The problem with SatNavs, or how feedback can prevent learning

I’m not an especially good driver, but I’m a truly terrible navigator. This used to mean that I would get lost. A lot. When I first moved to Bristol in 2001 I bought an A-Z of the city and when driving somewhere new I would have to stop the car periodically and try to align the map to the streets around me. Needless to say, I found this pretty stressful. Luckily, I’m a lot better at recognising landmarks than I am at reading maps. Slowly, through a process of trial and error, I started to learn how to find my way around. I’ve got to the point where I know the city fairly well.

Then, a few years ago I bought a SatNav. It was a boon. For the first time in my life I could set out on a journey with a fair degree of confidence that I would be able to make it to a new destination without getting horribly lost. I felt so happy following my arrow-shaped avatar along the purple path unfolding before me.

garmin_satnavAs you know, SatNavs are not perfect. Sometimes they suggest bizarre routes and sometimes they seem to freeze just when you need them most. I hate those moments of uncertainty; that helplessness as I flounder without the feedback I have become so accustomed to. The relief when the arrow pops back is palpable. Even when I make a mistake I can stay calm; the SatNav simply reroutes or points me back the way I came. I can safely say that my experience of driving has been revolutionised. SatNavs are just about perfect at giving feedback.

But I don’t learn any new routes. Why is that?

John Hattie says the following:

Effective feedback must answer three major questions asked by a teacher and/or by a student: Where am I going? (What are the goals?), How am I going? (What progress is being made toward the goal?), and Where to next? (What activities need to be undertaken to make better progress?) These questions correspond to notions of feed up, feed back, and feed forward. How effectively answers to these questions serve to reduce the gap is partly dependent on the level at which the feedback operates.

Well, my SatNav answers these questions very effectively;  the gap has been reduced, but I still don’t learn. The problem is I get too much feedback. I know where I am, where I’m going and what I need to do next all the time. I never have to struggle. And because I never struggle, I never learn. My contention is that this is a situation enacted all too often in schools. In our well-intentioned efforts to let pupils know exactly what they should be doing next we might be short-circuiting learning.

The ‘gap’ between where I am and what I should do next might be important. If someone fills the gap, I don’t have to think. And if I don’t have to think, I won’t learn.So maybe reducing the gap isn’t so overwhelmingly important? Hattie acknowledges this:

Simply providing more feedback is not the answer, because it is necessary to consider the nature of the feedback, the timing, and how the student ‘receives’ this feedback (or, better, actively seeks the feedback.)

As you may be aware, reducing feedback is also one of the ‘desirable difficulties’ advocated by Robert Bjork:

Empirical evidence suggests that delaying, reducing, and summarizing feedback can be better for long-term learning than providing immediate, trial-by-trial feedback… Numerous studies—some of them dating back decades—have shown that frequent and immediate feedback can, contrary to intuition, degrade learning.

Maybe this doesn’t sound quite so counter-intuitive when we think about the SatNav problem? The next time you’re considering giving a pupil feedback, maybe it’s worth letting struggle for just a little bit longer? It could be that immediate feedback negates the need for memorisation. Just as we outsource our memory of phone numbers and appointments to diaries and gadgets, we might be allowing pupils to outsource their knowledge of ‘what to do next’ to their teachers.

Of course the type of feedback also has a major effect. Giving pupils answers or complete solutions might be efficient, but it means they won’t have to think. It could be a lot more useful to offer hints or partial solutions – a nudge in the right direction rather than an arrow on a purple road.

But even the best feedback comes with baggage:

  • Providing feedback of success is a waste of effort
  • Students can become dependent
  • Slows down pace of learning

It is always worth considering the opportunity cost. What else could teachers do with all that time devoted to fetishising feedback?

Or maybe it’s all about practice? Maybe the reason I’ve stopped learning new routes is back I don’t practise them enough? Maybe if I moved to another new town I would, eventually, start to learn my way around? Maybe not. I suspect the problem is that the constant feedback I get from my SatNav means that all I would be practising is following the SatNav. The kind of practice I’d need to be engaged in would have to be deliberate; it would have to get progressively harder; at some point I’d have to try driving all on my own.

And here the metaphor breaks down. I don’t care about being an independent driver and am very happy relying on the wonders of technology. But this, surely, is not the fate we want for our pupils. We want them to flourish independently in the world. Of course we do. And just as we can never ride a bike until the stabilisers are taken off, maybe pupils will never learn to be independent until teachers stop giving them so much damn feedback?


I’ve tracked down Hattie’s use of the SatNav metaphor for feedback to a paper called “Using Feedback to Promote Learning” written with Gregory yates and published in this fabulous journal.

Related posts

Force-fed feedback: is less more?
Why AfL might be wrong
Getting feedback right

45 Responses to The problem with SatNavs, or how feedback can prevent learning

    • nappits1443 says:

      Enjoyed your post David as always, but I’m honestly wondering if, in this data/target/A*-C-driven world we work in, deep learning and a love of learning really is the aim of the modern teacher. Idealistically, of course; but to get a class its passes (and therefore keep the wolf off their backs), do teachers merely want their students to know stuff for a short period of time? – ie. in order to write the controlled assessment they’re working on tomorrow. The more I think about it, the more I think that shallow/quick recall for tomorrow’s assessment or next week’s exam is more attainable than a deep love of, in my case, English. Just thinking aloud…

  1. julesdaulby says:

    It’s not formative feedback though is it? It’s not telling you how well you drove. It’s not teasing out an answer from you – David what do you think? Should we got up there or turn left here? It’s telling you where to go so you don’t have to work it out? Dare I say, your analogy maybe better suited to explaining the limitations of direct instruction? (Note I am writing this late at night when no-one’s around) from a critical friend.

    • David Didau says:

      Maybe the problem is with anything done badly?

      • julesdaulby says:

        Yes you are probably right or doing something for the wrong reason – it’s that backwards working again – what do I want to achieve? what shall I use to achieve it? that could mean tech, various types of feedback, no feedback etc.
        There needs to be an informed reason behind the decision though – going back to your analogy it gives excellent feedback on telling you where to go but little else.

        I’m enjoying the feedback on your feedback question btw – that surely is real learning?

  2. Sean Harford says:

    I think you need to struggle for a bit longer before I feedback.

  3. Martin Harris says:

    The metaphor as you point out is stretched and has gaps in its logic. You have closed down the possibilities of sat nav use to help prove a point.
    However, regarding feedback the issue is the type, nature, extent of that then allows learning. Feedback can mean many things and not just A to B ‘this way only’.
    Good point I hope one that many will take note of.

    • David Didau says:

      Hi Martin – I’m not really trying to prove a point – more about suggesting a parallel. What are the possibilities of satnavs that have been closed down?

  4. wendy says:

    This is why maps are so much more effective they force you to think and review. The same ideas should permeate the feedback. Giving the answer (sat nav) is NOT feedback, feed forward is a better phrase, clues and pointers to the next stage. There is no sense of achievement with a sat nav.

  5. Does the analogy hold? Is it possible to give feedback to a child in a class in the way and frequency that a SatNav gives instant and continuous feedback to the driver?

    • David Didau says:

      I doubt it’s possible to give feedback continuously the way a satnav does. My point is that giving feedback is not the answer. It needs to be thought out in terms of structure & delivery if it is to be effective.

  6. Surely the satnav is not feedback at all. It’s the equivalent of asking a question and then giving the answer without waiting for a response. The art/craft (call it what you will) of classroom questioning is right at the heart of any definition of teaching, and is, in fact, a process of continuous feedback. Properly done, it challenges, informs and promotes learning across the whole group. People have written books about it (maybe you have, David) And, I might say, like any other art or craft it takes years to become proficient at it.

  7. I agree with JulesDaulby – it is not that the SatNav gives bad feedback, its that it isn’t giving feedback at all, beyond the implicit “Well done, you followed my last instruction”. The educational analogy of the SatNav would be that it is giving you assistance or scaffolding – that this is what needs to be reduced to enable the student to struggle and achieve cognitive load – not the feedback at the end of the struggle.

    I think you are confused by Hattie’s statement that good feedback should tell you where to go next. Surely that is what the SatNav is doing? But it isn’t – not in the sense that Hattie means.. Because what Hattie means is not that the feedback should give you the *answer* to the next question. Just that it should point you to the next question, the next problem that *you* have to work out.

    With respect to Peter Blenkinsop’s comment, I believe constant feedback is indeed possible with appropriate digital activities – thereby addressing the key deficiency of our modern education systems, that they are desperately feedback-poor. Track a child through a school day and tell me how much time is spent in meaningful conversation about work with a teacher – my guess is that it would very rarely be more than a minute and in most student-days, it would be zero. It is why personal tuition is so much more effective (and expensive) than class teaching – and points to the contribution that ed-tech needs to make to teaching.

    Thanks, Crispin.

    • David Didau says:

      Ah, a “You disagree with me so therefore you must be confused” comment! Thanks, I’ve not had one of those for a while. Hattie himself has used the SatNav metaphor in conversation – I suppose he’s confused about what he means too?

      • If he used the analogy in conversation and there is no record of what he said, then obviously I can’t really comment on whether I think Hattie is confused or not. And if you took the SatNav analogy from Hattie, it seems strange that you did not mention this fact in the body of the post.

        I never said that you are confused because I disagree with you. I said I disagree with you because you are confused. I have to say that it seems that you get confused pretty easily.

        If you think that I am wrong and you are right, then you need to make a substantive response to my argument, which you have not yet done. Do you not recognise a difference between assistance and feedback?

        • David Didau says:

          I didn’t attribute the analogy to Hattie as a third party reported this as a conversation they’d had with Hattie. We’re just going to have to take this person’s word for it (or not as you please.)

          Obviously you’re right and I’m confused; it was ever thus. Maybe try reading the post again with the assumption that I’m not confused?

          I have no idea whether you’re wrong; that is for you to say. But I shall do you the courtesy of assuming your position is not based on ignorance or confusion. If you feel unable to do the same then please refrain from providing any further ‘assistance’.

  8. muzzyizzit says:

    The POINT of the thing is the point. You can use the satnav to learn but that is not the point of satnavs. In fact, satnavs are, if you want to, exactly the thing to help you with deliberate practise: turn the thing on, go where you want to go and pay attention. You will very quickly learn the roads inside out. Learning is about motivation and expectation. As Noam Chomsky once said, ‘if you want to learn you will learn no matter how bad the methods are.’ Satnavs negate the need for learning your route:why bother? it is not because you are not struggling that you are not learning from the satnav, it is because you don’t care.

    Furthermore, feedback that is not immediate is basically useless. If you get the child to be interested, the best way to teach maths, for example, is to guide the child with immediate feedback as the lesson progresses. Try teaching integral calculus with delayed feedback.

    • David Didau says:

      “feedback that is not immediate is basically useless” – there’s a considerable body of research that suggests the exact opposite to that. This literature review is a good starting point if you want to challenge some of your beliefs.

      • Nigel W says:

        I think context is important. In some cases this is true, in others less so. Where a kinaesthetic experience is involved, allowing the learner to move from performing to reflecting is important, but feedback should ideally happen pretty swiftly; within a minute or two. Where skills are cognative, rather than physical, I believe this is less important.

      • Thanks for the interesting paper.

        With respect to feedback, it seems to deal with two sorts. Augmented or external feedback with reference to the learning of motor skills – which I understand to refer to teacher comments as opposed to the immediate, intrinsic feedback that the student gets from the physical environment. This makes sense to me, as having to interpret the results of your actions in the physical environment requires greater mental effort and is less emotionally laden (e.g. with feelings of failure, annoyance, pride etc) than being told how well or badly you just did.

        Second, it deals with feedback on multiple choice drilling exercises. One paper suggests that it is pointless giving feedback on correct answers – I should have thought this was fairly obvious. A second states that delayed feedback is more effective on incorrect answers. Again, intuitively, this seems to make sense to me as there is a danger of the correction getting muddled in your mind with the incorrect answer that you have just given. You know that two things just happened: one was that Paris was said to be the capital of France and the other was that Paris was said to the the capital of Germany but you can’t remember which was the incorrect answer and which was the correction.

        I think there is also evidence that delay is helpful for review and re-enforcement. So if you learn something, you should review it in the evening, 24 hours later, and then a few days later again. If you get something wrong, then it makes sense that the corrective feedback should be given according to a similar schedule. Maybe it is unwise to ask the question again when there is a risk of the student giving the same wrong answer and forming a habit. The way that I would do this, intuitively, in the context of a drilling exercise, is that if a student gets a question wrong, tell him the answer and ask him the question again immediately so that he repeats the right answer, ask him again 30 seconds later, then 1 minute etc, getting him into the habit of giving the right answer. My point here is that there are many different strategies other than just immediate or delayed feedback.

        But it strikes me that both these cases (augmented feedback when learning motor skills and drilling) are fairly special cases.I think in the case cited by muzzyizzit (teaching integral calculus) the immediate feedback given by the teacher is the equivalent for abstract knowledge of the immediate feedback given by the physical environment in the case of motor skills. I don’t think you would learn much about bowling if you couldn’t see what happened to the ball as you used different arm actions – that is the immediate feedback.

        I think it would also be useful to distinguish between different sorts of feedback: evaluative (“well done”, “that’s rubbish”); consequential feedback (if you throw it like that, it veers to the left”); corrective feedback (“don’t do it like that, do it like this”); progression feedback (ok – try this next). I am sure that too much evaluative feedback (praise or criticism) may well be unhelpful – and with respect to the other sorts, I am sure there will be much devil in the detail about which sorts suit which situation in which combinations.

      • muzzyizzit says:

        I can’t stand that Bjork fellow. His arguments are like ‘just-so’ stories. You can improve performance without long term learning (who knew?). I got a string of A*s in maths, English, bilology, physics, etc. You know how? By cramming and working on past papers. I forgot all of it about three minutes later, except the maths. This is old hat: of course you can improve performance without long term learning. The issue is what a given individual wants to achieve.

        • David Didau says:

          Of course (of course!) you can improve performance without long term learning. Does this mean you advocate exam cramming? This is certainly a successful strategy if you all you want is to pass exams. Is it?

          The point was (an is) that you can improve long-term retention and transfer by reducing performance in the short term.

          You can hate Bjork if it makes you feel better, but it’s probably my explanations that are at fault.

  9. Instead of critiquing the analogy, let’s consider Hattie’s three points.

    Firstly, where am I going? Having an end goal allows both teachers and students to at least point themselves towards the ‘so that’. My intitial reaction would be to accept this as essential. One thing that history students often do is fail to answer the question – they might have the knowledge or the skills, or even both, but use these inadequately. In this case I feel it necessary to point out those goals, if only because of constraints in the timetable – I’d love my students to engage more with problems at GCSE, thus having to unpick difficulties they find, but with five hours a week I really can’t afford it. At KS3, however, I can afford to delay this feedback (feed up) so that they’ll get it right later on. In an art lesson, though, I’d expect there to be a very different form of feedback; more questioning than repositioning.

    How am I going? Well, sometimes we need to tell students what’s up, and tell them soon. For example, what if they keep using the wrong ‘there’? However, if we have robust enough systems in place then perhaps the students can catch these themselves. Or maybe we feedback to the class: ‘One of the issues I’m finidng is the overuse of “they” – who are they? We must be more specific. Highlight every “they” now.’ I tend to ask questions that will extend the use of knowledge or critical thinking, such is my subject area, rather than, ‘You’ve not yet considered AO3 – refer to the mark scheme.’ I struggle with this, though, because I am a little scared about leaving students without any map at all. However, I’m also convinced that this will make them better historians and better writers over time. The issue of mapping their progress is, again, one that becomes more problematic with less time at GCSE and A Level – no one is going to leave their students to flounder! So I am trying to embed these questions at KS3.

    Finally, the feed up. I suppose I’ve covered this in the last paragraph. Again, I am trying to create a culture at KS3 of re-drafting, re-focussing and extension of the use of knowledge through written questioning. However, it’s here that I’m increasingly delaying the feedback/up. I’m doing this to, in part, develop independence and responsibility. This is tough. It requires everyone, not just in that class, to be on board. It means homwework has to be meaningful. It means students have to recognise that the finished product is something to be striven for, not completed last-minute. Students therefore have to feel as if what they’re doing has value and meaning, so delaying feedback is a tool to improve and reflect.

    Back to the satnav, then: I don’t own one precisely because I know that I won’t learn a route based on exact instructions. I’ll learn it by making mistakes. I actually (and genuinely!) make deliberate mistakes in order to learn a route better. I look for signposts (real and metaphorical) along the way and later use Google maps to work out where I’ve been. I delay the feedback for myself.

    So, I’d suggest that maybe it’s not about the amount of feedback, but the timing and the pointedness of it. That might mean that there is less feedback, or maybe that we give the same amount but vary its delivery, be that whole-class, in groups or individual.

  10. […] I’m not an especially good driver, but I’m a truly terrible navigator. This used to mean that I would get lost. A lot.  […]

  11. Mrbenney says:

    Another thought provoking post. Not providing or delaying feedback goes against every instinct I have as a teacher. However having read a bit of Bjork (and your blogs) I do question how effective my verbal feedback is. For me I think it depends on where the learner is. If they are “nearly there” then I can see that allowing them to struggle by delaying feedback (virtuous guidance as well as the blunter “this is the answer”) could be beneficial. However if the pupil is suffering from a misconception then I would struggle to see how instant feedback would not be beneficial. As always there are lots of variables in play but it does make me reflect on feedback, particularly verbal feedback, that I engage with in the classroom. I don’t doubt that sometimes I jump in with (virtuous) guidance where less might well be more.

    • David Didau says:

      I think that’s exactly right. The hard part is knowing when to jump in and when to hold back…

      • Martin harris says:

        Sat navs give direction not feedback . Poor analogy . Secondly what goal are you after David? To enable the passing of exams at the highest grades or something else? Have you considered that this continual rant about feedback and the inadequacy es of AfL in the end are not making a difference because I cannot see what your educational vision is? What are you aiming for- exam grades? Are children just being educated on a conveyor belt for work ? Or is there anything else for you? Maybe a post on your holistic vision for education into which all this fits would be a good idea. Your vision – not a set of possible questions.

        • muzzyizzit says:

          I am reading David’s book. He says the point of education is to make kids cleverer. He then spends chapter after chapter basically jumping from one thing to another. The main thing that I get from it is this: we should arm children with academic language they need to do well in life. This leads to cultural capital in the sense that kid will read the ‘right’ sorts of books and sound appropriately ‘educated’. He recognises that this is an ‘elitist scam’. I don’t know whether he forgot the ‘making kids cleverer’ bit by that stage or he thinks ‘cleverer’ just means sounding a certain way stuffed with academic language.

          Incidentally, in his book David has some hilarious examples that are intended to show how much better it is to use ‘academic language.’ In one case, I laughed out loud because the ‘every day language’ was much better.

        • David Didau says:

          Obviously I’d argue Satnavs do give feedback. But maybe you have developed a new definition which doesn’t include any kind of direction being given?

          Why do you want to caricature me as ranting? Does that make you feel better about something?

          As for my vision for education, did you read this? It’s not really a vision statement but it might give you some sort of idea:

          • Martin harris says:

            Direction and feedback are the same? Nope. They CAN BE but seldom are. Direction = obedience without questioning
            Feedback provides for thought, creativity, analysis, hypothesising etc etc.
            Furthermore this article does not give a vision but a small aspect of actions. Now if you communicated a vision that encompassed your many thoughts , then that would be something. A new book perhaps? 🙂

          • David Didau says:

            You’re very certain of your assertions Martin. I find it best to be wary of such certainty

          • Martin Harris says:

            I await your vision with genuine interest . It’s very important, I think, that you write it. It’s important because you are influential but also and more importantly are at the leading edge of educational thought in Britain. I challenge you to complete it.

          • David Didau says:

            Honestly, I don’t think I have what you’re after. Who would you point to as someone who has articulated what you consider to be a coherent vision of education?

  12. Debaser says:

    I would say that using a satnav is analogous to the ‘scaffolding’ process rather than the process of giving ‘feedback’. A satnav-aided journey is essentially a heavily scaffolded one: the satnav/teacher does most of the thinking for you and all you have to do is fill in the gaps between the prompts.

    Perhaps it could be seen as a warning against overusing scaffolds (such as the invidious PEE and its variants) and failing to ‘take them away’ and encourage independent learning (as per the cycle you recommend in your book).

  13. […] I’m not an especially good driver, but I’m a truly terrible navigator. This used to mean that I would get lost. A lot. When I first moved to Bristol in 2001 I bought an A-Z of the city and when driving somewhere new I would have to stop the car periodically and try to align the map  […]

  14. I read this yesterday and have been chewing it over since. I think the problem with the analogy, for me, is not with having a satnav stand in for classroom feedback, but in having David the driver stand in for a class of students, or an individual student.

    As a driver, you have a clear direction and the will to get there (I presume). You need to get the shopping/find the school you’re visiting and are (presumably) willing to keep looking until you find it. If you fail and call it a day, you’ll go home hungry – but you’ll presumably find a convenience store which will meet your immediate needs. How long will my students be willing to continue their search for a far further off, intangible goal, like being a great historian or even just getting an A…? I find the research detailed in books like Make it Stick compelling in terms of learning, but I think it sometimes overlooks the emotional dimensions which motivate students to begin work or keep going when it’s tough. Feedback – a nudge back in the right direction – is a key way to let students know either a) yes, it’s just around the corner, or b) you’re heading the wrong way, but if you just follow the road you were on before briefly, you’ll be back on track.

    As Damian and you have concurred, the critical question is not, then, that feedback is a bad thing as such – it is the judgment as to when it is best offered. Toby’s response is interesting on those grounds – we have to work out how best to build in productive struggle for all students… I’d be very curious to read more about your ideas of exactly how we can do that without undermining their commitment entirely.

  15. […] Didau’s […]

  16. […] The problem with SatNavs, or how feedback can prevent learning – David Didau […]

  17. […] too frequently. Too much input from teachers results in what I’ve come to think of as ‘satnav feedback‘ – the sort of continuous, immediate feedback provided by a satellite navigation […]

  18. Greg Yates says:

    Personally, I find the satnav analogy for feedback fascinating since the device provides feedback on a continuous basis, but as something you (as driver) has to actively control. Consider the visual display (coloured lines), where your position is shown in relation to where the machine wants you to move. You have full control over your glancing, and that is the point. To construe this process as ‘giving instructions without feedback’ is not right. My satnav gives feedback on several different levels: (a) where I am on the globe, (b) that I have successfully programmed in a genuine destination,and will get there in 3 hours god-willing, (c) the discrepancy between the road I am on, and where it wants me to be, (d) when I miss a turn it says ‘recalculating’ in a kind voice, and (e) that I am not exceeding the speed limit on this unfamiliar road.
    BTW I totally agree with David’s point that using the satnav, its very effectiveness as an external feedback tool, then prevents me from learning that route. As a casual visitor to Britain (from Adelaide), I may need to drive from Nottingham to Chichester once in this life. Trying to learn that route would be possible, but energy depleting. Cheers, Greg.

  19. […] Didau’s […]

  20. […] the past I’ve used the analogy of navigation to explain this. Using a map is effortful and it’s easier to memorise routes than to have […]

Constructive feedback is always appreciated

%d bloggers like this: