The impact of different competence levels of Care-Receiving Robot on children

A Care-Receiving Robot (CRR) receives care from the people around it. The concept of CRR was first proposed in 2009 [1], [2] by the second author of this paper. This is a novel concept developed in response to the need for continuing to involve human teachers and adult care-givers in raising and nurturing children even as we move on to the next step in technological development and gradually begin to incorporate robots in our daily lives and most importantly in our educational environment. In the past decade, the interest in the development of educational agents has gradually grown to include robots [3]. Educational robots so far have been seen in the teaching role [4], [5], [6]. In contrast, this concept explores the learner role for the robot even though its goal is to teach. The idea behind CRR is inspired by the concept of learning by teaching developed formally in the 1980s [7] and further tested in other works [8]. We have conducted a series of experiments to test the concept of CRR. We have formally launched an investigation into the feasibility, benefits, and requirements for implementing CRR in real classrooms [9]. In this paper, we will report our early findings about the environmental and circumstantial setup for successfully implementing CRR and its impact on students in terms of learning reinforcement. We have set CRR's competence level to two opposite values in the set of experiments we describe in this paper. We present here the comparison data from our analysis of the interaction rate during the two conditions. Our most prominent finding is that children are quite happy to play ‘teacher’ to a weaker student (CRR). The benefit of this opportunity is that it not only boosts the child's self-confidence about the topic, but also reinforces the existing knowledge of the child.