{"id":26867,"date":"2023-05-01T17:00:03","date_gmt":"2023-05-01T15:00:03","guid":{"rendered":"https:\/\/www.sciencefocus.com\/?p=142982"},"modified":"2023-05-01T17:38:25","modified_gmt":"2023-05-01T15:38:25","slug":"this-machine-can-read-your-mind-and-translate-how-you-see-the-world","status":"publish","type":"rss_feed","link":"https:\/\/c01.purpledshub.com\/bbcsciencefocus\/rss_feed\/this-machine-can-read-your-mind-and-translate-how-you-see-the-world\/","title":{"rendered":"This machine can read your mind and translate how you see the world"},"content":{"rendered":"

Neurotechnologists have developed a decoder that can reconstruct what you are seeing, thinking, and imagining \u2013 and put it into words. <\/p>

By Noa Leach\n \t\t<\/p>

Published: Monday, 01 May 2023 at 12:00 am<\/p>

\n\n

A machine that reads your thoughts is now more science than science fiction, thanks to neurotechnologists at the University of Texas at Austin (UT Austin).<\/p>\n

And for the first time ever, this machine doesn\u2019t require its subject to be wired up with implants and electrodes.<\/p>\n

Instead, this decoder is non-invasive, using functional magnetic resonance imaging (fMRI) to measure the changes in blood flow around your brain to translate ideas into words.<\/p>\n

\u201cWe are decoding something that is deeper than language<\/a>,\u201d Dr Alexander Huth<\/a> from UT Austin told BBC Science Focus<\/em> and other press. The decoder can grasp the intangible \u2013 the various shapes our thoughts take \u2013 and turn them into something understandable and, crucially, communicable.<\/p>\n

This means that the person who is having their mind \u2018read\u2019 does not have to verbalise their thoughts. For those who have lost the ability to speak \u2013 following a stroke, for example \u2013 the decoder could restore communication channels non-invasively.<\/p>\n

\u201cSpeech impairments can be highly debilitating,\u201d co-author Jerry Tang<\/a>, graduate student at UT Austin, said. \u201cProviding some sort of additional communication channel could be really valuable.\u201d<\/p>\n

Until now, non-invasive speech decoders have only been able to reconstruct single words or short phrases. This paper, published in Nature Neuroscience<\/em>, reveals a machine which can decode continuous, natural language<\/a>.<\/p>\n

The machine works by combining well-established decoding methods with modern machine learning techniques. Essentially, it works similarly to ChatGPT<\/a>, predicting the ends of sentences based on what has been learnt before.<\/p>\n

Akin to an AI chatbot, the decoder needs to be trained on a lot of data \u2013 in this case, MRI scans that measure blood flow in the brain. To amass enough neural data, each study participant listened to 16 hours of podcasts including The Moth Radio Hour <\/em>and The New York Times<\/em>\u2019s Modern Love<\/em>.<\/p>\n

Huth said: \u201cWhat we were going for was mostly what is interesting and fun for the subjects to listen to, because that\u2019s critical in actually getting good fMRI data \u2013 rather than boring subjects out of their skulls.\u201d<\/p>\n

Participants also watched silent clips of films \u2013 mostly those short stories played at the beginning of Pixar films that do not contain dialogue. Interestingly, this part of the experiment showed that the decoder translates something beyond language: ideas.<\/p>\n

Blood flow changes slowly in the brain, over the course of a few seconds rather than the nanoseconds of a neuron impulse. This means that the decoder doesn\u2019t translate the exact words we are reading, but how we interpret them. As Huth explains, how we make sense of something slowly changes \u2013 \u201cso we can see how the idea evolves, even though the exact words get lost.\u201d<\/p>\n

While MRI machines are not portable, in the future the researchers hope to test their methods using affordable and portable devices including functional near-infrared spectroscopy (fNIRS) which similarly uses non-invasive techniques to measure blood oxygenation in the brain.<\/p>\n

What does this mean for brain privacy?<\/strong><\/p>\n

For many, the prospect of a brain-reading machine rings alarm bells. But the researchers highlight how the current model can only be used on the individuals it has been trained on. It also needs a participant\u2019s cooperation to run, as the study found that participants could actively resist being \u2018decoded\u2019.<\/p>\n

They called this \u201csabotaging the decoder\u201d: by performing simple tasks like counting, naming animals or telling their own story, participants could \u2018distract\u2019 the decoder from their other thoughts.<\/p>\n

Of course, future technology<\/a> may be able to get round this \u2013 so the researchers underlined the importance of researching privacy implications going forward.<\/p>\n

\u201cWhile this technology is in its infancy, it\u2019s very important to regulate what brain data can and cannot be used for,\u201d says Tang.<\/p>\n

\u201cThere could be negative consequences if misused and misunderstood, so we think it\u2019s important to make sure that the decoder\u2019s capabilities are not misrepresented.\u201d<\/p>\n

Read more:<\/strong><\/p>\n