The role of primary audible cues in sound localisation

Abstract

Sound localisation is a complex perceptual process that involves the integration of information derived from multiple audible cues. Human sound localisation is mediated by these audible cues. The primary audible cues are the interaural time difference (ITD), phase difference (IPD) and intensity difference (IID) that result from diffraction of sound waves around the head and pinnae. The ITD and the IID have received much attention in the literature as they are considered to be the most important binaural cues. These cues result from a neural comparison between the signals at both ears. Although these cues are signal-dependent, they have not been investigated using day-to-day life signals due to the experimentation complexity. Many research studies employed either single frequency or unnatural signals to draw conclusions. Further, the IPD could not be distinguished from the ITD for multiple frequency signals. This thesis presents new experimental techniques which are developed to investigate the role of primary audible cues in sound localisation. These techniques independently control and investigate the cues with multiple frequency signals. Listening experiments are initially completed with single cues which are followed by experiments with cues in conflict. The stimuli are chosen to include representative day-to-day life signals. Two audible cues at a time are used simultaneously in conflict so that their relative importance can be determined. Experimental techniques included a mathematical approach to manipulate the phase of all the component frequencies in a multiple frequency signal, whilst leaving the amplitude structure unchanged. This approach is adopted to distinguish the IPD from an equivalent ITD for multiple frequency signals. It is therefore possible to investigate the IPD independently with multiple frequency signals. The experimental results indicate that the effect of the IPD can be compensated for by an appropriate opposing ITD. In general, localisations become centrally diffuse when either the ITD or the IPD is placed in conflict with the IID. A localisation model is developed to provide predictions for the effect of one cue against an alternative cue, similar to the experiments. The results from the model are in agreement with the experiments. It is not evident whether the ITD or the IPD is more effective. The new techniques and the localisation model provide the opportunity to investigate the primary audible cues with representative day-to-day life signals. It has become possible to independently assess these cues and to draw conclusions based on their role in sound localisation.

Date of Award	Apr 2017
Original language	English
Awarding Institution	University of Brighton