Thread: Scientists Extract Images Directly From Brain

This sounds good for us dreamers, but reading thoughts and feelings?

I don't know about you but I think that would be a little invasive.

More like one step closer to Orwell's "thought police"

Originally Posted by The Prodigy

More like one step closer to Orwell's "thought police"

I think when that happens a lot of people would get arrested lol. I know if someone got into my mind I would be too worried I'd think something weird causing me to think about it ironically.

About 4 people beat you to this

They've all been consolidated here:
http://www.dreamviews.com/community/...ad.php?t=69867

I wonder how accurate that can get. I also wonder if they can do the same thing for sound. Oh man, the possibilities!

Though I doubt that would work for dreams - usually our dreams follow unusual patterns. You may look at a blue and take it as the bluest blue out of any other blue you've seen, but the computer wouldn't be able to interpret that. I don't doubt it can interpret some very raw emotions, like anger, happiness, sadness, suspiciousness, boredom, stress and etc.

Originally Posted by Kromoh

Though I doubt that would work for dreams - usually our dreams follow unusual patterns. You may look at a blue and take it as the bluest blue out of any other blue you've seen, but the computer wouldn't be able to interpret that. I don't doubt it can interpret some very raw emotions, like anger, happiness, sadness, suspiciousness, boredom, stress and etc.

Maybe the people who claim they can see "other" colors can finally verify it.

I wonder how clear it can really get considering the peripheral vision is kind of blurry.

Originally Posted by TimeStopper

I wonder how clear it can really get considering the peripheral vision is kind of blurry.

I couldn't tell from the article where they're getting the input. See, it all depends on where in the visual pipeline they're getting their signals. For example, if they tapped directly into the optic nerve, they would see the raw, upside-down, mostly blurry image from the eye and would require very little processing. On the other hand, if they tapped the end of the line, they could see what the person literally perceives, but deciphering that data would be nigh impossible, given that individual uniqueness becomes very important at this stage, because of "qualia" and such.

Originally Posted by TimeStopper

Maybe the people who claim they can see "other" colors can finally verify it.

I don't think that's what's gonna happen. Dreams aren't about what people see, it's about what people believe they see. I think the machine wouldn't correctly interpret the "other" color at all.

Originally Posted by TimeStopper

Maybe the people who claim they can see "other" colors can finally verify it.
I wonder how clear it can really get considering the peripheral vision is kind of blurry.

Originally Posted by drewmandan

I couldn't tell from the article where they're getting the input. See, it all depends on where in the visual pipeline they're getting their signals. For example, if they tapped directly into the optic nerve, they would see the raw, upside-down, mostly blurry image from the eye and would require very little processing. On the other hand, if they tapped the end of the line, they could see what the person literally perceives, but deciphering that data would be nigh impossible, given that individual uniqueness becomes very important at this stage, because of "qualia" and such.

Hi all, I am new, was browsing and saw this thread. It is very interesting but there is a common misconception that I have observed.

Some of you are mis-interpreting what has happened here. I can explain it for you. The scientists are not directly reading the brain's code, and infact are nowhere near doing so. They are correlating certain images to different levels of neurological activity:

"The scientists were able to reconstruct various images viewed by a person by analyzing changes in their cerebral blood flow. Using a functional magnetic resonance imaging (fMRI) machine, the researchers first mapped the blood flow changes that occurred in the cerebral visual cortex as subjects viewed various images held in front of their eyes. Subjects were shown 400 random 10 x 10 pixel black-and-white images for a period of 12 seconds each. While the fMRI machine monitored the changes in brain activity, a computer crunched the data and learned to associate the various changes in brain activity with the different image designs."

For example, a picture of an apple is shown to John while his brain activity is measured. As he is shown the picture, a specific, unique instance of activity occurs in the brain (let us refer to this specific activation as Activition AB). I could conclude that in future, every time activation AB is observed in John's brain that the image/concept of an apple is being processed in some way (seen, imagined, dreamt e.t.c). This is the methodology behind these experiments.

Although the technique of doing this is very intelligent and pragmatic, it shows technological limitations and is rather rudimentary; let me give you an analogy:

Lets say that aliens visit our planet, find a vcr player, television and power supply; and yet they know nothing about the mechanisms of human technology. They discover that these three combined can do an amazing thing: play videos. So they try to reverse engineer it, and discover how it works. The endgame here is to be able to understand the code of the vcr player: DIRECTLY linking the nature of the black tape inside each video to the images on the television screen; and thus being able to predict solely from the black tape what will be shown. The aliens have not yet managed to so this, but, they have correlated the images in the video with time (by making a reference book with what is displayed onscreen, frame by frame, and the time that has elapsed at the beginning of the video. So they can predict what images will appear on screen even without looking. Yet this reference book can only work for this specific video. Also, it cannot only chart single stimuli, and encounters difficulty with sequences. e.t.c

But it is one step closer, perhaps, to understanding the code.

I understood the article, but my point stands. If you notice the picture of the word "neuron", a novel stimulus, you will see that they can use information from other images to learn about new images. So you haven't addressed my point at all.

Originally Posted by drewmandan

I understood the article, but my point stands. If you notice the picture of the word "neuron", a novel stimulus, you will see that they can use information from other images to learn about new images. So you haven't addressed my point at all.

Using the pixelated images, and a complex algorithm, they can kind of understand how the brain behaves when each single pixel is present, and how it behaves when cognizing the letters, and then use what they've learned from the person's brain behaviour to tell the next things he/she sees. It's like a mother is capable of predicting what her son is thinking, because of knowing him so well.