Title: Neural Representations of Speech, and Speech in Noise, in Human Auditory Cortex

Speaker: Dr. Jonathan Simon

Professor, Department of Electrical and Computer Engineering, University of Maryland College Park

Time: September 19 2016 (Mon) 10:00-12:00

Venue: Wangkezhen Building 1115

We investigate how continuous speech, whether presented alone, degraded with noise, or masked by other speech signals in a complex auditory scene, is represented in human auditory cortex. We use magnetoencephalography (MEG) to record the neural responses of individuals listening to continuous speech in a variety of listening scenarios. Comparisons of the dynamic cortical responses and the auditory stimuli reveal a common set of cortical representations of natural speech. Interestingly, we find that cortical representations of continuous speech are robust to noise under a wide variety of conditions, including clean speech, additive stationary noise, band-vocoded speech and noise, and speech with competing speakers. In the last case, individual neural representations of the speech of both the foreground and background speaker are observed, with each being selectively phase locked to the rhythm of the corresponding speech stream. Critically, the cortical reprsentation allows the temporal envelope of the acoustic speech stream to be reconstructed from the observed neural response to the speech. Finally, we show how these cortical represenations of speech are strongly affected by aging.

Host::Huan Luo