The Conversation: Hearing and Speaking
Three Dimensions from Two Ears
If language is an extremely complex system, hearing is another. H. Steven Colburn, professor of biomedical engineering and director of BU's Hearing Research Center, emphasizes, “Even putting aside the biology and just trying to solve it from a pure engineering standpoint, the interpretation of sound is a messy problem.”
After listening to a binaural simulation of sound processing with two cochlear implants, H. Steven Colburn discusses the experiment with Kelli Bechly, a senior who spent the summer conducting research in the Biomedical Engineering Department through the NSF-sponsored Research Experience for Undergraduates program.
“The auditory system,” he explains, “has no spatial dimension, other than the fact that we have two ears. Sound arrives in the form of two pressure waves that vary in time. We have to create a sense of space from these two temporal wave forms.”
Colburn's work focuses on the ways in which we exploit the two reception points of our auditory system—our binaural hearing—to interpret a three-dimensional universe. Our pair of ears help locate a sound in space in two ways. First, if a sound is on one side of us or the other, our head “shadows” the sound, making it less intense from the more distant ear. Second, there is also a slight timing difference between the moment that the sound arrives at the closer ear and the moment it arrives at the farther one, which allows us to determine where the sound lies in the horizontal plane.
“We can perceive a five or ten micro-second difference. That blew my mind when I learned it,” says Colburn, who, as a graduate student, arrived at the study of hearing from an interest in signal processing in the field of electrical engineering.
Colburn points out how challenging such fine analysis is in a noisy environment, where we not only have to “slice up” the pressure wave to identify the sources of various noises, but we also have to regroup the relevant parts of the sound in order to pay attention to, for example, one particular conversation. “This ability to sort out the sources of sound and control which one you listen to has long been described as 'the cocktail party effect.' It's the challenge I'm interested in, but I don't mean to imply that it's the only major challenge in hearing. However, it is the one hearing-impaired people complain about most.”
Colburn explains that modern life is organized in such a way that a sophisticated task is made even more difficult. “We've structured our environment today—with traffic noise, manufacturing noise, mechanical noise—so that our expectations are at the limits of what we can do,” he says.
In his lab, Colburn tests the limits of the auditory system in people with normal hearing, as well as in those with impaired hearing. His work has important implications for people whose hearing loss requires a cochlear implant. “Most people with a single cochlear implant do pretty well in simple environments, but they still have great difficulty in noisy environments.”
Colburn's work may eventually offer a strong argument for two implants and may also illuminate ways in which implants need to be improved in order to function together with something approaching the finesse of two ears. “Just as we have better depth perception with two eyes,” he says, “we have better spatial localization with two ears.”
For more information, see www.bu.edu/dbin/bme/faculty/?prof=colburn.
