well neurodivergence would be the main factor for me. It makes sense based on how perception is a multilayered process. Your eye sees only a small portion so most of your visual field is mainly from memory rather than live feed like it feels like. It also feels like not being able to acces the rawer data would be a wierd limitation, that the neurotypical way of seeing where you only experience the abstract deduction is a form of blindness. There is possible shift in focus that is I am not always aware of the mode where the noise is apparent. I would think this is similar with a synesthetic person that sees certain letters a certain color. When such a letter is written in a color that mismatches the brain probably has simultaneous opinions that it should both be color 1 and color 2 and emphasising text context or hue context could make one stand out more.
One of the theories for the type of divergence I have is context-blindness. That would explain that if a more typical brain has very strong magisteria for each kind of context they can’t cross-pollute as easily. Thus low-level pattern matching would be encapsulated to be invisible to the rest of the brain.
I guess with meditation black boxes become more white. The effect would depend a lot how how boxed things were to begin with. And it probably isn’t activity that is generated but just acknowledged. Thus it is not really hallucinations.
Althought even with proper hallucinations they have some strcture to them probably. Even the super crazy types could be understood by for example experiencing guilt by visual or auditory synesthesia. If it corresponds to a real brain state isn’t it in a sense accurate perception of a thing? One could think of a reinforcement learning agent that has one incentive strcture and suddenly shifts to a new one. Probably the old structures would be repurposed in a ad hoc way in service of the new goal. And while the abstraction used in that kind of zig-zag history would probably be weird and not happen in an agent trained directly for the later incentive structure it would probably be the most “directly ot the point” use of the old abstractions. In a similar way if the visual cortex is given raw visual data and then is tasked to produce any information that is useful to other brain parts there is no canon “right data” that would be the “real thing seen”. Rather there could be various kinds of information which could tell different useful stories. And what is deemed useful could depend what kinds of data other parts of the brain can utilise.
One of the theories for the type of divergence I have is context-blindness. That would explain that if a more typical brain has very strong magisteria for each kind of context they can’t cross-pollute as easily. Thus low-level pattern matching would be encapsulated to be invisible to the rest of the brain.
Thanks for pointing out “context-blindness”. Let me see if I’ve got this straight.
A neurotypical has these different contexts/magisteria where different rules and interpretations apply. Someone who is context-blind has trouble identifying different contexts and so applies a global set of rules and interpretations in all situation.(?)
And this relates to low-level patterns because these different contexts are actually just sort of arbitrary, or just social constructs, so they’re impossible to see when you’re only paying attention to low-level details (?)
Well having global set of rules is one feature of the context-blindness explanation for the kind of variation. One of the possible linkages is that appying this kind of universality to moral behaviour allows one to avoid being a hypocrite as you consistently and without fail apply the honed behaviour. In contrast a person with strong comparmentalization has trouble arriving at the generalization. You have “don’t steal at shop”, “don’t steal from family”, “don’t steal from classmates” ie you accumulate contexts where the behaviour is appropriate/inappropriate. If you get caught stealing and it feels punishing you form an opinion that “I should not steal here” (a very context sensitive person could go “I should not steal at the north end of this store” separate from “I should not steal at the south end of this store”) and do not form an opinion of “I should not steal”. In away you have to solve the same appropriateness problem all over in a new context. It need not be that the contextes are not recognised but their role in cognition is not so pronounced.
What I was more getting at I guess iddn’t write that explicitly is that context affects memory recall and attention too. That is when I try to see an object one doesn’t pay attention to a narrow group fo neurons but a whole big data dump most of which is probably irrelevant to the task at hand. However what you don’t recall can’t be used in the end product of the mental processing. For example in programming it is usual to declare most variables private. But you could declare all variables public. If you do the program constructs can use each others functionality. At the most extreme you could have the program function as a holistic whole where classes refer to each other public variables willy nilly. However if you a have code where there are lots of private variables you can be sure that those variables are refererenced from a narrow range that having the one class definition open you are aware of all the code that could influence it. Trying to do otherwise would not make the program compile or would raise a segfault.
But in a world where there is no segfaults if higher abstraction level is interested in the details it can explicitly go look at them. That is if I call a function and I know that as a side-effect of that the objects internal variables have changed if I want I can go read those variables.
[code]
result=fruit_detectorIsApple(blob)
curiosity=fruit_detector.pearness
[/code]
If pearness was declared private this would not be permitted
In brains it could be that if each brain region has a separate memory store that only it has access that would lead to a type of encapsulation. In the reverse if all functionalities dump their data to a common information store then they can interfere/cooperate. At one extreme all data could be sent all the time to all functionalities but each functionality only really digests a small portion of it. But while the production is made seeking for a particular important pieces of data a lot of secondary data would be floating around too. Or in reverse a brain that gets easily confused by garbage data might limit by only transitting information really required for the operations. And this involves hiding/destroying data that doesn’t directly answer questions it is asked. LIke in a math test you are supposed to show your midsteps but in this kind of arrangement the less steps revealed the better and preferably only the bottomline.
At one extreme all data could be sent all the time to all functionalities but each functionality only really digests a small portion of it.
Is this the context-blind extreme? and
Or in reverse a brain that gets easily confused by garbage data might limit by only transitting information really required for the operations
is the other extreme?
Rephrasing: All people need to filter data, but from which data? Context-blind filters global variables while context-sensitive filters from local variables.
Also, what about games/activities with explicit rules such as chess or programming languages? Wouldn’t everyone be able to identify those contexts and apply the right rules? (Assume they know the rules)
Well one of the other symptoms sensory-overload could be interpreted as not doing the filtering (I myself don’t exhibit that so much but it is connected). In that way it is not strictly neccesary. It’s also a multistage process so you might have a global-local-global-local alteration on different parts of the hierachy.
It isn’t that absolute and while everyone probably can manage to follow the rules there might be a difference how effortful it is. The theory might not be detailed enough to address questions on that level and i don’t have the most up to date familiarity with it (having wrong theory can do a lot of harm and it has fluxed quite a bit). While it is not context-blindness the related trait of literalmindedness would help with explicit rules as you don’t have to “apply common sense” but just “execute”. In a situation where there are literal rules to be followed and context-sensitive course of action context-blindness would drop the context sensitive option from being relevant. [What I think was such a conflict] (https://www.lesswrong.com/posts/G5TwJ9BGxcgh5DsmQ/yes-requires-the-possibility-of-no#oyoNqpuaanWcXC4uG) a context heavy person might not even realise that a literal interpretation was possible.
In a way justice is supposed to be blind in a very near sense. If law is being applied to persons differently it easily and quickly becomes unfair. But if there is no special adhereing to such principles the application tends to get uneven.
Thanks, I’ve honestly learned so much throughout our comment thread.
One thing I’m confused about it why/how local contexts recognized by neurotypicals.
Maybe “mimic high-status members of in-group” explains most of it(?), or “what’s other people doing?” or “what would someone else in my current role do?”
I think that’s confused because if I know what “role” I’m in, then I already have a context in mind, and I’m trying to figure out how that context is derived in the first place!
Maybe contexts feel more solid/real to neurotypicals. “School” feels like a real/solid thing (even though it’s just a building where kids …). “Money” feels like it’s real/solid (even though it’s just paper or a number in a database with a socially agreed upon value attached). Being a “Good Student” feels real/tangible (even though it’s just writing notes directly from the board and …)
Those 3 examples are definitely things I felt were real/solid/tangible and I didn’t connect the “even though it’s …” definitions until highschool/ undergrad.
It does not need to feel like context on the inside and arguably if you are recognising you are in a context you are thinking about the situation in a certain situation-independent way.
I don’t know if the analog hold but a typical reinforcement neural network upon error just backpropagates a weigth adjustment. One could think that weights that are moved a lot are interpret to be “very in context” and weights that are moved a lilttle are “somewhat out of context” which would lead a very fuzzy sense of context where there are no hard lines (well before they are reinforced into place). While it might not be realistic it would be computationally tractable to compare two neural networks which are more sharp or diffuse in the propagation weighting which could lead to a different structure in the high-fit state (or different times for reaching that high fit state).
There are a lot of prejudices so there tends to be hiding of these things if they are not strictly neccesary. It feels good to be seen and have that curiosity and openness be a positive interaction.
No wait, this doesn’t make sense framed this way. I think everyone isn’t context-blind when the rules are explicit. If we’re playing tag, or chess, or programming in Python, (I think) most people know which rules apply in this context because those rules are more explicit.
If so, maybe it’s contexts with implicit rules? And implicit rules are learned by mimicking other’s reactions?
I think I understand your first point connecting “not seeing rawer data” and a synesthetic person having a mismatched letter/color. I think your main point is: You do/don’t see rawer data depending on the context. (Also, can you choose to see the rawer data, or choose to only the abstract deduction?)
I guess with meditation black boxes become more white. The effect would depend a lot how how boxed things were to begin with. And it probably isn’t activity that is generated but just acknowledged. Thus it is not really hallucinations
What is “not really hallucinations” here? The 3 tests above? Also, what do you mean by hallucinations in this context?
Well my point was that the experience isn’t that your vision is replaced by another sensation. In the linked slatestarcodex there is a comparison picture. That kind of thing might suggest that the visual snow would appear the same as if there were mist or something. But it in fact superimposes or some relation which would only make sense in perceptual analysis. Like if you repeat the same word multiple times it can fail to seem like a word. But you are still aware of the all the phonemes/letters of the word. It would be weird if somebody could hear the word but could not hear the individual letters. And hearing the letters doesn’t interfere with hearing the word. Saying that hearing single letters would be “hallucinating things that are not there” would be really backwards. So in vision when I can see the rawer visual data I am not seeing stuff that isn’t there.
If you have a correctly working monitor and take it appart and study it’s function it will stay as a functional monitor. If you wire it differntly then it might function differently but if you refrain from rewiring it stays correctly working. If you look inside and see how your visual cortex works you might change your opinion on your visual cortex but it is unlikely that it started to act up just because you looked into it (in the cognitive sense). On the opposite pole if you intentionally set out to imagine a picture of a apple if your visual cortex complies and provides a red apple picture that would be a hallucination. But if it shares what it already has anyway there is no fraudulent component. If it happens during normal operation it is not an artifact even if you were not aware of it’s existence. There is some good quote that has parts to the effect of “People can handle the truth for they are already enduring it ”
I agree. These all feel like very real sensory information. This is in contrast to being in sleep paralysis and creating extra sensory information or in very vivid dreams, since in both of these cases I realize afterwards “Oh, those weren’t real” as in, I didn’t actually receive that sensory information.
Also, I made a mistake in my initial post, my correction is separating different things that might be confused with “visual snow” such as:
1. Visual Snow—Like a million very tiny dots. Very much like static/white noise in the wiki. More visible in low light conditions or when you’re tired. I saw it for the first time this (8/12) morning in low-light conditions.
2. Patterned lines (?) - Like the geometric/kaleidoscopic shape in this picture. Doesn’t have to be that consistent or patterned but is better described by “lines” than either of the other two. This is what I meant by “jumpy spiderwebs made out of light” and what I thought visual snow was.
3. Blue-sky Sprites—The picture is a nice animation (can be seen without looking at the blue sky but apparently it’s more prominent in that case). Dots and wisps the size of a mm or a little bigger. Maybe 5-100 at a time vs the million in “visual snow”. Resembles afterimages and the “black stars” when feeling faint.
4. (Also very possible there’s more that I’ve missed)
Visual snow is obvious.
Thanks for commenting that like I requested!
What are your thoughts on the connection to seeing these and neurodivergence, meditation, and psychedelic drug use?
well neurodivergence would be the main factor for me. It makes sense based on how perception is a multilayered process. Your eye sees only a small portion so most of your visual field is mainly from memory rather than live feed like it feels like. It also feels like not being able to acces the rawer data would be a wierd limitation, that the neurotypical way of seeing where you only experience the abstract deduction is a form of blindness. There is possible shift in focus that is I am not always aware of the mode where the noise is apparent. I would think this is similar with a synesthetic person that sees certain letters a certain color. When such a letter is written in a color that mismatches the brain probably has simultaneous opinions that it should both be color 1 and color 2 and emphasising text context or hue context could make one stand out more.
One of the theories for the type of divergence I have is context-blindness. That would explain that if a more typical brain has very strong magisteria for each kind of context they can’t cross-pollute as easily. Thus low-level pattern matching would be encapsulated to be invisible to the rest of the brain.
I guess with meditation black boxes become more white. The effect would depend a lot how how boxed things were to begin with. And it probably isn’t activity that is generated but just acknowledged. Thus it is not really hallucinations.
Althought even with proper hallucinations they have some strcture to them probably. Even the super crazy types could be understood by for example experiencing guilt by visual or auditory synesthesia. If it corresponds to a real brain state isn’t it in a sense accurate perception of a thing? One could think of a reinforcement learning agent that has one incentive strcture and suddenly shifts to a new one. Probably the old structures would be repurposed in a ad hoc way in service of the new goal. And while the abstraction used in that kind of zig-zag history would probably be weird and not happen in an agent trained directly for the later incentive structure it would probably be the most “directly ot the point” use of the old abstractions. In a similar way if the visual cortex is given raw visual data and then is tasked to produce any information that is useful to other brain parts there is no canon “right data” that would be the “real thing seen”. Rather there could be various kinds of information which could tell different useful stories. And what is deemed useful could depend what kinds of data other parts of the brain can utilise.
Thanks for pointing out “context-blindness”. Let me see if I’ve got this straight.
A neurotypical has these different contexts/magisteria where different rules and interpretations apply. Someone who is context-blind has trouble identifying different contexts and so applies a global set of rules and interpretations in all situation.(?)
And this relates to low-level patterns because these different contexts are actually just sort of arbitrary, or just social constructs, so they’re impossible to see when you’re only paying attention to low-level details (?)
Well having global set of rules is one feature of the context-blindness explanation for the kind of variation. One of the possible linkages is that appying this kind of universality to moral behaviour allows one to avoid being a hypocrite as you consistently and without fail apply the honed behaviour. In contrast a person with strong comparmentalization has trouble arriving at the generalization. You have “don’t steal at shop”, “don’t steal from family”, “don’t steal from classmates” ie you accumulate contexts where the behaviour is appropriate/inappropriate. If you get caught stealing and it feels punishing you form an opinion that “I should not steal here” (a very context sensitive person could go “I should not steal at the north end of this store” separate from “I should not steal at the south end of this store”) and do not form an opinion of “I should not steal”. In away you have to solve the same appropriateness problem all over in a new context. It need not be that the contextes are not recognised but their role in cognition is not so pronounced.
What I was more getting at I guess iddn’t write that explicitly is that context affects memory recall and attention too. That is when I try to see an object one doesn’t pay attention to a narrow group fo neurons but a whole big data dump most of which is probably irrelevant to the task at hand. However what you don’t recall can’t be used in the end product of the mental processing. For example in programming it is usual to declare most variables private. But you could declare all variables public. If you do the program constructs can use each others functionality. At the most extreme you could have the program function as a holistic whole where classes refer to each other public variables willy nilly. However if you a have code where there are lots of private variables you can be sure that those variables are refererenced from a narrow range that having the one class definition open you are aware of all the code that could influence it. Trying to do otherwise would not make the program compile or would raise a segfault.
But in a world where there is no segfaults if higher abstraction level is interested in the details it can explicitly go look at them. That is if I call a function and I know that as a side-effect of that the objects internal variables have changed if I want I can go read those variables. [code] result=fruit_detectorIsApple(blob)
curiosity=fruit_detector.pearness [/code] If pearness was declared private this would not be permitted
In brains it could be that if each brain region has a separate memory store that only it has access that would lead to a type of encapsulation. In the reverse if all functionalities dump their data to a common information store then they can interfere/cooperate. At one extreme all data could be sent all the time to all functionalities but each functionality only really digests a small portion of it. But while the production is made seeking for a particular important pieces of data a lot of secondary data would be floating around too. Or in reverse a brain that gets easily confused by garbage data might limit by only transitting information really required for the operations. And this involves hiding/destroying data that doesn’t directly answer questions it is asked. LIke in a math test you are supposed to show your midsteps but in this kind of arrangement the less steps revealed the better and preferably only the bottomline.
Is this the context-blind extreme? and
is the other extreme?
Rephrasing: All people need to filter data, but from which data? Context-blind filters global variables while context-sensitive filters from local variables.
Also, what about games/activities with explicit rules such as chess or programming languages? Wouldn’t everyone be able to identify those contexts and apply the right rules? (Assume they know the rules)
Well one of the other symptoms sensory-overload could be interpreted as not doing the filtering (I myself don’t exhibit that so much but it is connected). In that way it is not strictly neccesary. It’s also a multistage process so you might have a global-local-global-local alteration on different parts of the hierachy.
It isn’t that absolute and while everyone probably can manage to follow the rules there might be a difference how effortful it is. The theory might not be detailed enough to address questions on that level and i don’t have the most up to date familiarity with it (having wrong theory can do a lot of harm and it has fluxed quite a bit). While it is not context-blindness the related trait of literalmindedness would help with explicit rules as you don’t have to “apply common sense” but just “execute”. In a situation where there are literal rules to be followed and context-sensitive course of action context-blindness would drop the context sensitive option from being relevant. [What I think was such a conflict] (https://www.lesswrong.com/posts/G5TwJ9BGxcgh5DsmQ/yes-requires-the-possibility-of-no#oyoNqpuaanWcXC4uG) a context heavy person might not even realise that a literal interpretation was possible.
In a way justice is supposed to be blind in a very near sense. If law is being applied to persons differently it easily and quickly becomes unfair. But if there is no special adhereing to such principles the application tends to get uneven.
Thanks, I’ve honestly learned so much throughout our comment thread.
One thing I’m confused about it why/how local contexts recognized by neurotypicals.
Maybe “mimic high-status members of in-group” explains most of it(?), or “what’s other people doing?” or “what would someone else in my current role do?”
I think that’s confused because if I know what “role” I’m in, then I already have a context in mind, and I’m trying to figure out how that context is derived in the first place!
Maybe contexts feel more solid/real to neurotypicals. “School” feels like a real/solid thing (even though it’s just a building where kids …). “Money” feels like it’s real/solid (even though it’s just paper or a number in a database with a socially agreed upon value attached). Being a “Good Student” feels real/tangible (even though it’s just writing notes directly from the board and …)
Those 3 examples are definitely things I felt were real/solid/tangible and I didn’t connect the “even though it’s …” definitions until highschool/ undergrad.
It does not need to feel like context on the inside and arguably if you are recognising you are in a context you are thinking about the situation in a certain situation-independent way.
I don’t know if the analog hold but a typical reinforcement neural network upon error just backpropagates a weigth adjustment. One could think that weights that are moved a lot are interpret to be “very in context” and weights that are moved a lilttle are “somewhat out of context” which would lead a very fuzzy sense of context where there are no hard lines (well before they are reinforced into place). While it might not be realistic it would be computationally tractable to compare two neural networks which are more sharp or diffuse in the propagation weighting which could lead to a different structure in the high-fit state (or different times for reaching that high fit state).
There are a lot of prejudices so there tends to be hiding of these things if they are not strictly neccesary. It feels good to be seen and have that curiosity and openness be a positive interaction.
No wait, this doesn’t make sense framed this way. I think everyone isn’t context-blind when the rules are explicit. If we’re playing tag, or chess, or programming in Python, (I think) most people know which rules apply in this context because those rules are more explicit.
If so, maybe it’s contexts with implicit rules? And implicit rules are learned by mimicking other’s reactions?
I think I understand your first point connecting “not seeing rawer data” and a synesthetic person having a mismatched letter/color. I think your main point is: You do/don’t see rawer data depending on the context. (Also, can you choose to see the rawer data, or choose to only the abstract deduction?)
What is “not really hallucinations” here? The 3 tests above? Also, what do you mean by hallucinations in this context?
Well my point was that the experience isn’t that your vision is replaced by another sensation. In the linked slatestarcodex there is a comparison picture. That kind of thing might suggest that the visual snow would appear the same as if there were mist or something. But it in fact superimposes or some relation which would only make sense in perceptual analysis. Like if you repeat the same word multiple times it can fail to seem like a word. But you are still aware of the all the phonemes/letters of the word. It would be weird if somebody could hear the word but could not hear the individual letters. And hearing the letters doesn’t interfere with hearing the word. Saying that hearing single letters would be “hallucinating things that are not there” would be really backwards. So in vision when I can see the rawer visual data I am not seeing stuff that isn’t there.
If you have a correctly working monitor and take it appart and study it’s function it will stay as a functional monitor. If you wire it differntly then it might function differently but if you refrain from rewiring it stays correctly working. If you look inside and see how your visual cortex works you might change your opinion on your visual cortex but it is unlikely that it started to act up just because you looked into it (in the cognitive sense). On the opposite pole if you intentionally set out to imagine a picture of a apple if your visual cortex complies and provides a red apple picture that would be a hallucination. But if it shares what it already has anyway there is no fraudulent component. If it happens during normal operation it is not an artifact even if you were not aware of it’s existence. There is some good quote that has parts to the effect of “People can handle the truth for they are already enduring it ”
I agree. These all feel like very real sensory information. This is in contrast to being in sleep paralysis and creating extra sensory information or in very vivid dreams, since in both of these cases I realize afterwards “Oh, those weren’t real” as in, I didn’t actually receive that sensory information.
Also, I made a mistake in my initial post, my correction is separating different things that might be confused with “visual snow” such as:
Did you see visual snow as in #1? And the others?
Atleast 1 and I guess 3 too but no 2