Can progress be both rapid and sustained?

We start out with the aim of making the important measurable and end up making only the measurable important.

Dylan Wiliam

Does slow and steady win the race?

‘Rapid and sustained progress’ is Ofsted’s key indictor for success. Schools across the land chase this chimera like demented puppies chasing their own tails. But just when when you think you’ve gripped it firmly between your slavering jaws, the damn thing changes and slips away.

You see, the more I look into it, the more I’m convinced that progress cannot be both rapid and sustained. You cannot eat your cake and have it: we either focus on the long term goal of learning, or give in to the short term pressures of performance.

This last week has been a watershed. Over the past year or so I have become increasingly certain that making progress in lessons is a nonsense and any attempt to get students to demonstrate their progress is a meaningless pantomime that benefits no one. The past few days have seen any remaining doubts shattered.

The arguments laid out here should be adequate to convince even the most entrenched and wrongheaded champions of ‘progress in lessons’.

But there’s a further problem. Basically, slowing down the speed at which students learn increases long term retention and transfer of knowledge.  We know from the Hare and the Tortoise that travelling faster is not always better. And as in folklore, so in education; in our attempts to cover the curriculum we can sacrifice students’ learning. We’re all under increasing pressure to teach to the test and the idea of not cramming in the content is, frankly, a bit unnerving. We’re caught between a rock and a hard place: slow down and risk lack of coverage, or speed up and sacrifice depth of learning.

Relying on direct instruction would seem more efficient and predictable than messing around with enquiry and discovery learning and, unsurprisingly perhaps, this is borne out by research. In our efforts to make sure we cover the course engaging students in  time-consuming, cognitively demanding activities that nurture deep understanding appears an unaffordable luxury. In GCSE English courses, reading and analysing an entire book has become a relic of a half forgotten and happier past. Breadth trumps depth. And the more pressure you’re under, the more you’re likely to skip.

The idea of pacing, asks us to plan our programmes of study so that  learning is chunked and topics are arranged coherently, with a clear sense of how long different elements will take to teach. Obviously, we also need to allow for some unpredictability depending on the particular mix of kids in front of us: as teachers we need to keep our expectations high but keep a weather eye on areas in which our students struggle. In this way we can arrive at the most efficient way of rigorously covering our content while still allowing time for the experimentation and inquiry which which is so vital for long term retention. This is something Maurice Holt and the Slow Education gang have been bandying about for some time but I was fascinated to discover the work of cognitive psychologist, Robert Bjork which seems to bang the same drum.

Bjork describes conditions which slow the pace of learning but increase long term performance as ‘desirable difficulties’. Now in a world in which ‘rapid and sustained progress’ is sought we might have a problem: rapid progress may well be the enemy of sustained progress. And as such, techniques which favour sustaining progress at the expense of the speed at which this progress might well go unappreciated by a pitiless inspection regime.

But, as ever, we need to do what is right, hold our nerve and be ready to explain our thinking. Here’s an outline of some of the techniques we can use to concentrate on sustaining progress:

Variation – As we’re all aware, variety is the spice of life; a steady diet of the same-old same-old, no matter how delicious, is enough to put off anyone. So it should come as no surprise that using the same lesson structures will, eventually  start to pall. The research on variation in lesson design looks specifically at mixing up deep and surface learning strategies rather than trying to cram in as much deep learning as students can stomach. This may at first seem counter intuitive; surely we’re better off prompting students to make profound connections between the things they know and challenging them to make increasing abstract generalisations and hypothesises? No, apparently not. The theory suggests that getting students to remember facts and expand their knowledge base is just as important as getting them to creatively manipulate all the stuff they’re digesting.

The point is that if we are more interested in long term retention and processing we need to provide students with a balanced diet of deep and superficial knowledge. This may be less exciting in the short term, and certainly, messing about with hexagons can look really impressive to an observer in a way that learning facts doesn’t but we need to keep our eyes on the prize and remember that being able to perform spectacularly in a lesson is not the same as being able to perform well independently in an exam.

The effects of 'spacing' learning

The effects of ‘spacing’ learning

Spacing – The concept of designing your schemes of learning so that new concepts and important information is regularly revisited is nothing new. I first came across it several years ago and, ironically, promptly forget about it. I was reminded of it when reading Nuthall’s essential The Hidden Lives of Learners and stumbled across his insight that new information has be encountered on at least three different occasions in order to be retained. Bjork contends that spacing “is one of the most robust results in all of cognitive psychology and has been shown to be effective over a large range of stimuli and retention intervals from nonsense syllables to foreign language learning across many months.”  And if we increase the spacing between reminding students about new information this “enhances learning because it decreases accessibility of the to-be-learned information” Or, in other words, the harder you work at having to call something, the more likely you are to remember it in the future.

Here’s another clip of Bjork explaining the effects of spacing on retention:

So, we need to design our curriculum to cover and recover information. There are various competing theories on the optimum spacing of learning but as long as we work out in advance when and how we’re going to revisit what we want students to retain we should be OK. One piece of home spun common sense is that we should ‘input less, output more’. What this means is that having encountered some facts we will learn far more if we try in some way to recreate this knowledge rather than just reviewing what we’ve learned. Writing this post is far better for my retention of all this cognitive psychology than simply reading it over and over. Although this is something every teacher knows instinctively, it’s nice to have some of our biases confirmed by the boffins.

Interleaving vs blocking – If we accept that spacing works, then interleaving is a great way to design your scheme of learning. If we’re just hanging around for a few days waiting for the optimum time to have elapsed before reteaching what will we fill the intervening lessons with? Happily, interleaving provides the answer.

Traditionally we ‘block’ learning. This means that students exhaustively focus on one particular concept or type of problem until they are considered to have mastered it and then they move on to another, related topic and so until they have studied all the components of a course in discrete blocks. Interleaving, on the other hand, involves doing a bit of everything at the same time so that students might tackle several concepts or try to solves several different kinds of problems at once. Here’s the kicker: when students’ learning is ‘blocked’ they perform much better during lessons – it looks like their learning. But when they’ve finished studying all their blocks of knowledge and are tested at the end of a course, their score decrease fairly dramatically. When teaching interleaves knowledge students perform worse during lessons but their retention at the end of a couse appears to be dramatically better.

The observant among you will have highlighted a couple of problems: if we observe lessons looking for evidence of progress, we will encourage teachers to block learning so that students perform better at the time of the observation. But is a system which (increasingly) relies on terminal exams, teachers who interleave learning (an their students) should come out on top. If the research is accurate, this really is a no-brainer.

Here’s Bjork again:

Feedback– Apparently, delaying and reducing feedback promotes learning. But this can’t be right, can it? Surely feedback is the most effective thing a teacher can be doing? Well, yes it is, but sometimes less is more.

I recently signed up to Dr Will Thalheimer’s subscription service on feedback and he has this to say:

Feedback (in most learning situations) tends to be more effective if it is delayed. It works the same way as spaced repetitions. In general, the longer the delay the better, up to a point where the delay can reduce learning.

In addition, some research shows that reducing the frequency of feedback can actually increase learning. Giving feedback after only half of all activities had more impact on long term retention than giving feedback on all of them. There are three reasons for this this:

1. Frequent feedback makes students too dependent on external validation and prevents students from developing an ability to rely on their own judgement.

2. Feedback works by “facilitating next-response planning and retrieval. In this sense, frequent feedback might provide too much facilitation in the planning of the subsequent response, thereby reducing the participant’s need to perform memory retrieval operations thought to be critical for learning”.

I’d advise taking all this with a large pinch of common sense but it’s worth considering whether the way we give feedback might be preventing students from becoming sufficiently resilient and independent.


It seems that many of the things we’re told to do in lessons because they’re great for demonstrating progress may actually be getting in the way of deep learning. If we accept that performance is not a reliable indicator of learning then we may have a problem. Most current educational thinking is all about checking students’ performance in lessons to judge their learning so that we know what to teach them next lesson. We’ve been labouring under the misapprehension that we need to check progress in lessons otherwise we’ll have no idea what students have learned. But real learning takes time. As Nuthall points out, “learning is invisible and cannot be seen in the activities of the teacher or students”. The fact that learning and forgetting can happen simultaneously mean that it is “impossible for teachers to judge what their students are learning without much more detail and individually differentiated data than they have available in the classroom.”

So, instead of setting up activities which test students’ current performance to use as evidence of progress and then acting on this to plan future lessons in the belief that we know what our students have learned we should instead listen to cognitive psychologists about how the brain works and how learning happens and design curricula and lesson accordingly.

Let’s focus on learning rather then performance, and let’s focus on progress which is sustained rather than that which is rapid.

Next post: strategies for designing lessons based on what cognitive psychology tells us: lessons for learning

Related posts

Learning vs progress

What’s deep learning and how do you do it?