[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 113 KB, 974x727, DTI.jpg [View same] [iqdb] [saucenao] [google]
2403816 No.2403816 [Reply] [Original]

Dear /sci/

Why you no post good threads :_;

Here, have some neuroscience.

>> No.2403818
File: 58 KB, 976x732, DTI2.jpg [View same] [iqdb] [saucenao] [google]
2403818

>> No.2403820
File: 73 KB, 975x731, DTI3.jpg [View same] [iqdb] [saucenao] [google]
2403820

>> No.2403822
File: 88 KB, 961x846, brain2.jpg [View same] [iqdb] [saucenao] [google]
2403822

>> No.2403823
File: 24 KB, 360x357, Gage.jpg [View same] [iqdb] [saucenao] [google]
2403823

>> No.2403824
File: 44 KB, 458x440, association connections.gif [View same] [iqdb] [saucenao] [google]
2403824

>> No.2403828

Do you know somehting about the pictures you're posting? If so, would you like to expound on them? I'd love to red about neuroscience, not just stare at pretty pictures.

>> No.2403841

PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS PENIS

>> No.2403845

>>2403828
The first three images are the result of diffusion tensor white matter tractography. They show various fiber bundles imaged in vivo. The fourth image shows the topographical layout of a complex network for motor execution, goal selection and action perception of a chimpanzee (I think, could be a rhesus macaque). The fifth image shows the reconstructed damage to the brain that happened to one of the most famous case studies in neuroscience: phineas gage. The sixth image just shows a schematic representation of the general layout of association fibers in the cortex.

>> No.2403846

So...can a computer with sufficient complexity fully simulate a human brain?

>> No.2403851
File: 216 KB, 725x592, figure7-12.jpg [View same] [iqdb] [saucenao] [google]
2403851

This is the anatomical layout of the human visual system.

>> No.2403859

>>2403846
Sure
>with sufficient complexity
This is where the problem lies though. This, plus the fact that we don't know how to simulate everything exactly.

>> No.2403863

>>2403846
Yes of course, though practically it would need massive parallelism that current computers don't have. But absolutely yes.

>> No.2403867
File: 291 KB, 1117x1280, F1.large.jpg [View same] [iqdb] [saucenao] [google]
2403867

This image shows a computer generated projection of brain structures, used to determine the exact location of where intracranial electrodes will be placed (judging by the location, probably for the treatment of Parkinson's)

>> No.2403869

>>2403859

Super computers will reach human brain speeds in about 3-4 years. 10^16 flops is the lower bound.

So what are we missing?

>> No.2403874

>>2403869
The correct software. But we're getting there.

>> No.2403882

>>2403869
>So what are we missing?
Accurate knowledge of what exactly to simulate. We don't even have a complete anatomical record of the human brain. Post-mortem investigation can only give you so much, and in vivo anatomical mapping, but more importantly, functional mapping has only started to take off in the last two decades or so. There are parts of the brain we don't really have clear ideas of, like the medial frontal cortex. It seems to do a whole bunch of crap and projects towards pretty much any other brain region. There are parts of the brain we know a lot more about, like the primary visual cortex. I think simulations of cortical columns in V1 will have to be established first (and progress is being made in that field).

>> No.2403890
File: 31 KB, 400x299, epilepsy_spring05_fig5.jpg [View same] [iqdb] [saucenao] [google]
2403890

This image shows the placement of a mesh of intracranial electrodes for the purpose of locating epileptogenic tissue.

>> No.2403891

>>2403882
>>2403874

Would a computer brain, accurately simulated, develop it's own intelligence spontaneously?

>> No.2403897

>>2403891
If it's simulated accurately it would have the same properties as a human brain. A simulation will almost certainly be an approximation of the true complexity though. I don't think we'll see something as complex as an actual human brain within our lifetime.

>> No.2403900

>>2403891
It's all about the neural network. If you copied a person's neural network exactly it would already be intelligent. You would have copied that person.

>> No.2403913
File: 20 KB, 573x308, salmonfMRI2.jpg [View same] [iqdb] [saucenao] [google]
2403913

This image shows a salmon. Yes, an actual salmon. The paper that goes with it was meant as sort of a joke, to illustrate the importance of false positive correction.

>> No.2403919

I'm starting to run out of images.

>> No.2403933
File: 221 KB, 1077x690, cortex.gif [View same] [iqdb] [saucenao] [google]
2403933

This image shows a schematic representation of the typical 6 layers of the cortex, and the different types of neurons therein.

>> No.2403938
File: 658 KB, 191x138, 1292768822749.gif [View same] [iqdb] [saucenao] [google]
2403938

>>2403933
I understand completely

>> No.2403941
File: 152 KB, 650x712, duckataurus.jpg.jpg [View same] [iqdb] [saucenao] [google]
2403941

This image shows what abominations science has created.

>> No.2403948

>>2403938
Well, it might look sort of complicated, but the point is there are 6 layers and a bunch of different neurons. The type of neuron indicated by the letter 'P' for instance is a cortical pyramidal neuron. They're quite big and responsible for a big part of cortical processing.

>> No.2403952

>>2403900

If aliens came down every night when you were asleep and replaced you with an exact clone, from the clone's perceptive life goes on as usual. You could never know whether you are this clone or not. Since every cell in your body dies and is replaced over time, this is essentially happening.

So what would happen if your mind was fatally uploaded one night while you were asleep?

>> No.2403961

>>2403952
Well if you mean it's copied then you have created a copy of your mind which doesn't realize it's a copy (and it doesn't matter either, it is just as real).

Neuroscience is going to make life very hard for religious people and spiritualists.

>> No.2403963

>>2403952
Actually, the majority of neurons are never replaced.

In addition, I think this story will be much to your liking: http://www.newbanner.com/SecHumSCM/WhereAmI.html
It's a bit of a read but it's definitely worth it.

>> No.2403973

>>2403963
I started reading and quickly I was like "THIS IS FUCKING BULLSHIT RAGEEWFKGASFDBFUUU" but then I realized it's fiction.

>> No.2403978

>>2403961
>Neuroscience is going to make life very hard for religious people and spiritualists.
It's already being misinterpreted by lots of people like Beauregard (the guy from the article about the neural correlates of spiritual experience in nuns) and alternative medicine people that think the pineal gland is the seat of the soul.

Just when you think a bunch of horse crap has been disproved they misinterpret it and see it as evidence for their beliefs.

>> No.2403981

>>2403973
Lol, I had the same thing when I read it for the first time.

>> No.2403987
File: 7 KB, 224x217, 1251470300691.jpg [View same] [iqdb] [saucenao] [google]
2403987

>>2403978

>> No.2403994
File: 163 KB, 1127x845, brains.jpg [View same] [iqdb] [saucenao] [google]
2403994

doinitrite?

>> No.2403997

>>2403987
It's best just to ignore these people altogether.

>> No.2404002
File: 308 KB, 800x388, 1268859709104.png [View same] [iqdb] [saucenao] [google]
2404002

>> No.2404014

>>2403963
>Actually, the majority of neurons are never replaced.

Do you have a source for that? I thought all cells are replaced but the pattern persists.

>> No.2404016

>>2403845

Re: the first three images. I had no idea we could do that. Amazing. Is that how they determined the network's layout for the fourth picture? And I knew I recognized that fifth picture from something! I saw a documentary on it years ago.

I'm a lot more familiar with the next two pictures, thank you.

>> No.2404025

>>2404014
Actually, I don't have a source for that. Neurogenesis only happens in certain regions of the brain though, and generally only highly plastic structures like the hippocampus show signs that new neurons are created. It might be on the neurogenesis wiki page.

>> No.2404036

>>2404016
>I had no idea we could do that. Amazing. Is that how they determined the network's layout for the fourth picture?
Yeah, it's a pretty cool technique. It exploits the principle of magnetic resonance to track anisotropic diffusion (follow where the water goes) and construct fiber bundles from this signal.

There's no need to do this when studying monkeys though (the fourth pic) because you can invasively measure neural activity directly and map functional connectivity in this way.

>> No.2404055

>>2403890
The mere fact that we can do something like this blows my mind. It's fascinating.

>>2403913
I remember that study from a biostatistic class. As the teacher put it: "Only when we correctly, responsibly, interpret and arrange the data obtained do we acquire information. The rest is fish talk". Fish talk being Spanish slang for "talking nonsense".

>> No.2404066

I like this thread and the people that post in it. :3

This is why I visit /sci/.

>> No.2404076
File: 1.29 MB, 1796x2792, P.jpg [View same] [iqdb] [saucenao] [google]
2404076

>>2403938

That's the simplified version. Here's how a pyramidal neuron (and only a pyramidal neuron) looks under the microscope - after treating it to make it visible, of course.

>> No.2404083

>>2404002

IIRC a kid was killed in a similar accident. Someone left a fucking oxygen tank in the room while the kid was being scanned.

>> No.2404094

Can we 'upload' our mind to a computer and then 'download' that to a new body?

>> No.2404098

>>2404083
>left a fucking oxygen tank in the room
You do know you can't turn the B0 field off right? You can't leave an oxygen tank in the room because as soon as it enters the room it gets sucked into the scanner. But maybe you meant that instead. These accidents are always sad.

>> No.2404100

>>2404036

That's so clever, and so simple. Damn. I can't remember who said it, but they were right to say that you can usually spot true brilliancy when, after it's explained to you, you say, "Of course! How didn't it occur to me!?"

Usually accompanied by facepalming.

>> No.2404117

>>2404100
The people who developed the first machine that measures and images magnetic resonance got the Nobel prize for it. It's super awesome.

>> No.2404144

>>2404100
Lol, it sounds simple enough, but in reality is pretty complex.

>> No.2404161

>>2404098

Yes, sorry. I meant to type "into" and then I don't know, something happened to my brain. I think the incident was discussed in an article published by the AJR. I'll see if I can find it.

>> No.2404185

>>2404117

Super, super awesome!

>>2404144

Yeah, I can't imagine the actual process being anything but complicated beyond words, but the idea is simple enough to make sense to almost anyone, if explained with the right words.

>> No.2404232

>>2404161
Yeah these accidents (unfortunately) are somewhat common. I once heard of an incident where a guy was walking by a scanning room when holding a fire extinguisher and the thing got sucked through the door and into the scanner, killing the person in it.

>> No.2404349

Let's put this bitch back on the front page.

>> No.2404369
File: 23 KB, 684x537, Stomatogastric_Ganglion_Fig8.gif [View same] [iqdb] [saucenao] [google]
2404369

This is a diagram of the stomatogastric ganglion of a lobster.

Read more here:
http://www.scholarpedia.org/article/Stomatogastric_ganglion

>> No.2405616

might as well bump this