Mock exams: That time of year when you question everything you have ever said or done in a classroom. You pull your hair out over the quality of the scripts you dutifully read through and mark X’s with your red pen as you spot numerous factual errors. Why are their answers so devoid of specific information? Why do they not remember the order of things? Why didn’t they revise properly?! Well, after burying my head in numerous articles and blogs, I have come to the conclusion that it might actually be partly my fault. The way my curriculum is organised and how I deliver my lessons doesn’t take enough account of how memory works. Fortunately, there is plenty of time to put this right before the real exams in May and June.
The following blog is inspired mostly by an article by Daniel Willingham, Professor of Psychology at the University of Virginia, which you can read in full here. It is also informed by the work of Robert Bjork, Professor of Psychology at University of California, which can be viewed here. Other blogs related to memory and cognitive science that I found useful include David Fawcett’s, David Didau’s, Joe Kirby’s and Kris Boulton’s
Students (and probably most teachers) know little about how their memory works
The implications of this are huge: If we don’t know how our memory works then we will not know how to study effectively. I have not talked about the workings of memory in much depth in my lessons and I am pretty certain they haven’t covered it in other classes, so where are we expecting the students to pick this up from? Having said that, I do think it is unfair to blame teachers completely for this, as we were (most probably) not taught this either at school, university or teaching college, and it certainly hasn’t been a focus of any INSET I have sat through. What about you? It’s time to take this more seriously.
Robert Bjork here talks about good memory in terms of ‘high storage strength’ and high retrieval strength’. Cramming for a mock exam may give the information students are trying to recall ‘high retrieval strength’, however, it will have ‘low storage strength’, and will be forgotten by the time of the real exam. It is also unlikely to give students the depth of understanding to write high quality essays or complete the more complex tasks in the exam. If we study more effectively and commit it to long-term memory then it will have ‘high storage strength’.
Joe Kirby in his blog entitled ‘Why don’t students remember what they’ve learned?’ sets out Bjork’s ideas like this:
The Levels of Processing model was proposed by psychologists Craik and Lockhart (1972). They considered the structural and processing elements involved in memory. In particular they focused on the initial encoding of material. The idea is that in order to understand memory you need to be aware of what happens to information during initial encoding of information and then at retrieval. For Craik and Lockhart memory is a by-product of perception. If material is processed at a deeper level during initial encoding it will establish a longer lasting memory trace ready for retrieval.
The classroom setting requires creative delivery of content to ensure a semantic level of processing. The objective is to provide meaning and significance to the content being delivered. This can be done through different media, questioning and quizzes on a regular basis. The key is to ensure higher cognitive functioning and rehearsal by revisiting and reevaluating.
How can students improve their memory and how can we help them?
Willingham and Bjork give these suggestions to help students remember: distributive practice, overlearning, interleaving and frequent testing. This reminds me of my own guitar playing. I’ve been playing off and on for a long time now, with some sustained periods of practice and many gaps. However, because I put in the hours over the years there is high storage strength, meaning songs I learned to play can be recalled quite quickly when I sit down and re-learn them. When I was in a band I had high retrieval strength as well as high storage strength because I was engaged in distributive practice, overlearning, interleaving and frequent testing. In addition, the songs had meaning to me. Below are some ideas and strategies to improve what students remember:
1. We have to ensure they think about what we want them to remember
Unfortunately, wanting to remember something is not enough. In fact, much of what we remember from the past is stuff we didn’t need or want to remember. It is there because it is the thing that we devoted most of our attention and thought to at that moment. In short, the more we think about something, the more likely it is that we will remember it later. Willingham says we should think about memory as the ‘residue of thought’. Our day-to-day experiences are multi-sensory, but it is very unlikely we will recall the sound, smell, appearance and the emotions we felt during an experience. We will remember only the aspects we thought about the most; the ones we really devote some attention to.
During the experience of a school lesson, whether it be Science, Mathematics, Geography, etc. we typically want students to retain meaning. Thus, in the classroom we need to construct specific tasks that force students to think about meaning. These could be:
- After reading a sentence or paragraph of a textbook students should ask themselves ‘why’ what they have just read might be true or valid.
- Searching for and writing out the main ideas of a textbook chapter after they have read it, followed by consideration of how the author elaborates on these points.
- Constructing a hierarchical diagram with the main chapter ideas at the top of the diagram, and branching down to subordinate ideas that support the main ideas.
- Re-drafting their notes into another form such as a flow diagram or mind map, which forces them to consider how ideas relate to one another.
Some of the most ineffective strategies are ones that are most commonly used. Activities such as copying notes, highlighting and reading over the textbook will not ensure that the meaning has been considered. In addition, repetition does not guarantee it will stick, unless you keep thinking about meaning. Additionally, Nuthall in his book ‘The Hidden Lives of Learners’ discuss how students’ recollection of information can be affected by the type of activity we design. In other words, the task should not be the primary thing that students think about to the detriment of the subject content they were meant to be wrestling with. Furthermore, Willingham states that we need to ensure lesson plans are not a ‘long string of teacher explanations, with little opportunity for students to solve problems.’ Students should be challenged to create new meaning through cogitative activities that, when tackled successfully, create strong emotions, which also enhances the chances of retaining the information. In fact, if we look for ways to connect learning to the student’s own lives, things that matter to them or the world they see around them, then we may arouse an emotional response that embeds the information in the long term memory. In Demonstration 1 on page 8 of this article by Daniel Willingham there is an exercise you can do with your students to demonstrate how thinking about meaning really helps memory.
John Medina, author of the book ‘Brain rules,’ discusses the notion that ‘memory trace appears to be stored in the same part of the brain that perceived and processed the initial input. The idea behind this proposal is that the brain has no central point where memories go to be retrieved, rather they are spread over the surface of the cortex. Moreover he goes on to elaborate that at ‘retrieval memory is improved by replicating the conditions that surrounded the initial encoding’ (p113)
This would involve creating ‘context-dependent’ or ‘state-dependent learning’ (p114). Research has demonstrated that the more meaning information has when presented the more elaborate the encoding.
This can be done by presenting information with examples of a main theme. The more examples that are given result in greater encoding in memory. The more personal an example ensured deeper encoding is stored and is recalled easier.
Medina emphasizes the importance of introductions: ‘They are crucial as the memory of an event is stored in the same place that was initially recruited to perceive the learning event’. It is at this initial introduction that you establish a connection to ensure future access to information. (p116)
2. They can create ‘cues’ to aid recall
We have all had a moment when something someone says, a smell, a piece of music, or a sight triggers off a memory you long believed forgotten. Willingham calls these ‘cues’, and the quality of them will determine how well we remember something. The most effective way for a student to do this is to have cues that link to more cues, so that they can recall the large amounts of information needed to complete an essay or section of an exam paper. However, sometimes things we have to remember have no meaning, such as foreign words or dates of an event, thus there is no link to the cue. This is when we need to utilize mnemonics which make the meaningless more meaningful. Below are some memorization strategies from Willingham’s paper ‘How can we help student’s improve their memory?’
Other methods include the Major System and the Dominic System, which are aids to help us remember numbers. The Major System has associated consonants for each number which can then become words by inserting vowels.
If I wanted to remember that Germany invaded Poland in the year 1939 I would turn the last two numbers (I’m assuming we would know its in the 1900s) into a word using the letters m and p, for example ‘Mop’. I could then think about German soldiers carrying mops as they invaded Poland. You can read more about this system here. The Dominic System has different digit-letter associations and turns numbers into people who all have an action.
If we wanted to remember zinc’s atomic number is 30 we could think about Conan O’Brien coated in zinc. For a 4 digit word the first two digits are converted into a person, and the second 2 become an action, for example 3311 could be Charlie Chaplin swinging a tennis racket (Andre Agassi). More can be read about this here.
We should also utilize the power of story. Everyone loves a story and we can often retell large chunks of a tale even weeks after we were told it. The states of the US, the periodic table, events leading to a historical event are all examples of things that can be memorized with the use of story. When we are creating our revision stories we should try to use the following techniques, made famous in ‘The Memory Book’ by Harry Lorayne:
- Use “substitute words” to break down complex concepts, or words that are abstract and intangible into things we can visualize e.g. Arizona might become ‘air zone’
- Use your imagination to create vivid mental images of the ideas by exaggerating the proportion (make the item gigantic in your mind), exaggerate the number (think of millions of the item) and get action into the picture.
- Mentally associate the visual images you created to each other. These connections should almost always be formed by thinking of some kind of action that puts the images together.
Kris Boulton in his blog ‘Why is it that students always understand but never remember?’ gives an example of a story he told his students to help them remember Quadratic Formula using some of the methods described above.
Demonstration 2 (page 8 & 9) in Willingham’s article highlights the importance of using a variety of cues.
3. We (and they) should distribute study over time (don’t cram)
By studying at several different times you create different associations and strengthen cues, making it more likely the information will be stored in the long term memory. If you cram the night before you only have one association, meaning the facts are lost quickly. Bjork suggests spacing rather than massing; repeating something over spaced intervals. This means we need to revisit information at regular intervals throughout the year. Hattie, in his book Visible Learning, also rates this highly giving it an effect size of d=0.71. Some ideas of how this could be done are:
- Having review weeks when you revisit content covered earlier in the year. This shouldn’t just be when they are about to take an exam
- Use (what I call) ‘Back to the Future’ quizzes on content covered a week, month or year ago without any warning.
- Make explicit links between current content being taught to past topics. This might be in the form of ‘compare and contrast’ exercises.
The need to repeat information regularly is emphasized by Medina. ‘Memory is not fixed at the moment of learning and repetition provides the fixative’ (p146) This should be done over a period of time with new information introduced gradually to enhance what is already stored. It then needs to be revisited. A way to incorporate this into the classroom is to reissue work on a regular basis to be reworked and then re-grade. Students should be able to see a development in their ability to recall information and also embed new ideas and provide an enhanced piece of work as the mind has had time to reflect, absorb and then repeat.
4. We need to introduce ‘interleaving’ into our curriculum planning
This a suggestion from Bjork that has implications for how we plan our schemes of work. He says people actually learn content better when it is interleaved with other content. This is one of his ‘desirable difficulties’ that can slow down performance but lead to long term retention. However, this is uncomfortable territory because it goes against the conventional wisdom that topics should be taught discretely in blocks. Interleaving would involve making learning material less clearly organised, for example, by teaching topics together. However, if you interleave your curriculum the student’s learning will be much deeper at the end of the course, rather than just at the end of terms if you teach units in blocks.
When a student revises, no matter how effective their strategy is, there will be some forgetting. They may have been confident they knew the material the night before, however, by the morning some of it has gone. To counter this we need to encourage students to go beyond the point when they think they know something and, in the words of Willingham, ‘put in another 20 per cent of the time it took to master the material’. His article ‘How can we help student’s improve their memory?’ includes an exercise (Demonstration 3) to emphasize the problem of overconfidence on page 9.
The only way students can know what it is they know is by putting themselves through the same type of activity they will be undertaking in the exam. Alternatively, they should be encouraged to explain the things they will be tested on to other people, such as other students or their parents. In fact, parents may be the better option as they will only have partial if any knowledge of the content being revised and are in essence being taught by their son or daughter. Hopefully, parents would probe them with questions to test their grasp. How well they understand the material after that experience will be a fairly good gauge of their sibling’s own understanding. In addition, the act of teaching also forces the teacher to find the meaning in what they are trying to remember. I know I learnt most historical knowledge once I was in the classroom.
In the classroom, there is now very strong evidence that regular low stakes testing of students knowledge reaps huge benefits when it comes to later recall. In a recent article in the New York Times, Benedict Carey details how 900+ students in a University of Texas Psychology class were given online quizzes after every class. He explains, ‘the quizzes would be short and personalized — seven questions that the entire class would answer, and one tailored to each student, usually a question from another quiz that he or she got wrong.’ As well as outperforming previous classes, attendance was higher and drop out rates were much lower. Students had to develop good study habits and be at classes otherwise their scores would be immediately impacted, which was different to the past when they waited to mid-terms to be assessed. In addition, Bjork and Willingham both advocate frequent testing.
My old colleague Gavin Rayner is a skeptic when it comes to a focus on memory. After reading this blog after it was first published he said this:
Why are we testing memory rather than understanding? In history, is why not more important than when?
In response to his first question, we should do both because they are both interlinked. If a student’s recall of facts relating to a topic in science is weak then their understanding can only then be shallow. An full explanation needs specific information to demonstrate a deep understanding. Kris Boulton here, puts it this way:
I suggest that if we put all our thought and effort into building understanding, we do so at the expense of memory, and will nurture students who understood everything, once, rather than understand it, still.
As for the second question, addressing ‘why’ is certainly a large focus of what History teachers do. However, we can not explain causation without a clear grasp of chronology; the ‘whens’. Only by knowing exactly when something happens can we start to discuss with any sophistication why it happened when it did or how important a particular event is in the causal chain. If they get their ‘whens’ wrong in an exam then the whole essay could fall down.