#1
Imagine tapping into the mind of a coma patient, or watching one’s own dream on YouTube. With a cutting-edge blend of brain imaging and computer simulation, scientists at the University of California, Berkeley, are bringing these futuristic scenarios within reach.

Using functional Magnetic Resonance Imaging (fMRI) and computational models, UC Berkeley researchers have succeeded in decoding and reconstructing people’s dynamic visual experiences – in this case, watching Hollywood movie trailers.

As yet, the technology can only reconstruct movie clips people have already viewed. However, the breakthrough paves the way for reproducing the movies inside our heads that no one else sees, such as dreams and memories, according to researchers.



Full article: http://newscenter.berkeley.edu/2011/09/22/brain-movies/
(Click the pic for video.)


Hot damn
Quote by Ian_the_fox
You're not girly enough of a boy for me, and you're not man enough to take the top. So like, sorry bitch but you ain't mine! Sorry.
#2
DOOD

that's pretty awesome.
Quote by korinaflyingv
On the come up we were listening to Grateful Dead and the music started passing through my bowel and out my arsehole as this violet stream of light. I shat music. It was beautiful.
#4
Hm.

From what I understand, it's not really creating pictures using only the thoughts, it's using a mathematical algorithm to blend together different existing video clips that are most likely to produce the brain-patterns recorded.

Interesting, but a lot less of a breakthrough than what I initially believed it to be.

EDIT: and from what I can tell by the clips, it's still mostly guessing and has huge trouble telling different but similar clips apart.
Last edited by CoreysMonster at Sep 23, 2011,
#5
That IS awesome.

The video kinda looked like I expected it to, which is odd. It's not like a movie flashback, but instead it has lots of uncertainty and different/incorrect images flash up etc
Quote by Renka
OddOneOut is an Essex S&M mistress and not a pirate or a computer program.

#6
Different write up from Slate

Yours is a little more in-depth though. Slate's makes it seem a bit magical. Still quite neat.

[IN PHIL WE TRUST]


Quote by Trowzaa
I only play bots. Bots never abandon me. (´・ω・`)

#7
This is totally misleading.

What they are promising... yeah... good luck. There are all sorts of psychoactives and non-psychoactives that heavily alter dream content and we have a really shitty structural understanding of the brain currently in terms of function, we can just watch our similar insanely complicated sacks of reverberation light up in different ways, but these scientists are at least twice as delusional as me announcing crap like that.

What the hell... claiming that just because they can shape objective content to how we perceive it that they have a lead on how to accurately put somebody's dreams on a television screen? wat?
#8
Quote by CoreysMonster
Hm.

From what I understand, it's not really creating pictures using only the thoughts, it's using a mathematical algorithm to blend together different existing video clips that are most likely to produce the brain-patterns recorded.

Interesting, but a lot less of a breakthrough than what I initially believed it to be.

EDIT: and from what I can tell by the clips, it's still mostly guessing and has huge trouble telling different but similar clips apart.

Well the idea is that you make a robot that learns to get better at guessing what clip they were watching, and then why. Then you substitute clips for shapes. Shapes for more detailed shapes. So on and so forth?

It might work, just depends on what the mind's eye is actually doing.
#9
Quote by captaincrunk
Well the idea is that you make a robot that learns to get better at guessing what clip they were watching, and then why. Then you substitute clips for shapes. Shapes for more detailed shapes. So on and so forth?

It might work, just depends on what the mind's eye is actually doing.





We can never know what the minds eye is doing through measure, because it is always augmented by DMT! (But really, neuroscientists are cocky.)

That idea might work out for them, let us see.
#10
Do not want.
Quote by SomeoneYouKnew
You should be careful what you say. Some asshole will probably sig it.

Quote by Axelfox
Yup, a girl went up to me in my fursuit one time.

Quote by Xiaoxi
I can fap to this. Keep going.
#11
Quote by garden of grey
This is totally misleading.

What they are promising... yeah... good luck. There are all sorts of psychoactives and non-psychoactives that heavily alter dream content and we have a really shitty structural understanding of the brain currently in terms of function, we can just watch our similar insanely complicated sacks of reverberation light up in different ways, but these scientists are at least twice as delusional as me announcing crap like that.

What the hell... claiming that just because they can shape objective content to how we perceive it that they have a lead on how to accurately put somebody's dreams on a television screen? wat?

this
Quote by Night
wtf is a selfie? is that like, touching yourself or something?
#12
████████████████████████████
███████████████████████████
█████████████████████████
██████████████████████████
███████████████████████████
███████████████████████████
███████████████████████████
███████████████████████████
#14
Violating the privacy of a coma patient's mind doesn't seem to be that brilliant of an idea, tbh.
Quote by SlinkyBlue
I remember when I was really young, I had a wet dream in which i was being dragged along an urban countryside by a pickup truck.

Don't ask me I have no idea how the hell it happened.




To Me:

Quote by Son.Of.TheViper

I love you
#15
Quote by Psychedelico
Violating the privacy of a coma patient's mind doesn't seem to be that brilliant of an idea, tbh.


Ever seen that shitty movie Monkeybone before, lol?
#16
Quote by garden of grey
http://www.deviantart.com/download/113138270/Demon_with_third_eye__by_soeven001.jpg


We can never know what the minds eye is doing through measure, because it is always augmented by DMT! (But really, neuroscientists are cocky.)

That idea might work out for them, let us see.
Quote by Kensai
Quote by Psychedelico
#17
Quote by captaincrunk
what I'm talking about doesn't even require knowledge of what is happening inside. Just computing accuracy of guesses until you've got some kind of picture creation.


How many guesses by computers to catch up with brain activity in real time?

Get started! I strongly suspect using computers that are not conscious to catch up with such is impossible.
Last edited by garden of grey at Sep 25, 2011,
#18
I have no problem with this technique if realized is used to help a comatose patient but that's not what this is going to be used for. Say hello mind police.
Quote by SomeoneYouKnew
You should be careful what you say. Some asshole will probably sig it.

Quote by Axelfox
Yup, a girl went up to me in my fursuit one time.

Quote by Xiaoxi
I can fap to this. Keep going.
#19
Quote by garden of grey
How many guesses by computers to catch up with brain activity in real time?

Get started! I strongly suspect using computers that are not conscious to catch up with such is impossible.

At the very least, it would likely require quantum computing.
Quote by Jackal58
I have no problem with this technique if realized is used to help a comatose patient but that's not what this is going to be used for. Say hello mind police.

Your Zoot-esque paranoia is showing
--------------╯╰--------------
A SIGNATURE.
--------------╮╭--------------
#21
Quote by garden of grey
This is totally misleading.

What they are promising... yeah... good luck. There are all sorts of psychoactives and non-psychoactives that heavily alter dream content and we have a really shitty structural understanding of the brain currently in terms of function, we can just watch our similar insanely complicated sacks of reverberation light up in different ways, but these scientists are at least twice as delusional as me announcing crap like that.

What the hell... claiming that just because they can shape objective content to how we perceive it that they have a lead on how to accurately put somebody's dreams on a television screen? wat?

"I don't understand it, so it's bull"

[IN PHIL WE TRUST]


Quote by Trowzaa
I only play bots. Bots never abandon me. (´・ω・`)

#22
Quote by DonGlover
At the very least, it would likely require quantum computing.

Your Zoot-esque paranoia is showing

Please step into the witness chamber Mr. Glover. Ladies and gentlemen of the jury have you seen enough?
Quote by SomeoneYouKnew
You should be careful what you say. Some asshole will probably sig it.

Quote by Axelfox
Yup, a girl went up to me in my fursuit one time.

Quote by Xiaoxi
I can fap to this. Keep going.
#23
Quote by garden of grey
How many guesses by computers to catch up with brain activity in real time?

Get started! I strongly suspect using computers that are not conscious to catch up with such is impossible.

Consciousness has nothing to do with it. It's just a matter of getting closer and closer by trial and error. Something computers are really, really good at.
#24
Quote by captaincrunk
Consciousness has nothing to do with it. It's just a matter of getting closer and closer by trial and error. Something computers are really, really good at.

Not to mention it's not mapping consciousness anyway so this whole this is a red herring.

[IN PHIL WE TRUST]


Quote by Trowzaa
I only play bots. Bots never abandon me. (´・ω・`)

#25
That video creeped me out..
dirtbag ballet by the bins down the alley
as i walk through the chalet of the shadow of death
everything that you've come to expect


#27
Quote by captaincrunk
Well the idea is that you make a robot that learns to get better at guessing what clip they were watching, and then why. Then you substitute clips for shapes. Shapes for more detailed shapes. So on and so forth?

It might work, just depends on what the mind's eye is actually doing.

There's a big difference between seeing things, reading the brain activity, and then using observation to determine what parts of the brain are triggered by the visual impulses and guess what clip it is through that, and constructing a tangible image out of the "mind's eye", which is a thousand times more abstract and individual.
#28
Quote by CoreysMonster
There's a big difference between seeing things, reading the brain activity, and then using observation to determine what parts of the brain are triggered by the visual impulses and guess what clip it is through that, and constructing a tangible image out of the "mind's eye", which is a thousand times more abstract and individual.

You're not thinking about the same thing I'm thinking about. What I'm proposing is doing it on one individual until you can predict images that aren't on the database using shapes and colors and such. It may be possible with enough trial and error. Once you've done it for one, try a few more, until you start getting enough data to make predictions and such about any brain more generally, if it turns out that such a thing is possible.

You just need to give the computer access to all the relevant data without knowing what that is. So trial and error.
#29
Quote by captaincrunk
You're not thinking about the same thing I'm thinking about. What I'm proposing is doing it on one individual until you can predict images that aren't on the database using shapes and colors and such. It may be possible with enough trial and error. Once you've done it for one, try a few more, until you start getting enough data to make predictions and such about any brain more generally, if it turns out that such a thing is possible.

You just need to give the computer access to all the relevant data without knowing what that is. So trial and error.

I understand what you're saying, but my argument is it isn't that simple because our inner eye doesn't work like that. When we imagine things they don't have a concrete form, because in our minds we mix the abstract thought of an object or action with very, very unclear images. When you think of the word "elephant", a dozen different things are shooting through your mind, many of them without you even realising it. On top of you hearing the word elephant, associating an abstract idea of what the word means and putting together a very unclear image of an elephant in your mind, you might also automatically think of a bunch of things that you associate with the word elephant, that aren't elephants.
#30
Quote by CoreysMonster
Hm.

From what I understand, it's not really creating pictures using only the thoughts, it's using a mathematical algorithm to blend together different existing video clips that are most likely to produce the brain-patterns recorded.

Interesting, but a lot less of a breakthrough than what I initially believed it to be.

EDIT: and from what I can tell by the clips, it's still mostly guessing and has huge trouble telling different but similar clips apart.


~don't finkdinkle when ur supposed to be dimpdickin~
#31
Basically, it's a chatterbot but with pictures... Similar concept. Libraries, algorithms, barely-sensible outputs.

If they ever did discover a way to make thoughts/dreams perfectly rendered in some video format, I'd love to record some of my dreams. They're usually so screwed up they'd make Stephen King and John Carpenter crying insomniacs.
#32
Quote by CoreysMonster
I understand what you're saying, but my argument is it isn't that simple because our inner eye doesn't work like that.

To be clear, I meant inner eye metaphorically.
Quote by CoreysMonster
When we imagine things they don't have a concrete form, because in our minds we mix the abstract thought of an object or action with very, very unclear images.

Do we? Do we really? How do you know, if you claim it to be unknowable?
Quote by CoreysMonster
When you think of the word "elephant", a dozen different things are shooting through your mind, many of them without you even realising it. On top of you hearing the word elephant, associating an abstract idea of what the word means and putting together a very unclear image of an elephant in your mind, you might also automatically think of a bunch of things that you associate with the word elephant, that aren't elephants.

Do we think "elephant" or the word elephant? And I get what you're saying, but there are reasons to believe this could work. When we concentrate, the activity increases, this much is shown to be true. When this is the case, it may be that the relevant "images" are "louder" in the fMRI.

At the very least, we'd learn something.
#34
Quote by captaincrunk
To be clear, I meant inner eye metaphorically.

as a metaphor for what?

Do we? Do we really? How do you know, if you claim it to be unknowable?

Unless I am a freak who visualises things extremely differently than all other humans on the planet, then yes, I do know. I also never said it was unknowable, I don't know where you got that from.
Do we think "elephant" or the word elephant? And I get what you're saying, but there are reasons to believe this could work. When we concentrate, the activity increases, this much is shown to be true. When this is the case, it may be that the relevant "images" are "louder" in the fMRI.

No, there is no reason to believe this could work, not the way you were describing it. Doing what you were suggesting is pretty much the same as trying out every single imaginable formula in hopes of one of them being the grand unified theory of butt science.

At the very least, we'd learn something.

I'd love to know what new knowledge was gained from the experiment in the OP, because there quite obviously wasn't any new discovery about how our brains work described in the article. All they did was make a computer program use brain-scan data to guess a picture out of a pile.
#35
Pretty interesting, but I don't want to see what's in my head.
*-)
Quote by Bob_Sacamano
i kinda wish we all had a penis and vagina instead of buttholes

i mean no offense to buttholes and poop or anything

Rest in Peace, Troy Davis and Trayvon Martin and Jordan Davis and Eric Garner and Mike Brown
#37
Not impressed even one bit. Get it? One bit. That's the comparative resolution this shit is spewing out compared to actual brain activity, which we are too stupid to understand.
Quote by denizenz
I'll logic you right in the thyroid.

Art & Lutherie
#38
Quote by CoreysMonster
as a metaphor for what?

Who knows?

Quote by CoreysMonster

Unless I am a freak who visualises things extremely differently than all other humans on the planet, then yes, I do know. I also never said it was unknowable, I don't know where you got that from.

Your strange claims that scientists are "cocky".
Quote by CoreysMonster

No, there is no reason to believe this could work, not the way you were describing it. Doing what you were suggesting is pretty much the same as trying out every single imaginable formula in hopes of one of them being the grand unified theory of butt science.

Hardly. It's more like looking to see what happens when people view image X, and trying to find patterns. Using machines that can find patterns in encrypted data. I think it can find patterns in some brain functions given access to them. Maybe it's you that is cocky?

Quote by CoreysMonster

I'd love to know what new knowledge was gained from the experiment in the OP, because there quite obviously wasn't any new discovery about how our brains work described in the article. All they did was make a computer program use brain-scan data to guess a picture out of a pile.

Watch the video again. Learn how science is done, also. You seem to think it involves immediately knowing answers to questions you weren't even asking.

Also, you sound like one of those fools who says computer science isn't "real" science.
Quote by darkstar2466
Not impressed even one bit. Get it? One bit. That's the comparative resolution this shit is spewing out compared to actual brain activity, which we are too stupid to understand.

Yeah, look at those idiots doing science on things they don't yet understand! That's not science!
Quote by element4433
Pretty interesting, but I don't want to see what's in my head.

You already do.