Utilizing the new technology, the researchers were able to map brain patterns of subjects when they were watching videos from YouTube. The team then created another YouTube video based on the imaging and activity of the subjects' brain while they were watching the first video. The research was published on Sept. 22, in the journal Current Biology.
For the purposes of this research, the scientists combined functional Magnetic Resonance Imaging (fMRI) and computational models which decoded and reconstructed a person's dynamic visual experience. The researchers, who were themselves the subjects, spent hours lying inside an MRI machine and watched two sets of movie trailers while the machine recorded blood flow through the visual cortex, which processes visual information in the brain.
While watching the first trailer, a computer program recorded the brain activities of the subjects and created an algorithm against the video's visual patterns. Then the second trailer tests the computer's movie reconstruction algorithm.
The reconstruction algorithm resulted in a blurry but continuous 100-clip recreation of the original movie.
The technology can, potentially, help scientists to read human minds, dreams and memories; all of which, to date, lie beyond human comprehension.
In addition to mapping the human mind, the technology could also improve the lives of people with neurological diseases and will allow people with no motor skills to go into the MRI for two hours every day and communicate with their families, said Jack Gallant, Professor of Psychology at Berkeley and co-author of the study.
However, as the computational models are still in development, the date of implementation of the technology is still uncertain. The technology will definitely get there, the question is just when, said Gallant.