Boston Globe: MIT researchers find new ways into brain’s “music room.” The research by MIT’s Sam Norman-Haignere, Nancy Kanwisher, and Josh McDermott sounds straightforward: apply clustering to patterns of nerve activation to find the cells that respond to music. Turns out it’s a radical new approach that bridges magnetic resonance imaging and “machine learning,” or categories of math on sets that identify groups that are similar to each other.
For me the finding that’s most interesting is that the neural pathways for music are completely distinct from the ones for speech, though there’s a little activation overlap when listening to music with words. In fact, the research identified completely distinct neural pathways for music, speech, frequency, pitch, and “spectrotemporal modulation.”
I’d love to see the follow-on that evaluates these activation pathways in trained musicians vs. lay listeners.
Note: A more complete version of the article appeared in the New York Times.