Project Summary: Plasticity of Neural Integrator Dynamics<br/><br/> The oculomotor neural integrator converts brief input signals for rapid eye movements into sustained patterns of neural activity that maintain tension in eye muscles to hold the eye at a fixed angular position. The leading candidate mechanism for the production of this sustained activity is re-excitation of neurons through recurrent feedback loops, a mechanism that requires precise tuning of the amount of positive feedback for proper operation. Normally, visual signals produce the appropriate amount of tuning by changing the properties of the sustained activity in the neural integrator. Experimentally, the plasticity will be artificially controlled by a training paradigm consisting of a visual pattern rotating with an angular velocity dynamically controlled in real time by eye position. Optimal training protocols will be established and changes in the properties of neurons in the neural integrator will be quantitatively measured. These data will be used to incorporate learning rules into theoretical recurrent neural network models of the neural integrator. The conversion of a transient input into a sustained neural output is a form of short term memory that is widely observed in the nervous system. Thus better understanding of the role tuning mechanisms contribute to neural integrator function should have broader significance in neurobiology.