- Starts: 5:00 pm on Tuesday, July 10, 2018
Title: “Modeling Brainstem Neurons that Encode Auditory Space via Time Delay”
Committee:
Kamal Sen, PhD – BU BME (Chair, Co-Advisor)
David McAlpine, PhD – Macquarie University, Linguistics (Co-Advisor)
Steve Colburn, PhD – BU BME
Fernando Fernandez, PhD – BU BME
Abstract:
Spatial hearing comprises the acoustic cues and neural mechanisms we use to localize sounds. As sound from the side arrives first and more intensely at the near ear, binaural spatial hearing encodes and processes this interaural time difference (ITD) and interaural level difference (ILD). Understanding brain mechanisms of spatial hearing will help inform development of aids for selectively listening in noisy, everyday environments. In this proposed dissertation, ITD-encoding neurons are being computationally modeled from biophysics to a simple behavioral task. Results include a new biophysical-level model MSO neuron, and matching human ITD-discrimination performance by equivalent model neurons within established population models for ITD representation, which assume specific distributions of “best-ITD” (ITD at which a neuron responds most strongly). A primate-based distribution will be applied. In modeling neuronal delays, bilateral asymmetries in synaptic inhibition and axon location shifted best-ITD; dendritic asymmetries will be tested. Modeling synaptic depression in the cochlear nucleus emphasized ITD-encoding in the MSO during rising amplitude, which is when direct sound for localization is strongest compared with reverberation. This suggests that assistive listening devices should provide ITD during rising sound amplitude. Further simulations relevant to cochlear implants will relate envelope-ITD sensitivity in existing model LSO neurons to membrane speed, peripheral/processor filters, and human performance.
- Location:
- 44 Cummington Mall, room 401