Learning Self-Awareness: Lessons from Social Robotics

In this paper, we propose a theoretical model for implementing recognition of human qualia based on machine learning. Our model draws from the idea of mapping affects onto a multi-dimensional space of sensory stimuli. The idea has background in philosophy and cognitive science but we believe it can make a contribution to the fields of social robotics and social assistive robotics. The paper has two main parts. In the first part, we discuss some defining characteristics of consciousness referred to by contemporary scholars of robotics. We also explain the hard problem of qualia. In the second part, we present our own proposal of how to make qualia recognizable to a robot.