How-To Geek

Scientists Turn Memories into Actual Video Footage [Video]

In a rather stunning display of scientific inquiry and processing, Berkley scientists have successfully created video footage of memory.

What exactly does this mean and how does it work? From the Berkley press release:

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.

Check out the video above to see it in action and then visit the link below to read more about this extraordinary research. If you’re interested in reading the original white paper you can read it here.

Scientists Use Brain Imaging to Reveal Movies in Our Mind [UC Berkley News Center via Mashable]

Jason Fitzpatrick is a warranty-voiding DIYer who spends his days cracking opening cases and wrestling with code so you don't have to. If it can be modded, optimized, repurposed, or torn apart for fun he's interested (and probably already at the workbench taking it apart). You can follow him on if you'd like.

  • Published 09/23/11

Comments (12)

  1. Dan B


    This is unreal…

  2. Jason Fitzpatrick

    Completely unreal… and so very very creepy. I hope the outcome is amazing innovations for disabled people and not Minority Report/Total Recall hijinks.

  3. KayDat

    I remember an episode of House showed off some of this kind of tech.

  4. xana452

    I can’t believe this has finally happened… Technology is amazing.

  5. avenfisher

    just think of all the things people could do with this technology!

  6. Bill Randle

    And in a meditation state?

  7. Mutantone

    some of the mental pictures were very different from the video that they were supposed to be and there seemed to be text in some of them as well which makes me wonder what that said

  8. DJ Rose

    It’s probably a FRAUD cooked up to make people believe they’re capable of this.
    Think of the value in legal testimony, if they could create false memory testimony in court to suit any accusing party. Using technological B.S., to make a false conviction.

  9. George Ronald Adkisson

    You can employ the saccule of the ear and get to the eyes quicker and the reception is like a digital camera.The subject needs to be put into a magnetic field that excites the nerves in the saccule and then using the one tone hear thoughts, distinguishing which wavelength is from the eyes to match the thoughts. Thinking out loud just examples how our thoughts are processed in a language.It’s like using a karaoke program.
    Recall though … this procedure is not a new one and has been misused by the uS government.
    Hint … a speaker is the same thing as a microphone … so the one tone is heard using a microphone with a low impedance … real noisy.

  10. ¥en

    I don’t think it’s a fraud, in fact from what I understand, it is some quite amazing and extrexemly clever thinkboxing from these scientists.
    Let’s face it, some of such reputation would hardly april fools this late/early :p
    It seems that they monitor the subjects brainwaves whilst they watch clips they have been shown previous to the experiment. They then using a complicated algorithm convert the data into the most likely image from a database of youtube videos which would recreate that brain wave.
    IE –
    Clip 1, 3min 4sec = “brainwave pattern” > 010010110110
    A collection of this data could then be used to map or correlate images with brain wave patterns to the images in the database. Quite amazing!!

  11. TG

    The text you saw in the reconstruction is present because those images are created from a separate sample set of videos. If the only clips that contained the appropriate image (say a person in 3/4 view) generally had text along the bottom, this would show up in the reconstruction. This explains why some reconstructions aren’t perfect – they’re only drawing from a limited set to reconstruct images. For instance, I don’t think there were any images of elephants in the set of videos used to generate the image of elephants walking across the screen. It was all done by averaging other scenes that tended to produce similar brain activity.

    And no, it’s not fraud. It’s statistics and pattern classifiers – albeit very clever implementations of them.

  12. Gautam

    Some of the images have text when originals did not have and other blurs, probably because the subject was not paying attention to those pictures as much as he was to other pictures. So the varying attention span could be a reason for clarity or lack of it

More Articles You Might Like

Enter Your Email Here to Get Access for Free:

Go check your email!