Exploring How Primary School Children from a Low-SES Community Engage and Verbally Communicate with a Social Robot

Abstract

Social robots have the potential to support learning in the early years of primary school. Further research is needed to understand how children from low SES communities engage and verbally communicate with social robots. Children (N = 105; Mean age = 6.5 years) attending a primary school in Queensland, Australia, participated in this study. Each child was observed during a one-on-one session with a humanoid social robot (NAO). Children’s engagement and words spoken during a child-robot question-and-answer activity were examined. Nearly half the children (49.5%) responded to the questions asked by the social robot. Children spoke to the social robot an average of 47 words with an average length of four letters over the 4-minute activity. The number and length of words spoken by children was positively associated with children’s engagement with the social robot even after controlling for child age. Family SES was not related to any measures, suggesting that other factors may be more influential in predicting engagement and verbal communication with a social robot than SES status. The findings demonstrate that children from low SES communities do engage and verbally communicate with a social robot.