Cube Flipper

As a human being, the kindest thing you can do to your brain is to not think.

[Cube Flipper @cube_flipper]

it has been a little bit more than three years since i first started writing on https://t.co/o9UU8aBsq2. i would like to pause blogging for now! here is a retrospective to cap things off: https://t.co/AYXwxY0wqJ

[Cube Flipper @cube_flipper]

@ember_arlynx YES. the olfactory bulb is an issue. i have some ideas but they are too handwavey to tweet

[Cube Flipper @cube_flipper]

the way i thought about the brain changed quite significantly when i first encountered a diagram similar to this one (in a John C. Lilly book).

in mammals earlier than humans you (very very broadly speaking) had the back half of the brain devoted to sensory cortices and the front half the brain to motor output. input = green areas, output = purple areas.

as things got more complicated evolution tacked on more and more associative cortices adjacent to the sensorimotor ones

i imagine the sensorimotor cortices to be dealing with low-level shapes which are structurally isomorphic to the incoming information, and the associative cortices to be dealing with more and more abstruse recombinant correlation-based representations as they get further away from the sensory cortices

so, if you follow the cortical gradient/predictive processing gradient, i expect you also see a gradient from more shape-based sensory representations to more abstract/symbolic representations

[Cube Flipper @cube_flipper]

@tenobrus did you catch this

[Cube Flipper @cube_flipper]

this is relevant if you are still thinking about the abstraction fallacy paper

susan pockett suggested it was only the sensory cortices which would be conscious (in the holistic field behaviour sense), as they deal with more redundant, higher correlated information (i.e., bigger, more correlated LFPs), whereas the associative cortices further up the predictive processing stack would be less "conscious" given those neurons deal with sparser, higher dimensional, more tightly compressed representations which would be indistinguishable from white noise (https://t.co/wnVcedVkGZ)

now, perhaps this is implausible simply for the reason that the sensory cortices are far apart! how do you get a unified world simulation from those? those sensory cortices all have thalamocortical projections down to their corresponding thalamic nuclei, which basically are all in the same place, so perhaps it's the thalamus where the nice shapely world simulation lives

[Cube Flipper @cube_flipper]

@nosilverv hells yeah good question i don't think anybody truly answered it correctly anyway i lvoe reddit

[Cube Flipper @cube_flipper]

@nosilverv https://t.co/B3tVaWfMO8

[Cube Flipper @cube_flipper]

@MasterTimBlais i need to talk to this person

[Cube Flipper @cube_flipper]

thinking about this a second time – if @corsaren's piece was hypothetically rewritten in a way which leant upon relational rather than intrinsic qualia (as you have explored), i believe it would still wind up drawing the same conclusions about consciousness in digital minds. if not, then i don't understand your position, i would like to hear the explanation

[Cube Flipper @cube_flipper]

@jessi_cata @de_lagunez @corsaren like i mostly just feel like "herp a derp it's all fake qualia don't exist" is the wrong interpretation of this position, and is entirely dependent upon the language used

[Cube Flipper @cube_flipper]

@jessi_cata @foomagemindset insofar as qualia might not have intrinsic nature independent of physics, this would just mean that qualia and physics are two sides of the same coin.

[Cube Flipper @cube_flipper]

@kyleostboe see also, from the paper you sent

[Cube Flipper @cube_flipper]

@kyleostboe part of Ward's argument is that generally speaking cortical damage causes all kinds of remarkable degradations of experience but if the thalamus gets damaged it's lights out

[Cube Flipper @cube_flipper]

@kyleostboe oh that's very interesting. yeah it's really hard to probe the thalamus.

i do like this paper (Ward & Guevara 2022) https://t.co/zrpJeFt3oe as well as its predecessor, (Ward 2011) https://t.co/Az40ONjwpw

[Cube Flipper @cube_flipper]

@SpiritDirtbag this is of course a massive massive massive generalisation, if you go look at rat brain anatomy you might find the amount of real estate devoted to whiskers and smell striking

[Cube Flipper @cube_flipper]

@deepfates @1a3orn some more notes pertaining to this way of thinking here https://t.co/xbqijPIvKi

[Cube Flipper @cube_flipper]

the way i thought about the brain changed quite significantly when i first encountered a diagram similar to this one (in a John C. Lilly book).

in mammals earlier than humans you (very very broadly speaking) had the back half of the brain devoted to sensory cortices and the front half the brain to motor output. input = green areas, output = purple areas.

as things got more complicated evolution tacked on more and more associative cortices adjacent to the sensorimotor ones

i imagine the sensorimotor cortices to be dealing with low-level shapes which are structurally isomorphic to the incoming information, and the associative cortices to be dealing with more and more abstruse recombinant correlation-based representations as they get further away from the sensory cortices

so, if you follow the cortical gradient/predictive processing gradient, i expect you also see a gradient from more shape-based sensory representations to more abstract/symbolic representations

[Cube Flipper @cube_flipper]

i should expand on how i think about this model a bit, like the gist of it is that the field serves as a holistic guiding force for large scale synchronisation of neural oscillations

but broadcasting LFPs is expensive, ideally the neurons should learn how the global field behaves and mostly oscillate in sync with that. minimal friction flow state, minimal "consciousness" perhaps

if a prediction error comes in, that should create a sizeable disturbance in the force, neurons will notice, neurons will need to update their models somehow. more intense "consciousness"

(susan pockett also noticed that feed-back activity produces relatively large LFPs, whereas feed-forward activity does not)

i would also expect holistic field effects to dominate in younger organisms before their neurons have figured out how to predict an extremely wide amount of sensory input. drugs also mess with this, pushing dynamics out of homeostasis and causing field effects to dominate

[Cube Flipper @cube_flipper]

@deepfates @1a3orn the idea that a embodiment/sense organs are a requirement for consciousness is approximately as insane to me as the idea that complexity is required for consciousness. i should really write a takedown of this

[Cube Flipper @cube_flipper]

@kepe__ as i gather 5-meo can have a pretty strong bliss attractor for some people but i found it hard to get that to work

[Cube Flipper @cube_flipper]

@kepe__ yeah, it's interesting because unlike other drugs you have to sort of incline that way deliberately