ResearchEd Rugby 2019 – Memory #rEDRugby

Following my presentation at ResearchEd Rugby today I promised that I would get my slides online (see above download) and blog about the session. The slides won’t make a huge amount of sense without my notes, so here goes. I’ve tried to copy my notes across and make them make sense, so I hope it’s all ok. If you were at the session, there might actually be some extra content that I didn’t remember to say.

The session was based around memory and how we can use the findings of cognitive science research to help our students remember everything they need, in particular for the GCSE exam but these principles can of course be applied to all stages.

The aims of this session were to 

  • Understand the key learning strategies recommended by cognitive science.
  • Consider how we are already applying these in the classroom.
  • Discuss ways in which we can clearly plan lessons and lesson sequences with these in mind.

The key question that I kept referring back to over the course of the session was this:

Are our students having to think hard enough to get the learning to stick?

After all, memory is the residue of thought (Willingham, 2010).

I suggested that some of these quotes might be familiar:

  • ‘you should remember this, you learned it in year 7’
  • ‘We spent a whole lesson on this last week’
  • ‘Why don’t they remember anything I teach them?!’

It’s easy for us to assume this it’s just because we taught it badly or that they didn’t work hard enough, but that’s just not the case.

I then showed a demonstration of The Forgetting Curve by Ebbinghaus, I’m sure some of you have seen this before. It shows, depressingly for us, quite how quickly students forget newly learned information. There is a slight warning that comes with this, Ebbinghaus actually developed this curve by teaching himself nonsense syllables but it has been replicated and there are many similar experiments using things like English-Swahili pairs which come up with broadly similar results. 

Of course the nice graph is all well and good, but of course we are faced with a few challenges.

  • Timetabling often works against us, take for instance my year 10 that I see Weds, Thurs, Fri and then not again until the following Wednesday. Or the year 11 I once saw period 5 on Monday and period 1 on the following day. 
  • The amount of ‘content to be covered’. Depending on your schemes of work and your interpretation of what ‘covering the content’ actually means in the context of GCSE MFL this can definitely pose a challenge.
  • The vocabulary and grammar knowledge that is needed for GCSE, especially for getting the highest grades.
  • Homework policies and how to make them work well for MFL.
  • And of course the ever present perception that a GCSE in a language is a hard GCSE to do well in.

The issue is so many students believe that re-reading and highlighting is effective learning, think of all the beautiful revision notes that some students sit and read over and over. We’ve all had the experience of feeling like something we have done has worked because we feel more confident about it afterwards, and of course if it’s just before a test it probably paid off. But can we still remember it days, weeks or months later? Almost certainly not.

Our job, therefore, is to show students what is effective for long term learning, both through integrating it into our teaching and the homework that we set but also through the recommendations we make for their own revision. I think sometimes it can be too easy for us to ask students to ‘revise’ and expect them to know how do it well. I don’t know about you, but I’m often disappointed after an assessment when I ask students how they prepared. The answer is too often that they just read through their book, or if I’m lucky they did a couple of match-up activities on quizlet. That is of course if they did anything at all…

The extract on slide 9 comes from an article about desirable difficulties.

What strikes us as teachers is that the rapid progress we might want to see in a lesson, is not actually evidence of learning at all but an indication of student’s confidence in that moment, their familiarity with the topic having studied it for the whole lesson. The chance that that apparent learning is going to stick for long after the end of the lesson however,  is small. We have to do something that might go against what we naturally want to do and make things more difficult. 

So, our next issue is, how we can integrate these so-called desirable difficulties into our MFL teaching?

This is where the Learning Scientists come in. They are a group of academics who focus on how the findings of cognitive science can help us to teach and students to learn better. I went on a two day course with them back in January and I loved it. A good deal of what was discussed in the session has come from what I learned in those two days. We focused on the areas that I have found most useful in MFL myself (retrieval practice, interleaving and spaced practice), but you may want to take a deeper look at the others yourselves. There is loads on their website from blogs to videos and resources for both students and teachers. Just google The Learning Scientists.

According to researchers when information comes to mind easily and feels fluent, it’s easy to forget. In other words, just because we learn something quickly and easily does not guarantee we’ll remember it.

Retrieval practice is about bringing information to mind, and in the process helping to strengthen the relevant memories. For instance, attempting to recall a word or verb ending will be much more effective for helping long term memory than just looking back in an exercise book to find it. 

One form of retrieval practice that some students already use is flashcards. Flashcards can be used effectively when students force themselves to write out the answer or say it aloud, as opposed to what they frequently do – look at the card, decide they know the answer and flip it over straight away. Often we think we know something, but then struggle to precisely recall the answer, it is this struggle process that helps our learning and makes the answer come more automatically next time. This is just one of the so-called ‘desirable difficulties’ than can aid learning.

Retrieval practice can also be known as the testing effect, but these tests must be no/low stakes in order to have the desired effect.

The activities on this slide are all examples of how we can use retrieval practice in our lessons. They’re all designed as starter activities and can be downloaded from my blog. I can’t take all the credit, they came from various people who are all credited on the blog.

The key thing that all of these slides have in common is that they are requiring to think about their learning, retrieve previous learning and in some cases integrate it with newer learning.

There are many tech tools around that can help with retrieval practice, all of which are good for different purposes. Unfortunately by some these can be seen to be a gimik, but they can actually be great learning tools if used well.

For me, the key to good use of these apps is to know what you are trying to achieve. 

If you want your students to have repeated exposure to certain vocabulary and grammatical structures and get used to recognising them and understanding them, then quizlet live will be ideal. However, if you want students to get better at producing the language then this probably isn’t going to be the best way forward. Individual use of the ‘write’ and ‘spell’ modes would be time better spent. Use of these tools has to be done for the right reasons, at the right time. Without this level of thought they quickly become a fun time-filler but not much more than that.

Secondly, question construction is very important. Apps like Gimkit, Quizizz, Plickers and Kahoot all allow you to write the questions yourself. Whilst this will take longer, it really allows you to construct good questions that require students to think about the structures you want them to practice. For a multiple question to be worthwhile then all the options must be plausible.

That’s why, when practising adjective agreement the pink box is a more effective question than the black, but when recalling vocabulary for colours and clothing the black is more effective. Unfortunately when we leave Quizlet to write the questions for us, and my extension Gimkit when it creates using a Quizlet set, it doesn’t have the capacity to do this thinking, and so in certain situations is not as effective as the student engagement and high scores might suggest.  In the same way, students who are spending ages on the match mode on quizlet might get good scores, but have they really achieved what you wanted them to?

Another great thing about these apps is that because of the competition elemenet students don’t seem to realise how many times I am getting them to repeat the same knowledge over and over again. In fact, they’re quite pleased when I ask them to do a gimkit, quizziz or kahoot for homework!

Spacing is another important requirement for helping learning stick in long term memory. This can be applied to our lessons, lesson sequences, homeworks and the way our students choose to revise. Spacing practice over time is really important as it gives time for forgetting and therefore the opportunity to try to recall the knowledge and strengthen the memory of it for later on. 

Like I mentioned earlier, cramming loads just before a test may pay off at that moment, but is unlikely to lead to any long term learning. Besides which, students simply aren’t going to be able to cram the entire GCSE course in the day before the exam.

The really long study session is likely to make that student feel very confident at the end, but they’ll still forget what they have apparently learned just as quickly. If they’d spread this learning over time they’d be much more likely to remember it. That’s because each time they returned to studying that topic they would need to retrieve what they learned the last time. To begin with they may well struggle with this, but as time goes on this will get easier and easier for them. 

This can also be used as a way of approaching curriculum planning. There is a danger that in schools we get into the habit of teach the module, assess the module, move on. Come the end of the year, students can’t remember what they’ve done before, because they haven’t needed it since.

Luckily as learning a language is like building blocks, we tend to rely on knowledge of the basics all the time, so knowledge should naturally get re-visited. But it’s good to be aware of it and try to plan to bring back knowledge that we might not naturally have re-used otherwise.

For instance this summer I devised a revision plan to follow with my year 11s. This plan spaced all the modules from the textbook over different lessons and combined them with different elements of the exam that needed practicing. This helped me to structure my lessons a little as well as making sure that everything got covered.

I’m certainly not suggesting this is a perfect model, it was my first attempt. But it definitely gave a bit of structure to my thinking.

As part of the Chartered Teacher assessment I had to undertake a research project. This is the poster I produced at the end of the project. I’d been focusing on implementing retrieval practice and spacing to help support GCSE students. 

What I found didn’t really surprise me, it seemed that for the intervention to work students needed to have a positive attitude towards Spanish and have a positive attitude towards what I was asking them to do, presumably because those that saw it negatively were not actually putting in the effort needed. I found the research project really interesting and it certainly gave me more to think about in terms of exactly what I was asking students to revise, how, when and how I was testing this. From what I have seen there is definitely potential for making these principles work well with MFL learning with the right thought behind it.

The conclusion I reached is that this has promise, but that the real driver behind the success of such methods comes back to student attitudes towards the subject.

I thought the best way of explaining interleaving was to leave it to The Learning Scientists.  

Essentially interleaving is switching between topics in one session, going over ideas in a different order in each session and making links and contrasts between the ideas.

Of course, this has to be a balance as switching too often won’t help. Similarly, switching between unrelated topics won’t have the same positive effect.

As with all desirable difficulties, this is going to feel more difficult for the students than it would if they just studied the same thing in one block. But the interleaving should be much more effective.

For example, in MFL rather then teaching tenses individually, we would teach one tense, teach a similar tense and then practice both together. The important thing about interleaving is that students can make comparisons between the two areas to be learned. They can make links in their knowledge and see where the similarities and differences are. This can really help them to strengthen their long term memory when it comes to both of the concepts. 

This poster sumarises Rosenshine’s Principles of Instruction, a highly regarded paper on classroom practice based on research into how the brain acquires new information, research into effective classroom practice and findings from studies that taught learning strategies to students. Recently I’ve been working on some writing for the Chartered College that links Rosenshine’s principles to MFL teaching for their Accelerate course for early career teachers. I’ve found this process really useful for clarifying my own understanding of the principles and how to apply them and I thought they would be useful to share during this session. Some of the resources I have included here were found on Twitter and included in the writing I am doing with permission from the creators.

These two principles link back to the retrieval practice that we discussed earlier. Taking Daniel Willingham’s assertion that ‘memory is the residue of thought’ we know that in order for learning to take place students must have to think about what it is that you want them to learn. Therefore, by beginning the lesson with a short session of reviewing previous knowledge you are helping your students to create and strengthen memories. In addition to the types of retrieval task discussed earlier you could also try other tasks such as a Brain dump . Ask students to note down everything they can remember about a certain topic, maybe a particular area of grammar. You can do this completely unstructured or provide prompts to help students remember what you need them to. The important thing about this is that they are doing it without referring to their notes. Looking back in their books and just reading what they previously wrote will have a much much smaller impact on their learning, if any (as discussed in relation to the Dunloskey paper earlier). It is important that you don’t only review the knowledge needed for a particular lesson, but that you keep all knowledge in review regularly over the weeks and months.

The second of Rosenshine’s principles is to present new material in small steps. We know this as language teachers. You wouldn’t just present students with a comparative and expect them to be able to copy it. First they need to know how to give their opinions and have an understanding of adjective agreement. Once those building blocks are in place students are able to build the new knowledge on top in order to form a comparative. More often than not this is something that we just do naturally, but it’s important to remember this, especially when we’re short of time and might optimistically hope that students will just remember what they need to know, let’s face it, they probably won’t.

Principles 3, 6 and 7 are all closely linked, asking questions, checking student understanding and obtaining a high success rate. In order for learning to become long term students need to have been asked lots of questions, expected to explain their understanding and have practiced successfully over and over again. Again, this is something we do a lot of anyway, but the paper backs this up. In fact, Rosenshine’s research showed that the most successful teachers spent about half of their lesson time doing this. It’s something we shouldn’t feel bad about spending lots of time on, it’s really important. Again, the tech tools we talked about earlier can come in really useful here. Each have different uses but Quizlet live and Gimkit in particular will really make students answer a large amount of questions without even realizing it. The great thing is that students will happily play again and again with the same study sets and not even notice, in a way that you probably couldn’t get away with by using the same work sheet three times in a row. Recently I have used the same Quizizz set three times in one lesson, praising students for staying in the green the longest, encouraging quiet thinking and real focus in order to get the answers right or beat their previous score. Students were engaged in the activity and there was no doubt that they were really thinking through the process of making adjectives agree. This led to a fairly high success rate, at which point I knew we would be ready to move on next lesson. I’ve also set this class the same Quizizz set for homework to encourage a little spaced practice before our next lesson.

Giving students models to work with is also very important. It’s something we tend to do through reading exercises or when we are asking students to write something for themselves. Again, this is something we should never feel guilty about spending time on. It may feel like time would be better spent just letting the students get on with it themselves, but that’s not the case. You’re the expert in the room and shouldn’t be afraid to show that. More and more people now are going down the route of providing model parallel texts which include the features that they want students to be using themselves. Learning from these models and help students to understand the structures and, when used well, then manipulate the language themselves. 

Guiding student practice is of course really important. We all know that we can’t just teach students a structure and expect them to suddenly be able to use it. Guiding their practice from very structured activities to ones where they are eventually working independently is key. The examples I have used here were part of a booklet made by Miss Newnham on Twitter following activities such as sentence builders advocated by Gianfranco Conti. These are great examples of guided practice, from sentences that only use the builder to a paragraph integrating other knowledge and finally a written exercise using a structure strip. Of course, the final aim would be for the students not to need the structure strip at all. There are many ways of guiding how students practice, prior to this you might have some gap fill activities, some listening or narrow reading to expose students to the language they are going to need to complete the activities. 

The importance of scaffolding should never be underestimated. Just because students know how to talk about where they live, give opinions and past/future holidays across a series of lessons, we can’t assume that they’ll automatically be able to tackle the 90 word GCSE task.  This scaffold was created by @Senorcordero on twitter who kindly shared it and allowed it to be translated into French. 

By breaking the task down into translating the bullet points, changing key words such as verbs and possessives, retrieval of verb endings and finally retrieval of key vocabulary students are fully prepared to tackle the task. 

We can do this with anything, for example using Gianfranco Conti’s micro-listenings or narrow readings before tackling longer texts. This will of course take longer than going straight to the text that you want them to understand, but it will pay off in their understanding of the text. 

Visualisers are great for this. Recently my year 10 did particularly badly on a translation assessment, so I used the visualizer to highlight and annotate the text with them, show my thinking and look at the more difficult elements before students did the translation themselves again. All I needed was the visualizer and the worksheet with the translation. The rest was just my thinking aloud. Again, we shouldn’t be afraid of taking the time to show students things like this and encouraging them to replicate it in their own work. 

Of course, when the time is right students do need their own independent practice. Some quiet time to produce the best language that they can. This could be as long or as short as you want it to be, depending on what students have been preparing. This can be something some teachers shy away from, worried it looks like they’re not working and that it’s not time for the end of term writing assessment. 

This slide includes some related reading from names that I have mentioned during this session for anyone that is interested in reading a little further. None of it is MFL specific, but you will be able to see how I have applied the reading to what I have said today.

So, my closing message from this session is not to worry about spending too long doing things that have actually been shown to help learning. Lots of questions, closely guided practice, real thinking and lots of modelling and reviewing. I find it such an interesting area, and I’ll definitely be looking into it more in the future.

Leave a Reply