Michael Horn Photo: Alonso Nichols

I Robot, You Programmer

In a corner of Boston’s Museum of Science, the educational theories of Michael Horn, a doctoral student in computer science, are coming together in the form of interlocking wooden blocks. Young museum visitors connect the blocks in different sequences to enable a smiley-faced robot to go through a series of moves—forward, reverse, right, left, wiggle, spin—before lighting up a sign that says “Robot Park.” In the process, the kids are proving the point that Horn wants to make: programming a computer can be easy and fun.

The blocks are Horn’s antidote to the haphazard way in which schoolchildren typically are exposed to computers. In many schools, he points out, students must crowd around a few monitors and share a mouse and keyboard. Not only might the teacher not be able to see what the kids are doing, but some younger children are not ready to use a mouse. “I started thinking, What if kids could work on programming activities on their desk or on the floor?” Horn says. “What if writing a computer program became more of a physical activity?”

He came up with his simple, tangible system as part of his research at Tufts. Each wooden block describes a different action for the robot to perform. Once children finish arranging the blocks, they press a button, and a camera mounted overhead takes a picture of their work and sends it to a computer. Reading barcode-like symbols on the blocks, the computer can interpret the kids’ program and feed its commands to a robot in seconds. “When I write computer code, it’s with obscure symbols,” Horn says. “But the fundamentals of what I’m doing and what the kids are doing with the wooden blocks are the same: they’re making sequences of action, repeating sequences of action, making logical decisions.”

Horn, who comes from a family of teachers, has picked up some classroom experience of his own, working in a middle school in Boston’s Roxbury neighborhood as part of a National Science Foundation fellowship. His background makes him “very sympathetic to teachers,” he says. “I think a lot of technology that focuses on kids neglects the other side of that equation—the person who has to deal with classroom management and provide a productive and positive learning environment.”

Horn hopes to get children as fired up about programming as he was in his middle school days, when he learned Logo, a computer language designed to teach children about math and logic. “It changed my life dramatically, my approach to academics, even how I went about mowing the backyard,” he says. “When you learn how to program computers, you learn how to take a large problem and divide it into manageable subproblems.”

By learning to program, he says, children can even gain new self-confidence in their abilities. He has noticed that children at the Museum of Science—especially girls—are more likely to program the robots when the computer blocks are on the table. When only the mouse and monitor are available for programming, just 30 percent of girls participate. Yet 90 percent of them try the hands-on exhibit. “Maybe the girl who would never dream of programming a computer thinks, ‘This is interesting. I just programmed a robot and might try this later on,’ ” Horn says.

So far, Horn has been focusing on students of school age, but that will soon change. He and two Tufts faculty members—Marina Bers, an assistant professor of child development, and Robert Jacob, a professor of computer science—recently landed a three-year, $445,000 grant from the National Science Foundation to extend the robot-programming work to early childhood education.