A new angle on Memory?
March 24th 2011 17:10
“We’ve long been aware of the need for massed practice followed by distributed practice in order to achieve reliable recall, but this is the first time I’ve heard it expressed in terms of hours and minutes, as opposed to days, weeks & months. Our own teaching materials for early decoding and spelling use the latter pattern, for the simple reason that the complexities of the modern classroom are such that teachers lack the attention (to say nothing of the inclination) to attend to such minute details. Fortunately, educational psychologists in England understand the need for considered patterns of over-learning, so they frequently recommend our materials.
“However, we occasionally get feedback from teachers who have pupils with the most complex learning difficulties (such as extremely poor working memory) who still have great difficulty recalling the most basic letter-sound correspondences, although our Bearing Away programme Really Long Link does seem to work with most pupils in special schools. In such schools, staffing ratios are generally adequate to allow for the sort finely-distributed practice such as you suggest.”
I bypassed the issue of how to go beyond one-to-one instruction to enable a whole class to remember something, and answered:
“Here’s an angle of thought that helped me: What is the cellular unit of memory, the basic building block from which a whole body of knowledge is built? What is the single tiny piece, which if we get right, we can employ to construct the entire edifice of knowledge? If people even pose the question once, my bet is that they stop looking for a single answer because the actions of learning seem so varied–a little of this, and a little of that here and then there. Spaced repetition is one piece, but you have so many variables even to that.
“I reasoned this way: 1) If we want mastery rather than familiarization, we’re talking about recall of something rather than recognition of it, so we look at something actually retained, if only for a short time. One could think of it as a small piece installed at least firmly in at least working memory.
“2) But how big should that piece be? Is there a size that converts to everything else in multiples, that adapts to any subject? I answered that partly by acknowledging the brain’s limitation that extremely few people can retain seven separate items in simultaneous attention, and that even four requires a substantial degree of attention management but is possible for most people. I drew from that that the optimum size cellular piece should contain between four and seven novel bits. In the part where the new learning is contained, “bits” mean a new word or number. So a basic “point of knowledge” to remember as a unit contains four to seven novel bits probably expressed in just one or two sentences.
“3) The final piece is what to make of interval recall. The very idea of spaced recall, with the spaces steadily getting bigger, suggested that there should be an optimum moment at which recalling a point of knowledge is most effective–the deepest learning for the least amount of time spent. It matters when you choose to do the second, third, and fourth recalls of a piece, in other words. It appeared to me that just to save study time, you’d want to recall a piece before it begins to decay so you wouldn’t have to do any correcting or re-checking. The ideal time to recall a question with six parts to the answer must be sometime while you can still bring up all six “without looking.” Within that band, when is the most efficient moment likely to be? Typically, not immediately after you just did so (the issue of massed practice hinges on a different point), but rather just before you’re about to lose point 4, for example. The effect of this line of thinking is to find out how far you can stretch the interval and still not lose anything in the material you began with.
“4) This was the context in which much experiment with different types of material led me to the intervals of one minute, 3-5 minutes, 15-20 minutes, 45-60 minutes, and 3 hours or end of day—at each interval, having not looked at or thought about the material for that long. Using the method myself, I find it continually remarkable not to have thought about something for an hour (deliberately turning my attention away from it), posing the question, and realizing “Yes, I still know it perfectly.” The approach continually validates itself as people realize that they actually retain knowledge that previously seemed to disappear like water down a drain.
“There’s much more to talk about, such as the value of re-explaining the same material repeatedly to different people (a type of massed practice), engaging the social environment, and the expectation of ‘making sense to peers,’ etc. Although few people seem oriented to thinking about such issues, I believe that section 30 in my Silver Bullet book (cf. below) contains an extremely important distinction (i.e. ‘the practice element’ inherent in a particular learning activity), important enough to separate mediocre teaching practices from effective ones.”
I learned from Tom then that he’s involved in helping set up standards for a history curriculum in the UK, and that he values Daniel Willingham’s book Why Don’t Students Like School.
I agreed with him about the book, but noted “I think the problem lies in adapting ‘true’ principles into their most efficient form for classroom use” which has been my own focus, and commented then on the UK history curriculum:
“About the history curriculum, you may have a watershed moment at hand to open a different way of looking at learning. I’d suggest at least scanning Chapter 2 in my book Practice Makes Perfect (cf. below) about “Accumulating Knowledge.” I especially love the quotation from Ivan Pavlov. But also, consider this paragraph from Chapter 5. Practice by Explaining:
“Having the whole thing in your head” makes sense for students as well as teachers. We have plenty of time to make it happen, and it’s what students need later. This suggests a simple K-12 curriculum expressed on one page. Divide the universe of knowledge into, say, fifty categories. About each, ask the same question every year and answer it more thoroughly, incorporating everything learned before. Expressed in statement form, it’s “Tell all you know” about history, literature, mathematics, language, climate, geography, astronomy, psychology, religion, science, art, music, human relations, family, biology, physiology, physics, etc. Year by year master any regional occupational emphasis: “Tell all you know” about computers, finance, automobiles, manufacturing, mechanics, electricity, farming, mining, fishing, lumbering, and so on. We want students exiting formal schooling possessing accurate, extensive, competent knowledge–but about what? Why not about the world they’ll cope with as adults?
“I think that a great deal of concern about curricula really is the result of non-working instructional techniques. You have to parse carefully what you place in children’s minds precisely because you expect so little of it to stick! Once you realize you can teach them anything in a way that it sticks, why confine yourself? Learn everything!”
Students using the “expanding-interval approach” report a distinct sense of satisfaction at their objectively-measurable, obviously-increasing mastery of a piece of learning. What few stop to think, however, is that the method provides an objective and easily-applied measure of the depth of memory: the number of days since you last looked at a hard copy and can still remember it perfectly.
About massed practice, we might understand its usefulness better by separating two aspects. One is the sheer depth of memory of the material, which I believe is most efficiently obtained by the steadily increasing time intervals at which one can still recall the material perfectly.
A separate benefit arises, however, in the re-chunking that can be enhanced by repeatedly recalling the same material even within a short time frame. The effect we want presumes memory, but goes further to have entire segments of the material emerge as a single piece that’s totally mastered and can be employed spontaneously from several angles. What in a downhill skier means consigning increasing amounts of sensory data to automatic response, in a history professor means having a whole segment of history emerge before the inner eye as a single, consolidated, yet intricate piece–every aspect of which can be independently accessed but also that the whole piece can be treated as a single thing.
| 29 |
| Vote |
Add To: 
subscribe to this blog
