Date of Award

5-2017

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Program

Speech and Hearing Science

Track

Speech-Language Science

Research Advisor

Devin Casenhiser, Ph.D.

Committee

Ferenc Bunta, Ph.D. Ashley W Harkrider, Ph.D. Mark S Hedrick, Ph.D. Tim Saltuklaroglu, Ph.D. Antonio J. S. Teixeira, Ph.D.

Abstract

Models of speech perception suggest a dorsal stream connecting the temporal and inferior parietal lobe with the inferior frontal gyrus. This stream is thought to involve an auditory-motor loop that translates acoustic information into motor/articulatory commands and is further influenced by decision making processes that involve maintenance of working memory or attention. Parsing out dorsal stream’s speech specific mechanisms from memory related ones in speech perception poses a complex problem. Here I argue that these processes may be disentangled from the viewpoint of the temporal dynamics of sensorimotor neural activation around a speech perception related event.

Methods: Alpha (~10Hz) and beta (~20Hz) spectral components of the mu () rhythm, localized to sensorimotor regions, have been shown to index somatosensory and motor activity, respectively. In the present work, event related spectral perturbations (ERSP) of the EEG -rhythm were analyzed, while manipulating two factors: active/passive listening, and perception of native/nonnative phonemes. Active and passive speech perception tasks were used as indexes of memory load employed, while native and. nonnative perception were used as indexes of automatic top-down coding for sensory analysis.

Results: Statistically significant differences were found in the oscillatory patterns of  components between active and passive speech perception conditions with greater  alpha and beta event related desynchronization (ERD) after stimuli offset in active speech perception. When compared to listening to noise, passive speech perception presented significantly (pFDR<0.05) stronger alpha and beta ERD during and after stimuli presentation. When comparing native to nonnative speech perception, stronger alpha (8-14Hz) and beta (22-25Hz) event related synchronization (ERS) were observed before and during stimuli onset in the passive nonnative task. Passive native perception, on the other hand, presented stronger alpha and especially beta ERD before stimuli onset as well as stronger alpha ERD between presentation of the two syllables composing the stimuli (600-850ms), and during the presentation of the second syllable (1000ms).

Conclusion: These findings suggest that neural processes within the dorsal auditory stream are functionally and automatically involved in speech perception mechanisms. While its early activity (shortly after stimuli onset) seems to be importantly involved with the instantiation of predictive motor/articulatory internal models that help constraining speech discrimination, its later activity (post-stimulus offset) seems essential in the maintenance of working memory processes.

ORCID

http://orcid.org/0000-0002-1061-0801

DOI

10.21007/etd.cghs.2017.0431

Share

COinS