Holy crap. Of all the science fiction technology I've ever seen, this is near the bottom of the list of things I thought would ever be a reality.

*promote

Self promoting this video and sending it back into the queue for one more try; last queued Thursday, September 22nd, 2011 5:05pm PDT - promote requested by original submitter lucky760.

>> ^lucky760:

Holy crap. Of all the science fiction technology I've ever seen, this is near the bottom of the list of things I thought would ever be a reality.
promote


Anything is possible at Zombo.com!

Holy Fuckbuckets! That's amazing!

Sweet sassy molassy!

Science!

I think the right clip was analyzed from my brain while I was on acid

All you get from my brain scan are images of T&A.

How long until we have "thought police" using a refined version of this during suspect interrogations?

Also, wtf--some of those didn't match up at all (the birds flying looked like the person was psychically predicting the next image of the person talking...).

Here is a link to the article:
http://www.cell.com/current-biology/fulltext/S0960-9822%2811%2900937-7#Summary
also, this video has a breakdown of how the process works:
http://videosift.com/video/Brain-Scan-Movies-reconstruction-methodology

I wonder how good the reconstructions would look if the computer had every video on youtube in its library?

Not to be a downer (and this is very cool), it is still far from reading someone's mind. Even before reading up on what @grinter provided the fact that the people images were better defined tipped me off to the fact that the program "simply" compares the brain data with pre-existing clips.

Going from there to actually visualising what is seen in the "mind's eye" is still a long step, albeit surely not impossible.

^That's a bit of a pessimistic view of it. Sure, it's not (yet) seeing what someone is thinking, but it's incredible that they have even figured out how to read brain activity as visual images at all.

I never thought there'd ever be any way to get even a hint of an idea at what someone's brain was seeing in its mind's eye. This is mind-blowing.

*breakthrough

^Ditto on what lucky760 said, and @hpqp, I think you may be making too much of these videos being a composite of previously entered data. Think of the videos in the computer's library as Lego blocks. You can combine each block to build a rough approximation of some real object that you see (e.g. Chewbacca). Ok, so they only have about 50 Duplo sized blocks right now, and Lego Chewbacca kinda looks like a brown trashcan, but just give them more blocks and a greater range of shapes, and they'll build a wookie fit to tear your arm off.

OMG "Brainscan" will be a reality!

I don't understand how this works. I read the articles and I am a little skeptical. I've designed fMRI and DTI algorithms for years and I don't see why they keep talking about fMRI and brain waves. fMRI is an activity map that is related to the hot spots in the brain where the hydrogen protons aligned by the magnetic field resonate to the frequency of the emitter (TR/Echo Time) and only show consumption of glucose (hydrogen protons motility) during a designed paradigm, which in this case would be having the subject watch a video. Diffuse Tensor Imaging will help map the neurons going there in case a surgical procedure is necessary, and that's about it. Extrapolating fMRI (a very coarse k-space reconstruction) to brainwaves (an EEG signal) and images sounds very suspicious to me, and nothing published so far explains how this is technically done. I understand the excitement and it certainly would be possible in the future, but under the current state of the art, I don't see how this is possible, especially with fMRI or Fractional Anisotropy.

Think of the possibilities... all you have to do is find someone dreaming about natalie portman getting it on with Cheryl Cole and pray to god that tendrils dont start appearing. On the other hand nightmares will no longer fade away when you can replay them on Tivo.

Next stop, virtual brains!

Interesting how the faces have so much more clarity than the other random objects, especially text.

I'd like to see if more shocking images showed up more clearly. Since it's all based on feedback in the brain, I'd bet a scene of someone being stabbed in the face might show up more clearly.

When you read into the research, it's not quite as amazing as it first seems, but it's still pretty funky.

>> ^MonkeySpank:

I don't understand how this works. I read the articles and I am a little skeptical. I've designed fMRI and DTI algorithms for years and I don't see why they keep talking about fMRI and brain waves. fMRI is an activity map that is related to the hot spots in the brain where the hydrogen protons aligned by the magnetic field resonate to the frequency of the emitter (TR/Echo Time) and only show consumption of glucose (hydrogen protons motility) during a designed paradigm, which in this case would be having the subject watch a video. Diffuse Tensor Imaging will help map the neurons going there in case a surgical procedure is necessary, and that's about it. Extrapolating fMRI (a very coarse k-space reconstruction) to brainwaves (an EEG signal) and images sounds very suspicious to me, and nothing published so far explains how this is technically done. I understand the excitement and it certainly would be possible in the future, but under the current state of the art, I don't see how this is possible, especially with fMRI or Fractional Anisotropy.


...says "MonkeySpank", lol! Really, thank you for the insight, but the association between your well informed comment and your avatar is what did it for me.

This is what led me to believe that E = MC² --LoudBelcher78

So in the future I can record my "sexy dreams" and then rewatch my own mind porn? That's messed up!

What I do for a living and what I do in my bedroom are polar opposites...

>> ^AgentSmith:

>> ^MonkeySpank:
I don't understand how this works. I read the articles and I am a little skeptical. I've designed fMRI and DTI algorithms for years and I don't see why they keep talking about fMRI and brain waves. fMRI is an activity map that is related to the hot spots in the brain where the hydrogen protons aligned by the magnetic field resonate to the frequency of the emitter (TR/Echo Time) and only show consumption of glucose (hydrogen protons motility) during a designed paradigm, which in this case would be having the subject watch a video. Diffuse Tensor Imaging will help map the neurons going there in case a surgical procedure is necessary, and that's about it. Extrapolating fMRI (a very coarse k-space reconstruction) to brainwaves (an EEG signal) and images sounds very suspicious to me, and nothing published so far explains how this is technically done. I understand the excitement and it certainly would be possible in the future, but under the current state of the art, I don't see how this is possible, especially with fMRI or Fractional Anisotropy.

...says "MonkeySpank", lol! Really, thank you for the insight, but the association between your well informed comment and your avatar is what did it for me.
This is what led me to believe that E = MC² --LoudBelcher78

Shit if they can reproduce movies from your brain, we're all going to be sued into oblivion for seeding.