Interactions with personified technologies are increasingly common in children's everyday experience. Personified technologies are designed to communicate information and interact with people using a repertoire of highly social behaviors and human-like personality traits. Little is known about the social and moral developmental implications of children's interactions with such technologies. This dissertation presents a study of children's social and moral conceptions of a personified software agent. Two overarching research questions were addressed: (1) Do children generalize their understanding of distinctions between conventional and moral violations in human-human interactions to human-agent interactions? and (2) Does the agent's expression of harm and claim to personal rights influence children's judgments? A two condition between and within subjects study was conducted in which 60 eight and nine year-old children interacted with a personified agent and observed a researcher interacting with the same agent. Conditions differed only in the response of the agent to a verbal insult. A semi-structured interview was then conducted to investigate the children's judgments of the observed interactions. Results suggest that children do distinguish between conventional and moral violations in human-agent interactions and that the agent's expression of harm and claim to rights significantly increases the likelihood that children distinguish the two violations.
展开▼
