There is no accounting for matters of taste. Or is there? There are numerous, odd quizzes on Facebook and other social media which claim to help you determine which song you like or what song “you are”. While the simplistic, programme-generated answers are patently unscientific, scientists are moving closer to a technique, which can at least identify what music you are hooked to. According to research by D’Or Institute for Research and Education in Brazil and University Hospital Leipzig in Germany, a computer—in a remarkable feat of brain decoding—could guess with over 70% accuracy the genre of music that a person was listening to. The experiment required the computer to learn subject’s reactions to over 40 different kinds of music using MRI scans. Later, it could, with 85% accuracy, decode the right genre when given two options and with 74% accuracy when it came to 10 possibilities.
The study has a sounder basis, forgive the fun, than just decoding playlists by mapping against neuron activity, and can do more than just improving song delivery mechanisms. The high accuracy suggests that scientists are moving closer to the reconstruction of auditory imagination and “inner speech”, which would be helpful for patients who cannot communicate. More important, if scientists can link these results to emotion analysis, it can also pave the way for determining how humans make decisions.
Even if a world where computers can rightly predict what we want is too far-fetched, the study is indeed revolutionary when it comes to decoding what stimulates us. The possibilities are immense for the entertainment industry, especially as our attention span gets shorter and shorter with the glut of entertainment options made available today.