Last changed 25 Mar 2018 ............... Length about 2,000 words (20,000 bytes).
(Document started on 21 Mar 2018.) This is a WWW document maintained by Steve Draper, installed at You may copy it. How to refer to it.

Web site logical path: [] [~steve] [educ] [this page]
Other related pages: [Home] [My Educ. ref list] [Effect size] [Fail/succeed]
Linked circle of topics relevant to CCSE: [Adey&Shayer] [Papert's gear wheels] [Early informal learning] [spatial] [school curricula] [methodology]

Is success or failure what drives learning?

By Steve Draper,   Department of Psychology,   University of Glasgow.

This web page is related to a CCSE reading group meeting on Monday 26 March 2018 (1pm SAW-422).

photo Also attending the session is the person who stimulated the idea: Daisy Abbott (see:   1   2   3   4 )

This is a light relief session, discussing whether it is failure that drives learning (as some have claimed in print), or success (what rational child or adult would continue with an activity they fail at)?

Part 1: The essential preparation

The main preparation (homework) for the meeting is to spend 5 minutes with a pen and paper or your favourite digital note-taker and sketch out what your immediate response to that question is. (Possibly including the riposte that this is a bad question, and trying to articulate what is mistaken about it.) If you have done that, we'll have a good session.

Do not read anything further on this page until you have done that.

Part 2: Elaboration of the question

Whether success or failure drives learning is closely connected to the practice of giving learners feedback; to whether marks or feedback content promotes learning; etc. Just how it is connected is part of what we should be thinking about.

Learning from failure

  • "Step by tedious step, we stumble away from abject failure. And that's on a good day."
    Barth Netterfield, an experimental astrophysicist, as his team slaved away getting their balloon-borne telescope ready for launch.

  • If you want to increase your success rate, double your failure rate. Thomas Watson, Sr., founder of IBM
  • Mistakes are the portals of discovery. James Joyce Ulysses
  • Experience keeps a dear school, but fools will learn in no other — attributed to Benjamin Franklin

    The academic papers on this are by Manu Kapur. Kapur (2008) gives the idea (based on a substantial trial). His phrase is "productive failure" — he doesn't say that all failure is good, but does show that easy success does not lead to the most learning.
          What he reports is that (in part of school maths) if learners train only on well-structured problems they do not do as well even in a test on well well-structured problems, as those who train on ill-structured problems instead.

    Etkina (2015): her learning design for science labs ("ISLE") is described and justified. It is training on overcoming faiure = how to extract information from failure and so improve. Most science teaching fails to teach this, reducing "labs" to touching the equipment and getting the right answer, and NOT a) learning any new information; b) learning and practising how to overcome failure, which all experimental scientists do in their research.

    Learning without failure

  • Only a fool learns from his own mistakes. The wise man learns from the mistakes of others. Otto von Bismarck
  • Time is a great teacher, but unfortunately it kills all its pupils (Louis) Hector Berlioz

    In fact there are some things you cannot learn by failure because that would kill you, and that in turn means you wouldn't learn it and never could.

    Learning from success

    • Skinner's behavioral shaping lets humans train animals and does so by "differential reinforcement of successive approximations". Basically, if you want an animal to touch a target, you start by rewarding any movement towards it (i.e. the 180 degrees in that direction), and not rewarding any movement away (the other 180 degrees). Then you progressively narrow the angle that is rewarded. You could call this learning from success, because when the rewarding stops, the behaviour will gradually die away. (However, it is also true that when the failing stops, the learning but not the behaviour will stop at that stage of imprecise accuracy in the behaviour.)
    • When learners change their aims and their success criteria dynamically: i.e. the learning aims are NOT fixed and not determined by the teacher. That entails that learner and teacher do not agree on what success and failure are. But it frequently does mean that learner motivation is changed.
      Mitra and teacher-less education, self-organised by the learners.
      54 minute conference video.

      Note that Mitra's big reported successes at letting learners guide themselves contradicts Kirschner, Sweller and Clark (2006). But that, Mitra might riposte, is because they think success is learners having to learn what teachers dictate. Which is a perverse and dictatorial definition of success; and is diametrically opposed to Papert and Piaget's interest in learning without curriculum.

    • The education system is largely organised around specialisation, where learners decide to learn what they are good at, and to stop learning what they are unsuccessful at. This pretty much means that learners use failing exams as important feedback for deciding what NOT to try to learn.
      (This argument is explored in: Draper,S.W. (2009) "What are learners actually regulating when given feedback?" British Journal of Educational Technology vol.40 no.2 pp.306-315 doi:10.1111/j.1467-8535.2008.00930.x )

      You may argue that learners thus learn from failure what NOT to learn; but it implies that very few people actually choose to learn by failing, otherwise they would spend all their time trying to learn what they are useless at. Such learners (almost everyone) clearly believe they learn only from success, and change subjects, and are encouraged by teachers to do so, in response to this belief.


    Picture these two cases. In the first, a very young child can crawl or shuffle on their bottom vigorously and effectively, but also can stand and actually walk a few hesitant steps. She looks very pleased with herself in both modes of locomotion. An adult would regard crawling as a shameful failure, perhaps done only when very drunk, ill, or disabled.
    In the second, a slightly older child goes the first few yards on their first bike before putting down one leg. They look very pleased with themselves. They see that as a success; the adults see that as a failure compared to cycling from A to B in a purposeful way (however short).
    Can it be that a child and an adult disagree about whether the same event is a failure or a success? Does that mean we don't agree, and that success and failure are not agreed terms? (But if not, then how can anyone learn from failure, because success means no more improvement is needed?)

    "When you're still too young to shave, optimism is a perfectly legitimate response to failure" (Stephen King (2000); p.34). This was his response to his first rejection slip with a handwritten note added to it. After a while, a single nail would not hold all the rejection slips, and he bought a proper spike to hold them. After a bit more time the notes on rejections were more positive "This is good. Not for us, but good." After another 10 years, he had actually sold a couple of novels. In 2017 Stephen King's net worth was $400 million.

    The question here is whether you want to recommend failure like his as the way to learn. No success at all for more than 10 years. Elsewhere in his book he advised authors not to even think of being a writer unless they like the process of writing. The money (i.e. "success") just cannot be enough motivation to do it. I.e. success doesn't drive writing, and in particular, cannot drive learning to write. Or should we put it another way: publishers and the taxman judge success in one way; but while learning, King judged it in some quite different way.

    Part 3: Daisy Abbott's work

    Abbott designed and trialled a board game called "How to fail your research degree" as part of a course for postgrads. Her explicit rationale for this was learning by failing. Abstract of a talk about this

    My own view is different. Abbott created a course (in other places it might have been called "professional skills" or "reseach skills") to help research students with their degree programme. I believe it was effective. Its content was all the activities that a research student must do; and still more, must plan, manage, and complete in a timely manner.

    The main issue is that such a course threatens to be so scary it just creates negative emotion, depression, avoidance, and so ensures they fail their degree. Abbott attacked that top problem of negative affect by:

    • The course name is a big joke "How to fail your degree". You never see courses like that. Serious titles create the wrong effect e.g. "How to avoid failing your degree"; "Essential skills for a postgraduate". But it is like a once-famous book "How to lie with statistics".
    • The students must play a game about this. A game cannot be serious, can it? Especially a non-digital, low tech, board game. Thus the teaching method is also a joke.
    • In any case, failing at the game is not failing their degree. It is unthreatening as a learning method (unlike writing a "plan" which, like making new year resolutions, is full of making important promises you are almost certain to fail to carry out).

    Having succeeded at the main issue of defusing paralysing fear and counter-productive seriousness, Abbott nevertheless doesn't allow them to do nothing, but to engage with the game, its rules, and so with many detailed aspects involved. What can we say about this? Probably, an unusual student who "won" the game first time would not fail to learn; but those who "lost" the first times would continue and get better. Failing at the game may be useful, but leaves them free (and better prepared) to "win" at the real business of getting a research degree. Are these students learning by failing; or learning by succeeding?

    So personally, I don't think these students learned by failing. It may not be that they learned by succeeding exactly either. The most important thing (it seems to me) was that they learned not to fear the issue, yet to engage with it in a low-stakes environment that nevertheless exercised their thoughts about it and many of its details (tasks a research student must do).

    In another context, before I met Daisy, I devised and put on workshops for students with some similarities in the methods. With psychology students, whose experience of writing a critical review was doing a large piece of coursework, which counted towards their degree class, over three months. My complementary approach was to get them to do a 5 minute "review" in writing (this is a joke, isn't it?); to get them to mark each other's work on a simple scale about showing the markers of critical thought; then do a second one (on a different topic) so that those who didn't get a high mark in this funny task at the first go, would mostly succeed at the second go round. (And small, short work means you can do two exercises AND the marking easily in one 50 min. session.) I also used a similar approach to practising interviews by having students give a one minute talk/answer to a question with only 2 minutes notice; and to "mark" each other's talks by judging how good they were at "sounding like a psychologist".

    Part 4: Points

  • Nine tenths of education is encouragement. - Anatole France

    Part 5: Syntheses = Resolutions of the question

    • x
    • x
    • x
    • x

    Part 6: Further educational points and theories

    E.g. Constructivism.

    This is where the answer sheet will go, but it is too secret for you to know now, and it's for your own good that I don't demotivate you by giving you effortless success. But really I do know the answer, and of course, there is only truth and error; success and failure at understanding educational issues. Isn't there?

    Part 7: References

    Abbott, Daisy:

    Draper,S.W. (2009) "What are learners actually regulating when given feedback?" British Journal of Educational Technology vol.40 no.2 pp.306-315 doi:10.1111/j.1467-8535.2008.00930.x

    Etkina,E. (2015) "Millikan award lecture: Students of physics — Listeners, observers, or collaborative participants in physics scientific practices?" American Journal of Physics vol.83 pp.669-679 doi:10.1119/1.4923432

    Kapur, M. (2008) "Productive Failure" Cognition and instruction vol.26 no.3 pp.379-424 doi:10.1080/07370000802212669 [learning by failing]

    Kapur, M. & Bielaczyc, K. (2012) "Designing for Productive Failure" Journal of the Learning Sciences vol.21 no.1 pp.45-83 doi:10.1080/10508406.2011.591717 [learning by failing]

    Kapur, M. (2014) "Comparing Learning From Productive Failure and Vicarious Failure" Journal of the Learning Sciences vol.21 no.4 pp.651-677 doi:10.1080/10508406.2013.819000 [learning by failing]

    Kapur, M. (2016) "Examining Productive Failure, Productive Success, Unproductive Failure, and Unproductive Success in Learning" Educational Psychologist Vol.51 No.2 pp.289-299 doi:10.1080/00461520.2016.1155457 [learning by failing]

    King, Stephen (2000) On writing: a memoir of the craft (Hodder & Stoughton: London). The quote is from p.34

    Kirschner,P.A., Sweller,P. & Clark,R.E. (2006) "Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching" Educational Psychologist Vol.41 No.2 pp.75-86 doi:10.1207/s15326985ep4102_1

    Web site logical path: [] [~steve] [educ] [this page]
    [Top of this page]