ABC Seminar: Cortical mechanisms underlying perception in complex auditory scenes

2018-05-17 14:15:00 2018-05-17 15:15:49 Europe/Helsinki ABC Seminar: Cortical mechanisms underlying perception in complex auditory scenes Professor Shibab Shamma (U of Maryland) will talk about how we make sense of our auditory environment. His talk covers how we perceive complex auditory scenes, and how the brain processes the streams of pitch, timbre and location information. http://ani.aalto.fi/en/midcom-permalink-1e851ba37f7662e51ba11e892af11516c3da2e1a2e1 Otakaari 1, 02150, Espoo

Professor Shibab Shamma (U of Maryland) will talk about how we make sense of our auditory environment. His talk covers how we perceive complex auditory scenes, and how the brain processes the streams of pitch, timbre and location information.

17.05.2018 / 14:15 - 15:15
U3, Otakaari 1, 02150, Espoo, FI

Professor Shibab Shamma, Institute for Systems Research, Department of Electrical and Computer Engineering, University of Maryland

Cortical mechanisms underlying perception in complex auditory scenes


Abstract: Humans and many animals can easily attend to one of multiple similar sounds, segregate it, and follow it selectively over time even in extremely noisy environments. I shall review the psychological and neural underpinnings of this perceptual feat, explaining how it fundamentally depends on the representation of sound in the auditory cortex. I shall then outline how the attributes of simultaneous sounds (pitch, timbre, location) are disentangled into separate streams, and bound into unified sources, by the temporal coherence of the cortical responses they evoke and their rapid adaptive properties. Recent neurophysiological results in support of these ideas will be discussed, especially in animals that have been trained to segregate and attend to the speech stream of a target speaker in a mixture. Finally, these findings will be related to analogous tasks in other sensory systems (visual object segregation in crowded scenes), leading to biologically inspired computational algorithms that can perform these tasks with no prior information or training.

Coffee and pulla are served after the talk.