Organization of the neural system for pursuit eye movements.
Our long-term goal is to assemble our extensive knowledge of the neural basis for pursuit into a biomimetic neural circuit model that operates autonomously and dynamically on a time scale of milliseconds for a full 1-second pursuit response. We possess extensive data on the responses of neurons in sensory area MT, motor area FEFsem, the dorsolateral pontine nucleus, the nucleus reticularis tegmenti pontis, and the floccular complex of the cerebellum. Our neural circuit model will include local recurrent circuits at each node of the system, and long-range forward and recurrent connections among the nodes. We will explore what circuits and connections are needed so that the model neurons reproduce the diversity of neural responses in each area as well as the first- and second-order statistics of the neural responses and the behavior.
Our recordings are obtained with single electrodes, tetrodes, or multi-contact probes during multiple behavioral paradigms. These include the initiation of pursuit and steady-state tracking for different target speeds and the effects of reduced motion reliability through control of stimulus contrast or the coherence of a dot patch. We study the relationship between neural and behavioral responses during brief pulses of target motion under different conditions of modulation of the strength (“gain”) of visual-motor transmission. We explore how the neural system changes in relation to a reliability-weighted combination of sensory data and past experience. To complete our data set, a number of experiments are needed in the cerebellar flocculus, MT, FEFsem, and we plan to explore responses in the basal ganglia. Finally, we are open to the strategic use of optogenetics in monkeys to crack the neural circuit and explore the predictions of our circuit model.
Multiple components of learning in pursuit eye movements.
A single change in target direction causes a small change in eye movement in the subsequent trial. An important neural correlate of "single trial learning" occurs in the floccular complex of the cerebellum, where a single complex spike on one learning trial causes a properly-timed depression of simple spike response on the subsequent trial. Yet, there are multiple components of learning that have different time courses of acquisition. Preliminary data indicate that neural learning in the cerebellar cortex declines after a few hundred learning trials, even as behavioral learning continues to increase. We are investigating the neural expression of learning over short and long time courses in both the cerebellum and the cerebral cortex
Circuit mechanisms for cerebellar function and learning.
We are using multiple-contact electrodes to record simultaneously from multiple neighboring single-units in the cerebellar cortex of the floccular complex. Use of an improved spike-sorter designed for our use case, we can identify essentially all of the spikes of many neurons simultaneously and identify the layer of each recording. Through calculation of spike-timing cross-correlograms with Purkinje cells, we can determine the anatomical identify of interneuron recordings. After identifying the cell-types of isolated neurons and recording their activity during smooth pursuit eye movements and pursuit learning, then we will construct a circuit model of the cerebellar cortex and explain how the circuit works and learns.