In 1999, Sugata Mitra left a computer in a New Delhi slum and watched what the neighborhood kids would do with it. With no prior computer experience, they quickly figured out how to work it. In subsequent experiments, Mitra used computers and very little adult oversight (what we refer to as “education”) to teach children all sorts of different things.

Over the years, Mitra got more ambitious. For a study published in 2010, he loaded a computer with molecular biology materials and set it up in Kalikuppam, a village in southern India. He selected a small group of 10- to 14-year-olds and told them there was some interesting stuff on the computer, and might they take a look? Then he applied his new pedagogical method: He said no more and left.

Over the next 75 days, the children worked out how to use the computer and began to learn. When Mitra returned, he administered a written test on molecular biology. The kids answered about one in four questions correctly. After another 75 days, with the encouragement of a friendly local, they were getting every other question right. “If you put a computer in front of children and remove all other adult restrictions, they will self-organize around it,” Mitra says, “like bees around a flower.”

It’s tempting to conclude that the computer is the magical ingredient here: just add computers and children can learn anything. But if the story of Sergio Juárez Correa’s fifth-grade class is any indication, the secret is the kids organizing themselves to learn.

For Juárez Correa it was simultaneously thrilling and a bit scary. In Finland, teachers underwent years of training to learn how to orchestrate this new style of learning; he was winging it. He began experimenting with different ways of posing open-ended questions on subjects ranging from the volume of cubes to multiplying fractions. “The volume of a square-based prism is the area of the base times the height. The volume of a square-based pyramid is that formula divided by three,” he said one morning. “Why do you think that is?”

He walked around the room, saying little. It was fascinating to watch the kids approach the answer. They were working in teams and had models of various shapes to look at and play with. The team led by Usiel Lemus Aquino, a short boy with an ever-present hopeful expression, hit on the idea of drawing the different shapes-prisms and pyramids. By layering the drawings on top of each other, they began to divine the answer. Juárez Correa let the kids talk freely. It was a noisy, slightly chaotic environment-exactly the opposite of the sort of factory-friendly discipline that teachers were expected to impose. But within 20 minutes, they had come up with the answer.

“Three pyramids fit in one prism,” Usiel observed, speaking for the group. “So the volume of a pyramid must be the volume of a prism divided by three.”