Monaural adaptive mechanisms in human sound localization

Abstract

Human sound localization results primarily from the processing of binaural differences in sound level and arrival time for locations in the horizontal plane and of spectral shape cues generated by the head and pinnae for positions in the vertical plane. Both types of cues have to be weighted and mapped to form a single percept of sound location. The experiments in this thesis take a behavioral approach to elucidate several aspects of this cue convergence and mapping. This was done by observing the localization behavior of monaural listeners or by actively manipulating the various localization cues of normal-hearing binaural listeners. The monaural listeners strongly rely on the ambiguous head shadow effect, which may help them to cope with familiar acoustic environments. Only some of these listeners could incorporate spectral cues to localize the sound-source azimuth, but those listeners were also able to localize sound-source elevation. Normal-hearing binaural listeners seem to disregard the spectral cues and the head shadow cues as valid localization cues. However, when faced with degraded binaural cues (as when one of their ears is plugged, at low sound levels), spectral cues do seem to play a role in sound azimuth localization. At normal sound levels, when binaural cues are present, new spectral cues (through application of a mold in one ear) do not degradate sound-azimuth performance. Sound-elevation performance, however, is degradated at the mold-side. Listeners can adapt to these new spectral cues by relearning and remapping these cues to valid locations.