MRI scans of a sleeping person’s brain can help predict what’s seen in the land of Nod.
Researchers in Japan report in the journal Science on Thursday that they could predict images with 60 percent accuracy. First, they watched three subjects as they fell asleep and woke them in early sleep stages to ask them what they saw in their dreams. From those reports, the researchers built a database that grouped together the things the subjects saw in their dreams (e.g. “house” and “hotel” were grouped into a ”building” category). Later, when the volunteers were awake, the researchers showed them photos that matched these categories while recording their brain activity with MRI.
The researchers next built a computer model that linked patterns of brain activity to different types of images. This model was able to predict what kinds of images the subjects saw in their dreams in subsequent naps at a rate slightly higher than would be expected by chance.
According to the BBC, University of Oxford cognitive neuroscientist Mark Stokes says that the work brings us closer to creating dream-reading machines, although such devices are still something for the distant future. Stokes told the BBC that such machines would have to be customized for each individual, so it would be unlikely that your dreams could be read surreptitiously:
“You would never be able build a general classifier that could read anybody’s dreams. They will all be idiosyncratic to the individual, so the brain activity will never be general across subjects,” he said.
“You would never be able to build something that could read other peoples thoughts without them knowing about it, for example.”
A quick guide to the most important AI law you’ve never heard of
The European Union is planning new legislation aimed at curbing the worst harms associated with artificial intelligence.
It will soon be easy for self-driving cars to hide in plain sight. We shouldn’t let them.
If they ever hit our roads for real, other drivers need to know exactly what they are.
This is the first image of the black hole at the center of our galaxy
The stunning image was made possible by linking eight existing radio observatories across the globe.
The gene-edited pig heart given to a dying patient was infected with a pig virus
The first transplant of a genetically-modified pig heart into a human may have ended prematurely because of a well-known—and avoidable—risk.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.