The Science of Playing by Ear

When Pradeep Sharma wanted to learn to play the piano more than a decade ago, he bought a keyboard and began to plink. He didn’t have sheet music for his favorite Bollywood songs and decided to sound them out by ear. 

As he played around, he learned that, unlike him, his wife had the ability to hear a song and easily pick out the correct keys to play it — even though she had no formal musical training. 

“She could figure out the melody in two minutes flat,” says Sharma, dean of the University of Houston Cullen College of Engineering and Hugh Roy and Lillie Cranz Cullen Distinguished University Professor. “That made me wonder what is so special about her that she can hear music so much better than I do.” 

With that spark of curiosity, Sharma set out to learn why some people seem to possess a natural gift for hearing music. He approached the question from the perspective of a mechanical engineer and physicist who specializes in materials, computational nanoscience and electricity. 

He took a mechanistic approach (rather than, say, sociological or neurological) to investigate how physical differences in special cell structures change our ability to discern between two tones. The more sensitive a person is to subtle differences between similar tones, he reasoned, the better they can identify distinct musical notes. 

The result was a paper titled “A minimal physics-based model for musical perception,” published in January 2023 in Proceedings of the National Academy of Sciences, a peer-reviewed journal of the National Academy of Sciences. 

Then, more good news: This spring, PNAS recognized Sharma’s paper, from among thousands, with the prestigious Cozzarelli Prize in the award’s Engineering and Applied Sciences category. The prize honors researchers whose PNAS articles display scientific excellence and outstanding originality. 

The paper’s coauthors all studied under Sharma at UH. Qian Deng was a postdoctoral researcher; Fatemeh Ahmadpoor and Kosar Mozaffari were doctoral students. “We did not really have any funding for it,” Sharma says. “So we kind of just did it on the side, slowly. It took me about 10 years.” 

The crux of their model examined the mechanical, electrical and geometrical design properties of hair cells, which are vital to how ears gather sound.

“I wanted to make the study as physical as possible, to directly link it to the physical attributes of the cells,” Sharma says. He likened his approach to looking at physical differences in musculature that might separate merely strong sprinters from the Olympic gold medalist. “Your hair cells might have slightly different physical properties. All of that could actually lead to a pronounced difference in how well you’ll hear music versus me.” 

Sharma boiled down the question of how humans discern tones to the components of cells that are partly responsible for hearing. He knew that those specialized cells have bundles of tiny cylinders, called stereocilia, that react to sound waves that enter your inner ear.

Even though they’re vanishingly small — maybe as long as a human hair is wide — stereocilia nonetheless abide by the principles of physics.  

Sharma and his colleagues applied principles of fluid mechanics and electromechanics to build a new model that explains, in physical terms, how differences in stereocilia affect how the stimulus of sound moves them. Those microscopic movements are crucial to our ability to hear. 

One key to that understanding is a phenomenon called flex electricity, or the electricity generated when mechanical forces differ across the surface of any object. In tiny, sensitive structures such as stereocilia, that electricity in turn creates signals that our brains interpret as sound. Thus differences in the physical properties of the stereocilia mean differences in their flex electricity, leading to different levels of sensitivity to sound. 

“We are talking about things like bending the cell membrane,” Sharma says. “If you bend the cell membrane, you produce electricity because of that.” 

Even at the microscopic level, small divergences in structure matter. “How stiff is your cell wall? What is the average length and diameter and structure of those stereocilia? Of course, everybody’s values will be around the same average roughly, because we are all human,” Sharma says. “But there’ll be some differences. And those differences will account for the differences in our hearing.”