Learning Object Signi cance with an Emotion based

This paper presents an emotion based process generating cognition through learning. Two models are described which investigates the possibilities of designing cognitive systems based on emotion. In order to test these models through experiments, we used a mobile robot simulator, Webots, modeling the real Khepera robot. The rst proposed model is the simplest since it uses only positive and negative reinforcement signals as well as proximity sensors. The robot learns to avoid obstacles based on reeex reaction and on reinforcement signals. The second model relying on appraisal theory is more complex. It involves the use of reinforcement signals, proximity sensors and a simulated color camera to learn, in particular, to associate visual stimulus with the appropriate behavior. Interesting simulation results demonstrate the signiicance of this promising approach.