Event
Ph.D. Research Proposal Exam: Joshua P. Kulasingham
Friday, October 30, 2020
10:00 a.m.
https://umd.zoom.us/j/93608057822?pwd=WGJtWTVSQTBnN2hoUVowQ3hMWkZXUT09
Maria Hoo
301 405 3681
mch@umd.edu
ANNOUNCEMENT
Name: Joshua P. Kulasingham
Committee:
Professor Jonathan Z. Simon (Chair)
Professor Shihab Shamma
Professor Behtash Babadi
Date/time: Friday, October 30, 2020 at 10 a.m.
Location: Online https://umd.zoom.us/j/93608057822?pwd=WGJtWTVSQTBnN2hoUVowQ3hMWkZXUT09
Title: Time-Locked Cortical Processing of Speech in Complex Environments
Abstract:
Our ability to communicate using speech depends on complex, rapid processing mechanisms in the human brain. These cortical processes make it possible for us to easily understand one another even in noisy environments. Measurements of neural activity have found that cortical activity time-locks to the acoustic and linguistic features of speech. Investigating the neural mechanisms that underly this ability could lead to a better understanding of human cognition, language comprehension, and hearing and speech impairments.
We use Magnetoencephalography (MEG), which non-invasively measures the magnetic fields that arise from neural activity, to further explore these time-locked cortical processes. One method for detecting this activity is the Temporal Response Function (TRF), which models the impulse response of the neural system to continuous stimuli. Prior work has found that TRFs reflect several stages of speech processing in the cortex. Accordingly, we use TRFs to investigate cortical processing of both low-level acoustic and high-level linguistic features of continuous speech.
First, we find that cortical responses time-lock at high gamma frequencies (~100 Hz) to the low pitch segments of the acoustic envelope of speech. Older and younger listeners show similar high gamma responses, even though slow envelope TRFs show age-related differences. Next, we utilize frequency domain analysis, TRFs and linear decoders to investigate cortical processing of high-level structures such as sentences and equations. We find that the cortical networks involved in arithmetic processing dissociate from those underlying language processing, although both involve several overlapping areas. These processes are more separable when subjects selectively attend to one speaker over another distracting speaker. Finally, we propose novel algorithms that directly estimate TRF components, and seek to provide robust measures for analyzing group and task differences in these responses. Overall, this work provides insights into several stages of time-locked cortical processing of speech and highlights the use of TRFs for investigating responses to continuous speech in complex environments.
