Envoyer par SMS: Linear reinforcement learning in planning, grid fields, and cognitive control