MIT scientists have identified a population of neurons that responds to singing but not to other types of music
Listening to Billie Eilish singing is not the same as listening to Billie Eilish reciting the lyrics of her latest hit (without music) or listening to the melody (without vocals) of the same Billie Eilish song. That’s logical, but what neuroscientists at the Massachusetts Institute of Technology (MIT) have now discovered is that our brains respond differently to each of these experiences. Researchers have identified a population of neurons that ‘turns on’ only when listening to the specific combination of voice and music but not to instrumental music or normal speech.
The new work, published in Current Biology, builds on a 2015 study in which the same research team used functional magnetic resonance imaging (fMRI) to identify a population of neurons in the brain’s auditory cortex that responds specifically to music.
The researchers scanned the participants’ brains as they listened to a collection of 165 sounds, including different types of speech and music, as well as every day sounds like tapping fingers or a dog barking. For that study, the researchers devised a novel method for analyzing the fMRI data, which allowed them to identify six neuronal populations with different response patterns, including one that responded selectively to music and one that responded selectively to speech.
In the new study, the researchers used a technique known as electrocorticography (ECoG), which allows electrical activity to be recorded using electrodes placed inside the skull. This offers a much more accurate image compared to fMRI, which measures blood flow as an indicator of neural activity.
“There’s a population of neurons that responds to singing, and then very close by, there’s another population of neurons that responds broadly to a lot of music. At the fMRI scale, they’re so close together that they can’t be separated, but with intracranial recordings, we get extra resolution, and that’s what we think allowed us to separate them,” explains Sam Norman-Haignere, a former MIT postdoc who he is now a professor of neuroscience at the University of Rochester.
Epilepsy patients
Electrocorticography usually cannot be performed in humans because it is an invasive procedure, but it is often used to monitor epilepsy patients who are about to have surgery to treat their seizures. Patients are monitored for several days so doctors can determine where their seizures originate before operating. During that time, if patients agree, they can participate in studies that involve measuring their brain activity while they perform certain tasks. For this study, the MIT team was able to collect data from fifteen participants over several years.
For those participants, the researchers played the same set of 165 sounds that they used in the earlier fMRI study. Surgeons determined the placement of each patient’s electrodes, so some did not detect any response to auditory input, but many did. Using a new statistical analysis they developed, the researchers were able to infer the types of neural populations that produced the data recorded by each electrode.
“When we applied this method to this data set, this pattern of neural response that only responded to singing appeared,” says Norman-Haignere. “This was a finding that we really didn’t expect, so it pretty much justifies the point of the approach, which is to reveal potentially novel things you might not think to look for,” he adds.
That song-specific population of neurons had very weak responses to speech or instrumental music and is therefore distinct from the music- and speech-selective populations identified in their 2015 study.
Temporal lobe
In the second part of their study, the researchers devised a mathematical method to combine the data from the intracranial recordings with the fMRI data from their previous study. Because fMRI can cover a much larger portion of the brain, this allowed them to more precisely pinpoint the locations of the neural populations that respond to the song.
The song-specific hotspot they found is located in the upper part of the temporal lobe, near regions that are selective for language and music. That placement suggests that the song-specific population may be responding to features such as perceived pitch or the interaction between words and perceived pitch, before sending information to other parts of the brain for further processing, the researchers say.
The researchers now hope to learn more about what aspects of singing drive the responses of these neurons. They are also working to study whether babies have music-selective areas, hoping to learn more about when and how these brain regions develop.