Thomas M. Fischer - US grants

Neural networks are dynamic systems that enable organisms to adapt to and learn about complex and variable environments. At the core of this adaptive process is the phenomenon of synaptic plasticity, an alteration in the efficacy of a synapse due either to intrinsic processes (e.g. activity of the neuron) or extrinsic processes (e.g. neuromodulators released by other neurons). The functional significance of synaptic plasticity will be examined by focusing upon a specific set of identified inhibitory neurons within a well-defined neural circuit in the marine mollusc Aplysia. Inhibitory processes can have profound consequences on neural function, yet have received comparatively little experimental emphasis. By focusing on a specific synaptic process in a specific set of inhibitory neurons, these studies will allow for a simultaneous exploration of the function of both synaptic plasticity and inhibition in behavioral regulation. Two basic issues will be examined, integrating experiments at the level of synapses with the level of behavior: (1) The role of synaptic inhibition in behavioral regulation will be explored, examining the hypothesis that plasticity of inhibition may serve as a dynamic gain control mechanism that can regulate behavior across rapidly-changing environmental conditions. (2) The role of inhibition in learning will be explored, by examining the hypothesis that plasticity of inhibition may serve to dynamically regulate the capacity for neural networks to express learning-related changes.
Information obtained from this project could help to elucidate general computational principles utilized by a neural networks to generate adaptive modifications, which can potentially be employed in algorithms usable by other computational or information processing devices (e.g. smart machines). From a theoretical perspective, it could address fundamental questions concerning the nature of adaptive modifications within neural networks that are required for the expression of different forms of learning and memory.

The goal of this research is to achieve a quantitative description of network processes that underlie one of the most basic tasks performed by nervous systems: to use information from the environment to adaptively regulate behavior. These experiments will be carried out on the siphon withdrawal reflex (SWR) of the marine mollusc Aplysia californica, an experimental system that allows network-level events to be directly related to behavioral outcomes. The first major objective of this project is to achieve a quantitative description of cellular events that underlie behavioral regulation in response to ethologically relevant environmental change (calm to turbulent). A series of physiological experiments will be performed to characterize sensory processing of environmental information, as well as to measure subsequent changes in the intrinsic and synaptic properties of network interneurons and motor neurons. These measures will be used to construct a formal computational model of environmental information processing that will allow us to systematically explore how multiple network processes interact to regulate behavior. The second major objective of this project is to further develop and apply this computer model to aid our understanding of a simple form of learning, habituation. Incorporated into the model will be physiological measurements obtained within the SWR circuit in response to habituation training, and the model will be used to achieve a quantitative description of cellular processes underlying this basic form of learning.
The broader impacts of this research lie in the ubiquity of the basic synaptic and cellular processes analyzed, which are common to all neural systems and have been shown to be important in their regulation. The end result of this project should shed light not only on the neural basis of a specific behavior and its regulation, but also on the computational principles by which any network dynamically regulates its responsiveness. This research will provide a multi-disciplinary training environment for both graduate and undergraduate students seeking training in behavioral, physiological and computational methods. The project involves establishing a unique research-training pipeline with an outstanding liberal arts undergraduate institution, Kalamazoo College, which will allow summer research opportunities to be realized at Wayne State University.