Is the brain in your head a computer or not?

[SAMU]

Simple question. What is your opinion?

[Doodler]

My personal feeling, its a form of evolved firmware.

[The Supreme Canuck]

It can compute. Therefore, it is a computer, in the strictest.

[Moose]

No. It's a brain. Not a computer.

AI research has been trying to understand how to mimic the human brain's efficiency for thirty years now. They've inevitably had more success when they've tried things other than how they think the brain might process a problem.

They're just not the same thing at all.

[MAPNUT]

I'd say a computer is a machine that can process large amounts of information and can do multiple tasks at the same time and can do repetitive tasks over and over again without fatigue. I don't know about you, but that sure ain't me . . .

[Moose]

I'd say a computer is a machine that can process large amounts of information and can do multiple tasks at the same time and can do repetitive tasks over and over again without fatigue. I don't know about you, but that sure ain't me . . .

The irony is that you came closer to describing a brain than you did a computer. :)

A computer is a machine that can process up to two pieces of information of arbitrary length (although inevitably small), and can do one task and only one task* at a time with that information.

A brain can process vast amounts of data and perform multiple/many independant tasks simultaneously. Your fatigue may vary.

(* Multiprocessor hardware can either do multiple tasks on up to two pieces of data (a multi-processored PC, for example), or one task on multiple/many pieces of data (a supercomputer like the Cray) or multiple tasks on multiple pieces of data (a cluster, although this is stretching the definition of a computer somewhat.)


Edit: I know I'm going to get whapped over the head for the gaps in my definitions. There are computers engineered differently than what we commonly understand as a modern computer. For example, a machine that might process three pieces of data simultaneously. I can't imagine why one would create such a machine, what context where such a machine would be better than equvalent two-operand solutions, but such a machine could be created (and probably has... somewhere... to try out the concept.)

[Peter Wilson]

Although it can clumsily compute, the brain is not a computing device; it is a learning apparatus.

[Van Rijn]

Simple question. What is your opinion?

Actually, it is a very tricky question. What is your definition of "computer"?

The structure and operation of the brain is quite unlike conventional computers, so in that sense, it is not a computer. On the other hand, the brain does process information and change state based on that information. In that sense, it is a computer. Are you asking if the brain operates through physical processes, or some "supernatural" means?

[Cugel]

As Hannibal Lector showed us, you can eat a human brain.
You cannot eat a computer.
Surely, that must draw a line between the two?

[Van Rijn]

The irony is that you came closer to describing a brain than you did a computer. :)

A computer is a machine that can process up to two pieces of information of arbitrary length (although inevitably small), and can do one task and only one task* at a time with that information.

A brain can process vast amounts of data and perform multiple/many independant tasks simultaneously. Your fatigue may vary.

(* Multiprocessor hardware can either do multiple tasks on up to two pieces of data (a multi-processored PC, for example), or one task on multiple/many pieces of data (a supercomputer like the Cray) or multiple tasks on multiple pieces of data (a cluster, although this is stretching the definition of a computer somewhat.)


Edit: I know I'm going to get whapped over the head for the gaps in my definitions. There are computers engineered differently than what we commonly understand as a modern computer. For example, a machine that might process three pieces of data simultaneously. I can't imagine why one would create such a machine, what context where such a machine would be better than equvalent two-operand solutions, but such a machine could be created (and probably has... somewhere... to try out the concept.)

Yup. Even single processors today have a limited ability to do more than one task at a time, and in my view a multi-processor array that is (say) processing visual data can rightly be called a computer. It is performing many different tasks and operating on many different pieces of data, but it's still a digital computer. There are analog computers and other interesting computing devices that don't fit into the standard categories that are sometimes better suited to some tasks (object identification, fast sorting, and so forth).

Ultimately, it comes down to your definition of "computer."

[Lurker]

Actually, it is a very tricky question. What is your definition of "computer"?

The structure and operation of the brain is quite unlike conventional computers, so in that sense, it is not a computer. On the other hand, the brain does process information and change state based on that information. In that sense, it is a computer. Are you asking if the brain operates through physical processes, or some "supernatural" means?
I agree... this is a tricky question because the brain is by no means digital in nature. From what I understand parts of the brain make up an inference engine based on chaotic theory and strange attractors while other parts are much more hardwired and might be considered wet firmware... although that is a terrible an analogy.

[Moose]

Yup. Even single processors today have a limited ability to do more than one task at a time [...]

Pipelining? I'd thought of that, but I decided that pipelining was still doing only one task at a time, with some overlap between tasks.

[Moose]

Is it too late to change my answer? I understand how a computer works. I don't understand how a brain works. Because of this, they are clearly different. ;)

[Van Rijn]

Pipelining? I'd thought of that, but I decided that pipelining was still doing only one task at a time, with some overlap between tasks.

Somewhat. I was mostly thinking of superscaler (http://www.answers.com/topic/superscalar) design and speculative execution (http://www.answers.com/topic/speculative-execution). In a modern high end CPU, there are often a number of components operating in parallel. Effectively, they can often perform more than one task at the same time.

[LurchGS]

see, I think they are both computers - to me, a computer is a thing that processes information. Whether that's 01+01=10 or "Let's not jump off that bridge again", they both take input and produce a result.

Fatigue isn't a differentiator - both devices eventually go toes up
Parallelism might be - the brain is massively parallel, last I looked - but human-designed computers are getting that way. to a degree.

so, to me, it's a difference in structure: A Maebach is not a Hyundai, but they are both cars.

[zebo-the-fat]

Originaly, the word computer described a human performing computational tasks.

[SAMU]

I posted this question to gauge this board on its average understanding of the fundamental nature of computers and brains. The question is what is your opinion not what is some abstract definition.

Looks like this board is coming up about 50/50.

[Maksutov]

Simple question. What is your opinion?
01000100010011110100010101010011001000000100111001 00111101010100001000000100001101001111010011010101 00000101010101010100010001010010110101010000010011 00010001010100000101010011010001010010000001010010 01000101010001010100111001010100010001010101001000 1000000101000101010101010001010101001001011001

0100001001010010010000010100100101001110

[Gillianren]

I posted this question to gauge this board on its average understanding of the fundamental nature of computers and brains. The question is what is your opinion not what is some abstract definition.

But opinions are based on the definition of terms, at least in this instance. As was mentioned, "computer" did originally mean a person who computes. Definitions change, and without defining your terms at the outset, an answer is meaningless.

[Moose]

Which might explain why the tally is running pretty close to 50-50. Without defining the question any more meaningfully, it's pretty much a coin-toss.

[Van Rijn]

Which might explain why the tally is running pretty close to 50-50. Without defining the question any more meaningfully, it's pretty much a coin-toss.

You beat me to it, I was going to say the same thing. I could answer either way, depending on the details of the definition for reasons previously stated. So, no answer.

[Lurker]

I don't mind calling the mind a computer... just don't get it confused with the thing on your desk... they have nothing in common except function.

[Lord Jubjub]

The computer on your desk (or lap) has nothing much in common with the neurons inside your skull.

[Lurker]

The computer on your desk (or lap) has nothing much in common with the neurons inside your skull.
Hey... this is a family site!! I don't want to be hearing about the computer in your lap!! :naughty:


:)

[yuzuha]

A computer is that which computes, be it analog, digital or neural network.

[Ronald Brak]

Is the brain a computer? Yeah sure, just worserer, as I have so amply demonstrated.

[sarongsong]

I don't mind calling the mind a computer...Uh-uh; one term at a time.

[SAMU]

I have my opinion based on my understanding/definition and you have yours.

What is the likelyhood that I or anyone can comunicate with this board on this topic with a common base of understanding and without debating opinion and definition? 50/50.

[Maksutov]

I have my opinion based on my understanding/definition and you have yours.

What is the likelyhood that I or anyone can comunicate with this board on this topic with a common base of understanding and without debating opinion and definition? 50/50.Well, you could always take the middle ground.

http://img137.imageshack.us/img137/566/iconwink6tn.gif

[Ronald Brak]

How can the brain not be a computer, unless you are a silicon chauvanist? Unless we go back to ancient ideas about the brain being a source of nasal excretions or a device to cool the blood, I don't see how it can be anything else.

[Argos]

The functioning of the brain departs so radically from that of a computer that both can´t be said to be equivalent. An artificial brain is still beyond the horizon.

[Gillianren]

I have my opinion based on my understanding/definition and you have yours.

What is the likelyhood that I or anyone can comunicate with this board on this topic with a common base of understanding and without debating opinion and definition? 50/50.

The issue, here, is that we don't have a common base of understanding without some terms defined. There are other subjects where we can discuss things without having to define everything first, because they're clearly defined already; this isn't one of them.

[Manchurian Taikonaut]

In some ways it is very like a computer core, but it is also very different. The biological brain is self commanding while connections are made between the computer's chips only when the programmer directs the connections to occur, it is much easier to understand a computer chip but to understand all the reactions in the brain and map the neural and chemical processes in the brain completely has been far more difficult to achieve, you might think computer's a very clever but computer technology has a long way to go before it will match the human brain's processing power.

[Lurker]

Uh-uh; one term at a time.
I have no issue with calling the mind a computer... ;)

[Lurker]

The functioning of the brain departs so radically from that of a computer that both can´t be said to be equivalent. An artificial brain is still beyond the horizon.
The functioning (physical architecture) of the brain in indeed radically different from a digital computer. This is not to say that there there are not functions (abstract functions) that both share in common. Both can perform basic arithmetic... both can perform image recognition...

The digital computer can perform calculation faster... the brain is much more efficient at image recognition and what might be poorly defined as intelligence.

Computer has not been defined in this context... I therefore chose abstract function... in this way there are many similarities between the two...

[Peter Wilson]

The brain is like 99% computer; it can be looked at as a huge if-then processor. Yet there is "something else" going on in my brain that's not happening in the computer on my desk.

[Lurker]

The brain is like 99% computer; it can be looked at as a huge if-then processor. Yet there is "something else" going on in my brain that's not happening in the computer on my desk.
There a lot of things going on in your brain that are not going on inside a computer. There are a lot of things going on inside the space shuttle that are not going on inside your car, but they are both transportation devices...

the function may have some commonality even though the form may be very different...

The space shuttle and the automobile are by no means equivalent... your car is very unlikely to reach orbit even with high octane gasoline... but they do both provide transportation. The human brain and a digital computer are extremely different and in many ways they do very different things. They do, however, share the capacity to compute... so it may be said that they are both computers...

[Moose]

Quick, Lurker, as quickly as you can, tell me the answer to 2+2, then tell me the answer to 5497 + 24037. No cheating by using the computer or paper. Do it in your head.

Come on, come on. Quickly now.

Okay. Got it?

How did you figure the answer to 2+2? By drawing on your memory of rote memorization.

You probably had to rely on a process to figure out 5497 + 24037, with each step based on that original rote memorization you did in elementary school.

And that's simple addition, one of the three simplest tasks you can build into an ALU. Both require only one clock cycle (one step) on a 2-byte ALU (pretty much been the standard for decades, although we have 4 and 8 byte archetectures now in common use. My "one cycle" claim is true for anything 16 bits/two bytes and longer. )

And the computer actually computed 2+2, all at once. It also computed 5497 + 24037 all at once. It didn't need to divide the problem, it didn't need to remember arithmatic, all it needed to know is how to add one and how to carry one, and that's all in the wiring.

Computers and human brains don't even perform elementary school arithmatic using anything even remotely resembling the same strategy. No, they're simply not directly comparable.

Lurker, your shuttle and car are not the same thing. You say they both provide transportation, but that's true only in the very vaguest definition of transportation. They really have nothing else in common. It's why we have two names for them: car, and shuttle.

Yes, computation is one task a brain is capable of performing, with effort, but it's not a skill the brain is especially adapted to. We're very good at processing visual data. We're very good at judgement in unexpected situations. Computers cannot do either of these things, except in limited ways by following strict pre-programmed instructions that may only number in the low dozens (in the case of RISC chipsets.)

Anyway, it's a lot of talk to say that brains and computers are different things, but for which there is some very small, very vague overlap in functionality.

[SAMU]

So lets adopt a common understanding for discussion of this topic.

The synapse of the neuron is equivalent to the gate or transister in a computer. Any objection to this and why?

The synapse and gate both produce an electronic signal that is transmitted to other synapses in the brain or gates on the RAM chip. Any objection to this and why?

This is called communication. A brain can communicate with another part of the brain through some means. A computer can comunicate with other parts of itself through a known language.

The computer, if used as a model of the comunication within the brain, shows that the brain uses a language to communicate with itself.

A computer communicates with other computers. The most familiar language it uses is the internet protocol, Hyper Text Markup Language, Super Hyper Text Markup Language Java and many others. Brains communicate with other brains verbaly, visually and by many other means such as all the media including the internet, politics, diplomacy and war.

The desktop computer obviously doesn't have the power and complexity of the human brain but neither does a cat's brain but a cat's brain is undeniably a brain.

A difference in quantity is not a difference in quality.

[Moose]

So lets adopt a common understanding for discussion of this topic.

The synapse of the neuron is equivalent to the gate or transister in a computer. Any objection to this and why?

Objection.

A synapse is simply a gap between one neuron and a receptor on another. Neurotransmitters (chemicals) are excreted from the neuron (a living cell) and reacted to by another. Brains can grow new neurons and form new connections. Neurons must be stimulated in order to "fire" usefully, but keep functionning when not actively stimulated.

A logic gate is a group of transistors that direct electricity in useful, but strictly predetermined ways. A computer cannot grow new transistors or arrange itself into new logic gates. Logic gates aren't stimulated, and if there's no current, the logic gate is non-functional.

The synapse and gate both produce an electronic signal that is transmitted to other synapses in the brain or gates on the RAM chip. Any objection to this and why?

Objection.

The synapse produces nothing. It's merely a gap. The neuron passes a chemical signal to connected neurons via a synapse when stimulated.

A logic gate directs electricity according to the placement of transistors and the input of its current.

This is called communication. A brain can communicate with another part of the brain through some means. A computer can comunicate with other parts of itself through a known language.

Objection. This paragraph contains both factual errors and a serious amount of hand-waving. There's no useful meaning to this paragraph.

The computer, if used as a model of the comunication within the brain, shows that the brain uses a language to communicate with itself.

Objection. Not so much. You seem to have a great many misconceptions about the basics of how a computer functions below the component layer. And the brain for that matter.

A computer communicates with other computers. The most familiar language it uses is the internet protocol, Hyper Text Markup Language, Super Hyper Text Markup Language Java and many others. Brains communicate with other brains verbaly, visually and by many other means such as all the media including the internet, politics, diplomacy and war.

Objection. We're getting into overvague abstraction again.

The desktop computer obviously doesn't have the power and complexity of the human brain but neither does a cat's brain but a cat's brain is undeniably a brain.

Objection. Non-sequitor.

A difference in quantity is not a difference in quality.

Objection. Non-sequitor.

SAMU, have you ever taken a course in either computer archetecture or biology at the college level?

[Lurker]

So lets adopt a common understanding for discussion of this topic.

The synapse of the neuron is equivalent to the gate or transister in a computer. Any objection to this and why?

Nope... this is no good. The neuron and the transister do not server anything like the same function.


The synapse and gate both produce an electronic signal that is transmitted to other synapses in the brain or gates on the RAM chip. Any objection to this and why?

yup... you are comparing a digital system to a chaotic system that is governed by completely different principles... no good...


This is called communication. A brain can communicate with another part of the brain through some means. A computer can comunicate with other parts of itself through a known language.

There are no data buses, bridges, etc. in the human brain. It is not digital in nature...


The computer, if used as a model of the comunication within the brain, shows that the brain uses a language to communicate with itself.

I would like to see some references on this...


A computer communicates with other computers. The most familiar language it uses is the internet protocol, Hyper Text Markup Language, Super Hyper Text Markup Language Java and many others.

nope... people use these languages to communicate with each other not computers...


Brains communicate with other brains verbaly, visually and by many other means such as all the media including the internet, politics, diplomacy and war.

This is not even close to the same sort of communication...



The desktop computer obviously doesn't have the power and complexity of the human brain but neither does a cat's brain but a cat's brain is undeniably a brain.

Both the brains in this example are built around organic tissues... the desktop computer is a digital device. No connection...


A difference in quantity is not a difference in quality.
The differences are fundamental...

[Lurker]

Quick, Lurker, as quickly as you can, tell me the answer to 2+2, then tell me the answer to 5497 + 24037. No cheating by using the computer or paper. Do it in your head.

Come on, come on. Quickly now.

Okay. Got it?

How did you figure the answer to 2+2? By drawing on your memory of rote memorization.

You probably had to rely on a process to figure out 5497 + 24037, with each step based on that original rote memorization you did in elementary school.

And that's simple addition, one of the three simplest tasks you can build into an ALU. Both require only one clock cycle (one step) on a 2-byte ALU (pretty much been the standard for decades, although we have 4 and 8 byte archetectures now in common use. My "one cycle" claim is true for anything 16 bits/two bytes and longer. )

And the computer actually computed 2+2, all at once. It also computed 5497 + 24037 all at once. It didn't need to divide the problem, it didn't need to remember arithmatic, all it needed to know is how to add one and how to carry one, and that's all in the wiring.

Computers and human brains don't even perform elementary school arithmatic using anything even remotely resembling the same strategy. No, they're simply not directly comparable.

Lurker, your shuttle and car are not the same thing. You say they both provide transportation, but that's true only in the very vaguest definition of transportation. They really have nothing else in common. It's why we have two names for them: car, and shuttle.

Yes, computation is one task a brain is capable of performing, with effort, but it's not a skill the brain is especially adapted to. We're very good at processing visual data. We're very good at judgement in unexpected situations. Computers cannot do either of these things, except in limited ways by following strict pre-programmed instructions that may only number in the low dozens (in the case of RISC chipsets.)

Anyway, it's a lot of talk to say that brains and computers are different things, but for which there is some very small, very vague overlap in functionality.
I think it is far from this cut and dried... but I am not really that passionate either way...

[sarongsong]

I have no issue with calling the mind a computer... ;)But are you calling brain mind?

[SAMU]

(paraphrased) Objection to the comparison of gate to synapse

"The on or off state of a logic gate corresponds to the binary values." http://www.webopedia.com/TERM/L/logic_gate.html



"A synapse is a neural junction used for communication between neurons."
http://en.wikipedia.org/wiki/Synapse_%28disambiguation%29

"synapses allow the neurons of the central nervous system to form interconnected neural circuits"
http://en.wikipedia.org/wiki/Synapse
In effect on off logic gates.



"This is called communication. A brain can communicate with another part of the brain through some means. A computer can comunicate with other parts of itself through a known language."


Objection. This paragraph contains both factual errors and a serious amount of hand-waving. There's no useful meaning to this paragraph.


"A synapse is a neural junction used for communication between neurons."
http://en.wikipedia.org/wiki/Synapse_%28disambiguation%29




"The computer, if used as a model of the comunication within the brain, shows that the brain uses a language to communicate with itself."


Objection. Not so much. You seem to have a great many misconceptions about the basics of how a computer functions below the component layer. And the brain for that matter.

Are you sure it is I who has the misconception? Maybe you should digest the links I posted and then maybe we would have a comon understanding. But...



"A computer communicates with other computers. The most familiar language it uses is the internet protocol, Hyper Text Markup Language, Super Hyper Text Markup Language Java and many others. Brains communicate with other brains verbaly, visually and by many other means such as all the media including the internet, politics, diplomacy and war."

Objection. We're getting into overvague abstraction again.


that one there sounds like a deal breaker to me.




The rest would require a comon understanding to even discuss

[mugaliens]

The term "computer" applied to people long before the physical machine ever existed. These computers dove through trigonometry tables, reconciling ballistics data throughout much of the 20th century.

My HS Trig teacher was one of those hired by the DoD during WWII to do just that.

So of course, anything that "computes" anything is a "computer."

[Moose]

Are you sure it is I who has the misconception?

If I wasn't before, I am now. Sorry SAMU.

Look, this is the thrust of your argument: "An orange and a bowling ball are the same thing because they both are round and can be used to knock bowling pins over." And that would be an absolutely silly argument to make, right? While they are indeed both round, and indeed both can knock bowling pins over, it simply isn't what oranges do except in the most artificial of situations.

Computers are 'bowling balls', designed to 'knock down bowling pins' (manipulate numbers). Oranges (brains) can 'knock down pins too', but not as well, and it's not really what they were meant to do.

Does anybody here think that mathematics are a natural function of the brain? Before you answer yes, how many animal species do you know to be able to do even basic addition?

Even Clever Hans couldn't add. But he could process visual data and make judgements from subtle behavioral cues dropped by the creatures around him, and realize that doing so benefits him. Just like all critters including homo sapians. But Clever Hans was especially good at it. This, not math, is what a brain evolved to be able to do.

Brains aren't computers. Shuttles aren't cars. Bowling balls aren't oranges.

[SAMU]

If I wasn't before, I am now. Sorry SAMU.

Did you even glance at the links I posted or am I just wasting my time with you?



Does anybody here think that mathematics are a natural function of the brain? Before you answer yes, how many animal species do you know to be able to do even basic addition?

Well since we can't seem to come to common terms of communication for such basic functions as performed by switches and relays (they perform the same function regardles of whether they are made of hardware or wetware) it is unlikely that you would understand that brains perform millions of calculations per second. While it does take a longer way around to perform arithmatic calculations in human language (the computer has the language part predigested) it does perform mathematical calculations of a much higher quality and a much higher quantity than today's computers.



Even Clever Hans couldn't add. But he could process visual data and make judgements from subtle behavioral cues dropped by the creatures around him, and realize that doing so benefits him. Just like all critters including homo sapians. But Clever Hans was especially good at it. This, not math, is what a brain evolved to be able to do.


So you would agree that he is doing functions of calculating subject matter merely different than arithmatic but not functions different in kind as a computer has a different kind of function than a car?



Brains aren't computers.

I dissagree and further say that enough of their functions are similar enough to adopt a common ground of communication.



Shuttles aren't cars. .
But enough of their functions are similar enough to adopt a common ground of communication.



Bowling balls aren't oranges.

But their shapes are similar enough to adopt a common ground of communication, at least with regard with their shape. I know I can train a dog to chose to pick up a ball or a stuffed cat at my command and then communicate that I want it to pick up an orange when the ball is not present and he would understand their similarities well enought to pick up the orange almost immediately even if he had never picked up or even seen an orange before.



We should be beginning to gain some common ground for understanding, Unless you really do understand all this already, are just debating for the sake of debate and don't have or want to contribute anything meaningful to this discussion.
Your last response was nonspecific enough to be almost unworthy of a reply.
What's your game pal?

[Moose]

Did you even glance at the links I posted or am I just wasting my time with you?

Uh, yeah, I did. You badly misinterpreted just about everything you pointed at.

Your last response was nonspecific enough to be almost unworthy of a reply.

Meh? My last response was as specific as it gets. You're simply reaching and twisting other peoples' published definitions to try and make them fit your preconceptions. They simply don't fit, logic gates aren't synapses, and the Emperor is feeling unusually drafty today.

What's your game pal?

Look, whatever SAMU, I've said my peace. You want to pretend to be the Bionic Man or Locutus of Borg or whatever it is you're trying to do, you go right ahead and imagine whatever you like.

Life's too short for me to waste this much time on something that doesn't even interest me all that much, especially when put next to really cool things like this morning's shuttle launch. Think I'm going to go take a walk to the waterfall with my camera before supper.

Peace out.

[Robert Andersson]

SAMU, I'm no neuroscientist, but a logical gate is more akin to a neuron. The synapse could perhaps be likened to copper wire, or whatever allows current to flow from a gate to another.

[Lurker]

SAMU, I'm no neuroscientist, but a logical gate is more akin to a neuron. The synapse could perhaps be likened to copper wire, or whatever allows current to flow from a gate to another.
Nope... the synapse does not work as a binary device. It has the ability to assume an infinite number of states depending on the both the amount of neurotransmitter that is released, and the efficiency of the reuptake system that reabsorbs the neurotransmitter.

I do not have an issue with calling the brain a computer, but it is not a digital device and trying to map the architecture of a digital system to the organic system of the brain is an exercise in distortion. The brain is not a digital device...

for sarongsong :)

I do not have an issue with calling the brain a computer... :p

I am sure that I have not have heard the last yet... :)

[sarongsong]

( http://www.bautforum.com/images/icons/icon7.gif "Now hold on, Mr. President!")
You seemed to use mind and brain interchangeably back there, that's all.

[yuzuha]

This is kind of daft. Everyone seems to have their preconcieved ideas of what a computer is and some of you seem to be talking right past each other.

Oranges are not bowling balls, but they are both spheres. "Computer" is a generic term, like sphere, that was originally applied to people who sat around calculating tables of figures used to accomplish some goal, and later grew to encompass devices (hardware or wetware) to produce similar results.

Not all computers are digital and not all of them even have logic gates. Some are total analogue devices that use operational amplifiers with logarithmic transfer functions (our sense of hearing is a logarithmic function) as well as integral and differential functions. Not all digital computers add in binary (you had to load add and multiply tables into the IBM 1620 and it did arithametic by table lookup). Your credit rating is often determined by a computer that is a neural network which operates on fuzzy logic much like your brain. And, synapses are not a chemical "wire"... one neuron releases transmitters, whether the next neuron fires and how strongly it does so is determined by how many of those neurotransmitters arrive at its receptor sites, and that is also a function of the concentration of inhibitors and transmitters which is a function of adjacent neurons in the network, which gives it a bit of built-in fuzzy logic. The brain also needs training and gets stuck in false minima just like a silicon neural network.

Human language is not stored digitally in the brain but you can also make the analogy that it is like programming since language can be treated both as data or as instructions, in much the same way a stored program digital computer can treat blocks of machine code.

You can make all kinds of good analogies between the brain and specific bits of hardware that may be "spot on" in some instances while being totally "off" in regards to other details. The original question did not ask if the brain was a "digital" computer. So, I'll have to answer "yes." If you want to get more granular, then I'll have to change that to a "maybe."

[Ronald Brak]

To me the question is similar to asking, "Do you use your brain for processing infomation, or do you just carry three pounds of pink tissue around in your skull for fun?"

[Maksutov]

To me the question is similar to asking, "Do you use your brain for processing infomation, or do you just carry three pounds of pink tissue around in your skull for fun?"Fun is the one thing that money can't buy.She's Leaving Home Sgt. Pepper's Lonely Hearts Club Band The Beatles.

[Lurker]

( http://www.bautforum.com/images/icons/icon7.gif "Now hold on, Mr. President!")
You seemed to use mind and brain interchangeably back there, that's all.
Hey... I am trying to convince the people I work for that I know something about kernel software... posting here is just a side sport. You're right... I was far too loose with my terminology, and you caught me flat footed... :)

However, you are talking to a Buddhist mystic... I do not consider the brain and the mind to be one and the same... even when I thoughtless let my fingers post such things... :o

and yes... sometimes my fingers do that sort of thing behind my back...
[well that's my story anyway and I'm stickin' with it...} :razz:

[Fortis]

If you are talking about a Universal Turing Machine, then there is a rather striking image that I recall from an old copy of Scientific American. There was an article which described how you could build a "tinker toy" UTM (one of those contraptions made from connected wooden rods, etc.) This then posed the question that if a sufficiently complex UTM could be described as being conscious, then we are left with the slightly uncomfortable notion that a "tinker toy" structure built from wooden rods could also be described as being conscious (though it may be slower than a solid state implementation).

[ggremlin]

The human brain is not a single computer or even a super scalar parallel processing platform. A better analogy would be the current model of the internet.

[Van Rijn]

This is kind of daft. Everyone seems to have their preconcieved ideas of what a computer is and some of you seem to be talking right past each other.
[SNIP]
You can make all kinds of good analogies between the brain and specific bits of hardware that may be "spot on" in some instances while being totally "off" in regards to other details. The original question did not ask if the brain was a "digital" computer. So, I'll have to answer "yes." If you want to get more granular, then I'll have to change that to a "maybe."

The reason I wanted to know SAMU's definition is that some people flatly don't think the brain is a computer of any sort. They think it works by non-physical means. So in that case, the question could mean: Do you think the brain operates by physical principles or is it by some kind of supernatural "magic"? And my answer is that it operates by physical means, and it is a computer of a sort.

But the question could also mean, do you think the brain operates like a conventional digital computer? In that case, my answer is, no, it is quite different from conventional computers.

I'm not going to get into an argument on details of physical design, but I could see different ways the question could be interpreted.

[Robert Andersson]

Nope... the synapse does not work as a binary device. It has the ability to assume an infinite number of states depending on the both the amount of neurotransmitter that is released, and the efficiency of the reuptake system that reabsorbs the neurotransmitter.
My point was not that they were identical. However, I just pointed out that the synapse was more like copper wire, rather than a gate, which SAMU suggested. (A wire is typically not in itself "binary".)

A brain is an adaptive neural network, while a (typical) computer just executes predetermined instructions. Our brain can emulate a computer, although *very* inefficiently, and a computer could potentially emulate a brain. To say that our brain *is* a computer is a tough call, although there are similarities, depening on how fuzzy one is allowed to be.

[Lurker]

My point was not that they were identical. However, I just pointed out that the synapse was more like copper wire, rather than a gate, which SAMU suggested. (A wire is typically not in itself "binary".)

A brain is an adaptive neural network, while a (typical) computer just executes predetermined instructions. Our brain can emulate a computer, although *very* inefficiently, and a computer could potentially emulate a brain. To say that our brain *is* a computer is a tough call, although there are similarities, depening on how fuzzy one is allowed to be.
Bold mine... nope... the computer is a digital device and the brain is definitely not... digital electronics will never emulate the human brain...

[Robert Andersson]

Bold mine... nope... the computer is a digital device and the brain is definitely not... digital electronics will never emulate the human brain...
I hope we're misunderstanding eachother here. I did not mean the computer as-is, but software running on it. Assuming there is nothing "unnatural" about the brain and it obeys the laws of physics as we know them (now or in the future), I cannot see why we couldn't model a brain in software with arbitrary accuracy, when we understand it enough and have a *really* powerful computer.

What I said was in no way an argument for brain==computer, but rather the other way around.

[Lurker]

I hope we're misunderstanding eachother here. I did not mean the computer as-is, but software running on it. Assuming there is nothing "unnatural" about the brain and it obeys the laws of physics as we know them (now or in the future), I cannot see why we couldn't model a brain in software with arbitrary accuracy, when we understand it enough and have a *really* powerful computer.

What I said was in no way an argument for brain==computer, but rather the other way around.
I suggest you read:

The Emperor's New Mind
by Roger Penrose
Oxford University Press, 1990

The issue is far less straight forward than you might think...

There has never been an artificial intelligence created on a digital system, or any system to date, and there is good reason to believe that a digital system could never host such an intelligence. Digital systems deal poorly with chaotic systems... weather forecasting is great example.

The digital computer is a powerful system, but whether it can truly replicate all natural systems is very doubtful.

[Van Rijn]

I suggest you read:

The Emperor's New Mind
by Roger Penrose
Oxford University Press, 1990

The issue is far less straight forward than you might think...


Penrose is effectively arguing for a "supernatural" basis of the mind. Until there is some real evidence for this nonalgorithmic quantum consciousness, I'll consider it a fun idea, but that's it.


There has never been an artificial intelligence created on a digital system, or any system to date, and there is good reason to believe that a digital system could never host such an intelligence.


Just because we haven't managed it yet means nothing. Current computers have far less computational capacity and memory than even modest animal brains. Also, there is the issue of understanding the structure and design of a functioning brain in detail. An analogy is genetic engineering and genetic science. We have some ideas of where things will go, we can build genes if we want, but it will be some time before we understand genome function in detail.


Digital systems deal poorly with chaotic systems... weather forecasting is great example.

The digital computer is a powerful system, but whether it can truly replicate all natural systems is very doubtful.

Efficiency of emulation is an issue. Most likely you would need to do hardware optimizations for practicality. However, in principle, I've seen no reason to believe that a digital computer could not perform the functions of a biological brain.

[yuzuha]

Efficiency of emulation is an issue. Most likely you would need to do hardware optimizations for practicality. However, in principle, I've seen no reason to believe that a digital computer could not perform the functions of a biological brain.

I don't think it would work on a pure digital "off the shelf" system. I think you'd need something with a body and sensors so you could build in the hysteresis loops and feedback systems that biological critters use to interface with the world (without them, there is no sense of "self" and "other"), and you'd need hardware neural networks and associative memories. Sure, parts of an artificial brain could be digital, but some things, like visual pattern recognition are too difficult to do digitally, but neural networks can be trained to do such things. So, I think AI would require some specialized hardware.

[Donnie B.]

So, I think AI would require some specialized hardware.Positronic brains? :D

I voted yes (the brain is a computer). I base that on my acceptance of the physical vs. supernatural basis of mind. I do agree that the brain is not analogous to a digital computer -- a synapse is not a logic gate, and neither is a neuron (though the latter is closer).

Finally, and rather loosely: mind is to brain as software is to hardware.

[gwiz]

There has never been an artificial intelligence created on a digital system, or any system to date, and there is good reason to believe that a digital system could never host such an intelligence.

There's a guy over on Apollo-hoax who uses this for his sig:

If our brains were simple enough for us to understand them, we'd be so simple we couldn't

I think this has relevance to the poor progress in artificial intelligence. Modelling something much simpler than our own brains - a worm's, for instance - would seem more achievable.

[Van Rijn]

I don't think it would work on a pure digital "off the shelf" system. I think you'd need something with a body and sensors so you could build in the hysteresis loops and feedback systems that biological critters use to interface with the world (without them, there is no sense of "self" and "other"), and you'd need hardware neural networks and associative memories. Sure, parts of an artificial brain could be digital, but some things, like visual pattern recognition are too difficult to do digitally, but neural networks can be trained to do such things. So, I think AI would require some specialized hardware.

As a practical matter, you are almost certainly correct that it would require specialized hardware (not to mention a whole lot of information we don't have yet). That is why I mentioned the issue of efficiency of emulation. But in principle, I don't see any reason that a digital computer couldn't perform the functions. And, true, it probably isn't going to be useful for much without the proper interfaces.

[Robert Andersson]

I suggest you read:

The Emperor's New Mind
by Roger Penrose
Oxford University Press, 1990

I'll add it to my ever growing to-read list... :)
The issue is far less straight forward than you might think...

There has never been an artificial intelligence created on a digital system, or any system to date, and there is good reason to believe that a digital system could never host such an intelligence. Digital systems deal poorly with chaotic systems... weather forecasting is great example.

The digital computer is a powerful system, but whether it can truly replicate all natural systems is very doubtful.
Yet again, I think you misunderstand me. I do not really mean AI, so emulation might not be the best choice of word. All I meant was that if we understand the brain in terms of physical laws, we should in principle be able to model it, utilizing a computer for the calculations. Not saying this is something trivial or feasible or that we will be able to do it anytime soon or ever.

[Lurker]

I'll add it to my ever growing to-read list... :)

Yet again, I think you misunderstand me. I do not really mean AI, so emulation might not be the best choice of word. All I meant was that if we understand the brain in terms of physical laws, we should in principle be able to model it, utilizing a computer for the calculations. Not saying this is something trivial or feasible or that we will be able to do it anytime soon or ever.
Yup... just like the weather... I have my personal doubts about the ability of digital systems to model chaotic systems.

Anyway... we're a long way from understand the chemical processes that govern the brain's activity. I truly wish we were a lot closer... then I could have some nice designer neuro-transmitters or perhaps an adjustment in my reuptake system and be normal like other people...

[mugaliens]

Main Entry: com·put·er
Pronunciation: k&m-'pyü-t&r
Function: noun
Usage: often attributive
: one that computes; specifically : a programmable usually electronic device that can store, retrieve, and process data

Of course the brain is a computer. I can't believe this is still being debated.

There are many electronic computers that are much faster than the brain at some tasks (math), but the brain is much faster and more capable than electronic computers at other tasks (visual recognition). Furthermore, the brain is capable of tasks that other computers are not (self-programming...).

Interesting discussion, but the question as to whether or not it's a computer is no longer a question.

Is it a digital computer? No.

Is it an analog comptuer? Yes.

As for adding two numbers, I knew a guy in college who could add six digit numbers in his head as rapidly as you could fire them off to him.

[Robert Andersson]

Yup... just like the weather... I have my personal doubts about the ability of digital systems to model chaotic systems.
Just want to emphasise that I didn't suggest predicting a specific brain - lack of precise measurements and QM will quickly stop that - just simulating a brain.

[Lurker]

Just want to emphasise that I didn't suggest predicting a specific brain - lack of precise measurements and QM will quickly stop that - just simulating a brain.
We can't even simulate weather based on given data and models. When data is used in current software, the results diverge quickly from what actually occured. It's a much more simple to model weather than the brain.