The Brain-Scanning Algorithm That Can Read Your Thoughts
It seems that once again, science doesn’t know when to stop – because even though most of us would say that having someone read our thoughts would be an actual nightmare, scientists have built an algorithm that can do just that.
This “decoder” analyzes brain scans and uses them to reconstruct people’s thoughts by relying on functional magnetic resonance imaging (fMRI) recordings.
Image Credit: iStock
Neuroscientist Alexander Huth from UT Austin spoke with The Scientist about how the non-invasive technique works.
“If you had asked any cognitive neuroscientist in the world twenty years ago if this was doable, they would have laughed you out of the room.”
Even though the breakthrough has not been peer-reviewed, Huth and his team believe their decoder will be incorporated into many brain-computer interfaces in the future.
Huth’s method uses fMRI to observe how blood flow around the brain can map users’ thoughts – a process that could prove an important communication aid for those unable to speak (among others).
“This decoder generates intelligible word sequences that recover the meaning of perceived speech, imagined speech, and even silent videos, demonstrating that a single language decoder can be applied to a range of semantic tasks.”
Image Credit: iStock
This was accomplished even though we’re not 100% sure which cortical circuits represent language. They focused the decoder on three separate brain networks: the classical language network, the parietal-temporal-occipital association network, and the prefrontal network.
They found that each could be used to decode word sequences.
The system isn’t perfect, though, as the authors note.
“While our decoder successfully reconstructs the meaning of language stimuli, it often fails to recover exact words.”
The system struggles with pronouns and distinguishing first versus third person – so it knows what’s happening, but not who.
Image Credit: iStock
They do claim that invasion of privacy is not a concern, as the algorithm was incapable of reconstructing users’ thoughts when the participants simply thought of something mundane, like the names of animals or a range of colors.
For now, at any rate.
