BIOPHYSICAL CHANNEL MODELS OF ASSOCIATIVE LEARNING

点击翻译按钮获取中文摘要

This is a request for a Scientist Development Award. The goal of the proposed research is to begin to understand, in quantitative detail, mechanisms of learning and memory in Hermissenda crassicornis, an invertebrate marine snail that shares behavioral characteristics and biophysical and biochemical mechanisms of learning and memory with mammals. Hermissenda demonstrates associative learning in response to paired presentations of light (Conditioned Stimulus) and turbulence (Unconditioned Stimulus). Acquisition and storage of this memory occurs m the medial B photoreceptor cell, the site of convergence of the light-evoked signal and the turbulence-evoked signal. Although the long term changes in the B cell membrane corresponding to memory storage have been characterized essential details of some of the biophysical mechanisms underlying associative learning are still unknown, and the CS and UCS interactions leading to the behavioral characteristics of associative learning are not completely understood. To achieve an in-depth understanding of some of these essential issues, the research plan describes an inter-related series of biophysical experiments and mathematical modeling organized around the following three questions: (1) What properties of the light induced currents makes stimulus coding sufficiently sensitive to permit the degree of stimulus specificity seen in Hermissenda? (2) Are there pre-synaptic mechanisms, as well as post-synaptic mechanisms, underlying the learning induced change in the effect of hair cell stimulation on the B photoreceptor? (3) What is the functional role of each of the currents, and of interactions among currents with respect to behavioral characteristics of classical conditioning? The biophysical experiments and mathematical modeling required to address these issues are necessary and sufficient training to enable me to become an independent investigator and to make significant contributions in the field of computational neuroscience. Quantitative understanding of biophysical and biochemical interactions underlying association learning in Hermissenda produced by the proposed research also will contribute to understanding of associative learning in mammals. Development of concise mathematical descriptions of the channel models would make possible the formulation of improved learning rules, network architectures and neuronal elements in artificial neural networks.