Relational AI

Relational AI

Creating long-term interpersonal interaction and shared experiences with social robots 

Many of our current projects explore the use of social robots as a technology to support young children's early language development. In this project, instead of focusing on how to make social robots effective as an educational tools, we ask why they are effective. Based on our prior work, we hypothesize that a key aspect of why social robots can benefit children's learning is their nature as a relational technology—that is, a technology that can build long-term, social-emotional relationships with users.

Thus, in this project, our goals are twofold. First, we aim to understand how children conceptualize social robots as relational agents in learning contexts, and how children relate to these robots through time. Second, we explore the core nature of autonomous relational technologies, that is, relational AI. We will examine how adding features of relational AI to a social robot impacts longitudinal child-robot learning interactions, including children's learning, engagement, and relationships.

As part of this project, we are taking a second look at work we have done so far, this time through the lens of children's relationships. We are creating assessments for measuring young children's relationships. We are developing a computational relational AI model, which we will test during a longitudinal study with a social robot.

Read more about children's relationships with robots here!

This work is part of Jacqueline Kory Westlund's dissertation. Her committee includes Cynthia Breazeal, Rosalind Picard, and Paul Harris.