TEDGlobal: Sugata Mitra, beyond Hole in the Wall

Sugata Mitra tells us that there are places on Earth, in every country where, for various reasons, good schools cannot be built and good teachers cannot or do not want to go. And those places, as it turns out, is often where trouble comes from.

In 1999, Mitra embedded a computer in a wall in a slum in New Dehli, connected it to high speed internet and left it there – the Hole in the Wall Experiment. He repeated this experiment in other parts of India and discovered that children learn what they want to do. We see a Rajastani village where children were recording music and playing it back for each other, only four hours from seeing the computer for the first time. His conclusion – children could learn to use computers by themselves.

An experiment in Hyderabad asked children who spoke English with a strong Telugu accent to use a voice recognition system on a computer. Two months later, their accents had changed and were closer to the neutral British accent of the speech synthesizer.

Mitra had a conversation with the late Arthur C. Clarke where Clarke said, “If a teacher can be replaced with a machine, he should be.” And Clarke told him that student interest is the most important thing in education.

As children begin to Google their homework, teachers in India are reporting that their English is improving… and they’re becoming surprisingly deeper thinkers. Mitra believes that this might be a shift from memorization to exploring information online.

How difficult a task can students take on. In Kalikuppam, a small village, Mitra decided to see if Tamil speaking children could learn about biotech in English on their own. After two months, the students sheepishly told him they’d learned nothing. He asked whether they’d learned nothing at all, and a twelve year old girl told him, “Apart from the fact that improper copying of genetic molecules could cause disease, we’ve learned nothing.”

Students took biotechnology exams and scored a 30, while they’d scored a 0 before… “an educational impossibility.” He asked one of the best students to teach the others and improve their schools. She asked how she could possibly teach them, and Mitra suggested “the grandmother method” – stand behind, admire, act fascinated and praise. After two months, the class score was up to 50.

Mitra is now conducting experiments in the UK, with students at Gateshead school. Students work in groups of four, using one computers, and can change between groups. One group started solving GSCE questions within 20 minutes – the least successful group took 45 minutes. They were using Google, Ask Jeeves and other sources. Teachers asked, “Is this deep learning?” Mitra sees evidence that test scores rise over time with groups like these, and believes that students have almost near photographic recall because children discuss what they’ve learned together.

He’s got a great new idea – the granny cloud. He’s recruited hundreds of British grandmothers who donate their time over online video connections and answer questions for children. In both India and the UK, he’s teaching children using groups, Google and the granny cloud.

Maybe the most amazing experiment comes from Turin, where Mitra went to a primary school and started writing questions on the white board in English for students who speak only Italian. Using Google translate, students were answering questions like “Who was Pythagoras and what did he do?” in a few minutes.

Mitra tells us that he future of education is self-organized learning environments. They let students learn together, use resources and people they can access online and explore on their own, and he plans on testing this going forward.