Auditory perceptual organization

University Of Virginia, Charlottesville VA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Deformations in atmospheric pressure vibrate an observers' eardrum; this causes waves to be formed in the cochlea, and as a result we are able to perceive sounds in the environment. Obviously, a qualitative model of this nature is insufficient as a model of auditory perception - and yet it is precisely this type of qualitative model that researchers have been forced to use when dealing Gestalt grouping principles. For example, we say that sounds that are similar are more likely to group together, or that sounds close together in time or pitch are more likely to group together, but the question is: How likely? What we desire is a way to quantify these grouping principles. In this research proposal, we outline new theory and an experimental paradigm, which will allow us to do just that. This proposal outlines a methodology for the quantification of Gestalt grouping principles in auditory perception. We have developed a new type of stimulus, the tone lattice, which is perceived by observers in one of two exclusively competing ways. By systematically manipulating pitch relations within a tone lattice, we are able to change the likelihood that observers will hear it organized in one way or the other. Measuring these changes in observer percept probability with respect to the experimentally manipulated pitch relations allows us to calculate functions, attraction functions, which can be used to predict observer perceptions. Attraction functions are measures of the strength of grouping. We are then able to introduce other features into our tone lattices, such as similarity of elements, to see how the attraction functions change. What we have in fact developed, is a sort of "yardstick" for quantitatively measuring the effect of Gestalt grouping principles on the segregation and formation of auditory objects.