Prednáška

Abstrakt

The neural mechanisms that underlie spatial processing in simple auditory scenes are well described by a series of bottom-up processing steps. Much less is known about auditory spatial perception in complex scenes. For example, few previous studies have examined listeners’ ability to localize a talker in a complex mixture of sounds, or looked at how their localization strategy changes when the listener has some a priori information about the spatio-temporal arrangement of the target and distractor objects in the scene. Similarly, it has been previously shown that misaligned audio-visual inputs can cause a lasting recalibration in the representation of the auditory space (the “ventriloquism aftereffect” illusion). However, it is not known how visual and auditory spatial representations are aligned in order to enable such visually guided recalibration. Here I present results of a series of behavioral and neuroimaging experiments that investigated human sound localization and speech perception in complex auditory and audio-visual scenes. I show that the contribution of the top-down factors is considerable in such scenes, and that top-down factors can have both beneficial and detrimental effects on localization performance. In this talk, I will also provide a brief overview of current and recent projects running in the lab.