On the trail of amazingly fine specializations: The auditory cortex of the brain not only contains a kind of music center – there is also a group of nerve cells that is specially adapted to the perception of singing, researchers report. These neurons respond to sung versions of human sounds, but rarely to speech or instrumental music. This is the result of investigations using functional magnetic resonance imaging in combination with the recording of nerve activity through electrodes in the brain of epilepsy patients. How and why the “singing neurons” react in such a special way is now to be clarified by further investigations.
The brain is and remains the most mysterious of all organs: Scientists are still puzzled by how this structure, made up of innumerable and complex interconnected nerves, produces our mental abilities. What is clear is that there are specific areas of responsibility—regions and groups of neurons that perform specific tasks. For example, movement sequences are controlled in the motor cortex, while the auditory cortex is responsible for processing auditory impressions. In addition to these rough classifications, researchers have also been able to identify increasingly finer specializations on certain aspects in different brain centers in recent years. The team led by Nancy Kanwisher from the Massachusetts Institute of Technology in Cambridge is also dedicated to this challenge.
On the trail of the neuronal echo of music
Specifically, the researchers are investigating how music is processed in the auditory cortex. The current publication builds on an earlier study in which the team was able to identify a population of neurons that respond specifically to music. This study used functional magnetic resonance imaging (fMRI), which can show activity in specific areas of the brain. To do this, the researchers scanned the brains of test subjects while they were listening to 165 sounds of different categories – including different types of speech and music as well as auditory impressions such as a dog barking. Using a special analysis method of the fMRI data, the scientists were able to identify neuronal populations with different reaction patterns – including one that reacts to speech and another that responds specifically to music.
The researchers are now showing that the neuronal specialization can be broken down even further using a new combination method. To do this, they used data generated by the procedure known as intracranial electroencephalography (EkoG). Here, activity patterns of nerves are recorded by electrodes placed in the skull. The invasive procedure is normally used to examine epilepsy patients as part of their treatment. The probes can provide more detailed insights than fMRI, which uses blood flow in the brain as an indication of neuron activity. “The data we have been able to collect so far tells us that this is a part of the brain that is doing something. We wanted to know more about what is being processed where,” says Kanwisher.
For their study, the scientists gained the support of 15 epilepsy patients. When their brain activity was captured by the EkoG method, they played them the same set of 165 audio samples that they used in the earlier fMRI study. With the help of statistical analyses, the researchers were able to deduce the neuronal populations that generated the data recorded by the individual electrodes. “This neuronal reaction pattern emerged that only arose with singing,” says lead author Sam Norman-Haignere. There appears to be a song-specific neuron population that responds only very weakly to speech or instrumental music. “That was a result that we had not expected,” says the researcher.
targeting vocal neurons
To further investigate the finding, the researchers then developed a data processing method to combine the information from the intracranial recordings with the fMRI data from their previous study. Since fMRI can cover a much larger part of the brain, this allowed them to more accurately pinpoint the locations of the neuronal populations that respond to song. It turned out that the vocal hotspot is located in the upper part of the temporal lobe, in the vicinity of the areas that respond to speech and more generally to music. “This kind of combination of EKoG and fMRI is a significant methodological advance,” emphasizes co-author McDermott.
The location of the song-specific neuron population suggests that they may respond to features such as perceived pitch or the interaction between words and perceived pitch before relaying the information to other parts of the brain for further processing, the researchers say. “The results thus provide evidence of a multi-layered separation of functions within the auditory cortex in a way that is consistent with an intuitive distinction within music,” says Norman-Haignere.
According to the researchers, however, the insights into the neuronal code of music have now raised a few more questions: What exactly do the “singing neurons” react to – are pitch and timbre or certain structures decisive? And how does selectivity develop over the course of life? The scientists now want to pursue these questions with more detailed investigations.
Source: Massachusetts Institute of Technology, professional article: Current Biology, doi: 10.1016/j.cub.2022.01.069