how much?
QUOTE (DavidD+Oct 20 2008, 12:47 PM)
how much?
Infinity.
What is this "memory" of yours?! Would you clarify what it is supposed to be?
I am used to thinking that there is no physical end to memory and the reason is very simple: asking how much memory there is to be had is like asking how many numbers the digit "1" fits.
If memory had a "size" then there would be an end to the number of places where you can fit the "1". Therefore, both can not be true.
Similarly, how much forgetfulness can exist? As much forgetfulness as you can delete the digit "1" from all numbers.
(This is a philosophical consideration, of course. It doesn't address brain function and doesn't need to since my idea of "memory" is abstract. My line of reasoning: "memory" is immutable and stopped in time whether or not I have brainly access to it.
Infinity.
What is this "memory" of yours?! Would you clarify what it is supposed to be?
I am used to thinking that there is no physical end to memory and the reason is very simple: asking how much memory there is to be had is like asking how many numbers the digit "1" fits.
If memory had a "size" then there would be an end to the number of places where you can fit the "1". Therefore, both can not be true.
Similarly, how much forgetfulness can exist? As much forgetfulness as you can delete the digit "1" from all numbers.
(This is a philosophical consideration, of course. It doesn't address brain function and doesn't need to since my idea of "memory" is abstract. My line of reasoning: "memory" is immutable and stopped in time whether or not I have brainly access to it.
Been a while since I posted, but I enjoyed reading the post and reply.
I assume a conventional answer would estimate some number of possible configurations each neuron could be in and compute an approximate number of bits per neuron and potentially rule out some rather useless configurations to estimate the total possible storage (the results could vary depending upon the possible configurations a neuron could assume that would be interpreted as computational components versus storage components with a large possible variance depending upon what extent learned processes are considered to be memory versus algorithms computing or decoding interpretations of those memories. Memory, in terms of binary digits, means nothing without something to interprete what they mean within some context, which is similarly defined by a structure that we could interprete to represent various combinations of genetically inherited features or learned specifics).
Anyway, if we simply ignored all those details and just asked how much memory the entire brain possesses, then we could theoretically compute this if we knew how many possible distinct configurations the brain could be configured as and then we simply compute the logarithm base 2 to convert this to an equivalent number of bits.
But the question of how many possible configurations a human brain could be in is a rather interesting one because it requires defining which configurations would represent a human brain and which wouldn't. It may not be too difficult to get some approximation when we're simply referring to some unit of mass that contains 'brainlike' structures, but if we're referring to a specific brain, then there's a question regarding what alternate possible configurations would still represent the same brain.
For example, if someone was asking the question about their own brain, then the potential bits or binary digits of memory is defined by the number of binary alterations that could be made, but with the requirement that it still be the same brain ... I guess as long as the only altered "memories" were those that had no influence on anything in the present or future, then those altered locations could potentially be considered bits of memory, but would obviously be useless and unprovable in their storage capacity (trying to prove their quantity would violate this requirement that they not alter the present or future operation of the brain.
So if we look at memories that are influencial in thought processes and attempted to alter them, we'd have different thoughts and actions. Now would that still be the same brain of the same person? From an external perspective, small changes in such a manner could be considered unnoticeable, but to truly know for certain this is the case, you'd need to have the "correct" version to compare things with and which version would that be?
If it was your own brain, then it would seem that any version would be the "correct" one and there would effectively be no memory because there would be no possible range of alternate configurations that were still you. You couldn't store anything except a representation of yourself, which has no "free bits" left over to store anything, except as a potential source of 'to-be-configured' growth.
As a possible clearer explanation - if we have a single decimal digit, there are 10 possible values it can store - 0 to 9. A specific digit itself, such as 2, does not store anything (except with respect to the space in which it was located, but given only a single decimal location to write and read from, the possibility of reading 1 of 10 symbols gives us log2(10)~=3.3 bits of information).
Now if we had 1 of 10 possible people behind a door, and we assumed an equal probability for any of them to be there, we'd theoretically gain ~3.3 bits of information by knowing who it was, but if we already have a specific person, there's no more information gained by knowing for a second time, who it is.
Now consider that if someones own brain can store information, then that means there are configurable states that are unknown, until witnessed, sampled, or acted upon etc., and if, for example, my brain had 3 bits of storage, then there should exist 2^3=8 possible "me"s, and if my brain could store a million bits, well there would likely be more "me"s than the universe could contain ... just doesn't sound right. Most likely that storage is a physical unknown and not part of "me". There's only one of me, and that single possibility leaves no room for anything else to be stored within it, though I guess it's possible that the potential storage could be defined in terms of the number of possible binary alterations that could be made before I'd be defined as something else (in other words, if by the time you made a trillion binary alterations to yourself it was almost inevitable that you'd no longer be yourself, then you potentially represent a trillion binary digits, as that's the equivalent volume of an information space that the definition of yourself encompasses ... but it seems difficult to ever think that changes made to oneself would ever result in anything other than oneself, so the number could be infinite on one hand, yet because nothing else could be yourself, then on the other hand you have no available storage for anything else and there's no "free" memory available)
Don't you love how questions that seem rather simple at face value can become convoluted, paradoxical or unintuitive when you dig down further than usual?
I assume a conventional answer would estimate some number of possible configurations each neuron could be in and compute an approximate number of bits per neuron and potentially rule out some rather useless configurations to estimate the total possible storage (the results could vary depending upon the possible configurations a neuron could assume that would be interpreted as computational components versus storage components with a large possible variance depending upon what extent learned processes are considered to be memory versus algorithms computing or decoding interpretations of those memories. Memory, in terms of binary digits, means nothing without something to interprete what they mean within some context, which is similarly defined by a structure that we could interprete to represent various combinations of genetically inherited features or learned specifics).
Anyway, if we simply ignored all those details and just asked how much memory the entire brain possesses, then we could theoretically compute this if we knew how many possible distinct configurations the brain could be configured as and then we simply compute the logarithm base 2 to convert this to an equivalent number of bits.
But the question of how many possible configurations a human brain could be in is a rather interesting one because it requires defining which configurations would represent a human brain and which wouldn't. It may not be too difficult to get some approximation when we're simply referring to some unit of mass that contains 'brainlike' structures, but if we're referring to a specific brain, then there's a question regarding what alternate possible configurations would still represent the same brain.
For example, if someone was asking the question about their own brain, then the potential bits or binary digits of memory is defined by the number of binary alterations that could be made, but with the requirement that it still be the same brain ... I guess as long as the only altered "memories" were those that had no influence on anything in the present or future, then those altered locations could potentially be considered bits of memory, but would obviously be useless and unprovable in their storage capacity (trying to prove their quantity would violate this requirement that they not alter the present or future operation of the brain.
So if we look at memories that are influencial in thought processes and attempted to alter them, we'd have different thoughts and actions. Now would that still be the same brain of the same person? From an external perspective, small changes in such a manner could be considered unnoticeable, but to truly know for certain this is the case, you'd need to have the "correct" version to compare things with and which version would that be?
If it was your own brain, then it would seem that any version would be the "correct" one and there would effectively be no memory because there would be no possible range of alternate configurations that were still you. You couldn't store anything except a representation of yourself, which has no "free bits" left over to store anything, except as a potential source of 'to-be-configured' growth.
As a possible clearer explanation - if we have a single decimal digit, there are 10 possible values it can store - 0 to 9. A specific digit itself, such as 2, does not store anything (except with respect to the space in which it was located, but given only a single decimal location to write and read from, the possibility of reading 1 of 10 symbols gives us log2(10)~=3.3 bits of information).
Now if we had 1 of 10 possible people behind a door, and we assumed an equal probability for any of them to be there, we'd theoretically gain ~3.3 bits of information by knowing who it was, but if we already have a specific person, there's no more information gained by knowing for a second time, who it is.
Now consider that if someones own brain can store information, then that means there are configurable states that are unknown, until witnessed, sampled, or acted upon etc., and if, for example, my brain had 3 bits of storage, then there should exist 2^3=8 possible "me"s, and if my brain could store a million bits, well there would likely be more "me"s than the universe could contain ... just doesn't sound right. Most likely that storage is a physical unknown and not part of "me". There's only one of me, and that single possibility leaves no room for anything else to be stored within it, though I guess it's possible that the potential storage could be defined in terms of the number of possible binary alterations that could be made before I'd be defined as something else (in other words, if by the time you made a trillion binary alterations to yourself it was almost inevitable that you'd no longer be yourself, then you potentially represent a trillion binary digits, as that's the equivalent volume of an information space that the definition of yourself encompasses ... but it seems difficult to ever think that changes made to oneself would ever result in anything other than oneself, so the number could be infinite on one hand, yet because nothing else could be yourself, then on the other hand you have no available storage for anything else and there's no "free" memory available)
Don't you love how questions that seem rather simple at face value can become convoluted, paradoxical or unintuitive when you dig down further than usual?
neural fibers grow what are called spines that hold memory, you can't run out of memory with age but you can run low on nurons to process the information, unless you inject stem cell neurons taylored to your DNA into you bloodstreem. Then they will settle into your brain and be baby nurons that you have to toilet train, otherwise you will have verbal diahrea on the boardroom table lol.
16 bits in games I think is 5 bits for each colour and 1 bit for visibility through like through smoke or through colored glass and you can see that this visibility is in points some in old games. So there is 2^15=32768 colourse in 16 bits, so maybe there enough just 20 bits with 5 bits alpha instead 1 bit? But I think if somthing moving then much harder to see diference between colourse so colourse are better recognitional only looking long and kinda precisly...
Maybe even with daster connection to monitor there is possible to invent faster architecture of cpu - much faster... But still don't need to give very fast each frame, but maybe need long wiating and counting how much those frames go, but no, because processor cpu speed is 1 GHz and this for about 1 cm speed of light and going from cpmputer to monitor is about 1 m distance so there is like 10 MHz, but is only about 100 Hz, so 100000 times - so pixels, signal analog. So maybe counting how much wait to next frame because limited speed of induction power of energy string vector... So even those Hz imposible to trust, because who actualy calculating and counting them? There still traveling electricity like in one dimension and actualy computation is in one dimension, maybe more parallel (?), but then harder to configurate those computed data. Supercomputers kinda 3D rendering rendering some not moving object splitted or frame, but in frame is physics and who knows how must be physics need feed all this physics and lights data to over computer or not? Anyway more processors rendering kinda is maybe good in not real time processings like 3D games...
I have kinda physics becnchmark for 3D games kinda like in 3D mark, how to see how good 3D video card simulating physics. So need to put 2 or more spheres-balls into say qubic empty box and then move fast this box or met those balls and they kinda have some reflectibility and box don't moving and those balls bouncing from box walls and still some time will bounce to each over and this balls have same density in each they point of mass... So there seems only simple newton 3D physics and video card seems should be able this to do without any problems if it realy simulating physics not only for one object (and over don't moving)... To hard huh? Imbosible!?
Maybe even with daster connection to monitor there is possible to invent faster architecture of cpu - much faster... But still don't need to give very fast each frame, but maybe need long wiating and counting how much those frames go, but no, because processor cpu speed is 1 GHz and this for about 1 cm speed of light and going from cpmputer to monitor is about 1 m distance so there is like 10 MHz, but is only about 100 Hz, so 100000 times - so pixels, signal analog. So maybe counting how much wait to next frame because limited speed of induction power of energy string vector... So even those Hz imposible to trust, because who actualy calculating and counting them? There still traveling electricity like in one dimension and actualy computation is in one dimension, maybe more parallel (?), but then harder to configurate those computed data. Supercomputers kinda 3D rendering rendering some not moving object splitted or frame, but in frame is physics and who knows how must be physics need feed all this physics and lights data to over computer or not? Anyway more processors rendering kinda is maybe good in not real time processings like 3D games...
I have kinda physics becnchmark for 3D games kinda like in 3D mark, how to see how good 3D video card simulating physics. So need to put 2 or more spheres-balls into say qubic empty box and then move fast this box or met those balls and they kinda have some reflectibility and box don't moving and those balls bouncing from box walls and still some time will bounce to each over and this balls have same density in each they point of mass... So there seems only simple newton 3D physics and video card seems should be able this to do without any problems if it realy simulating physics not only for one object (and over don't moving)... To hard huh? Imbosible!?
I think memory is analog and if you forget some part then you can lear it again from nature or by reading, for example I don't know how to write cousnisciouse, but I know that is mean and if I would read this word I would be able to recognise it. Kinda in 3D analog memory is not wrong, because there many things to remebereing and making concept clear again connecting forgeted parts and in 1D (like computers thinking) loosing one bit without error correction braking all computing process... But in HDD no errors correction, because there already one bit is about 100nm*100nm or 1000*1000 nm^2 surface so errors correction even kinda 7 bits have 3 bits still prety complicated but maybe... There very hard to read this memory and to rotate with such big precision, except maybe if in hard drives memory is saved into 8 analog levels in one space 10000 nm^2 kinda 8 bits. Maybe more analog levels, but errors then seems uncorrected...
Even each animal can have same inteligence like human, but human due some reasons maybe have more brains mass and then become looking, hey we have more brains than any over animals, maybe we can do somthing more? And then he become to think that brain is very interesting and powerfull computing thing and become to think and monkeys maybe also have same inteligence, but they thinking "waterver" so hard and long to work, better to live in nature shorter... And elephants similar, but kinda gorils and chimps I guess in nature living on average (don't including babies up to 5 years) 50 years and elephants maybe about 60-80 years. So they thinking, go away with your stupid brains and mass proportion... And human thinking that he is smartest and that need to live so long and to work and so on and live so comfortable because he smartest one. And kinda all those monkeys thinking, how it is unrealistic that we will do like they... And humans also don't excepts this... There monkeys can talk etc, to computer, but maybe sciencists just giving to primitive tasks for them and they don't have big motives because still nobody will want to accept them because of they skin with antens and size diferent and to them would be hard to adapt... http://www.youtube.com/watch?v=JoKiTs67J4k&feature=related
Even each animal can have same inteligence like human, but human due some reasons maybe have more brains mass and then become looking, hey we have more brains than any over animals, maybe we can do somthing more? And then he become to think that brain is very interesting and powerfull computing thing and become to think and monkeys maybe also have same inteligence, but they thinking "waterver" so hard and long to work, better to live in nature shorter... And elephants similar, but kinda gorils and chimps I guess in nature living on average (don't including babies up to 5 years) 50 years and elephants maybe about 60-80 years. So they thinking, go away with your stupid brains and mass proportion... And human thinking that he is smartest and that need to live so long and to work and so on and live so comfortable because he smartest one. And kinda all those monkeys thinking, how it is unrealistic that we will do like they... And humans also don't excepts this... There monkeys can talk etc, to computer, but maybe sciencists just giving to primitive tasks for them and they don't have big motives because still nobody will want to accept them because of they skin with antens and size diferent and to them would be hard to adapt... http://www.youtube.com/watch?v=JoKiTs67J4k&feature=related
Hi All,
Humans are not digital computing devices. We are parallel analog signal processors. We are also "holographic" in that what we are is more than the simple sum of the parts. Parts of the brain are reflections of the whole and the whole of the brain can reflect on it's parts. A memory or an idea is not easily "localized"... it is most probably a distributed phenomenon some of it encoded even in our genes. The mental processes of analysis is in some way an "emergent" process.
One human lives a perfectly ordinary life and have graduated as a Mathematician from a Sheffield University without the substantial portion of th brain present.
Strange Case Of A Man With No Brain
This is no urban legend I have seen scans of this man's brain previously and he is in no way "subnormal" by not having a complete brain (IQ = 125). I am sure these scans are still on the web somewhere.
I think the spinal cord and the distributed network of signals throughout our body which communicates with all of our cells provide "awareness". This awareness is a "shared experience" and is not actually our brain "thinking" (it is a specialized "device" constructed by our cells acting in cooperation for their mutual needs and overall awareness... "your" awareness!) All the cells of our body are in cooperation "communicating together" to continue the existence of your body... the way we should be cooperating as a world and as a society. The internet is something that roughly translates to this process on planet earth "connecting" people together the way our nervous system connects our cells together. It is still evolving as a "machine" but we are not "evolving" as integrated parts of the whole. One day we may need to join an "extended internet" that span the stars. Will we be ready? Or are we moribund?
Cheers
Cheers
Humans are not digital computing devices. We are parallel analog signal processors. We are also "holographic" in that what we are is more than the simple sum of the parts. Parts of the brain are reflections of the whole and the whole of the brain can reflect on it's parts. A memory or an idea is not easily "localized"... it is most probably a distributed phenomenon some of it encoded even in our genes. The mental processes of analysis is in some way an "emergent" process.
One human lives a perfectly ordinary life and have graduated as a Mathematician from a Sheffield University without the substantial portion of th brain present.
Strange Case Of A Man With No Brain
This is no urban legend I have seen scans of this man's brain previously and he is in no way "subnormal" by not having a complete brain (IQ = 125). I am sure these scans are still on the web somewhere.
I think the spinal cord and the distributed network of signals throughout our body which communicates with all of our cells provide "awareness". This awareness is a "shared experience" and is not actually our brain "thinking" (it is a specialized "device" constructed by our cells acting in cooperation for their mutual needs and overall awareness... "your" awareness!) All the cells of our body are in cooperation "communicating together" to continue the existence of your body... the way we should be cooperating as a world and as a society. The internet is something that roughly translates to this process on planet earth "connecting" people together the way our nervous system connects our cells together. It is still evolving as a "machine" but we are not "evolving" as integrated parts of the whole. One day we may need to join an "extended internet" that span the stars. Will we be ready? Or are we moribund?
Cheers
Cheers
Why all nonsense going on in USA, like instant combustion, half brain human, brain division, narconeurons and so on?
I think one of treaded pacients for experiment leave messege for us, in brain with electricity to him with electrods find place, that this place make this human to made gund into ball with fingers - fist, so he saying "doctor I think your electricity is stronger than my will", I think he mean, that he don't have will and choise and is like experimental rat 
I'd vote Infinity (Though that's not quite correct from my limited understanding)
But IIRC we're more Analog than Digital.
But IIRC we're more Analog than Digital.
QUOTE (PGP_Protector+Oct 23 2008, 02:46 PM)
I'd vote Infinity (Though that's not quite correct from my limited understanding)
But IIRC we're more Analog than Digital.
The more I think about it, the less I am able to say what everyone already knows and doesn't say it.
Let me try to restate my puny case. We are digilog. (Or is it, uh, analal? Anal?
) 
There is and there can be no END to the information we can access even though we may after all have a finite, and in theory, *countable* (to a satisfactory approximation) number of internal states.
There isn't a limitation to the number of bits (as in physical states) we can access because information isn't a bit. It's an event, therefore it has no theoretical end. That is why you can say "me hungry" and Einstein could say "E=mc squared".
When I see that sort of information I have no desire to "understand" any of it. They are two statements of equality, two declarations of global or local (in)variable. Structurally that is what they are, anyway. How you use them is something else: they are different events. I will never lose a second's sleep over their "meaning".
Also, there is no difference to me if you say 2, 4, 8, 16, 32... theoretically, at least in my mind, there is no difference because there are no powers of numbers. What that is supposed to mean philosophically is that no matter where you throw a number at me, it will be something that I can **access** even though I don't particularly care to **understand** it because I am sticking close to my "internal structure"... whatever that is.
Within any system a least number of bits can random-access anything.
But IIRC we're more Analog than Digital.
The more I think about it, the less I am able to say what everyone already knows and doesn't say it.
Let me try to restate my puny case. We are digilog. (Or is it, uh, analal? Anal?
) There is and there can be no END to the information we can access even though we may after all have a finite, and in theory, *countable* (to a satisfactory approximation) number of internal states.
There isn't a limitation to the number of bits (as in physical states) we can access because information isn't a bit. It's an event, therefore it has no theoretical end. That is why you can say "me hungry" and Einstein could say "E=mc squared".
When I see that sort of information I have no desire to "understand" any of it. They are two statements of equality, two declarations of global or local (in)variable. Structurally that is what they are, anyway. How you use them is something else: they are different events. I will never lose a second's sleep over their "meaning".
Also, there is no difference to me if you say 2, 4, 8, 16, 32... theoretically, at least in my mind, there is no difference because there are no powers of numbers. What that is supposed to mean philosophically is that no matter where you throw a number at me, it will be something that I can **access** even though I don't particularly care to **understand** it because I am sticking close to my "internal structure"... whatever that is.
Within any system a least number of bits can random-access anything.
Human memory will have "N" bits / 8 Bytes !!
I think others have made the same point before, but the brain isn't digital so "bits" doesn't apply.
Comparing the human brain to computers is at best an apples-to-oranges comparison, because there are things that humans (even dumb ones) can do that supercomputers can't, like vision, and few humans, even the smartest ones, can keep up with pocket calculators at performing math out to 13 decimal places.
In certain specific limited tasks, the biggest supercomputers these days can do things that the smartest people can do, like play chess at the Grand Master level, so we're about on-par with supercomputers. Roughly speaking.
Then again, note how agile a housefly is, and try to guess when aerospace engineers will be able to replicate something that small flying under autonomous control.
Comparing the human brain to computers is at best an apples-to-oranges comparison, because there are things that humans (even dumb ones) can do that supercomputers can't, like vision, and few humans, even the smartest ones, can keep up with pocket calculators at performing math out to 13 decimal places.
In certain specific limited tasks, the biggest supercomputers these days can do things that the smartest people can do, like play chess at the Grand Master level, so we're about on-par with supercomputers. Roughly speaking.
Then again, note how agile a housefly is, and try to guess when aerospace engineers will be able to replicate something that small flying under autonomous control.
From new research notice if you will that human memory is an increasing exponential as we continue to evolve.
PhysOrg scientific forums are totally dedicated to science, physics, and technology. Besides topical forums such as nanotechnology, quantum physics, silicon and III-V technology, applied physics, materials, space and others, you can also join our news and publications discussions. We also provide an off-topic forum category. If you need specific help on a scientific problem or have a question related to physics or technology, visit the PhysOrg Forums. Here you’ll find experts from various fields online every day.
To quit out of "lo-fi" mode and return to the regular forums, please click here.