Scientists try to develop mind reading device

A mind-reading device that can picture what a person is seeing or thinking may not be a sci-fi fantasy for much longer.

New research could soon enable real or imagined images in the brain to be shown on a visual display screen. It may even be possible to take pictures from dreams.

Scientists in California have taken the first step towards developing such a machine by producing an analytical tool which can identify seen images from brain activity patterns.

The mathematical model relates perceptual brain activity to points in three dimensional space called "voxels"- 3D versions of the pixels used to construct digital photos.

It helped the researchers pick out with 92 per cent and 72 per cent accuracy which of 120 images two volunteers had been viewing. Left purely to chance, their performance would have only been 0.8 per cent accurate.

Writing in the journal Nature, the scientists led by Dr Jack Gallant from the University of California at Berkeley said: "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.

"Imagine a general brain-reading device that could reconstruct a picture of a person's visual experience at any moment in time."

The study began with two of the scientists acting as volunteers and undergoing a functional magnetic resonance imaging (fMRI) brain scan to record activity data from three areas of the brain's visual cortex. This was done while each of them viewed a total of 1,750 natural images.

The data obtained was used to construct mathematical descriptions of "voxels" in the brain - the "points" that make up a 3D thought image.

In the second stage of the experiment, each volunteer viewed a new set of 120 images while having their brains scanned.

Combining the new data with the mathematical descriptions created earlier, the scientists were able to identify which images had been seen.

Because fMRI data are "noisy", voxel activity patterns were averaged across 13 repeated trials.

However, a practical brain-reading device would ideally respond to activity triggered by just one perceptual event.

To test how possible this might be, the experiment was conducted again using data from single brain scans.

This yielded performance ratings of 51 per cent and 32 per cent for each of the volunteers. The results were good enough to suggest that it may be possible to decode perceptual experiences "in real time", said the researchers.

The scientists said a "general visual decoder" that could unscramble nerve data to picture images in the brain would have "great scientific and practical use".

They added: "For example, we could use the decoder to investigate differences in perception across people, to study covert mental processes such as attention, and perhaps even to access the visual content of purely mental phenomena such as dreams and imagery.

"The decoder would also serve as a useful benchmark of our understanding of how the brain represents sensory information."