Researchers devise technique to 'read' a person's mind
06 February 2018
Using a new technique, scientists can 'read' a person's mind to identify the songs they are listening to.
According to commentators, the technique developed by researchers at D'Or Institute for Research and Education in Brazil and University Hospital Leipzig in Germany, would likely lead to new research on reconstruction of auditory imagination and inner speech. It can also be useful in enhancing brain-computer interfaces in order to establish communication with locked-in syndrome patients.
The experiment involved six volunteers who heard 40 pieces of classical music, rock, pop, jazz, and others.
The neural fingerprint of each song on participants' brain was captured by an MRI machine while a computer was learning to identify the brain patterns produced by each musical piece even as it factored in musical attributes like tonality, dynamics, rhythm and timbre.
Following this, researchers expected the computer to be able to work out in reverse order ie - identify which song participants were listening to, based on their brain activity - a technique known as brain decoding.
When there were two options, the computer was able to identify the song with up to 85 per cent accuracy, which is a great improvement on earlier studies.
Researchers then made it harder providing 10 options to the computer instead of two and the computer could correctly identify the song in 74 per cent of the decisions.
According to commentators, in future, studies on brain decoding and machine learning will create possibilities for communication not limited by spoken or written language. ''Machines will be able to translate our musical thoughts into songs,'' said Sebastian Hoefle from D'Or Institute and Ph.D student from the Federal University of Rio de Janeiro, Brazil, IANS reported.
Hoefle expects to find answers to a host of questions, in future, like what musical features make some people love a song while others don't? Do our brains prefer a specific kind of music''?
In clinical settings, this technology could find application in enhancing brain-computer interfaces in order to establish communication with locked-in syndrome patients.