Ten years of Test-enhanced learning #retrieval practice

More than ten years

Back in 2011, I encountered the research of Dr Jeff Karpicke via a Purdue University press release. At the time I added a note to the blog post.

“One to follow up on. Testing promotes learning. If the practice of testing promotes learning, how can we make testing part of learning?”

I did go on to follow up. I casually read research papers that explored the impact of retrieval practice on learning, spacing, interleaving and to a lesser extent test-enhanced learning. 

3 Years

After returning from teaching overseas in 2019, I found myself back in the classroom full-time teaching English. This opportunity reignited my interest in testing as a “part of learning.”

Back in 2011, I encountered the research of Dr Jeff Karpicke via a Purdue University press release.

8 months

I’ve spent 8 months writing and reflecting upon all that I had read, all that I have discussed with researchers and fellow test-enhanced learning practitioners and road tested at the chalkface of teaching.

For what it is worth…

How has my teaching pedagogy evolved? How has my teaching pedagogy evolved? How has my understanding of learning developed? 

With a deeper understanding of cognitive architecture, memory and remembering, I am fastidious about pupils being attentive and being precise about what they attend to. That includes pupils paying attention to one another.

“Next, I am going to ask an important question. [Pause]. Following that, I am going to select three pupils for their considered response. The fourth pupil needs to be ready with their opinion on which pupil’s answer was best – and why.”

As I have developed a deeper understanding of learning/encoding, retrieval and successive relearning, I am more deliberate in what I select to teach. I now teach less and I actually expect pupils to learn and remember less. When introducing new knowledge, I teach even less, at a time, than I once did. And as I told the SecEd Podcast on memory (June 2022), I never expect to teach something just once.

I see test-enhanced learning as teaching. I purposefully quiz (self-marked) every lesson. Not solely for teaching and relearning, but to establish robust routines that demand every pupil’s full attention and I often find myself reteaching what I previously taught following a quiz.

I quiz every lesson, almost always at the start, impromply, and quite frequently to close the lesson.

“Consistent routines, build retrieval opportunities into every lesson (not just at the start).”

Kirby Dowler

“…lesson routines to help make the habits stick.”

Helen Webb

I test (peer or teacher marked) every week and test every teaching cycle. I now recognise and use the spacing opportunities afforded to me by my teaching timetable, to promote spaced retrieval practice. That includes using personalised spaced retrieval as homework. I recognise the value of sleep. I see teaching as much more of a process and much less of an event.

I aim to use personalised spaced retrieval to promote acquisition of key knowledge ahead of the learning. For example, learning character names or contextual information about the text we are about to study. I will never forget that first experience of teaching Othello.

I also sometimes utilise testing for learning, particularly pretesting to spark curiosity and interest in a topic, as well as priming future learning moments.

I know that relearning is super efficient. Investing in testing for learning, for encoding, through retrieval practice at the start of the unit can be pedestrian. But I know I need to persist. Relearning rapidly accelerates learning and pupil success has a significant impact on pupils’ motivation. Using self-assessment in class empowers pupils and offers significant metacognitive benefits, too.

Promoting and forewarning pupils about upcoming “end of unit” or “end of year” tests provides a very valid reason for pupils to direct their attention to your teaching in the first place. And then I build long-term, medium-term, short-term retrieval opportunities into the teaching.

Taking the time to explain to your pupils why you are investing in test-enhanced learning is crucial. That said, I know that allowing pupils to fail the first test is a valid strategy as it shows up pupils’ ineffective revision strategies and then allows us a more powerful opportunity to convince them that there is a better way…. It is a strategy recommended by Dr John Dunlosky, although it may be more appropriate for older pupils.

And finally, as teachers, you are required to explicitly teach these strategies to your pupils and “model them for your subject in your class”. Remembering the majority of pupils will loathe to take their “cough syrup” or eat their  spinach without good reason.

As for the actual use of test-enhanced learning and retrieval practice, I am more aware of the different modes of recall, the power of cues and hints to underpin pupils’ success and how to manipulate the level of difficulty of retrieval prompts. Item difficulty is a factor. Difficult-to-learn information should be reviewed at shorter time intervals.

I have learnt that the shallow question or flashcards banks with high repetition and high success rates in the early stages of learning build confidence and motivation that pays a weighty dividend in the long run.

When pupils acquire and retain knowledge, the way they interact and think about learning evolves and the climate in class changes. So when new knowledge surfaces in lessons, notice it. Celebrate it.

One notable change is how I employ homework – moving from general learning tasks and activities to personalised spaced retrieval practice flashcards. Homework commitment is measured in time invested. Every minute counts. The personalisation of retrieval practice and the efficiency of relearning, its impact on motivation, is tangible. 

“Without question, the most efficient schedule is a personalised one, accounting for the learner’s rates of forgetting and prior knowledge.” (Latimier et al, 2021)

“Retrieval” is misleading. It implies a terminal strategy. Pre-test, testing during, post-test. Testing is, and informs, learning.