Here's digital video of what we see inside our brains
In extraordinary film, UC Berkeley scientists show how they found a way to depict the images we see in our brains.
I don't know what kinds of things you see inside your head, but I do worry about it.
As for the things I see inside my head, well, if only I could show you. Actually, there are scientists at UC Berkeley who believe that they can show you.
I haven't let them into the house yet. But I can show you video of their work. I am grateful to the inner brains at Gizmodo, who first revealed this footage to me.
You will, naturally, be wondering whether the scientists created this footage, well, naturally.
In a way. They put their human guinea pigs (themselves) into a functional MRI machine for several hours. Which sounds uncomfortable.
They then showed themselves two different Hollywood movie trailers. As they watched these trailers, the blood-flow through their visual cortex was being recorded.
Then it was down to a computer program that divided into voxels-- three-dimensional cubes otherwise known as volumetric pixels.
"We built a model for each voxel that describes how shape and motion information in the movie is mapped into brain activity," Shinji Nishimoto, lead author of the study, explained on the UC Berkeley site.
The computer program, on a second-by-second basis, began to learn what sort of visual patterns appeared in the brain as subjects saw particular types of visual stimulus in the first set of trailers.
Then a second set of trailer information was used for the computer--which had been fed 18 million seconds of random YouTube footage--to correlate.
In the end, what you see is a merger of the 100 clips most similar to the first clip the subject saw, but blurred together to form an entity.
Clearly, this is both complicated and imperfect. However, wouldn't you ultimately like to know what your Congress sees inside its collective head when it confronts, say, the economic crisis?
Wouldn't you want to see what your lover sees when he or she suddenly sees, well, red?
Surely, one day, these images might help us understand each other. At least a little.