Neural Representation of Auditory Temporal Structure

Computation and Neural Systems

Lewicki, Michael Samuel ; Konishi, Masakazu

University:California Institute of Technology

Department:Biology

关键词: Computation and Neural Systems;   

Others  :  https://thesis.library.caltech.edu/7592/1/Lewicki_ms_1996.pdf

美国|英语

来源: Caltech THESIS

【 摘 要 】

Neurons in the songbird forebrain nucleus HVc are highly sensitive to auditory temporalcontext and have some of the most complex auditory tuning properties yet discovered. HVcis crucial for learning, perceiving, and producing song, thus it is important to understandthe neural circuitry and mechanisms that give rise to these remarkable auditory responseproperties. This thesis investigates these issues experimentally and computationally.

Extracellular studies reported here compare the auditory context sensitivity of neuronsin HV c with neurons in the afferent areas of field L. These demonstrate that there is asubstantial increase in the auditory temporal context sensitivity from the areas of field Lto HVc. Whole-cell recordings of HVc neurons from acute brain slices are described whichshow that excitatory synaptic transmission between HVc neurons involve the release of glutamateand the activation of both AMPA/kainate and NMDA-type glutamate receptors.Additionally, widespread inhibitory interactions exist between HVc neurons that are mediatedby postsynaptic GABA_A receptors. Intracellular recordings of HVc auditory neuronsin vivo provides evidence that HV c neurons encode information about temporal structureusing a variety of cellular and synaptic mechanisms including syllable-specific inhibition,excitatory post-synaptic potentials with a range of different time courses, and burst-firing,and song-specific hyperpolarization.

The final part of this thesis presents two computational approaches for representingand learning temporal structure. The first method utilizes comput ational elements that areanalogous to temporal combination sensitive neurons in HVc. A network of these elementscan learn using local information and lateral inhibition. The second method presents amore general framework which allows a network to discover mixtures of temporal featuresin a continuous stream of input.