Retrieval practice: recalling the end goal

Sherrington (2022) wrote a brilliant blog called “When daily quiz regimes become lethal mutations of retrieval practice”. In it, he suggests that retrieval practice regimes may not be as useful as teachers and leaders think. As he puts it –

“Rather than it being a consolidatory reinforcement of their learning, plugging a few gaps, students feel they are being quizzed on the thin air of vaguely-encountered wisps of disconnected factoids from a dim past.”

We’re pretty confident that retrieval practice has the potential to improve learning. So why might the way it’s used change how effective it is?

Here’s my hunch:

I think sometimes, we lose sight of the end goal.

What’s the end goal?

You might say, the end goal is pupil learning. You wouldn’t be wrong.

But I think the word ‘learning’ can be misleading.

I think we might have conflated learning with just retention of individual bits of information. A lot of the research treats learning this way because it’s easier to measure.

But this isn’t the type of learning we’re after in schools. Retaining distinct pieces of information over time doesn’t help pupils do the things we actually want them to be able to do.

And, mandating ‘5-question do now quizzes’ at the start of the lesson may, unless we are careful, encourages this same view of learning.

What’s gone wrong here?

We’ve forgotten what we really mean by learning.

The best way we can understand the learning we want for pupils is to start with what we want pupils to do with the knowledge they have.

Amongst other things, I think we want pupils to be able to –

  • Retain important knowledge in subjects they study.
  • Access relevant priorknowledge so they can use what they know.
  • Transfer their prior knowledge to new contexts to understand new ideas and build on what they know.

To do all these things, pupils don’t benefit much from isolated facts.

They need their knowledge to be stable, accessible and well-organised. This comes in the form of bodies of knowledge (Ausubel, 2012). Bodies of knowledge can be seen as hierarchically connected, related ideas.

Pupils need to form bodies of knowledge within and across domains.

Forming bodies of knowledge enables pupils to do all the good stuff listed above:

  • Retention: newer ideas in a body of knowledge are connected to more stable, older ideas. By virtue of this link, they are anchored and retained for longer.
  • Accessibility: bodies of knowledge are well organised. This means knowledge is more readily available when pupils need to use it.
  • Transfer: we only learn something new when we connect it to what we already know. This is transfer – using what we already know in a new situation to understand something new. If pupils have a body of knowledge in a domain, the ideas are connected. It’s easier for them to recognise and make links between new ideas and prior knowledge.

If bodies of knowledge are the end goal, what does this mean for using potentially-potent techniques like retrieval?

Content is king

Retrieval practice isn’t the first thing we need to think about. Retrieval needs to be viewed as a potential catalyst for building bodies of domain knowledge and links between them.

The first thing we need to think about is the content of the lesson and the integrity of the subject we are teaching. Retrieval of what?

At Ambition Institute, we call this ‘the what before the how’. Content is king. This comes first.

With this mindset, we start by defining which knowledge in the subject is important enough to warrant retrieval by pupils. Retrieval questions should then target this core knowledge.

These decisions should be made in departments/phase teams. Teachers work to determine what is important enough to spend precious lesson time having pupils retrieve and giving feedback on.

Done this way around, retrieval is a catalyst. And content, i.e. the body of knowledge, is king.

Build the body of knowledge

It goes beyond just the selection of content though. The bodies of knowledge we want pupils to form determine the types of questions we should ask. Different types of questions will get pupils to access knowledge in different ways and recognise different connections.

As an English teacher, I might be teaching a Shakespeare play with a complicated plot and characters with similar-sounding names. I may decide pupils should do some factual recall of characters’ names and relationships at the start of the unit. Short answer or multiple-choice questions could do the job.

Later in the unit, I may want pupils to recall and make links between the use of themes across plays. This will help create connections, curating the bodies of knowledge I want them to build. I may need one, long-answer question.

I need the flexibility to use retrieval practice in a way that suits the content and helps pupils build bodies of connected knowledge.

If I’m forced to use five quick questions in a certain format and on random topics (e.g. so I can say I’m ‘spacing’ learning), I may not be helping pupils make connections and build the bodies of knowledge they need.

To be clear, I’m not an expert in every subject. Randomly spacing content may be useful in some subjects: I’m open to debate. What I am saying is that we need the end goal in mind: how do we best help pupils build the bodies of knowledge they need?

Tight but loose

So what’s the alternative to mandating 5 retrieval practice questions? Is losing the routine a problem? Routines are almost always handy. They save time and help manage cognitive load. We also have to take into account the idea that a do now routine can motivate pupils and lead to a calm start to the lesson.

One way to manage this is for leaders to think about being ‘tight but loose’ (Thompson & Wiliam, 2007). ‘Tight’ on important things teachers must do. ‘Loose’ on things where they have more scope to decide.**

For example, something ‘tight’ could be the carefully considered use of retrieval practice each lesson. Something ‘loose’ could be the format and perhaps when dedicated retrieval is carried out in the lesson. These loose decisions can be devolved to department/phase team judgement based on the ideas presented above around the importance of content and building bodies of subject knowledge.

Key takeaways:

  • Keep the end goal in mind. We want pupils to build bodies of knowledge. To do this they need to make connections. Peppering them with random, quick-fire questions may not be the best way to achieve the goal.
  • Respect the content. The best way to teach the subject is to understand what the important knowledge is and how it is connected. Retrieval practice needs to target this important knowledge and help pupils form these connections.

*These aren’t mutually exclusive. For example, accessibility enables transfer.

** Making the decisions on what is tight and what is loose is tough and we do not always get it right. We can look to evidence to support us and also our experience. These decisions may evolve as our understanding evolves.


Ausubel, D. P. (2012). The acquisition and retention of knowledge: A cognitive view. Springer Science & Business Media.

Thompson, M., & Wiliam, D. (2007, April). Tight but loose: A conceptual framework for scaling up school reforms. In annual meeting of the American Educational Research Association. Chicago, IL.


3 responses to “Retrieval practice: recalling the end goal”

  1. […] Retrieval Practice: Recalling the End Goal – Sarah Cottingham […]


  2. longsandscpd avatar

    Reblogged this on Longsands LPD.


  3. […] Retrieval practice: recalling the end goal […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: