In a study recently published in Communications Biology, auditory neuroscientists from the University of Pittsburgh uncovered a machine learning model that helps us to understand how the brain processes and recognizes different communication sounds. This is important for understanding and treating disorders that can affect people’s ability to understand speech, and improving hearing aids.
The research team examined the process of vocal communication in both animals and humans and found that communication sounds, such as animal calls and words, are made up of small characteristics. This is similar to how our brain recognizes and identifies faces, where we are able to pick up on subtle features like the eyes, nose and mouth.
To test this theory, the team ran experiments with guinea pigs by exposing them to different categories of vocal communication sounds. They monitored the guinea pigs’ brain activity while they responded to the sounds and compared the responses with the machine learning model. The results showed that the guinea pigs’ neurons responded to the sound categories, indicating that the machine learning model was able to accurately process the sounds.
The team also tested the ability of the animals to understand modified versions of the vocal communication sounds. They made the vocal sounds unnaturally high or low in pitch, added noise and echoes, and sped them up or slowed them down. Despite this, the animals were still able to recognize the modified sounds and respond to them in the same way they did when they were unaltered.
Lead author Satyabrata Parida, Ph.D., a postdoctoral fellow at the University of Pittsburgh’s department of neurobiology, explained the implications of the research. “From an engineering viewpoint, our model is unique because it has a close correspondence with behavior and brain activity, giving us more insight into the biology,” he said. “In the future, these insights can be used to help people with neurodevelopmental conditions or to help engineer better hearing aids.”
The University of Pittsburgh is well-known for its world-class research, education, and healthcare. Founded as the Pittsburgh Academy in 1787, the university has grown to become one of the most prestigious institutions of higher learning in the United States.
Srivatsun Sadagopan, Ph.D., is an assistant professor of neurobiology at the University of Pittsburgh who led the research team. He has been an active researcher in auditory neurosciences for many years, and his research focuses on understanding the biological processes behind sound recognition. Additionally, he hopes to one day use what they’ve learned to help those with neurological conditions affecting their ability to understand speech.
