November 6, 2018

  • Don’t teach kids – I’ve shown their hive mind can learn on its own

    murdos_segata_14690-800x533Sugata Mitra believes knowledge should be learned, not taught.  by Murdo MacLeod
    Educationalist Sugata Mitra's pioneering experiments suggest teaching facts doesn't work in the internet age – fostering creativity and collaboration is the key.

    By Bob Holmes of NewScientist

    IN 1999, an inquisitive physicist named Sugata Mitra installed a computer in a slum in New Delhi, India, and then walked away. Local children congregated and began trying to use the unfamiliar device. When Mitra returned a few days later, they had already taught themselves to surf the internet.

    Mitra is now a professor of educational technology at Newcastle University in the UK. In the decades since the “hole in the wall” experiment, he has found that groups of children aged 8 to 12, left alone with the internet, can teach themselves even technical subjects such as evolutionary biology to a level several years ahead of their school age. In 2013, he won a $1 million TED prize to help his work.

    How can children learn on their own, without guidance?

    The learning I’m talking about appears spontaneously in response to a query, which may be posed by an adult, or by the children themselves. I need the internet to be available on large, publicly visible screens in a safe space. I need to have mixed groups of children – boys and girls of different ages together, not each child on a different computer. Then I need to remove all supervision. We have got to keep the adults away.

    Everybody says, who taught them? How clever they are! But this doesn’t have anything to do with cleverness or teaching. It has to do with a hive-like mind with a common desire.

    But are the children really learning, or just repeating phrases they don’t really understand?

    I could give them questions on quantum mechanics, and they would come back, not with the understanding that a physicist like me would have, but with a child’s understanding. I teased 9-year-olds by saying something could be in two places at the same time. They said no, if you have a pencil, it is here when you put it here and there when you put it there. I said there are some things that can be here and there at the same time – then I left. They came back with quantum entanglement, and said it means that the two particles know about each other, but we don’t know how those particles know.

    Indian children

    Children in India worked together to teach themselves when provided with a “hole in the wall” computer

    Philippe Tarbouriech

    Can children discriminate between good and bad information on the internet? Many adults have trouble with this.

    The children get as confused as you and I. However, when they work in groups, they can detect extreme points of view and avoid those. Working in groups is key: the hive mind is more discriminating and less gullible than an individual in front of the screen.

    If I show kids a picture of the Colosseum in Rome and ask “what’s that building? And why is it broken?”, they come back with a lot of information about the Roman Empire. But now I go on to ask “how do we know what you said is true?” They start by saying that every website says so, which is good, but then they start going deeper, toward understanding historical evidence.

    Is this method effective even for very young children?

    You can push it down to the point where the child is beginning to read, which might be age 4 or 5. And it accelerates reading development. With the TED prize, I set up labs in seven schools – five in India and two in England. We have got all the data now, and there is a definite, measurable improvement in reading comprehension over and above what would be expected.

    What role does this leave for teachers, and what do they think of your approach?

    It’s not that you don’t need a teacher. You just don’t need the teacher to tell children things they could look up for themselves. You need the teacher to ask them questions that make them sit up and wonder. My teacher friends tell me it is harder than preparing for a lesson where they just read out facts.

    In the 20 years I’ve been doing this work, the opinion of teachers has changed dramatically – from saying this is rubbish to saying, isn’t what you say obvious?

    “You just don’t need teachers telling children things they can look up on their own”

    But isn’t there a set of basic knowledge that needs to be learned by every child for them to get by in society?

    I’m very glad that you said needs to be learned, not needs to be taught. All my work is about making that distinction. My answer is yes, but I’m not able to define that basic set very well. How much should I keep in my head, and how much should I rely on the internet? I tend toward thinking that we should know how to make the information infrastructure answer a question just in time, when we need it to. This is as opposed to the old system, which is to teach the child everything we think they need, just in case.

    Does that mean I can go through life without knowing what the solar system is? That’s a horrifying thought. So perhaps we can agree on the big questions that children should engage with at various ages, then give teachers the liberty to pose the questions as they wish.

    Should we allow internet access during exams, then?

    I think we should. But then we have a problem: who is going to evaluate the answer, and how? I don’t know yet – that’s what I’m working on now. But for example, when you are learning to play an instrument, the exam is to play the instrument, and judges assess how good you are. Maybe we should move toward that kind of assessment in an internet-assisted world.

    If knowledge is in the cloud, not our heads, what happens if the internet fails?

    If the internet goes down, we will live very uninformed. But that doesn’t mean that we should learn how to live without the internet – no more than we should learn how to tell the time of day without looking at a watch, just in case watches disappear. If I don’t have a watch, I won’t be able to tell the time. Sorry.

    You have faced scepticism because you haven’t published much comparative data in top-tier journals. What has stopped you?

    I find it hard to publish – I get one rejection after another. I’m not from the social sciences. In the natural sciences, if results are unexpected, others repeat the experiment and report whether they got the same results. In the social sciences, people just seem to point at the holes in your work, but nobody ever says they repeated the experiment and got something else. So I have a humble request to social scientists: replicate my work and see if you get the same results.

    This article appeared in print under the headline “Hey, teacher! Leave those kids alone”