Spatial Hearing in Complex Sound Fields

University Of Massachusetts Amherst, Amherst MA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): In typical situations the acoustic signals reaching our ears do not possess the simple and unambiguous directional cues that characterize those produced by a single source in an anechoic environment. Rather, the received signals are most often created from a mixture of direct sound from multiple sources as well as reflections from walls, floors, ceilings, and other surfaces. In order to make sense of the acoustic environment, the signals arriving at the ears must be analyzed to find separate auditory objects arising from different sources, such as a person talking, music from a television, or an air conditioner. To add to the complexity, reflections must not be perceived as separate sources, but instead be integrated with the direct sound and associated with the correct individual source. It is truly remarkable that we are capable of forming and maintaining auditory objects - and localizing them correctly -- in the face of such complex inputs. To succeed in these situations listeners benefit greatly from having two ears separated by the head and with each ear providing input to a central auditory system capable of binaural analysis. Failure of communication can result in consequences ranging from significant long-term issues with classroom learning and achievement, to immediate serious misunderstanding of what is said during emergencies. Most importantly, people with hearing losses appear to have a significantly lower tolerance for such situations, and the prostheses that help in many other ways unfortunately do little to increase this tolerance. We do not have a good understanding of what is involved in successful listening in complex environments, even by normal-hearing listeners. The goal of this project is to make major inroads toward that understanding. The first of three specific aims is to characterize the processes that allow us to form a single perceptual image from sources and reflections. These studies will investigate the hypothesis that fusion is enhanced as the listener constructs an internal model of the acoustic spatial environment. The second aim is to discover the auditory mechanisms that determine where this single sound image is localized. Three distinct mechanisms appear to be involved. The goal is to isolate and understand these separate processes and learn how they work together to facilitate localization of everyday sounds. The third aim is to understand how the individual, correctly-localized images described above are maintained in the presence of competing sound sources, and how this ability is used in conjunction with knowledge of linguistic context to improve listener performance in complex listening environments.