Reinforcement learning accounts for zone allocation behavior between two competing mice

Abstract

Animals show various response when competing over a limited amount of resources. In the case of humans, due to a long history of being a part of great societies, altruistic and cooperative behaviors are easily observed. However, in non-human animals, aggression and selfish behaviors are more readily observed. We observed two mice showing cooperation-like behavior (Henceforth referred to as ‘Cooperative zone allocation behavior’) in competitive situations over a reward. If we interpret this observation from the viewpoint of ethology, we can say that they made a heuristic rule of reciprocity. However, this can also be interpreted as a result of reinforcement learning of each individual mouse. Here, we attempt to explain this zone allocation behavior from the aspect of reinforcement learning and compare it with the aspect of ethology.