A computational theory for a smooth pursuit model based on the population dynamics in the MST area

摘要

Several studies for oculo-motor system have shown the possibility that the cerebellum has the inverse dynamics model of the eye plant. However, it has been unknown where and how the desired trajectory signals for the inverse model are calculated. In this paper, we investigate the possibility that the NIST area in the higher visual cortex has the internal model to calculate the desired trajectory. We propose the smooth pursuit model which has recurrent neural networks in the NIST area based on the physiological data, and we show that our model can reproduce the smooth pursuit eye movement plausibly. Then, we explore the computational theory for the proposed model. Our results suggest that the MST area possibly estimates and predicts the target motion.