Memory method

(The following is edited and elaborated from my comments in this BrillKids Forum discussion. The BrillKids Forum is awesome. The method we follow is greatly updated in this post.)

I had an interesting conversation with Dr. Miles Jones, an accelerated learning and memorization expert. I asked him for his advice on how we might best remember the things that we are learning in our book-reading. After all, H. and I read an awful lot of books, and while I’m sure some of it does sink in, most of it becomes implicit memory (the sort of memory that makes you go, “Oh, yeah, I remember that,” when someone reminds you of a fact–but which you can’t articulate when someone asks you about the fact). This strikes me as a perennial problem for education, one that teachers seem to assume is solved by exams and finals. As homeschoolers, we don’t have to do exams, so I am free to explore other, possibly more efficient ways to solve the problem.

Now, apart from those exams, most of us go through our schooling with very little memory work, and we don’t think it’s pointless to read books just because we’ll forget most of them. If we didn’t read them, we’d be really ignorant. So if that’s how it has to be with my boys, I’m resigned to their fate. They’ll still be well-educated.

But what if there is a way to retain more of what we learn?  Obviously, always re-reading books after reading them once will help do the trick. But ultimately, you can’t read as much that way and it’s not clear that you would learn a lot more that way.

Anyway, Dr. Jones gave me an intriguing answer. He said that you’d review the information one hour, one day, one week, one month, and one year later. Seems this is something that people in the field often say. He recommended that I highlight the info I want H. to retain as we read, then read the highlighted portions into a recorder, then we simply listen to the recording a day, a week, a month, etc., later.

So, never one to pass up trying out some easily testable idea, and since I have a nice handheld recorder and am an unimaginably fantastic user of it (not really, but I know how to edit sound files and stuff), I decided to give it a try.

For a couple of days, I was doing 12 minutes of summaries per day. I quickly calculated that, while continuing on that way might seem admirably ambitious, it’s really just plain crazy.  I mean, it’s OK as long as you’re reviewing just one of these recordings per day. But suppose you limit the recordings to 10 minutes a day, and each day you are reviewing recordings from a day ago, a week ago, a month ago, a season ago, a year ago, and two years ago. That would be 60 minutes per day of listening to yourself summarize stuff that you read…that long ago.  Have you ever heard of anyone doing such a thing?

Now, H. expressed great enthusiasm after listening to our first recordings. We talked about it and he said basically that it was a lot of fun, because it reminded him of what he knows. His reaction surprised me. I thought it was pretty interesting, most of the time, but maybe not that exciting. I did see two great advantages aside from the long-term memory aspect. First, it immediately reinforces what we just read. That alone might make the practice worth doing. Second, H. (when he pays attention to my summary while I make it–he doesn’t always) gets a very lively idea of what a good summary “narration” looks like. I flatter myself that I am good at summarizing things I’ve just read quickly and accurately, and picking out the important points.

But I just can’t imagine that we’ll want to listen to a full hour of this stuff every day. Eventually, I saw reason, as I’ll explain further down.

Then I thought, suppose it were (somehow) practical. But is it really desirable? Well, there are several considerations here. The first is knowledge. Knowledge is good, and we want to maximize it. Second is love of knowledge, or motivation–a different thing. We don’t want to burn out kids (or parents) by requiring too much of anything or more than tolerable of what is tedious. Third is pleasure, we want life fun, especially life for children. Fourth is opportunity cost–even if it is in all a benefit to do, would the time we spend on this be better spent on something else (like reading more books)?

Would it be worthwhile in this sense? I tend to believe Dr. Jones. I’ve heard his advice before, I don’t know where, and it’s very plausible that jogging your memory according to that pattern would help you retain information that otherwise would be forgotten. Well, if so, it would be extremely valuable. It might even be worth some unpleasantness, or at least foregoing of more intense pleasures. After all, what we’re talking about is remembering a hell of a lot more stuff than you would remember otherwise. Suppose that someone could wave a magic wand and suddenly you’d remember, instead of 20% of what you learned in school, more like 60%. (I’m picking numbers out of the air, but you get the idea.) You know that, right now, you’d be a lot better-informed than you are. Having a handle on all that information would in turn enable you to draw connections and make insights that are unavailable to you because, well, you’ve forgotten so much. Reflecting on this makes me wonder if this is something that we should all be doing, even as adults. Should we be spending an hour each day simply reminding ourselves of what we have already learned? I don’t know. It sounds like a fascinating idea to consider, though. (I have since been informed that some people actually do this.)

Another consideration I’ve been thinking about, however, is that we might very well achieve a similar effect simply by reading increasingly difficult books on the same subjects that we’ve already studied. In this way, maybe we don’t have to review, and we get a similar effect. Perhaps–but I don’t think so. Even someone who revisits the same fact four times in his education, in increasingly difficult contexts, might still forget it because it never, on any of the passes, makes it into long-term memory. But the method Dr. Jones describes is designed specifically to get those facts into long-term memory. Reviewing info a day and a week later, in particular, seems important to getting it into long-term memory. My guess is the month and three-month reviews will set the neural pathways quite well. Then one will likely come across the information later in one’s education, and if one is still using the same memory technique, and one has forgotten it, then it will be revived all over again. Of course, that assumes that the technique would be used long-term…

So anyway, I was off and running. That was about five weeks ago. After about five days, H. was very excited about the memory method. He said twice that he was grateful to Dr. Jones for suggesting it. (H. is a very weird kid.) But after a week, H. started resisting, a little. Then he went back to being fully supportive, then resisting, and so forth. A couple times, I’ve started talking like maybe we’ll quit and he quickly backtracked and said he likes recordings, he learns a lot from them, and so forth. It effortful to listen to these recordings, but it’s a “good hurt.” He and I both are learning a lot.

A week in, I wrote, “I am 90% sure that within a week to a month, we’re going to either give up on this or completely change it.” Well, it’s been a month, and we’ve made only one significant change: I made a big effort to reduce the recording amount per day to no more than eight minutes. Rarely, it’s more and often it’s closer to five and sometimes three. Let me put this in terms of the length of time spent summarizing individual readings: instead of three minutes to summarize a reading that took 20 minutes, I summarize in 2 or 1.5. Shorter readings get just a minute or less. There are some books that are just really hard to summarize so quickly, and I indulge in longer summaries. Anyway, using shorter summaries makes the whole thing much more doable. This did require that I reduce the amount of information I put into recordings, but really that’s OK; the big important points are the ones we really want to remember anyway, of course.

I have tried to get H. to do summaries. I have tried to train him, give him examples, explain about the main idea, etc. We have been practicing outlining stories, and he’s not bad at doing that, with some help from me. But none of this training makes it feasible for him to do summaries on the tape recorder. Actually, there was one that he did that was actually a summary rather than a word-for-word reading with commentary, out of a half-dozen tries. So, if we want to do this memory method, then for now, I have to make the recordings. I expect I’ll continue to have to do them for at least another year or two. When he starts doing more reading to himself (when I start reading more to the baby at the table instead of him–which I suspect will be reasonably soon), either he won’t do summaries of those readings, or he’ll have to learn to do them quickly and accurately (seems like a non-starter right now), or we’ll have to figure out something else, I don’t know what.

In case someone is interested in technique/how-to, here are a few more notes.

I summarize only nonfiction and poetry, not fiction (except for myth–I treated Norse myth like nonfiction). We’ll read, say, 20 minutes during breakfast, and then I’ll wander away from the noisy family and take a few minutes to make the recording (often yelling to H. and others to be quiet). I don’t mark up the books, because it isn’t necessary. I keep a thumb on the Pause/Record button, and record only when I know what I want to say, and pause as soon as I run out of things to say. Whether I summarize or simply pull out interesting facts depends on the kind of nonfiction. For example, with the Usborne Encyclopedia of World History, there’s no summarizing it because it’s not a narrative, so I just record the most important facts. But with the Look-It-Up Book of Presidents, a summary usually hits the factual high points too. When we read poems (Piping Down the Valleys Wild, these days), we usually read from 5-10 poems, depending on how many we really try to analyze, and then what I do is simply ask H. which one he wants to record. If we read a famous poem, I will record that one in any case, which means that yesterday I recorded two poems; that’s OK, because kid poems are short and are nice to listen to in the recording. He often picks or goes along with a famous one. I find I have to speak with a very clear voice, i.e., not slurring and aspirating consonants, or the recording can be difficult to understand. Speaking too quickly is a mistake, but so is using lots of um’s and pauses. The most efficient way all around (to keep recording times down but also to make a listenable recording) is to speak clearly and deliberately, but without long pauses. Brief pauses are necessary for the listener to be able to mentally process the recording. It also helps for listenability if you keep the recording as dramatic or at least as interestingly worded as possible. It would be nice if the recordings word made by a professional voice person, but since that’s not possible, just try to sound clear, convincing, engaged, maybe excited, but definitely not bored.

I use a Sony Digital Voice Recorder, I think it cost me something like $200 back when I was recording music sessions. It comes with some pretty good software (“Digital Voice Editor”), which makes it easy to save recordings quickly, organize them, and convert recording formats if necessary. At first I was saving recordings to wav format, but this increased the file size so I decided to stick with Sony’s proprietary .msv format, which combines high quality with low file size. As long as I have the software, it’s easier to go with this than to convert to wav. I also used to combine the day’s recordings automatically (which the software makes it easy to do), but H. didn’t like that, and I also found it somehow satisfying (as opposed to annoying) to start each short recording individually.

As to how I organize the recordings: I have a “Recordings” folder. Each day gets its own new folder, where I put the day’s three to six recordings. It’s easy enough to find the recordings from a week ago. After we’ve listened to the 1-week recording, I transfer it to the “To review one month” folder. After we’ve listened to the recording made one month ago (I go with the same numbered day last month, resigning myself to some confusion as there are greater or fewer days at the end of months…it’ll all work out), I transfer that to a “To review three months” folder. It’s usually easy enough to find the recording to review–it’s the oldest one in any given folder (together with yesterday’s recording).

Sometimes we don’t get around to listening to yesterday’s recording after dinner (the reserved time) so we do it early the next day.

I have noticed that it takes a considerable amount of attention to stay focused throughout what is now 15-25 minutes of recordings to listen to. The recordings are, after all, nothing but information. The only reason we can stick with it is that it does, after all, remind us of something we read, and that usually has some interest for us. Sometimes I have to settle H. down…he is only five, after all…I’m not meaning to brag, but I am sure that most kids wouldn’t be able to do this. As hyper and independent as he can be, H. actually has a remarkably good attention span and he’s a very geeky kid. Just as I thought he would be. (To illustrate that further, we recently started learning/playing with Scratch, a programming language learning program from MIT, and he just plays with it endlessly as he would a game. I’m not sure if he’s learning much, but, well, he’s learning something…) Baby E. is very different, we can and do read him books, but not like I read to H. I simply can’t sit down with him anytime, the way I did with H., and read any number of books–even though I try pretty often. He has to be in the mood and it’s only one or two board books at a time. He’s also more extraverted and sociable. So if we can do this with baby E. when he is five, I’ll be very surprised.

As to the “success” of the memory method, I still haven’t tested H. on much since we first started the memory method a month ago. I am very sure that he does not remember 100% or even 80% of the facts, in the sense that I might ask a straightforward short answer question based on some statement made in the recordings, and he would give me a more or less complete and correct answer. My guess is that the number is closer to 50%. Still, I do have the distinct impression that he’s learning significantly more because of these recordings than if we had not used them.

I have had one confirming bit of evidence. We were going through the Story of the World (Vol. 2, we’re 2/3 of the way through this…along with the other 3+ history sources we’re covering at the same time) test book orally. We did some quizzes over material we read over a month ago, and some quizzes over material we covered in recordings. He didn’t get 100% on the latter (several of the questions weren’t even covered in the recordings) but he definitely did better.

I also had the impression that the questions he did best on (i.e., he was most confident on, or said the answer before the multiple choice options were given, or was confident of the answer and turned out to be correct when I wasn’t sure, etc.) were questions over information that appeared in multiple sources we’ve read. Whether this better performance is due to sheer repetition, or having the information embedded in different, complementary contexts, or something else, I do think it’s a good idea to use multiple sources when learning history, science, geography, and other such subjects. We find it indispensible to use multiple math systems concurrently as well, he simply wouldn’t understand math nearly as well if we weren’t using Singapore and MEP and 2+2 is not 5 (almost done with this).

Still, doing recordings definitely seems to be highly beneficial memory training, so far. We’ll see if we can keep it up for the “one season” and “one year” reviews–doesn’t seem likely, but it’s possible. Then we’ll really have a good idea of how beneficial this method is. One thing seems fairly sure to me already: it won’t give you perfect memory of everything you’ve read. It will simply improve your recall and ultimately deepen your understanding. While doing this, we started reading about physics in depth. We’ve read a lot about Newton’s three laws. The combination of reading about them in 4-5 different books (really! It’s amazing how many accessible, simple explanations of physics are out there!) and listening to summaries of all of these as well has really helped them to sink in. The proof will come in a year–we’ll see if H. (or I!) can remember them well enough to explain them then, long after we’ve moved on to chemistry.