Carnegie Mellon University
November 17, 2015

Carnegie Mellon Building Educational Software To Teach Children Basic Skills Without a Teacher

Student-Faculty Team Competing for Global Learning XPRIZE

By Byron Spice / 412-268-9068 /

PITTSBURGH—Can children learn to read, write and do basic arithmetic without a teacher or classroom, relying only on tablet computers, each other and some intelligent software? A team of educational researchers from Carnegie Mellon University aims to find out in the $15 million Global Learning XPRIZE competition.

The question is a pressing one in much of the developing world, where teachers and schools are in short supply, if they exist at all, said Jack Mostow, a research professor emeritus in the Robotics Institute and the leader of CMU's "RoboTutor" team.

"At Carnegie Mellon, we take pride in solving big problems, and the lack of formal schooling is a truly immense problem in many parts of the world," Mostow said. "If we can develop educational technology to fill that gap, we can significantly improve the lives of the 250 million children who today can't read, write, or do basic math."


The Global Learning XPRIZE, announced last year by XPRIZE, will award a grand prize of $10 million to the team whose open source software proves best able to help children learn basic literacy and numeracy skills during a field test in East Africa. Almost 200 teams from 40 nations have registered.

Carnegie Mellon, with its expertise in data-driven learning science, is uniquely suited to such a challenge, Mostow said. Researchers have a long history of developing computerized tutors, such as a Reading Tutor developed by Mostow that listens to students read aloud and helps them with pronunciation. Likewise, cognitive tutors pioneered by Ken Koedinger, professor of human-computer interaction and psychology, teach algebra to hundreds of thousands of U.S. students each year. Educational computer games, another CMU strength, will be an important part of the RoboTutor software.

The teams have until November 2016 to develop their solutions. An expert panel will select five finalists in 2017, with each of those teams receiving $1 million as they prepare for the critical field test in at least 100 African villages in 2017-2018.

A gift of $250,000 from an anonymous donor will enable the RoboTutor team to complete the initial development of its software package, Mostow said, "though we still are seeking funding for additional components that we believe are important." The team was launched with seed funding from the Simon Initiative, a university-wide effort to measurably improve student learning outcomes by harnessing CMU's learning research and engineering ecosystem.

Students and post-doctoral researchers across campus "have come out of the woodwork in droves" to work on the project, Mostow said, with more than two dozen actively involved so far.

"Part of the appeal is the possibility of having an impact on society," said Ran Liu, a post-doctoral researcher in CMU's Human-Computer Interaction Institute and Department of Psychology. But part of it has to do with the immensity of the challenge. "A lot of my research involves small iterations to improve a curriculum, but not designing a curriculum from scratch," she explained. "It's just not like anything else I've been involved in."

A unique challenge for the team is testing the educational software, Mostow noted, which targets children, ages 7-10.

"We don't have an ideal proxy for that population," he said. "It's hard to find American kids who have no access to formal schooling." So the team is testing its software at the Children's School, CMU's laboratory school that provides early childhood education for children, ages 3-5. Children of that age are at a different developmental level than the target population, he acknowledged, but their responses will help determine if the curriculum designs are valid.

In anticipation of making the finals and mounting a field test in East Africa, the team already is working with experts in Swahili and CMU faculty members who study cultural differences in learning, as well as consulting with Joash Gambarage, founder of the Mugeta Children's School in Tanzania.

"I'm excited as I can be about this project and the students are too," Mostow said. "This is the most excited I've been in years."


About XPRIZE: Founded in 1995, XPRIZE is the leading organization solving the world's Grand Challenges by creating and managing large-scale, high-profile, incentivized prizes in five areas: Learning; Exploration; Energy & Environment; Global Development; and Life Sciences.  Active prizes include the $30M Google Lunar XPRIZE, the $20M NRG COSIA Carbon XPRIZE, the $15M Global Learning XPRIZE, the $10M Qualcomm Tricorder XPRIZE, and the $7M Barbara Bush Foundation Adult Literacy XPRIZE. For more information, visit