Researchers Aim to See What Others See
Results of a study conducted by a team at the University of California Berkeley may make it possible to decode brain signals and track attention as well as see how people dream.
Jack Gallant and colleagues reported their research in the journal Nature.
The team sought to overcome limitations of recent functional magnetic resonance imaging (fMRI) studies that decoded orientation, position and object category from activity in visual cortex.
“We show that these receptive-field models make it possible to identify, from a large set of completely novel natural images, which specific image was seen by an observer.”
"Imagine a general brain-reading device that could reconstruct a picture of a person’s visual experience at any moment in time, and perhaps even provide access to the visual content of phenomena such as dreams and imagery."
This is something the team hopes to achieve in the future by using fMRI to predict which photograph a volunteer was looking at.
"Our results suggest that it may soon be possible to reconstruct a picture of a person’s visual experience from measurements of brain activity alone," they wrote.
The first step was to scan two members of the team using fMRI as they looked at a set of 1,750 photographs.
"The content of the photographs included animals, buildings, food, humans, indoor scenes, man-made objects, outdoor scenes, and textures," they wrote.
Then, the two researchers viewed 120 new images as the research team attempted to determine which photograph each one had been looking at.
The answers for one researcher were correct 92 percent of the time while answers for the second researcher were correct 72 percent of the time.
They noted that their recent study is one simple step in a progression of man others toward their goal of mapping brain activity to see what others are seeing.
"Identification of novel natural images brings us close to achieving a general visual decoder," they wrote.
"The final step will require devising a way to reconstruct the image seen by the observer, instead of selecting the image from a known set."
—
On the Net: