Retrieval Practice and the Art of Schema Building

Blake Harvard, a Psychology teacher and a regular in the Research ED community suggest “You would be hard-pressed to find a more impactful and widely applicable learning strategy than retrieval practice”. Having spent 2 years implementing and researching retrieval practice I could not agree more.

My favourite statement to use with all my pupils is “Imagine your favourite athlete or musician, they would not begin practicing the day before their main event, they practice continuously, identifying gaps and doing what is necessary to reduce them.”

The teacher I have become over the last 2 years dislikes the teacher I was for the first 7 years of my career. Having trained in an era of league tables, performance-related pay and quality assurance I inevitably succumb to an approach of teaching to the exam spec. I mean, all things considered, is it a bad thing to want your pupils to achieve good GCSE grades? I would argue NO as surely at least one of our motivations to become teachers in the first place is to broaden the life chances of the young people we teach. However, ‘teaching to the test’ can result in the manifestation of blinded and misguided practice.

In my early years as a teacher and a middle leader, I was obsessed with ensuring the GCSE specification was taught by the Christmas before year 11s were due to sit their exams. Even the areas I line managed in addition to Geography; History and MFL were being encouraged by me to ‘have the course finished’ by Christmas. Why? I hear you all asking. “Because it gives us more revision time”, I would preach. Hindsight being 20/20 I can see now just how blinded I was. We didn’t spend time revising, we spent time re-teaching, because putting it simply pupils are human beings, and as human beings they forget things. Imagine being asked to recall very specific information from 3 years ago, which has only been revisited on 2 other occasions and even those 2 occasions were to prepare for an end of year test or mock exam.

Now before I continue, it would be worthwhile to point out that the results of my department and the faculty I managed were fairly good. But I would argue that for the amount of time I and my team were investing in the above approach, they weren’t good enough. We might as well have been doing our jobs with one hand tied behind our back, as without being aware of it we continued to face issues of our own making. 

So what changed? Well aside from the fact I, rightly lost my obsession for headline figures in favour of teaching the subject I love, I was introduced to retrieval practice. I owe a lot to @GeographyJake here. He introduced me to and is arguably the catalyst behind my fascination into research-informed practice. This post will not cover anything that more than likely hasn’t been covered previously, given the plethora of books, websites and blogs that have already been written about retrieval practice. But what I hope to do is provide a different lens in which to view retrieval practice and how to get the most out of its application.

When delivering a CPD session earlier in the year I used the analogy of a computer and its ram and internal hard drive. I asked my colleagues “Imagine using a computer, you have 10+ applications and windows open, what would happen?” I got the response I was looking for “the computer would slow down” and this is because it is overloaded and cannot process all of the desired information at once. This I liken to our working memory (WM) and I describe this as being finite, meaning our brains can only process so much new information at any one time. However, our computer has another component for storing data, such as documents, photos and music, and we call this a hard drive. Our brains have their very own hard drive, called long term memory (LTM).

So our brains have computer ram called the working memory (WM) and a hard drive called long term memory (LTM). Retrieval practice aims to strengthen the connection between these so that rich schemas can be established, linking new learning to prior knowledge. This will strengthen our brain’s ability to recall knowledge on-demand from the LTM, freeing up our WM to be utilised for more complex tasks, in which the knowledge can be more effectively used. Retrieval practice is supported by extensive research into cognitive science, and therefore we can have a high degree of confidence that retrieval practice does what it says on the tin, and is not some unsupported hypothesis, such as ‘we all have desired learning styles’.

Getting the Most Out of Retrieval Practice

Best examples will activate prior knowledge to be built on in the current lesson, creating authentic and organic links. Therefore, when curriculum mapping or producing SOL Tom Sherrington advises you consider what prior knowledge is required and also how you will check the pupils have this required knowledge. Tom Sherrington speaks of how knowledge builds on knowledge (schema building), and a gap in an individual’s schema can prevent them from learning new material. An example I can offer here is I am currently planning a new SOL, using India as an example of a Newly Emerging Economy, with a focus on rapid economic growth. This presents me with the opportunity to revisit knowledge on the fabled global atmospheric circulation model, plate tectonics concerning the formation of the Himalayas, rivers and resource management when thinking about how the 3 main rivers of India are essential for farming and industrial growth. When checking what knowledge, the pupils can recall, it allows me to see where the gaps are and reteach areas and or address misconceptions.

For retrieval tasks to be effective they must test the knowledge of all pupils and be generative, meaning pupils have to generate their ideas from memory and not be allowed to look through their exercise books. A case could be argued that when first implementing retrieval practice you could use visual stimuli, but this is different to pupils looking for the answers in their books. Shimamura talks about how we strengthen our memories when we generate information from our memory and use our own words. His research revealed that if we tell someone what we’ve learned we can improve our memory by 30 – 50%.

It is also important to foster a culture whereby pupils are not afraid to get things wrong. Retrieval practice should be done to find out who got answers wrong not who got them right and then taking the necessary steps to address the gaps and/or misconceptions in knowledge.

Kate Jones’ book Retrieval Practice: Research & Resources for every classroom offers a wealth of resources, which are free of context, making them useable across all key stages and subjects. She also outlines the research and science behind retrieval practice, in a much better way than my computer analogy, described earlier. An easy way to check the knowledge of all pupils is through low stakes quizzing. Kate Jones, along with Mark Enser and Tom Sherrington speak of how quizzing can be used to strengthen the connection between the WM and LTM.

In Mark Enser’s book, Making Every Geography Lesson Count, he offers the following as method for quizzing; 10 questions, some taken from the previous lesson, some from the last term, some from further back. Quizzes should be generative and low stakes, meaning pupils should not be afraid to get questions wrong and should not be looking for the answers in their books. I always begin my quizzes by reminding pupils it is not about getting 10/10, but about the practice of delving deep into the long term memory, identifying what we don’t know, so that we can do something about it. This has enabled me to create a culture and environment whereby all pupils can benefit from retrieval practice.

As well as being low stakes, quizzes should be self-assessed, providing instant feedback. For this to be achieved you must have some sort of resource for which pupils can check their answers against. Whilst quite polarising as a resource, I like the use of a knowledge organiser for this. A knowledge organiser used effectively can be a prompt for a much richer narrative and schema and they set out all of the knowledge pupils must know, meaning pupils never feel like you are trying to catch them out.

There should be zero reluctance to quiz pupils and to do so regularly. Whilst a quiz, might lend itself more to closed or shorter responses, it is important to remember that if pupils can retrieve knowledge with a high degree of fluency, their WM is free to be used for more complex tasks. Quizzing can also be used to promote self-regulatory learners. With a knowledge organiser or alternative resource at the ready, it would be easy to foresee a scenario whereby pupils are responsible for quizzing each other, whilst the teacher navigates the room listening to pupil responses.

There is more to retrieval practice than simply quizzing and in his CPD sessions, Tom Sherrington suggests it is important to vary the diet, in terms of the activities being used, to avoid the loss of student agency. He states that as well as the initial encounter of new knowledge, pupils need to experience it in different ways to generate their ideas of it. There are a variety of activities that have been produced/shared by brighter minds than my own, and rather than going into great detail about them here I think it would be more favourable for you, the reader, to signpost where to find these.

Blake Harvard’s superb blog, the Effortful Educator has some great resources and research into multiple-choice questioning, and a particular favourite of mine, the ‘memory dump’.

Kate Jones’ book is a must for the research and then context-free resources. Katy also talks about how parental engagement can be used to strengthen the impact of retrieval practice.

Tricia Taylor’s book, Connecting the Dots is another great resource for retrieval practice. Tricia suggests how best to model retrieval practice, and how to train pupils to get the most out of creating flashcards.

The Learning Scientists website presents a lot of research on how learning happens. Their research is very accessible and supported by resources.

Adam Boxer’s superb retrieval roulette excel documents in science are a great way to ensure regular and structured retrieval practice. Adam’s blog cited below discusses how he implements this and his thinking behind it.

There are very few negatives in the use of retrieval practice, but it is important to consider these. The first is, done badly it can fortify misconceptions and errors. Think about a swimmer, who practices daily, but picks up bad habits and demonstrates a poor technique, their coach will spend a lot of time trying to address this and improving their form. Errors and misconception that are left unaddressed can begin to cement further in the mind of a pupil. The second is if retrieval practice is done sporadically and isn’t well thought out. A quiz to test knowledge is far more impactful if what is being asked is in somehow connected to the intended learning of that lesson and the links are made explicit. Where teachers can get it wrong is by simply shoehorning a quiz into the start of a lesson as they realise they have not done one for a while or haven’t asked questions on a particular topic for a while.

Final Thoughts

Our WM is finite and our ability to retrieve knowledge for its application is something we need to practice. When at its best retrieval practice can help to establish a rich schema, connected by well thought out and authentic links. To get the most from retrieval practice pupils must be forced to generate their ideas by completing retrieval tasks without looking for the answers in their books, they should also be responsible for assessing their work against quality resources. Finally, variety is important so that student agency is not lost and pupils can encounter the same knowledge through different experiences, fortifying authentic links.


Boxer, A. (2018) Retrieval Roulettes! (Blog) Available at:

Harvard, B. (2020) The Effortful Educator. (Blog) Available at:

Jones, K. (2019) Retrieval Practice: Research & Resources for every classroom. Melton: John Catt Educational Ltd.

Learning Scientists. (2020) Retrieval Practice Resources. (Website) Available at

Sherington, T. (2020) Rosenshine Masterclass V Stages of Practice and Conclusion. (YouTube Video) Available at:

Taylor, T. (2019). Connecting the Dots. The Collective Power of Relationships, Memory and Mindset in the Classroom. Melton: John Catt Educational Ltd.

4 thoughts on “Retrieval Practice and the Art of Schema Building

  1. Interesting analogy and so well structured. I believe if alongside this we gave more opportunity for really purposeful application being developed, including teaching others but for a reason beyond the test, we would get deeper understanding and effortless recall. It’s in our muscles.


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 )

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s

Create your website with
Get started
%d bloggers like this: