The most important part about what I said is that it isn't possible while remaining the way we are now. If we are not the way we are now, then IP law would no longer apply: it would have changed as the people changed.
I wasn't simply declaring the biological change alone as impossible: I was declaring the whole thought experiment as impossible, combined together. If you change people, you change society and hence the laws would be changed rather naturally. Nobody would force a law that doesn't apply anymore, in the same way laws about witches are no longer applied. It would be irrelevant.
Focusing on the biological aspect in an attempt to preserve the thought experiment is ignoring half of the argument I made.
I also acknowledged a possible situation, that isn't really biological, where an artificial implant that stores the song and activates the relevant neural connections could be invented. This preserves the thought experiment, if somewhat revised.
This I think is more likely and probably more pertinent than attempting to rewire the human psyche with the full willingness of the population. I can't see that happening myself.
Fair enough, I'm mostly interested in the possible science behind such a project. But yes, fundamentally by having people undergo such a procedure where they could remember anything they wish at will would render a hell of a lot of human practices nowadays incredibly redundant and meaningless. Because if we'd figured out how to create perfect recall we would also have figured out how to read this perfect recall without any difficulty so any trial or court procedure would just be "read the perfect memories" "he did it, next case".
The car analogy is very flawed. Again, you cannot compare external machine technology with physiological technology. Your brain doesnt work like a fucking car, for reasons other people have already pointed out.
Look, you are never going to have special abilities, get over it.
Also, Daxton, it is absolutely impossible to read minds, and always will be. The actual physical mainfestations that produce thoughts are simply chemical and electrical pulses, they cannot be decoded.
Cliff on
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
edited January 2010
I was mostly concerned with the assumption about how memory works that was kind of ignored. No current data recording device on the planet works like our brain when it comes to storing information: any such analogies, be they simplistic hardware software or whatever, they don't really help you understand it.
Rather than this kind of bioengineering solution, what you can do right now is learn mnemonic strategies to help you remember things better. This will give you a very good memory and with practise you could memorise your favourite song if you really wished. Just like someone can learn to play it themselves if they like it.
There's no such thing as a natural photographic long term memory but there are very good strategies to willfully make the storage mechanism we do have remember what you want it to remember. It's just not going to happen without effort and it's not going to be completely perfect.
If anyone wants to remember something some simple advice is to associate as many possible things to it as you can. Make as many possible links as you can to what you want to remember when you are trying to memorise it. The more links you make to what you want to remember the easier it will be to reconstruct it later with better accuracy.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
I was mostly concerned with the assumption about how memory works that was kind of ignored. No current data recording device on the planet works like our brain when it comes to storing information: any such analogies, be they simplistic hardware software or whatever, they don't really help you understand it.
Rather than this kind of bioengineering solution, what you can do is learn mnemonic strategies to help you remember things better. This will give you a very good memory and with practise you could memorise your favourite song if you really wished. Just like someone can learn to play it themselves if they like it.
Which is why bards memorised stories with song and poetry as this rhythm naturally lends itself to easier recall, thinking you'll be able to listen to a song through modification of your memory is just daft though.
Just get some ear implants that play music for you or something, there's dozens of ways that would be easier than trying to create perfect recall memory.
Daxon on
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
edited January 2010
Yeah rhyme is an excellent way of remembering something since it gets associated with all the information that has the same rhythm.
Also, Daxton, it is absolutely impossible to read minds, and always will be. The actual physical mainfestations that produce thoughts are simply chemical and electrical pulses, they cannot be decoded.
Some visual and motion elements are set up in a very structured way in most people. It's possible to read someones imagined movements, for example.
Abstract meanings not really tied to motion or visual, are probably impossible to decode because everybody has a different neural pattern for them so you can't compare them to other people to find the regularities. It's like everybody speaking a completely different language without a rosetta stone.
I'm not sure what you mean by thoughts since the word is pretty much a catchall for "internal mental activity". But imagined visual and imagined motion can be classified under thought and they could be read. I know there's a lot of work in the human interface fields doing just this, even if it's clumsy right now. So that's at least some part of "internal mental activity" that can be decoded.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
Also, Daxton, it is absolutely impossible to read minds, and always will be. The actual physical mainfestations that produce thoughts are simply chemical and electrical pulses, they cannot be decoded.
Well of course I know this, I don't know if you noticed but I just spent two long ass posts saying creating perfect recall memory would be impossible.
But in bizarro world where this is possible then you bloody well have figured out how to read memories as well meaning in such a society they could do that.
The things you want perfect recall of are of VASTLY different areas of the brain. There's enough difference between the region you process auditory information and the region where you process visual information, but then you add skill sets which involves the cerebellum and the ENTIRE central nervous system (CNS).
Well, I'm sure it's obvious right now but I'm not a neurologist (or any kind of doctor/biologist for that matter) and only have a very basic understanding of the brain, memory, etc. So I apologize if I'm making this frustrating for you. I am very curious about this though; So I do appreciate getting to hear more of the details from someone who's well versed with this stuff.
From what I do know, yes, there are many types of memory stored in different ways, in different regions for different things. I understand that to augment any one of those would require a lot of varied changes across different regions of the body. I would think that this just means that more intensive re-engineering or re-organizing of things would need to be done to facilitate change.
So I do agree that with the changes we might want to make we'd essentially be creating a new design of brain from scratch. What I meant when I spoke of consciousness was that I would think, that while incredibly difficult, it would at some point be possible to create something differently that still was able to produce the same state of self-aware consciousness that our brain does, although possibly via different methods and that if such a thing existed that I would still consider it a "human" because I believe the results are what define us, and not the methods.
In programming you often have times where you have to re-factor your code and you may essentially have to throw the entire old program out and start over. However so long as the data it returns still remains consistent with the old program (except for what you wanted to change) you've still for all intents and purposes got the same program, just a better one. Obviously the brain and the human consciousness is infinitely more complex than a simple computer program but that seems to me like what would be a barrier posed by understanding and technology rather than a fundamental one. (When I say a fundamental problem I mean something being logically impossible. Such as it being logically impossible for a person to always lie)
The brain doesn't have bits that go "okay stimulate this and your memory will be enhanced" or "this cell takes care of processing all the BLUE that you've seen in the last ten years". What you have is populations of cells that process information as a physical property of these cells.
Sounds sort of like what happens when you use an evolutionary algorithm with an FPGA. I can find the story if you'd like but I was reading about an attempt in creating learning software. The person involved programmed the FPGA to learn to do better frequency analysis and after countless generations of evolution he decided to experiment by manually altering the configuration. What he found was that even by disabling inert pins that never were used during the program's run it still caused the program to fail because the chip was making use of the magnetic properties of those inert pins as a part of how it performed analysis.
So you had the physical properties of the chip entirely linked to its function. Now I understand that the human brain and biology are different and definitely more complex. However what I'm trying to ask with this anecdote is that even if the data, and functions to process it are stored as physical properties of neurons what specifically makes it logically impossible for us to ever be able to reverse engineer that information and build something similar?
There is no distinction between "software" and "hardware" in the brain, they are inherent to one another and to create some kind of perfect recall you would essentially have to rewire every single neuron individually then perhaps add another 2 or 3 billion neurons and find out how they're meant to be wired in order to create this "perfect recall" you so massively desire.
Okay, so it may be impractical or impossible to do safely with how the brain is currently composed. I can accept that. Sometimes something just is built so poorly for what you need it to do you have to just "start over." Still, I'd think that it should be possible at some point to be able to understand enough about the brain to be able to build something that while made differently performs the same desired functions with at least 99% accuracy, and then we could build from that.
People have taken hardware and software completely foreign to them, and with no support from any official source, successfully and completely reverse engineered it and built 100% compatible devices, and in some cases devices that can do those same things better, with added functionality. History has shown that every time someone said something like that couldn't be done, it has been. Of course sometimes it can take a long time, and the solutions found may not be practical or elegant. Nobody's cracked a StarForce protected game on its release date, or been able to create a master crack that works for all StarForce protected games, but people have been able to break it.
So again, from what I understand it sounds like the problem is still just a matter of details. We may have to do something very drastic but it still seems like something we could eventually do.
Perhaps you should go pick up a nice short book on neuroscience, I suggest "Neuroscience at a Glance". It's nice and concise and is easily understood. Of course in talking about this we haven't even thought about how the CNS interacts with the rest of the body with its hormones and growth factors and immune system.
Do you know of any good online resources for this stuff? I'd definitely be interested in having a read, I just don't exactly have money to spend right now.
What you are demanding is that scientists essentially create life from scratch (we're getting there!)
Yay!
then somehow create a highly complex multicellular organism that works somewhat like a computer. It's preposterous, it just is, I truly don't think you understand the level of complexity that our brains exist at. We haven't even bloody well figured out how thought works! We just know that sections of the brain activate in strange ways and are constantly changing in sensitivity, growing, degenerating etc.
I understand enough to know that I obviously can't begin to fathom how complicated the brain is, and that we do barely know anything about it. That's why my arguments all go back to the concept of whether or not it's logically impossible. If something's logically impossible then I'd agree there's no way to ever do it since that would go against the very foundations of science. (Although from what little I overhear about quantum mechanics things we once thought were simpler may be a lot more complex.)
However if something is logically possible, then while it may be unlikely it still can happen. It just may not be possible to do right now, or be impractical because of the time or effort required. Centuries ago ideas of human flight, especially spaceflight, or being able to communicate with people, instantly, hundreds of miles away were considered impossible and ridiculous. However none of those ideas violated basic logic.
The Cerebellum, the bit which coordinates all our movements is a minibrain onto itself. It contains more neuron cell bodies in it than the other parts of the brain combined and it is constantly working. It's not like you learn a movement like "I can now reach out my hand a grab an object which I can see, there that skill has been learnt - save in the memory banks and retrieve when necessary" it works more in the way of the motor cortex sending information down the spine to the appropriate motoneurons of the arm to tell it what to do, simultaneously the cerebellum is sent a copy of the information of what the motorcortex intended to do and the cerebellum modifies the output of the motorcortex to the arm making sure it is coordinated and then the muscles of the arm feedback their positions in space and your eyes feed information to the cerebellum and premotor cortex to get the movement right. The cerebellum is meanwhile constantly working making sure the movement is done appropriately, even deviations as small as a millimeter are noted and corrected for in future movements.
Sounds like a complicated setup with a lot of inputs and outputs required to do something relatively simple. From what you've said though it still seems like, in the end, all of this is a process you can reduce and break down into logical components. They may be very complicated but the process still is one of data going in and out, as well as being modified during the process, even if you have a lot of inputs, outputs and things running in parallel it still doesn't seem inherently impossible to replicate by its very nature.
Wow, that last bit ended up being a bit of a ramble, but it was a fairly simple bit of brain workings.
I actually enjoy learning about this stuff. I may not spend days pouring over source material but whenever someone brings something up in a discussion/debate I'd rather spend some more time to hear the background behind something.
I am unsure of what these technical problems are versus the fundamental problems. To me they are exactly the same, your technical difficulties will arise because it is fundamentally a terrible idea.
Fundamental Problem: Logically impossible, therefore the very existence of the desired result cannot ever happen in any form.
Technical Problem: We can't produce the results we want with the tools/methods we have. (It's not possible to achieve successful spaceflight in a turboprop airplane, but it is possible with a shuttle and large booster rockets.)
Oh god I haven't even thought of the implications of this perfect recall business on the body's metabolism and hormonal control. Jesus fuck, you'd end up spending so much more energy, even if this didn't kill you think of all the extra food you'd have to eat. Various other processes would have to be much more efficient and quicker at absorbing food to keep up.
I'm sure there are a whole realm of implications, both biological, psychological, and sociological none of us has ever even dreamed of yet. But that's why we start small with carefully controlled experiments and if it's safe, work from there.
The car analogy is very flawed. Again, you cannot compare external machine technology with physiological technology. Your brain doesn't work like a fucking car, for reasons other people have already pointed out.
Look, you are never going to have special abilities, get over it.
Also, Daxton, it is absolutely impossible to read minds, and always will be. The actual physical mainfestations that produce thoughts are simply chemical and electrical pulses, they cannot be decoded
The point is that both objects are reducible to a series of logical components. Something is put in, processed, and something is created or put out. The entire basis of modern science is in reducing complex systems into their components, analyzing their interactions and then predicting future results based on what we've found.
Cars, computers, and even the body are all complex systems made up by a huge number of components. Some systems are more complex then others but when given an unlimited time frame what else is there to prevent us from understanding these things?
In regards to reading minds, that reminds me of what I hear a lot whenever someone claims that their copyright protection is unbreakable. It's impossible for any system to be completely secured if at some point someone has to be allowed to decrypt the data. If our eyes process light, then obviously at some point something has to make sense of that information. If at any point one thing can interpret the data, what makes it logically impossible for something else to do the same thing?
If the best you can come up with against someone who's patently ignorant is to yell back at him, "Yeah? Well there's BOOKS, and they say you're WRONG!"
Then honestly you're not coming out of this looking great either.
Fall out man: it is impossible because one cannot break down the electrical pulses that make up thoughts any further. We can map what areas of the brain take care of certain functions, you can define chemicals that produce certain effects (I.e. serotonin, dopamine, etc.), but you cannot decode actual thoughts. For example, I could be strapped to the most advanced brain scanning technology possible, and you could tell the basic structures of my brain, which nuerons are firing, and all sorts of things. However, you cannot tell what I am thinking or gain access to any of my memories. It doesn't matter if I was thinking about red trucks or if i have memories of a beautiful autumn day, all youll see are chemicals and electricity.
Fall out man: it is impossible because one cannot break down the electrical pulses that make up thoughts any further. We can map what areas of the brain take care of certain functions, you can define chemicals that produce certain effects (I.e. serotonin, dopamine, etc.), but you cannot decode actual thoughts. For example, I could be strapped to the most advanced brain scanning technology possible, and you could tell the basic structures of my brain, which nuerons are firing, and all sorts of things. However, you cannot tell what I am thinking or gain access to any of my memories. It doesn't matter if I was thinking about red trucks or if i have memories of a beautiful autumn day, all youll see are chemicals and electricity.
If I were to tap into a properly encrypted connection all I would see would be garbled nonsense. I may never be able to make sense of it by simply tapping in that way. However does that mean that it's impossible to ever break the encryption and turn the nonsense back into understandable information?
We know a lot about the building blocks but not the context, and yes that's definitely an obstacle. But if one device can decode the signals then it is possible in theory to do it, it just may be horribly complicated.
If the best you can come up with against someone who's patently ignorant is to yell back at him, "Yeah? Well there's BOOKS, and they say you're WRONG!"
Then honestly you're not coming out of this looking great either.
I do not think you quite understand. It is not encryted. There is no code to crack. You cannot extricate thoughts, they do not exist as readable data.
There is no analogy for this because there is nothing that works like a brain. There are some things that are just not possible, no matter how much time or technology is available.
IP problem of the near future Reselling your MP3s. Haven't see anything about it and their site is basically just a page and a year old blog post, but the idea is not without merit. I can resell old CDs I have. Especially considering how much stuff get re-released in ever expanding boxed set formats; how many copies of the same recording does anyone need.
I do not think you quite understand. It is not encryted. There is no code to crack. You cannot extricate thoughts, they do not exist as readable data.
There is no analogy for this because there is nothing that works like a brain. There are some things that are just not possible, no matter how much time or technology is available.
I cannot find this a convincing argument, to be honest. I'm not going to assert that it's just around the corner, or even that it's possible – but the claim that something is in principle impossible should require significantly more evidence in support than complexity. Given sufficient data (and what a phrase that is!), it does not seem in principle impossible to correlate brainstates with 'thoughts' with reasonable accuracy*. It seems to be in principle possible to read brainstates directly and at range, and it is possible that some components of this state correlate with thoughts (obviously with variation in individuals)
Does this make it possible? Well, no. But it seems to imply to be that the assertion that to transform this form of information into that which arises from it† is impossible should be considered a far stronger claim than you appear to be treating it as.
Excuse my babbling! I just wanted to present a response of some form. My apologies if it is difficult to follow.
* consider the extreme: we have a database of every possible thought along with the corresponding brainstate for a given individual
† I'm assuming that you grant this, although I accept that it's been a problem in past threads
Linden on
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
I do not think you quite understand. It is not encryted. There is no code to crack. You cannot extricate thoughts, they do not exist as readable data.
There is no analogy for this because there is nothing that works like a brain. There are some things that are just not possible, no matter how much time or technology is available.
I cannot find this a convincing argument, to be honest. I'm not going to assert that it's just around the corner, or even that it's possible – but the claim that something is in principle impossible should require significantly more evidence in support than complexity. Given sufficient data (and what a phrase that is!), it does not seem in principle impossible to correlate brainstates with 'thoughts' with reasonable accuracy*. It seems to be in principle possible to read brainstates directly and at range, and it is possible that some components of this state correlate with thoughts (obviously with variation in individuals)
Does this make it possible? Well, no. But it seems to imply to be that the assertion that to transform this form of information into that which arises from it† is impossible should be considered a far stronger claim than you appear to be treating it as.
Excuse my babbling! I just wanted to present a response of some form. My apologies if it is difficult to follow.
* consider the extreme: we have a database of every possible thought along with the corresponding brainstate for a given individual
† I'm assuming that you grant this, although I accept that it's been a problem in past threads
When you want to work out what part of a brain does what, you can't just look at an individual person. you have to take multiple people and compare them to each other. If the part of the brain is relatively the same across multiple people, as macro structures such as "the hippocampus" tend to be, then you can make statements about them through this process of comparison.
However a persons individual thoughts are the function of their brains own particular wiring within these macro structures. You can only isolate a particular piece of brain activity from another if many people tend to activate the same pattern when engaging in the same activity.
There is always neuronal activation happening within the brain. Left alone with no stimulus, brains have a kind of idle pattern of spontaneous firing all over the entire structure. This still happens when someone is fully engaged in an activity. It's kind of like a huge amount of random static you have to deal with in order to be able to isolate any activity from another.
Then lets say you can isolate this: now you have to isolate which part of the many different pieces of neuronal activity relating to cat are not also related to remembering any animal, or remembering furry animals, or remembering ears and so on. All of these are activated when you remember something: semantic meaning is heirarchical with higher order details such as "is an animal versus a plant" being activated before you get to "is a furry animal". (This was deduced from studying semantic dementia, who symptoms are a progressive loss of semantic meaning that correlates with the brain damage caused by semantic dementia)
In this situation, everybodies neural setup of this structure within the macro structures that store these semantic meanings (because they're in specific parts of the brain) is different. One persons "is an animal" may be activating in the same place where anothers is activating "is a furry animal". Or more extreme but equally likely, one persons "rhodendron" is the same pattern of activation as another persons "coffee cup". Everybodies learning of semantic meaning is an individual thing in terms of what patterns of neuronal activation correspond to a particular piece of semantic meaning.
So for some things like this, it becomes impossible, logically impossible, to deduce any further. You can't separate the activity from other nearby activity and you can't use other people in comparison to achieve a benchmark for comparison. There's no way to tell what is what.
Some things you can because they are relatively stable across different people. For example your visual field is set up roughly the same and even corresponds to a 3d mapping of what an eye would see. So you can say that field a1 of this persons brain will be in the same region as this other persons brain and will be activated by light from this section of this persons eye. Hence you can make much stronger statements about what kinds of shape a person is seeing just by looking at their brain activity. But you can't do that with all "thought", only visual and some parts of movement. You can't do it for semantic meaning beyond the point of knowing that semantic meaning is generally stored in this part of the brain.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
No, no, it is not a matter of complexity. There is no data. You can map the stimulus of every brain cell and every electrical pulse and chemical change and you still won't be able to tell what someone is thinking. It is not impossible because of complexity, it is impossible because it cannot be done, ever, in any way.
I think some of you fundamentally do not understand the brain. It is not like a computer. Memories and thoughts are not stored as physical data.
Cliff on
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
edited January 2010
Cliff I know you agree with me that it's impossible but I do not agree with your reasons for why it is impossible.
As reasons go they are limited to "it can't happen because it can't happen" which is not a very strong argument dude.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
I do not think you quite understand. It is not encryted. There is no code to crack. You cannot extricate thoughts, they do not exist as readable data.
There is no analogy for this because there is nothing that works like a brain. There are some things that are just not possible, no matter how much time or technology is available.
I cannot find this a convincing argument, to be honest. I'm not going to assert that it's just around the corner, or even that it's possible – but the claim that something is in principle impossible should require significantly more evidence in support than complexity. Given sufficient data (and what a phrase that is!), it does not seem in principle impossible to correlate brainstates with 'thoughts' with reasonable accuracy*. It seems to be in principle possible to read brainstates directly and at range, and it is possible that some components of this state correlate with thoughts (obviously with variation in individuals)
Does this make it possible? Well, no. But it seems to imply to be that the assertion that to transform this form of information into that which arises from it† is impossible should be considered a far stronger claim than you appear to be treating it as.
Excuse my babbling! I just wanted to present a response of some form. My apologies if it is difficult to follow.
* consider the extreme: we have a database of every possible thought along with the corresponding brainstate for a given individual
† I'm assuming that you grant this, although I accept that it's been a problem in past threads
When you want to work out what part of a brain does what, you can't just look at an individual person. you have to take multiple people and compare them to each other. If the part of the brain is relatively the same across multiple people, as macro structures such as "the hippocampus" tend to be, then you can make statements about them through this process of comparison.
However a persons individual thoughts are the function of their brains own particular wiring within these macro structures. You can only isolate a particular piece of brain activity from another if many people tend to activate the same pattern when engaging in the same activity.
There is always neuronal activation happening within the brain. Left alone with no stimulus, brains have a kind of idle pattern of spontaneous firing all over the entire structure. This still happens when someone is fully engaged in an activity. It's kind of like a huge amount of random static you have to deal with in order to be able to isolate any activity from another.
Then lets say you can isolate this: now you have to isolate which part of the many different pieces of neuronal activity relating to cat are not also related to remembering any animal, or remembering furry animals, or remembering ears and so on. All of these are activated when you remember something: semantic meaning is heirarchical with higher order details such as "is an animal versus a plant" being activated before you get to "is a furry animal". (This was deduced from studying semantic dementia, who symptoms are a progressive loss of semantic meaning that correlates with the brain damage caused by semantic dementia)
In this situation, everybodies neural setup of this structure within the macro structures that store these semantic meanings (because they're in specific parts of the brain) is different. One persons "is an animal" may be activating in the same place where anothers is activating "is a furry animal". Or more extreme but equally likely, one persons "rhodendron" is the same pattern of activation as another persons "coffee cup". Everybodies learning of semantic meaning is an individual thing in terms of what patterns of neuronal activation correspond to a particular piece of semantic meaning.
So for some things like this, it becomes impossible, logically impossible, to deduce any further. You can't separate the activity from other nearby activity and you can't use other people in comparison to achieve a benchmark for comparison. There's no way to tell what is what.
Some things you can because they are relatively stable across different people. For example your visual field is set up roughly the same and even corresponds to a 3d mapping of what an eye would see. So you can say that field a1 of this persons brain will be in the same region as this other persons brain and will be activated by light from this section of this persons eye. Hence you can make much stronger statements about what kinds of shape a person is seeing just by looking at their brain activity. But you can't do that with all "thought", only visual and some parts of movement. You can't do it for semantic meaning beyond the point of knowing that semantic meaning is generally stored in this part of the brain.
The obvious solution seems to be to be to do individual mappings based on correlations to external stimuli.
It also seems that the patterns have to be similar even if their spacial location is different. In other words, a structure corresponding to "is an animal" has to send the same output based on the same input no matter where it is or what the exact details are for it to actually be informationally identical.
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
I do not think you quite understand. It is not encryted. There is no code to crack. You cannot extricate thoughts, they do not exist as readable data.
There is no analogy for this because there is nothing that works like a brain. There are some things that are just not possible, no matter how much time or technology is available.
I cannot find this a convincing argument, to be honest. I'm not going to assert that it's just around the corner, or even that it's possible – but the claim that something is in principle impossible should require significantly more evidence in support than complexity. Given sufficient data (and what a phrase that is!), it does not seem in principle impossible to correlate brainstates with 'thoughts' with reasonable accuracy*. It seems to be in principle possible to read brainstates directly and at range, and it is possible that some components of this state correlate with thoughts (obviously with variation in individuals)
Does this make it possible? Well, no. But it seems to imply to be that the assertion that to transform this form of information into that which arises from it† is impossible should be considered a far stronger claim than you appear to be treating it as.
Excuse my babbling! I just wanted to present a response of some form. My apologies if it is difficult to follow.
* consider the extreme: we have a database of every possible thought along with the corresponding brainstate for a given individual
† I'm assuming that you grant this, although I accept that it's been a problem in past threads
When you want to work out what part of a brain does what, you can't just look at an individual person. you have to take multiple people and compare them to each other. If the part of the brain is relatively the same across multiple people, as macro structures such as "the hippocampus" tend to be, then you can make statements about them through this process of comparison.
However a persons individual thoughts are the function of their brains own particular wiring within these macro structures. You can only isolate a particular piece of brain activity from another if many people tend to activate the same pattern when engaging in the same activity.
There is always neuronal activation happening within the brain. Left alone with no stimulus, brains have a kind of idle pattern of spontaneous firing all over the entire structure. This still happens when someone is fully engaged in an activity. It's kind of like a huge amount of random static you have to deal with in order to be able to isolate any activity from another.
Then lets say you can isolate this: now you have to isolate which part of the many different pieces of neuronal activity relating to cat are not also related to remembering any animal, or remembering furry animals, or remembering ears and so on. All of these are activated when you remember something: semantic meaning is heirarchical with higher order details such as "is an animal versus a plant" being activated before you get to "is a furry animal". (This was deduced from studying semantic dementia, who symptoms are a progressive loss of semantic meaning that correlates with the brain damage caused by semantic dementia)
In this situation, everybodies neural setup of this structure within the macro structures that store these semantic meanings (because they're in specific parts of the brain) is different. One persons "is an animal" may be activating in the same place where anothers is activating "is a furry animal". Or more extreme but equally likely, one persons "rhodendron" is the same pattern of activation as another persons "coffee cup". Everybodies learning of semantic meaning is an individual thing in terms of what patterns of neuronal activation correspond to a particular piece of semantic meaning.
So for some things like this, it becomes impossible, logically impossible, to deduce any further. You can't separate the activity from other nearby activity and you can't use other people in comparison to achieve a benchmark for comparison. There's no way to tell what is what.
Some things you can because they are relatively stable across different people. For example your visual field is set up roughly the same and even corresponds to a 3d mapping of what an eye would see. So you can say that field a1 of this persons brain will be in the same region as this other persons brain and will be activated by light from this section of this persons eye. Hence you can make much stronger statements about what kinds of shape a person is seeing just by looking at their brain activity. But you can't do that with all "thought", only visual and some parts of movement. You can't do it for semantic meaning beyond the point of knowing that semantic meaning is generally stored in this part of the brain.
The obvious solution seems to be to be to do individual mappings based on correlations to external stimuli.
It also seems that the patterns have to be similar even if their spacial location is different. In other words, a structure corresponding to "is an animal" has to send the same output based on the same input no matter where it is or what the exact details are for it to actually be informationally identical.
What do you mean by individual mappings and how do you differentiate these simultaneous activations from each other. How do you determine wether it's animal or cat that is being activated when they are all activated. For all intents and purposes animal and cat activate at the same time, as one big neural pattern. How do you separate that pattern?
Also what are you basing your second assumption on? Because all I see is a giant leap into space but you seem to think it's obvious. I don't see it at all, can you explain where that came from please? I've played around with computer modelling of neural connections and you can represent the same information in different ways on such a network even for extremely simple things like approach behaviors. Let alone abstract semantic meaning.
In fact that second assumption is just flat out wrong based on what I know of neural network modelling. What corresponds to cat is the activation that occurs at the time each individual sees a cat. It's this activation that gets stored as cat and activated the next time it is seen. This activation is not identical, it's entirely dependent on the overal setup of the individual. It's pattern is dependent on the makeup of the persons brain and the situation it is seen in. All the patterns would be different. If you then upgrade this mental model to include the earlier point that it's all essentially one big neural pattern of animal, furry, legs, ears, paws, eyes, meow, etc including details that correspond to cat it is guaranteed that this pattern would be different.
Unless you have a very good argument for why they are all going to be identical that is based on some solid evidence and reasoning you are going to have to take the "obvious" out of your statements, as things are not as clear cut as you are assuming.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
Cliff I know you agree with me that it's impossible but I do not agree with your reasons for why it is impossible.
As reasons go they are limited to "it can't happen because it can't happen" which is not a very strong argument dude.
Well, nueroscience isn't exactly my specialty, I just know the end result of these "mind reading" labours.
Cliff on
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
edited January 2010
This is very hard to explain and understand in words, so tomorrow I'm going to try and find some images in my notes or maybe something online that is an example of a neural network that I can link to to help explain it.
It's just infinitely easier to understand when you use diagrams in something like this. I remember I actually had to go down after the lecture and talk to her afterwards to get it all through my head, and she had to bring up the diagrams again to do so.
So I feel like I wont be able to explain it unless I do that. I'm too tired tonight though.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
IP problem of the near future Reselling your MP3s. Haven't see anything about it and their site is basically just a page and a year old blog post, but the idea is not without merit. I can resell old CDs I have. Especially considering how much stuff get re-released in ever expanding boxed set formats; how many copies of the same recording does anyone need.
I fail to see how someone else having paid for a piece of content somehow gives you the right to acquire that content without paying anything to the creater/owner of the content, when you did not have that right before.
Either reselling is morally wrong, or piracy isn't.
IP problem of the near future Reselling your MP3s. Haven't see anything about it and their site is basically just a page and a year old blog post, but the idea is not without merit. I can resell old CDs I have. Especially considering how much stuff get re-released in ever expanding boxed set formats; how many copies of the same recording does anyone need.
I fail to see how someone else having paid for a piece of content somehow gives you the right to acquire that content without paying anything to the creater/owner of the content, when you did not have that right before.
Either reselling is morally wrong, or piracy isn't.
And sometimes they're basically the same thing. Fun fact: My local Gamestop wouldn't accept returns on the Fallout 3 expansions, due to obvious piracy concerns (they were one-time install deals). Yet they sold the expansions used. At their customary tremendous markup.
Also, Daxton, it is absolutely impossible to read minds, and always will be. The actual physical mainfestations that produce thoughts are simply chemical and electrical pulses, they cannot be decoded.
Some visual and motion elements are set up in a very structured way in most people. It's possible to read someones imagined movements, for example.
Abstract meanings not really tied to motion or visual, are probably impossible to decode because everybody has a different neural pattern for them so you can't compare them to other people to find the regularities. It's like everybody speaking a completely different language without a rosetta stone.
I'm not sure what you mean by thoughts since the word is pretty much a catchall for "internal mental activity". But imagined visual and imagined motion can be classified under thought and they could be read. I know there's a lot of work in the human interface fields doing just this, even if it's clumsy right now. So that's at least some part of "internal mental activity" that can be decoded.
There was a recent project with fMRI which was able to, after training, decode low-resolution images of what someone was seeing directly from the visual cortex. While I highly doubt remote telepathy can ever be a reality, barring say, us getting The Culture's hyperspace Minds, currently it seems like the limiting factor is our sensor technology.
I keep a look out for anything coming up which might let us see electrical activity in the brain, in real-time with spatial resolution.
IP problem of the near future Reselling your MP3s. Haven't see anything about it and their site is basically just a page and a year old blog post, but the idea is not without merit. I can resell old CDs I have. Especially considering how much stuff get re-released in ever expanding boxed set formats; how many copies of the same recording does anyone need.
I fail to see how someone else having paid for a piece of content somehow gives you the right to acquire that content without paying anything to the creater/owner of the content, when you did not have that right before.
Either reselling is morally wrong, or piracy isn't.
And sometimes they're basically the same thing. Fun fact: My local Gamestop wouldn't accept returns on the Fallout 3 expansions, due to obvious piracy concerns (they were one-time install deals). Yet they sold the expansions used. At their customary tremendous markup.
so you could still just buy a used one and resell it and end up with a huge discount...
What do you mean by individual mappings and how do you differentiate these simultaneous activations from each other. How do you determine wether it's animal or cat that is being activated when they are all activated. For all intents and purposes animal and cat activate at the same time, as one big neural pattern. How do you separate that pattern?
By also showing them a dog, a fish, a monkey and so forth and comparing them to identify specific elements by overlap. Like a Venn diagram.
Also what are you basing your second assumption on? Because all I see is a giant leap into space but you seem to think it's obvious. I don't see it at all, can you explain where that came from please? I've played around with computer modelling of neural connections and you can represent the same information in different ways on such a network even for extremely simple things like approach behaviors. Let alone abstract semantic meaning.
In fact that second assumption is just flat out wrong based on what I know of neural network modelling. What corresponds to cat is the activation that occurs at the time each individual sees a cat. It's this activation that gets stored as cat and activated the next time it is seen. This activation is not identical, it's entirely dependent on the overal setup of the individual. It's pattern is dependent on the makeup of the persons brain and the situation it is seen in. All the patterns would be different. If you then upgrade this mental model to include the earlier point that it's all essentially one big neural pattern of animal, furry, legs, ears, paws, eyes, meow, etc including details that correspond to cat it is guaranteed that this pattern would be different.
Unless you have a very good argument for why they are all going to be identical that is based on some solid evidence and reasoning you are going to have to take the "obvious" out of your statements, as things are not as clear cut as you are assuming.
To act as information storage it cannot be entirely arbitrary. It has to produce an appropriate output from the input. In other words, the range of reasonable and likely behavioral responses to seeing a cat is small (by comparison). So however the information about cats is stored, it has be a way such that when confronted with a cat, it will trigger an appropriate behavioral response.
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
0
Options
MorninglordI'm tired of being Batman,so today I'll be Owl.Registered Userregular
What do you mean by individual mappings and how do you differentiate these simultaneous activations from each other. How do you determine wether it's animal or cat that is being activated when they are all activated. For all intents and purposes animal and cat activate at the same time, as one big neural pattern. How do you separate that pattern?
By also showing them a dog, a fish, a monkey and so forth and comparing them to identify specific elements by overlap. Like a Venn diagram.
Also what are you basing your second assumption on? Because all I see is a giant leap into space but you seem to think it's obvious. I don't see it at all, can you explain where that came from please? I've played around with computer modelling of neural connections and you can represent the same information in different ways on such a network even for extremely simple things like approach behaviors. Let alone abstract semantic meaning.
In fact that second assumption is just flat out wrong based on what I know of neural network modelling. What corresponds to cat is the activation that occurs at the time each individual sees a cat. It's this activation that gets stored as cat and activated the next time it is seen. This activation is not identical, it's entirely dependent on the overal setup of the individual. It's pattern is dependent on the makeup of the persons brain and the situation it is seen in. All the patterns would be different. If you then upgrade this mental model to include the earlier point that it's all essentially one big neural pattern of animal, furry, legs, ears, paws, eyes, meow, etc including details that correspond to cat it is guaranteed that this pattern would be different.
Unless you have a very good argument for why they are all going to be identical that is based on some solid evidence and reasoning you are going to have to take the "obvious" out of your statements, as things are not as clear cut as you are assuming.
To act as information storage it cannot be entirely arbitrary. It has to produce an appropriate output from the input. In other words, the range of reasonable and likely behavioral responses to seeing a cat is small (by comparison). So however the information about cats is stored, it has be a way such that when confronted with a cat, it will trigger an appropriate behavioral response.
Why can't it be arbitrary? Where are you getting these information rules from? You are working from some kind of axiom regarding information that I've never heard of.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
IP problem of the near future Reselling your MP3s. Haven't see anything about it and their site is basically just a page and a year old blog post, but the idea is not without merit. I can resell old CDs I have. Especially considering how much stuff get re-released in ever expanding boxed set formats; how many copies of the same recording does anyone need.
I fail to see how someone else having paid for a piece of content somehow gives you the right to acquire that content without paying anything to the creater/owner of the content, when you did not have that right before.
Either reselling is morally wrong, or piracy isn't.
And sometimes they're basically the same thing. Fun fact: My local Gamestop wouldn't accept returns on the Fallout 3 expansions, due to obvious piracy concerns (they were one-time install deals). Yet they sold the expansions used. At their customary tremendous markup.
so you could still just buy a used one and resell it and end up with a huge discount...
And meanwhile some gamestop executive is yelling "c-c-c-combo breaker!" because they're earning a large multiple of their normal margin by buying these "used" games cheap and selling them high. And though each of these repeat sales earns gamestop cash, the devs see none of it.
What do you mean by individual mappings and how do you differentiate these simultaneous activations from each other. How do you determine wether it's animal or cat that is being activated when they are all activated. For all intents and purposes animal and cat activate at the same time, as one big neural pattern. How do you separate that pattern?
By also showing them a dog, a fish, a monkey and so forth and comparing them to identify specific elements by overlap. Like a Venn diagram.
Also what are you basing your second assumption on? Because all I see is a giant leap into space but you seem to think it's obvious. I don't see it at all, can you explain where that came from please? I've played around with computer modelling of neural connections and you can represent the same information in different ways on such a network even for extremely simple things like approach behaviors. Let alone abstract semantic meaning.
In fact that second assumption is just flat out wrong based on what I know of neural network modelling. What corresponds to cat is the activation that occurs at the time each individual sees a cat. It's this activation that gets stored as cat and activated the next time it is seen. This activation is not identical, it's entirely dependent on the overal setup of the individual. It's pattern is dependent on the makeup of the persons brain and the situation it is seen in. All the patterns would be different. If you then upgrade this mental model to include the earlier point that it's all essentially one big neural pattern of animal, furry, legs, ears, paws, eyes, meow, etc including details that correspond to cat it is guaranteed that this pattern would be different.
Unless you have a very good argument for why they are all going to be identical that is based on some solid evidence and reasoning you are going to have to take the "obvious" out of your statements, as things are not as clear cut as you are assuming.
To act as information storage it cannot be entirely arbitrary. It has to produce an appropriate output from the input. In other words, the range of reasonable and likely behavioral responses to seeing a cat is small (by comparison). So however the information about cats is stored, it has be a way such that when confronted with a cat, it will trigger an appropriate behavioral response.
Why can't it be arbitrary? Where are you getting these information rules from? You are working from some kind of axiom regarding information that I've never heard of.
I don't understand people's insistence on comparing the brain to computers and computer programming either. They have absolutely nothing in common, hell, they're not even made up of the same materials.
Have you not seen this, from 2007, or more recently, this?
One is based on analysing areas related to planning (the prefrontal cortex and other areas of the frontal lobes) as well as areas related to emotional impulses (I assume anyway, there's no source there for me to look up) and another is based on analysing the place cells in the hippocampus that is responsible for forming memories of locations. Which is pretty much memory of where you have moved before.
I've already said somethings can be read this way so it's not a big surprise to me. I was talking about semantic meaning, which is located in different parts of the brain to what they would have been scanning.
The articles also got a heavy reporter spin on it. You'd be advised to downgrade claims of "reading minds" perfectly. This is a specific thing they are doing in both cases.
Using place cells like that is pretty ingenious though, I'm impressed by that.
I'd already heard of the other one, I can't remember where from though. It was familiar.
The brain is not at all homogenous when it comes to the way each structure within it performs it's function, so claiming to be able to generalise from what they are doing in each case is actually kind of silly of them. Just because they are an expert in one group of structures doesn't mean they know everything about the whole brain. They shouldn't be making claims like that. That's if the reporter didn't sensationalise what they really said, which is much more likely.
I wish they'd report their goddam sources for these things, it's really irritating when an article shoots off it's mouth about something interesting but never cites any published work. It makes it a real pain to look it up to see what they're actually doing.
Morninglord on
(PSN: Morninglord) (Steam: Morninglord) (WiiU: Morninglord22) I like to record and toss up a lot of random gaming videos here.
Hmm... since all brains develop differently it does seem impossible to create any way to read humans minds as a whole, but assuming you could filter out the static, with an individuals cooperation, and perpetual testing, couldn't you eventually be able to read their specific mind?
Hmm... since all brains develop differently it does seem impossible to create any way to read humans minds as a whole, but assuming you could filter out the static, with an individuals cooperation, and perpetual testing, couldn't you eventually be able to read their specific mind?
Hmm... since all brains develop differently it does seem impossible to create any way to read humans minds as a whole, but assuming you could filter out the static, with an individuals cooperation, and perpetual testing, couldn't you eventually be able to read their specific mind?
Sorry but that just, sounds like a massive waste of time.
Well done! After years of studying you we can finally know when you want to go get a meal and what you will cook! Huzzah for science!
The day we start taking a machine's word over a human's when dealing their own brains I'm gonna go live in a cave somewhere . I love technology, but I'd give it all up if the alternative is to have a religious like faith in it.
Posts
Fair enough, I'm mostly interested in the possible science behind such a project. But yes, fundamentally by having people undergo such a procedure where they could remember anything they wish at will would render a hell of a lot of human practices nowadays incredibly redundant and meaningless. Because if we'd figured out how to create perfect recall we would also have figured out how to read this perfect recall without any difficulty so any trial or court procedure would just be "read the perfect memories" "he did it, next case".
IP laws would be the least of your worries
Look, you are never going to have special abilities, get over it.
Also, Daxton, it is absolutely impossible to read minds, and always will be. The actual physical mainfestations that produce thoughts are simply chemical and electrical pulses, they cannot be decoded.
Rather than this kind of bioengineering solution, what you can do right now is learn mnemonic strategies to help you remember things better. This will give you a very good memory and with practise you could memorise your favourite song if you really wished. Just like someone can learn to play it themselves if they like it.
There's no such thing as a natural photographic long term memory but there are very good strategies to willfully make the storage mechanism we do have remember what you want it to remember. It's just not going to happen without effort and it's not going to be completely perfect.
If anyone wants to remember something some simple advice is to associate as many possible things to it as you can. Make as many possible links as you can to what you want to remember when you are trying to memorise it. The more links you make to what you want to remember the easier it will be to reconstruct it later with better accuracy.
Which is why bards memorised stories with song and poetry as this rhythm naturally lends itself to easier recall, thinking you'll be able to listen to a song through modification of your memory is just daft though.
Just get some ear implants that play music for you or something, there's dozens of ways that would be easier than trying to create perfect recall memory.
I'm a poet and I didn't know it.
Some visual and motion elements are set up in a very structured way in most people. It's possible to read someones imagined movements, for example.
Abstract meanings not really tied to motion or visual, are probably impossible to decode because everybody has a different neural pattern for them so you can't compare them to other people to find the regularities. It's like everybody speaking a completely different language without a rosetta stone.
I'm not sure what you mean by thoughts since the word is pretty much a catchall for "internal mental activity". But imagined visual and imagined motion can be classified under thought and they could be read. I know there's a lot of work in the human interface fields doing just this, even if it's clumsy right now. So that's at least some part of "internal mental activity" that can be decoded.
Well of course I know this, I don't know if you noticed but I just spent two long ass posts saying creating perfect recall memory would be impossible.
But in bizarro world where this is possible then you bloody well have figured out how to read memories as well meaning in such a society they could do that.
Well, I'm sure it's obvious right now but I'm not a neurologist (or any kind of doctor/biologist for that matter) and only have a very basic understanding of the brain, memory, etc. So I apologize if I'm making this frustrating for you. I am very curious about this though; So I do appreciate getting to hear more of the details from someone who's well versed with this stuff.
From what I do know, yes, there are many types of memory stored in different ways, in different regions for different things. I understand that to augment any one of those would require a lot of varied changes across different regions of the body. I would think that this just means that more intensive re-engineering or re-organizing of things would need to be done to facilitate change.
So I do agree that with the changes we might want to make we'd essentially be creating a new design of brain from scratch. What I meant when I spoke of consciousness was that I would think, that while incredibly difficult, it would at some point be possible to create something differently that still was able to produce the same state of self-aware consciousness that our brain does, although possibly via different methods and that if such a thing existed that I would still consider it a "human" because I believe the results are what define us, and not the methods.
In programming you often have times where you have to re-factor your code and you may essentially have to throw the entire old program out and start over. However so long as the data it returns still remains consistent with the old program (except for what you wanted to change) you've still for all intents and purposes got the same program, just a better one. Obviously the brain and the human consciousness is infinitely more complex than a simple computer program but that seems to me like what would be a barrier posed by understanding and technology rather than a fundamental one. (When I say a fundamental problem I mean something being logically impossible. Such as it being logically impossible for a person to always lie)
Sounds sort of like what happens when you use an evolutionary algorithm with an FPGA. I can find the story if you'd like but I was reading about an attempt in creating learning software. The person involved programmed the FPGA to learn to do better frequency analysis and after countless generations of evolution he decided to experiment by manually altering the configuration. What he found was that even by disabling inert pins that never were used during the program's run it still caused the program to fail because the chip was making use of the magnetic properties of those inert pins as a part of how it performed analysis.
So you had the physical properties of the chip entirely linked to its function. Now I understand that the human brain and biology are different and definitely more complex. However what I'm trying to ask with this anecdote is that even if the data, and functions to process it are stored as physical properties of neurons what specifically makes it logically impossible for us to ever be able to reverse engineer that information and build something similar?
Okay, so it may be impractical or impossible to do safely with how the brain is currently composed. I can accept that. Sometimes something just is built so poorly for what you need it to do you have to just "start over." Still, I'd think that it should be possible at some point to be able to understand enough about the brain to be able to build something that while made differently performs the same desired functions with at least 99% accuracy, and then we could build from that.
People have taken hardware and software completely foreign to them, and with no support from any official source, successfully and completely reverse engineered it and built 100% compatible devices, and in some cases devices that can do those same things better, with added functionality. History has shown that every time someone said something like that couldn't be done, it has been. Of course sometimes it can take a long time, and the solutions found may not be practical or elegant. Nobody's cracked a StarForce protected game on its release date, or been able to create a master crack that works for all StarForce protected games, but people have been able to break it.
So again, from what I understand it sounds like the problem is still just a matter of details. We may have to do something very drastic but it still seems like something we could eventually do.
Do you know of any good online resources for this stuff? I'd definitely be interested in having a read, I just don't exactly have money to spend right now.
Yay!
I understand enough to know that I obviously can't begin to fathom how complicated the brain is, and that we do barely know anything about it. That's why my arguments all go back to the concept of whether or not it's logically impossible. If something's logically impossible then I'd agree there's no way to ever do it since that would go against the very foundations of science. (Although from what little I overhear about quantum mechanics things we once thought were simpler may be a lot more complex.)
However if something is logically possible, then while it may be unlikely it still can happen. It just may not be possible to do right now, or be impractical because of the time or effort required. Centuries ago ideas of human flight, especially spaceflight, or being able to communicate with people, instantly, hundreds of miles away were considered impossible and ridiculous. However none of those ideas violated basic logic.
Sounds like a complicated setup with a lot of inputs and outputs required to do something relatively simple. From what you've said though it still seems like, in the end, all of this is a process you can reduce and break down into logical components. They may be very complicated but the process still is one of data going in and out, as well as being modified during the process, even if you have a lot of inputs, outputs and things running in parallel it still doesn't seem inherently impossible to replicate by its very nature.
I actually enjoy learning about this stuff. I may not spend days pouring over source material but whenever someone brings something up in a discussion/debate I'd rather spend some more time to hear the background behind something.
Fundamental Problem: Logically impossible, therefore the very existence of the desired result cannot ever happen in any form.
Technical Problem: We can't produce the results we want with the tools/methods we have. (It's not possible to achieve successful spaceflight in a turboprop airplane, but it is possible with a shuttle and large booster rockets.)
I'm sure there are a whole realm of implications, both biological, psychological, and sociological none of us has ever even dreamed of yet. But that's why we start small with carefully controlled experiments and if it's safe, work from there.
The point is that both objects are reducible to a series of logical components. Something is put in, processed, and something is created or put out. The entire basis of modern science is in reducing complex systems into their components, analyzing their interactions and then predicting future results based on what we've found.
Cars, computers, and even the body are all complex systems made up by a huge number of components. Some systems are more complex then others but when given an unlimited time frame what else is there to prevent us from understanding these things?
In regards to reading minds, that reminds me of what I hear a lot whenever someone claims that their copyright protection is unbreakable. It's impossible for any system to be completely secured if at some point someone has to be allowed to decrypt the data. If our eyes process light, then obviously at some point something has to make sense of that information. If at any point one thing can interpret the data, what makes it logically impossible for something else to do the same thing?
If I were to tap into a properly encrypted connection all I would see would be garbled nonsense. I may never be able to make sense of it by simply tapping in that way. However does that mean that it's impossible to ever break the encryption and turn the nonsense back into understandable information?
We know a lot about the building blocks but not the context, and yes that's definitely an obstacle. But if one device can decode the signals then it is possible in theory to do it, it just may be horribly complicated.
There is no analogy for this because there is nothing that works like a brain. There are some things that are just not possible, no matter how much time or technology is available.
I cannot find this a convincing argument, to be honest. I'm not going to assert that it's just around the corner, or even that it's possible – but the claim that something is in principle impossible should require significantly more evidence in support than complexity. Given sufficient data (and what a phrase that is!), it does not seem in principle impossible to correlate brainstates with 'thoughts' with reasonable accuracy*. It seems to be in principle possible to read brainstates directly and at range, and it is possible that some components of this state correlate with thoughts (obviously with variation in individuals)
Does this make it possible? Well, no. But it seems to imply to be that the assertion that to transform this form of information into that which arises from it† is impossible should be considered a far stronger claim than you appear to be treating it as.
Excuse my babbling! I just wanted to present a response of some form. My apologies if it is difficult to follow.
* consider the extreme: we have a database of every possible thought along with the corresponding brainstate for a given individual
† I'm assuming that you grant this, although I accept that it's been a problem in past threads
When you want to work out what part of a brain does what, you can't just look at an individual person. you have to take multiple people and compare them to each other. If the part of the brain is relatively the same across multiple people, as macro structures such as "the hippocampus" tend to be, then you can make statements about them through this process of comparison.
However a persons individual thoughts are the function of their brains own particular wiring within these macro structures. You can only isolate a particular piece of brain activity from another if many people tend to activate the same pattern when engaging in the same activity.
There is always neuronal activation happening within the brain. Left alone with no stimulus, brains have a kind of idle pattern of spontaneous firing all over the entire structure. This still happens when someone is fully engaged in an activity. It's kind of like a huge amount of random static you have to deal with in order to be able to isolate any activity from another.
Then lets say you can isolate this: now you have to isolate which part of the many different pieces of neuronal activity relating to cat are not also related to remembering any animal, or remembering furry animals, or remembering ears and so on. All of these are activated when you remember something: semantic meaning is heirarchical with higher order details such as "is an animal versus a plant" being activated before you get to "is a furry animal". (This was deduced from studying semantic dementia, who symptoms are a progressive loss of semantic meaning that correlates with the brain damage caused by semantic dementia)
In this situation, everybodies neural setup of this structure within the macro structures that store these semantic meanings (because they're in specific parts of the brain) is different. One persons "is an animal" may be activating in the same place where anothers is activating "is a furry animal". Or more extreme but equally likely, one persons "rhodendron" is the same pattern of activation as another persons "coffee cup". Everybodies learning of semantic meaning is an individual thing in terms of what patterns of neuronal activation correspond to a particular piece of semantic meaning.
So for some things like this, it becomes impossible, logically impossible, to deduce any further. You can't separate the activity from other nearby activity and you can't use other people in comparison to achieve a benchmark for comparison. There's no way to tell what is what.
Some things you can because they are relatively stable across different people. For example your visual field is set up roughly the same and even corresponds to a 3d mapping of what an eye would see. So you can say that field a1 of this persons brain will be in the same region as this other persons brain and will be activated by light from this section of this persons eye. Hence you can make much stronger statements about what kinds of shape a person is seeing just by looking at their brain activity. But you can't do that with all "thought", only visual and some parts of movement. You can't do it for semantic meaning beyond the point of knowing that semantic meaning is generally stored in this part of the brain.
I think some of you fundamentally do not understand the brain. It is not like a computer. Memories and thoughts are not stored as physical data.
As reasons go they are limited to "it can't happen because it can't happen" which is not a very strong argument dude.
The obvious solution seems to be to be to do individual mappings based on correlations to external stimuli.
It also seems that the patterns have to be similar even if their spacial location is different. In other words, a structure corresponding to "is an animal" has to send the same output based on the same input no matter where it is or what the exact details are for it to actually be informationally identical.
What do you mean by individual mappings and how do you differentiate these simultaneous activations from each other. How do you determine wether it's animal or cat that is being activated when they are all activated. For all intents and purposes animal and cat activate at the same time, as one big neural pattern. How do you separate that pattern?
Also what are you basing your second assumption on? Because all I see is a giant leap into space but you seem to think it's obvious. I don't see it at all, can you explain where that came from please? I've played around with computer modelling of neural connections and you can represent the same information in different ways on such a network even for extremely simple things like approach behaviors. Let alone abstract semantic meaning.
In fact that second assumption is just flat out wrong based on what I know of neural network modelling. What corresponds to cat is the activation that occurs at the time each individual sees a cat. It's this activation that gets stored as cat and activated the next time it is seen. This activation is not identical, it's entirely dependent on the overal setup of the individual. It's pattern is dependent on the makeup of the persons brain and the situation it is seen in. All the patterns would be different. If you then upgrade this mental model to include the earlier point that it's all essentially one big neural pattern of animal, furry, legs, ears, paws, eyes, meow, etc including details that correspond to cat it is guaranteed that this pattern would be different.
Unless you have a very good argument for why they are all going to be identical that is based on some solid evidence and reasoning you are going to have to take the "obvious" out of your statements, as things are not as clear cut as you are assuming.
Well, nueroscience isn't exactly my specialty, I just know the end result of these "mind reading" labours.
It's just infinitely easier to understand when you use diagrams in something like this. I remember I actually had to go down after the lecture and talk to her afterwards to get it all through my head, and she had to bring up the diagrams again to do so.
So I feel like I wont be able to explain it unless I do that. I'm too tired tonight though.
Keeping things legitimate
Heard a song you really liked?
The gremlin will get rid of it
I fail to see how someone else having paid for a piece of content somehow gives you the right to acquire that content without paying anything to the creater/owner of the content, when you did not have that right before.
Either reselling is morally wrong, or piracy isn't.
And sometimes they're basically the same thing. Fun fact: My local Gamestop wouldn't accept returns on the Fallout 3 expansions, due to obvious piracy concerns (they were one-time install deals). Yet they sold the expansions used. At their customary tremendous markup.
There was a recent project with fMRI which was able to, after training, decode low-resolution images of what someone was seeing directly from the visual cortex. While I highly doubt remote telepathy can ever be a reality, barring say, us getting The Culture's hyperspace Minds, currently it seems like the limiting factor is our sensor technology.
I keep a look out for anything coming up which might let us see electrical activity in the brain, in real-time with spatial resolution.
so you could still just buy a used one and resell it and end up with a huge discount...
By also showing them a dog, a fish, a monkey and so forth and comparing them to identify specific elements by overlap. Like a Venn diagram.
To act as information storage it cannot be entirely arbitrary. It has to produce an appropriate output from the input. In other words, the range of reasonable and likely behavioral responses to seeing a cat is small (by comparison). So however the information about cats is stored, it has be a way such that when confronted with a cat, it will trigger an appropriate behavioral response.
Why can't it be arbitrary? Where are you getting these information rules from? You are working from some kind of axiom regarding information that I've never heard of.
And meanwhile some gamestop executive is yelling "c-c-c-combo breaker!" because they're earning a large multiple of their normal margin by buying these "used" games cheap and selling them high. And though each of these repeat sales earns gamestop cash, the devs see none of it.
I doubt it would stand up anyway, but it's a gambit I'm surprised hasn't been attempted.
Pluto was a planet and I'll never forget
I don't understand people's insistence on comparing the brain to computers and computer programming either. They have absolutely nothing in common, hell, they're not even made up of the same materials.
Gamestop would send assassins.
One is based on analysing areas related to planning (the prefrontal cortex and other areas of the frontal lobes) as well as areas related to emotional impulses (I assume anyway, there's no source there for me to look up) and another is based on analysing the place cells in the hippocampus that is responsible for forming memories of locations. Which is pretty much memory of where you have moved before.
I've already said somethings can be read this way so it's not a big surprise to me. I was talking about semantic meaning, which is located in different parts of the brain to what they would have been scanning.
The articles also got a heavy reporter spin on it. You'd be advised to downgrade claims of "reading minds" perfectly. This is a specific thing they are doing in both cases.
Using place cells like that is pretty ingenious though, I'm impressed by that.
I'd already heard of the other one, I can't remember where from though. It was familiar.
The brain is not at all homogenous when it comes to the way each structure within it performs it's function, so claiming to be able to generalise from what they are doing in each case is actually kind of silly of them. Just because they are an expert in one group of structures doesn't mean they know everything about the whole brain. They shouldn't be making claims like that. That's if the reporter didn't sensationalise what they really said, which is much more likely.
I wish they'd report their goddam sources for these things, it's really irritating when an article shoots off it's mouth about something interesting but never cites any published work. It makes it a real pain to look it up to see what they're actually doing.
It would be interpreted by your mind though.
What would be interpreted by whos mind?
Their own thoughts would be interpreted by themselves?
Sorry but that just, sounds like a massive waste of time.
Well done! After years of studying you we can finally know when you want to go get a meal and what you will cook! Huzzah for science!
For a disabled person, that could be absolutely huge.
Also... think bigger picture, if it's possible for one, then it's possible for all.
Imagine if as a child going to school included an hour brain test every day, you could have individual profiles on all students!
It would have absolutely horrifying possiblities to be used badly, and amazing possibilites for humans if used correctly.