Featured Research Projects

A Family-oriented Bilingual Conversational Agent for Co-reading

In collaboration with the Joan Ganz Cooney Center at Sesame Workshop, we developed a bilingual conversational agent embodied in the Sesame Street character, Rosita, along with a story reading interface, Rosita Reads with My Family. The goal of this project is to support culturally-engaging shared reading in Hispanic families. Unlike most existing conversational agents that only supports one-to-one dialogue with individual children, our conversational agent, Rosita, not only engages children in back-and-forth conversations about the story plots but also facilitates parental involvement by providing discussion prompts that connect the story to children’s life experience.


Analyzing Children’s Verbal Engagement with a Conversational Agent

This project is part of a larger study on using conversational AI to suppport children’s science learning. We partnered with PBS KIDS to develop interactive science video programs in which the main character, powered by artificial intelligence, engaged children in conversations by asking them questions and providing responsive feedback. As part of the study, I analyzed children’s verbal engagement with the conversational agent, including the relevance, accuracy, response rate, and response latency of their responses to the agent’s questions and how it was related to children’s science learning.