ASIA: Attention driven pre-conscious perception for socially interactive agents

The development of socially interactive agents emerged as a challenging task due to paradigm shift from behavioral to motivational based agents. In current nascent scenario, machines may have their own motives and goals like humans, which subsequently enhance their capability to improve their communication during any social interaction. Moreover single general purpose learning is not applicable in every social interaction. These conditions improve attention and perception based subjective learning mechanism for agents to make better cooperation with human during any social interaction. In this paper, we propose a cognitive model of attention driven pre-conscious perception with subjective learning and social-human interaction for machines to improve their communications.