Neuroevolution of a recurrent neural network for spatial and working memory in a simulated robotic environment

We evolved weights in a recurrent neural network (RNN) to replicate the behavior and neural activity observed in rats during a spatial and working memory task. The rat was simulated using a robot simulator to navigate a virtual maze. After evolving weights from sensory inputs to the RNN, within the RNN, and from the RNN to the robot's motors, the robot successfully navigated the space to reach four reward arms with minimal repeats before the timeout. Our current findings suggest that it is the RNN dynamics that are key to performance, and that performance is not dependent on any one sensory type, which suggests that neurons in the RNN are performing mixed selectivity and conjunctive coding. The RNN activity resembles spatial information and trajectory-dependent coding observed in the hippocampus. The evolved RNN exhibits navigation skills, spatial memory, and working memory.