An adaptive neural model for mapping invariant target position.

摘要: We perceive a constant target in space as even though the registration of that on our senses is continuously shifting. This article derives and stimulates neural network model represents visual spot targets, invariant with respect to any combination egocentric measures. The terms signals used move space. learns maintains precise sensory-motor calibrations starting only loosely defined relations. It adaptive physical changes eye muscles well internal system parameters. Its performance noise fault tolerant. Computer simulations show average error orientation after learning about 1% total field extent. good accuracy many different parameter choices. most related function posterior parietal cortex. Testable predictions are made for columnar topography brain structure.