Scientists have found a way to ‘listen’ to what people are thinking. (Or saying to themselves, at least.)
As New Scientist reports, researchers from the University of California, Berkeley recorded the brain activity of people having epilepsy surgery while they saw the text of either the Gettysburg Address, John F. Kennedy’s inaugural address, or Humpty Dumpty projected onto a screen. (I guess this was one of those wide awake brain surgeries that sounds utterly terrifying, so maybe it was a nice distraction…)
The subjects were asked to read the text aloud, read it silently to themselves, and then do nothing. The researchers saw which neurons reacted when people spoke and used an algorithm trained to recognise speech to turn this into a spectrogram, a way to present sound visually that shows the frequency of different sound waves and which, because each word has a unique signature, can be used to recreate what someone’s said.
When the scientists repeated this process while participants read silently, they were able to use the same algorithm to create a spectrogram that could recreate at least some of the words participants thought, just from picking up on their neural activity. And it worked even though the pronunciation of some words was slightly different internally than externally.
The team, led by Brian Pasley, doesn’t just want to overhear our thoughts to be nosy. They’re hoping that this technology could one day be used by people who are unable to speak due to illness or paralysis. But they say it needs some fine-tuning before they’ll be ready to build a device for regular use. They’re currently conducting further research into how the brain processes sound so they can understand as much about it as possible, in order to build a viable product down the road.
Image via dobroide’s Flickr.
If you’re more interested in getting a new camera than new clothes this season, you might be interested in these camera accessories for your iPhone, which beg the question: Do you need a camera if you own a camera phone?