Date

2015

Document Type

Dissertation

Degree

Doctor of Philosophy

Department

Integrative Biology

First Adviser

Burger, R. Michael

Other advisers/committee members

Haas, Julie S.; Iovine, M Kathryn; Xu-Friedman, Matthew A.

Abstract

Auditory stimuli are processed in parallel frequency-tuned circuits, beginning in the cochlea. Auditory nerve fibers impart this ‘tonotopic’ organization onto nucleus magnocellularis (NM), a division of the cochlear nucleus specialized for maintaining and improving temporal representation of the acoustic waveform. Auditory neurons primarily perform this task by phase-locking, whereby spikes occur at a particular phase of the auditory stimulus. NM’s sole target is nucleus laminaris (NL), which performs binaural comparisons of input timing in order to localize sound stimuli. Therefore, NM’s sole function is to preserve and improve its spike timing precision over a broad range of stimulus frequencies and intensities.Though all NM neurons perform this task, synaptic and membrane properties differ along the tonotopy, and thus may confer computational specificity. Low characteristic frequency (LCF) neurons receive many subthreshold auditory nerve fiber inputs and have a low spike threshold, while high characteristic frequency (HCF) neurons receive only 1-2 large endbulb synapses and have much higher thresholds. In this study, I investigate the tonotopic distribution of synaptic and membrane properties in NM, and demonstrate a functional interaction that enables temporally optimal output to NL. First, I show that auditory nerve inputs to LCF neurons are less effective at driving high stimulus rates, and are more susceptible to short-term synaptic depression than inputs to HCF neurons due to a higher vesicle release probability. Then, I demonstrate that the integrative properties of the postsynaptic membrane are tonotopically distributed in a way that improves phase-locking. HCF neurons select for more temporally synchronous input than their LCF counterparts, and are thus better able to discriminate between inputs associated with successive stimulus periods.Together, these data suggest that NM neurons preserve temporal precision by using different computational strategies along the tonotopy. LCF neurons integrate many converging subthreshold inputs and improve phase-locking to low frequency stimuli, while rapid membrane properties likely allow HCF neurons to relay the spike timing of a single robust input in order to represent high frequency stimuli.

Share

COinS