Rosenshine’s Principle 9: Require and monitor independent practice: Students need extensive, successful, independent practice in order for skills and knowledge to become automatic.

“Research Findings

In a typical teacher-led classroom, guided practice is followed by independent practice – students working alone and practising the new material. This independent practice is necessary because a good deal of practice (overlearning) is needed to become fluent and automatic in a skill. When material is over-learned, it can be recalled automatically and doesn’t take up any space in working memory. When students become automatic in an area, they can devote more of their attention to comprehension and application… (Rosenshine)”

I’m rubbing my hands together as I do this because Rosenshine’s 9^th Principle leads us to juicy topics that can divide educators. I’m going to tackle this (self-indulgently) by hitting it from three sides:

1. How teacher-led should independent practice be?
2. How much independent practice is necessary, and what about the kids who catch on fast?
3. What does this all have to do with Cognitive Load Theory (seeing everyone talking about it)?

The clear downside to my enthusiasm on this topic is that you’ll need to settle in for a slightly longer-than-usual post. Here we go!

How teacher-led should independent practice be?

In a word – heaps!

Like always, Rosenshine’s other principles send their tentacles into this one (remember, think of these principles more as a Venn diagram than a list), so you’ll be thinking about Rosenshine’s 7^th and 5^th, 4^th and even his 2nd Principle as you read about this one.

Anyone who’s ever taught anything to anyone knows what teacher (expert) )-led practice is. It’s a person with expert knowledge breaking skills up into teachable pieces, teaching those pieces and helping the novice get enough practice so they store those pieces in their long-term memory. I know this is covered in Rosenshine’s 2^nd and 4^th Principles, but I do feel compelled to revise them for you. You’re welcome!

You’ll recall that the novice (learner) is borrowing from the expert’s (teacher’s) long-term memory and putting new piece(s) of declarative or procedural knowledge into their long-term memory. The expert is on hand to carefully watch how these skills are being learned, with an eager eye, always watching for errors.

The tennis coach (expert) teaching a topspin forehand knows that the student will need to practice correct foot placement and racket position, both on the backswing and follow through. Watching the ball, hitting it at the top of its bounce, and having a proper grip are all sub-routines that must be practised by watching the coach and replicating the sub-skill. The coach will watch students as they practice, correcting as soon as they see an error and reinforcing the correct technique before the students are expected to consistently return a 150+ km/h serve on the forehand.

This is highly teacher-led because there are so many places where the skill of a topspin forehand can break down through a mistake in any of those component skills. An error in footwork will set off a domino effect that will see the ball hitting the net or flying out. Even once students are released to have a go at the entire skill, they are watched because the teacher knows what errors often crop up and when. They can be quickly corrected before bad techniques are locked in.

Many see education – teaching as somehow being exempt from these same principles. It is as if, somehow, a teacher can provide one or two demonstrations/explanations or incomplete instruction and then set students off to practice independently and successfully. It’s just not how learning works.

Do you remember watching a teacher model one, maybe two examples of finding the value of ‘x’ in the equation 3x +5 = 15? This has many steps; just like a forehand, one error will impact the final result. There are many sub-skills that a student must have automated in long-term memory to tackle this. Understanding what an inverse operation is and in what order to carry them out (BEDMAS) on this equation is one of them, and how to deal with negative numbers is another. But before all this, a student needs to understand that what sits on each side of the = sign has the same value; they’re just expressed differently.

Many pieces of declarative and procedural knowledge are needed to tackle this problem, and one or two complete run-throughs will not cut the mustard for most students in the room. If you’re experiencing shortness of breath as you read this, we share some trauma about algebra.

If you’re interested, here’s how this one gets solved!

In a classroom informed by Rosenshine, the teacher will be starting a long way back and tackling each of the sub-parts of this algebra problem one at a time; students will be using personal whiteboards to work on and show the teacher their work as they go. I suspect this is going to take two lessons at the very least to have most students solving these types of multi-step equations.

The teacher will also ensure that when students have a go at the whole procedure on their own (the independent practice part), the practice problems will be identical or nearly identical to the type of worked example, all tackled together as a class. Any differences will be well thought out by the teacher. To send students off to try something entirely different to what was modelled as a class under the guise of ‘inquiry’ or ‘productive struggle’ would be like asking students to watch the tennis pro forehand hit two or three forehands and then go off to another court to experience success. Bonkers!

When the independent practice comes, and students are released to have a go at a series of similar problems, the teacher will make sure adequate in-class time is allocated to this independent practice. If they set it only for homework, the student will probably not be monitored, and if they are (unless they have a maths teacher as a parent), the monitoring will be non-expert. I’m not saying all parents are like mine; I’m just trying to emphasise that it’s best to have expert teachers watching the most error-prone stages of skill development.  

Independent practice needs to be plentiful and carefully monitored. Otherwise, we create casualties because of low student success rates, and more content needs to be retaught from scratch…

“All right, everyone, I’ll show you all one more time, pay attention to this time!…(sigh)”.  

In EDI™, a popular framework I’ve mentioned throughout this series, the ‘I Do, We Do, and finally ‘You Do’ sequence helps teachers structure and monitor independent practice in a highly effective way. EDI™ has built on research from traditions of Rosenshine’s ‘d.i.’ and Engleman’s ‘D.I.’ and has probably hit the sweet spot between them both.

How much independent practice is necessary, and what about the kids who catch on fast?

Ah… the elephant in the room. We all know that some kids need more practice than others. Teachers worry a lot about not giving some kids enough time and practice as much as they do about the possibility of the top end of kids getting bored by practising something they can already do. Differentiation is sometimes offered as a simple solution by folks who were in classrooms too long ago to remember that it’s impossible to design entirely different instruction and at-level practice for other students. Still, they’ve convinced themselves it was easy when they were taught! Don’t get me wrong, differentiation is possible, but when done well, it involves slight and manageable variations away from core instruction delivered to the whole class, not thirty different learning programs for thirty students. Again, bonkers.

So, where does this leave us? We know students need different amounts of independent practice to become proficient at something. That’s a fact! We also understand that just because you can do something doesn’t mean you can’t get even better (faster/more automatic) at it. Here comes analogy number two! Away from tennis to driving a car. If you started driving at sixteen, by the time you got off your Probationary licence, say around your nineteenth birthday, you thought yourself to be a proficient driver. You were good at getting from A to B, even in traffic or rain, without endangering your or other road user’s lives. I’m approaching fifty, and I would say I’m a much better driver than I was at nineteen. I can better anticipate other driver’s behaviour. I know to gently ease the car back to the tarmac if the wheels happen to hit gravel, and insurance companies certainly know I’m a better driver, as reflected by my premiums when compared to what a nineteen-year-old would be paying! 

How is this possible? I’ve practised driving to the point of overlearning, and as long as I’m able, I’ll keep overlearning. Experts in any field have crossed into overlearning. It’s a thing, and it’s what puts a considerable distance between me and a Formula One driver or a professional tennis player.  

So, is more practice at something you can do already damaging? No! Most students will tell you that it’s pretty satisfying to do something that comes relatively easy. Becoming fluent and automatic in a skill feels excellent, and I suspect that grownups spend more time worrying about students practising something they can already do than students do! “I’m bored” usually means “This IS hard, and I have low frustration tolerance” or “I miss my phone, and I want it back”. Not always, of course, but a lot of the time.

Schools using structured literacy programs, characterised by lots of independent practice of previously learned material, see benefits for high-achieving students as well as their lower-achieving peers. This isn’t much help when little Billy goes home and complains to Mum that school is boring (to get some attention), but many kids overlearn this skill, too!

So, does practising something you can already do harm you or your learning? Nope. It moves you closer to an expert, and that’s never a bad thing in my book!

What does this all have to do with Cognitive Load Theory (seeing everyone talking about it)?

Now, the one I’ve been waiting for! Cognitive Load Theory has risen like the sun for educators. It is fast becoming an important, even unifying set of ideas that makes sense of what teachers have always known works and simultaneously dispels the many nonsense ideas and myths that have littered the teaching landscape, shredding the learning of far too many children. Cognitive Load Theory was developed by Australian (oi, oi, oi) Professor John Sweller. Rosenshine says:

“When material is over-learned, it can be recalled automatically and doesn’t take up any space in working memory. When students become automatic in an area, they can devote more of their attention to comprehension and application…”

This is precisely what Cognitive Load Theory (CLT) explains. Ollie Lovell’s book ‘Cognitive Load Theory In Action” is a very nice summary of CLT. Still, if you’re a little stretched, you’ll find heaps on the internet, including clips of Sweller taking no prisoners as he explains the implications for CLT to classroom teachers.

In a nutshell, CLT goes a little like this. We, humans, can only think about and work with a minimal number of things at one time. Working Memory is the tiny place we do this! Some suggest we can only simultaneously deal with (process) about 3-5 pieces of novel (new) information. This is a startling constraint imposed by a memory system that has to be very selective about what is allowed to get into and modify our long-term memory  (aka – learning) and what isn’t.

We all develop an intuitive way of managing life with this tiny attentional workspace by doing things like resisting ducking off to do another job while our drink bottle fills at the filter tap. If we don’t, most of us end up cursing ourselves as we mop up a puddle of water, lamenting that we exceeded the limits of our working memory and forgot the water bottle was filling.

Life is full of moments when we exceed the limits of working memory. Any time you think about too much and feel like a fog has descended on your ability to think, that’s cognitive overload. Cognitive overload is terrible for learning.

So, the question begs, with this seriously limited mental workspace, how can we think complex thoughts and carry out complicated tasks (like driving) when we can only deal with a handful of pieces of information at a time? The answer lies in our brains’ ability to take sequences of tasks or strings of information (like facts) and bundle them into chunks that move to our long-term memory and attach themselves to other chunks of similar information. These form what the brain nerds call ‘schema’ or ‘schemata’. When we need to do or think something complex, we draw on a schema from long-term memory (retrieval), and vast bundles of information from these schemas come into working memory. The incredible trick of evolution is that these densely packed bundles of information from long-term memory take up next to no space in working memory but allow us to process new information.

Rosenshine tells us quite accurately, and CLT agrees with him.

“When material is over-learned, it can be recalled automatically and doesn’t take up any space in working memory. When students become automatic in an area, they can devote more of their attention to comprehension and application…”

So, before information becomes tightly bundled and popped into a schema in long-term memory, we have to do the hard work with it! This requires lots of practice (rehearsal) and retrieval. You might remember my retrieval metaphor of the grassy, foggy field in Principle 5. Rehearsal (and retrieval) of overlearned information is often automatic and takes no great mental effort and, thus, puts next to no load on working memory that needs to be unencumbered to process the new information.

So, lots of carefully monitored independent practice, after a great deal of guided practice (see principle 5), is essential to allowing students enough rehearsal/retrieval to chunk and automate new information in long-term memory.