feel it's appropriate to associate a soul with a mind, you should have no objections to associating a soul with a mind on a platform other than biological.

Originally posted by: Azulsky
We must also consider creative capacity diminishing, as stimuli are cutoff and experience terminated.

I'm not sure why this matters, nobody suggests to cut off stimuli.

Originally posted by: Azulsky
Also to note that if the circuitry of the brain could be emulated by a computer, it still wouldn't accomplish anything. A humans capacity for logic is based on experience, aka memory, and bases decisions on past experiences.

Flawed logic, I can't even imagine how you came up with that considering that most people know by now a computer is capable of storing memory in any arbitrarily defined pattern.

Originally posted by: Azulsky
We dont truly understand the brains capacity for information storage,

Actually, we do. Each brain cell is capable of variating their action potential, which causes memory formation. You simply identify the information storage necessary to represent the state of each cell, and multiply that by the number of cells. I believe the calculated value was just over a Petabyte if I remember correctly. However that does not necessarily indicate how well the brain makes use of it's alloted storage, we do know it adjusts what is remember based on what it judges valuable information in the scenario to be. Which brings me to the point of savants that you mention. It is known that their condition is caused by the brain's decreased ability to distinguish significant information.

 

[firefly2442]

There are ~100 billion neurons in the average brain.

That's a lot of storage. And we don't even use all of it.

 

[Gannon]

This thread proves that no one has enough understanding of the brain to say much about it, because most scientific advances do not come from people, they come from tools by which to better observe in more detail what the structure looks like, what is there, what it is made of and how it functions.

 

[kmalo]

If people could transfer brain functions/thought into a computer powered system, I don't think they'd ever come back.

 

[blackllotus]

Originally posted by: Gannon
If neurons are responsible for learning in multi-celled animals it is hard to explain how a one-celled animal with NO neurons can learn. The theory is that the cytoskeleton is the nervous system of the paramecium and the cytoskeleton is a miniature computer. The gist of the article quoted above is to explore ways of doing computing using microtubules without considering quantum mechanical effects. The authors of that article also estimate that a paramecium (or a neuron or some other cell) could move around bits at the rate of 10^13 bits per second. Thus every cell with microtubules (this is almost all cells) may contain a computer. A human brain counting only the use of 10^11 neurons and allowing for some redundancy would move around 10^23 bits per second according to this article. Another estimate I've seen is 10^28 bits per second. In either case this is rather a lot more than a digital computer can manage at this time and if this is what is going on then it will be a while before digital computers can compete with the brain. (There is an online article by Joel Henkel where he speculates how quantum effects could account for learning in paramecia, again more than I can follow but for the sake of people who may be interested I think I should list it.)

According to a recent book I read by Ray Kurzweil we will have supercomputers with the computational capacity of the brain around 2013 (10^16 instructions per second) and [$1000] computers around 2025. Even if 10^28 is correct, we should have [$1000] computers with that computational capacity around 2050.

 

[Azulsky]

Originally posted by: NanoStuff

Originally posted by: Azulsky
Empathy is something that couldnt be accounted for by binary, it defies logic, just as logic defies it.

Everything that can be accounted for can be accounted for in binary. Binary is the fundamental representation of truths and non truths, it's the core element to an entire turing model. As such, empathy can be accounted for in binary once the information processes responsible for it are known, even though not necessarily understood.
I suppose so, but the process of currently converting emotion into logic analysis beyond us. Could argue that point into the ground i suppose

Originally posted by: Azulsky
It isnt really known whether the brain is the seat of the "soul" or "spirit" or just a interface, because such cannot be measured.

Such cannot be measured because such does not exist. If you feel it's appropriate to associate a soul with a mind, you should have no objections to associating a soul with a mind on a platform other than biological.

I don't but I don't think its as simple as the just transplanting a brain into binary logic, the brain acts as part of a system, putting it standalone would affect in in a way we wouldn't know until we did it, unless we simply emulate stimuli. But that could be a corrupting factor as well. Too much speculation without information i suppose

Originally posted by: Azulsky
Also to note that if the circuitry of the brain could be emulated by a computer, it still wouldn't accomplish anything. A humans capacity for logic is based on experience, aka memory, and bases decisions on past experiences.

Flawed logic, I can't even imagine how you came up with that considering that most people know by now a computer is capable of storing memory in any arbitrarily defined pattern.

I don't see a flaw, the ability to distinguish between something like good and evil(first example i could think of) is something every competent mind can do, but people are taught how to respond. And yes that does result in flawed logic in people, hence why society is in its current situation. Its exactly how a computer is programmed, just that we skip the process that allows for the computer to define its own rules and define them ourselves

Originally posted by: Azulsky
We dont truly understand the brains capacity for information storage,

Actually, we do. Each brain cell is capable of variating their action potential, which causes memory formation. You simply identify the information storage necessary to represent the state of each cell, and multiply that by the number of cells. I believe the calculated value was just over a Petabyte if I remember correctly. However that does not necessarily indicate how well the brain makes use of it's alloted storage, we do know it adjusts what is remember based on what it judges valuable information in the scenario to be. Which brings me to the point of savants that you mention. It is known that their condition is caused by the brain's decreased ability to distinguish significant information.

You quoted that out of context, computers have perfect memory, they store precisely what they are told too and recall it exactly. Human memory degrades and warps over time. In the end we agree on that point

 

[NanoStuff]

Originally posted by: Azulsky
You quoted that out of context, computers have perfect memory, they store precisely what they are told too and recall it exactly. Human memory degrades and warps over time. In the end we agree on that point

I don't think we do. Again, a computer can 'compute' any bayesian system. A system can also evolve with genetic algorithms. Everything about what you propose of the human mind can be replicated on a computer. As a matter of fact, in many cases we do allow the computer to define it's own rules. Sometimes we allow it to develop it's own software, software that can be probabilistic, not deterministic as you so suggest it must be. Software we understand no better than that of the human brain.

In fact if you'd care to know, odds are that the automated software control system for the power grid to which your home is hooked up is exactly this, a sort of neural network no human could possibly write, but could only set the conditions for it's evolution. It also happens to be that the interest on your life savings depends on such software.