“Sir, have we done this before!”

“I have no idea what Newton’s second law is.”

“What is the function of the stigma?”

“What does ‘s’ stand for in that equation?”

Any of these sound familiar? Well, it’s common, no, it’s likely that this has happened to you in a recent lesson, perhaps even your last lesson. So what’s the all too common problem? Retention, retrieval and spaced study techniques. All of which need improving so that students can take ownership of their learning and improve, thereby reducing the load on themselves during lessons when asked to recall information.

Students are busy, their brains are overloaded, there are a million and one distractions in their lives so why would they remember what we taught them a day or two ago. Perhaps the statements are different but the sentiment is the same. The daily frustration for educators is real!

We teach material and a few lessons, perhaps even one (we’ve all been there), later it’s gone and we find ourselves re-teaching previously visited content which makes time even more of a premium when teaching exam groups. What is the direct influence of this and how much time do we waste weekly, monthly or yearly as a result? The frustration is real and it serves as a good a reason as any to pursue this as an action-research project that puts Ebbinghaus to the test.

Taking CPD into your own hands is powerful, there are more and more routes that enable staff to do this now too compared to ten years ago when most CPD was done in house and just occasionally you were allowed out the building to go on a course. I believe this is why so many educators are adamant that Twitter is such a valuable tool to have at our disposal, at the tap of a button you can access experience/evidence/researchers/experts plus share ideas/resources and evidence thanks to the ability to tap into a greater breadth of fellow educators than our school staff rooms can afford us. Collective knowledge and experience further enable us to develop and move forwards our own practice while in turn benefitting our charges.

Each year my school runs a CSI, a Collaborative School-Based Inquiry. Teachers are given the chance to run action-research projects, in an area of their choice, over the course of four months and then share their findings in a learning symposium at the end of February. This is a powerful tool that enables staff to “find their tribe” (Godin) and work on a specific area of interest, road-test strategies for the future or delve deeper into research to improve an aspect of education whether it be as an individual, department or across the school. To give you an idea of the vast array of sessions from this year, 26 in total, check out the image below.

This year was particularly special as the school was visited by the Director of Education for Nord Anglia, Andy Puttock, along with an excellent keynote on the importance of ‘Power and Love in Education’ from Steve Munby.

Before I briefly reflect and share my teams project, I’d like to take the chance to thank Chris, Alan, Rana, Saleem and Darren for their efforts over the course of the project. I think we did excellent work and shared some great findings with the delegates at our sessions and I’m sure our work will not only continue in our STEAM subjects but also be evident across more departments in the school.

So we wanted to test out some evidenced strategies and research to the test, specifically how we could further implement strategies suggested by The Learning Scientists (see their website here) across our subjects (Science, Maths and Computer Science). How could we use retrieval and spaced practice to our advantage to ensure that students could recall information when called upon while also developing their independent study skills. This led us to our research question; “Do retrieval and spaced practice really improve memory like Ebbinghaus suggests?”


Ebbinghaus’s forgetting curve has been somewhat of a hot topic in recent years across education, and for good reason, as it has tacit links to much of the work around self-regulation and metacognition. If we can teach students the skills to monitor and evaluate their own learning, while also using strategies in the classroom to aid learning, then we should be able to improve their retention curve and avoid those typical questions at the start of lessons.

We chose to focus on retrieval and spaced practice as there is an obvious link between them, plus we knew our end of term 1 and mid-year exams would serve as a good data point for analysis. We had also discussed them frequently since September and they formed a part of my previous T&L newsletters in issues 1 and 2. It is worth noting here that while the project ran for four months, we agreed that we should be adopting this approach continuously even after the project ended. We need to build habits in our students, and ourselves, if there is to be a lasting impact as a result of our findings, plus we want the strategies to become part of their ongoing self-evaluation toolkit.

The research began by looking at memory, how Ebbinghaus found what he found and the strategies that we could deploy to enable our students to improve their retention rate. Thanks to the work of some outstanding educators our work was made much simpler as they’ve written some poignant material on the topic. Willingham said that “memory is the residue of thought” and “understanding is memory in disguise.” While Kirschner says that “Learning, in turn, is defined as a change in long-term memory.” Both point towards there having been some acquisition of knowledge and skill, whilst then being able to express what it was that has been gained by the learner.

Eysenck (1994) stated that “Where the information being learned has a framework or structure that can be used to organise both the learning and the retrieval then the memory is often considerably improved.” This got us thinking about the project as a whole and brought us to the conclusion that we needed to teach the students both how to organise their revision, structure how we teach topics so that it reduces the load on students and make links between concepts across curricula whilst using a few key strategies to aid them to ensure we were not cognitively overloading the students.

We chose to use a few simple strategies such as brain dumps, low-stakes quizzes and mind mapping. Primarily we decided on these strategies as the required low workload on the teacher end, after all the students should be the ones working hardest and they formed a part of what many of us were already doing within lesson with exam year groups. These strategies work as they afford students the chance to encode the information from their short-term memory into their long-term memory, enabling greater access to the memory store due to the repetitive recalling of information. Coupled with this the strategies also enabled us to give immediate feedback to students, this was either as a whole class or as individuals depending upon the task.

We are fortunate enough to be a technology-rich school, with a bank of iPads for each department, so naturally, we used this to our advantage thanks to tools such as Quizizz, Quizlet Live and Mentimeter. These tools allowed low-stakes testing and the opportunity for students to retrieve and elaborate on their learning. Each week students were given a task using these tools so that they could retrieve prior learning, coupled with this we used low-tech strategies such as structure strips and brain dumps. There was an emphasis here that this shouldn’t be onerous on staff or increase workload, thus we worked on these live in class and feedback to students immediately as this is proven to have the greatest effect.

Thanks to the #edtech tools mentioned above this was easy to achieve as the results are instantly generated and feedback given to the students, the low-tech options were student-led in class with the final answers also being shared with students on their classes Microsoft Teams for future reference during their own independent study. We also started to develop retrieval practice challenge grids, linked to specification points, which use as homework tasks and then mark in class thus further reducing workload. These resources will also prove useful later in the course as a revision resource once the course has been covered, you can see an example of one here.

We specifically liked the approach of Impact Wales towards retrieval practice, see their sketchnote here, as their suggested brain dump activity requires no preparation on the teacher end while putting the students’ retrieval of information to the test. The Mathematics department also uses tarsia puzzles extensively, either at the start or end of lessons, these proved to be an excellent tool for retrieval of knowledge when problem-solving.

The delegates that came to our session were not passive participants, we ran a trial on them too! We split them into groups, A and B, and through the course of the session, we gave them a task to remember as they would need to retrieve it at the end of the session. Group A was given two minutes to read a set of instructions as to how to build a simple motorized Lego car, they had to retain this information until the end of the session. Group B was given four sets of 30 seconds, spaced throughout the session. This activity mirrored what we did with our students and highlighted that it was much more challenging for group A to successfully build the cars compared to group B.

The build time at the end proved this pretty clearly, group A found it much more challenging having been overwhelmed with information for over forty minutes since they saw the instructions whereas group B could recall the instructions more clearly as the chance of it being stored in their long-term memory was greater having had repetitive exposure to the information. Over the two sessions we ran, there were six group A’s and six group B’s; no group A managed to build the car whereas four group B’s successfully built a working model car.

To ensure we checked our strategy from differing angles we applied our project to the following year groups:

Year 10 – two similar attainment groups based on average CAT4 target grades, 1 group we did not use any of the mentioned strategies with while the second group we used a mixture of the strategies mentioned over a 10 week period. We chose this as they are both large mixed attainment sets.

Year 9 – set one (highest attainting CAT4 score by two whole grades compared to set two) we used the strategies mentioned for 10 weeks, set 2 we used none of the strategies. We decided to do this, rather than see if set two could ‘catch-up’ set one, to see if it would further increase the attainment gap.

Year 7 – we followed one mixed attainment class and used the strategies mentioned from week 5 of the trial onwards, to see if it would have a positive effect on their mid-year exam results. By this point, they had completed two assessments and we had a large amount of material to review in the build-up to the mid-year exam whilst studying topic three of the year.

You can see the attainment results for each of the groups in the presentation slides (see below), they are colour coded based on if the student is below (red), on (amber) or above (green) their target grade. Our main findings were that there was a positive impact for each group as a whole who we used the strategies with however it didn’t work in every individual case. The over-arching viewpoint was that if we were to repeat the trial we would do it with a larger effect size over a longer duration; an entire year group and over the course of half an academic year or more. Initially, we would teach the students the strategies we intended to use over a course of a few weeks and then run the trial with that cohort over a whole year to see if the difference is more or less significant compared to a cohort that we did not use the strategies with. Yes, there is an ethical issue here in the sense that we should be giving every student the chance to use these strategies in order to improve their learning capacity, and we would look to do just that next time around.

As our assessments are standardized, we can compare year group attainment as students transition through each of the coming years and we will likely see a positive long term impact as a result of embedding these strategies even further. This will likely be a continuing departmental project, and future blog post, in the coming year to see how much of an improvement can be made to the forgetting curve of our students now that we are more research-informed as a team. We also plan to incorporate knowledge organisers towards the end of this year and into next year, along with spending this summer’s department time looking at our curriculum offering at key stage 3 to allow for greater retrieval and spaced practice while further developing skills and knowledge in our students.

You can see the presentation slides here:

Not all of our references were used in this blog post or indeed referred to in the presentation slides however you can see the full list below.
Reference List


  • Understanding How We Learn  – Learning Scientists
  • Why don’t students like school?  – Willingham
  • Change in Long Term Memory – Kirschner
  • Applying Cognitive Science to Education  – Reif

Blogs, Websites & Articles:

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.