Treading of dangerous waters, here. But the following is a distillation of what I have read and hopefully understood about memory.
At some point your children will take exams and the more they can remember, facts and processes - know what and know hoy - the better they are likely to do in the exam, other things being equal.
It is important to be clear about two stages relating to memory.
The first is one that many teachers will already do well. Getting the stuff in. This starts with the teacher being very clear about what the student is to learn. Do they really need to hear, and potentially learn, that interesting story you always tell when teaching the topic? Really? I know you enjoy it and your children may well laugh but is it time well spent? Would it have been better to repeat the work in a different way rather than add a piece of narrative that is loosely connected to the thing you want them to learn?
That does not mean there should be no stories. Stories have a very special place in relation to learning. Stories seem to be privileged as routes into the brain. Stories place facts in a sequence and relate items together. Stories can change facts into understanding - or at least begin that process for our learners.
So the first feature of improving the memory of students is to be sure they receive the information they need. As rich as it needs to be and no more. If we think of this as a process model we want to refine the input so that what is presented is as accurate and complete as can be.
I do not mean spoon feeding nor rote learning. Both have their place in very specific setting but in general they will not lead to quality learning per se. There are reasons to make the learning a little more difficult that one might at first choose. Make the learners work a little harder to comprehend and their learning will be enhanced. Perhaps this works by priming their brain to activate the areas they already know which will relate to the new learning. There is some evidence that asking students to try to solve problems that they do not quite have all the required knowledge for enhances the learning hen that new knowledge is presented. Which knowledge we should present in this way and how often we should do this is a matter of professional judgement. I can see it would work for those students who were well motivated and somewhat tantalised by the puzzle aspects of such learning. But it could also be the case that some did not really try, knowing that they did not yet have enough information to resolve the puzzle. It might annoy some students who feel that the teacher should be providing the knowledge first.
Given the teacher is providing the knowledge to students the ability to explain clearly is paramount. Great teachers give great explanations. They pace the explanation that provides pace, possibly just a little faster than the students can easily manage but not so fast that they are lost. Make them work for it. Great explanations may well be a knife edge and that is difficult to define. Presenting work that is at the time just outside the student yet expecting them to keep up could be seen as an exemplification of high expectations. 'I know you can do this' is an attitude great teachers display to their students. Not dumbing down as this could be come just spoon feeding where the students do little thinking about the new knowledge as it is presented.
Robert Bjork describes these as 'desirable difficulties'. It is very well worth reading the details and watching the videos.
So we have carefully selected what is to be learned and we have constructed the explanation phase using the ideas of Robert Bjork. Our children surely know what they need to know. Sadly no. In most schools the above is often well done. In some classrooms too much emphasis is placed on discovery type learning. What these teacher probably believe is that children who 'discover' the knowledge will understand if better that 'being told'. The evidence is that this is not true. At best children understand as well and in many cases they will not know as well. The major disadvantage for all children is that the discovery process takes so much longer. The second issue is that disadvantaged children do considerably less well than more advantaged children.
The Matthew Effect prevails.
The second matter is to do with what, in my view, is less well done by many teachers. Once learned, ie in one's brain, there is a need to practise recall. if we cannot recall then we cannot truly say something, in the academic sense, has been learned. The recall needs to be, as dog trainers know, proofed. I am in no way suggesting children are dogs other than to say that dogs trainers and good owners know that their dog needs to be taught to, say, sit in as many different environments and with as many different distractors as possible.
In multiple choice question parlance the wrong answers are called distractors. They are there to distract. The learning must be as secure as possible. So present children with answers that are similar, more similar as they become more secure. Make sure they are answering from their knowledge rather than using other clues. (Other clues are fine once we know they are secure in their learning.)
Testing, children testing the extent to which they know something, is an underused technique to aid memory. Not an important test like the one to decide which set you are in for GCSE physics but low stakes, the outcome is for you to know what you now need to relearn type of tests. We need lots of these. Short tests where the feedback allows a child to realise what he/she knows and does not know. If possible to know why they don't know whatever has been tested. Proper diagnostic testing. Teachers can write these tests. They do not have to be something external but it does require some thought if the test outcome is to be more than a simple list of what the child has answered correctly and what incorrectly. Write the test so that the outcome is for the child and not for the teacher to be able to record a mark.
So clarity of input and checking of what has been properly learned will support memory and recall.