The Retention Toolkit #2: Spacing Effect

klsgfx-Swiss-Army-Knife-2The Retention Toolkit is a series of posts describing research based tools which help improve long-term knowledge retention.

In my first post of this series, I presented research on test-enhanced learning: using testing not as an evaluation technique, but as a learning technique to improve retention.  Testing is a far more efficient use of time than study techniques such as rereading, reviewing lecture notes, or making outlines.

But how is test-enhanced learning best implemented?  Research in human memory shows that timing is everything.  It’s called the spacing effect.

Information that is spaced over time is better remembered than the same amount of information massed together.(1)

forgetting-curve_en

Take a look at this graphic of memorization over time.  New material is presented on day 0.  It is then studied at gradually increasing intervals.  The intervals double approximately after each successful retrieval (test) from memory.  Over time, the rate of forgetting (slope of the blue line) approaches zero, even though the material is being retrieved less and less often.

Further, retention is boosted the most when a memory is just about to be forgotten. (1,2) Simply reviewing on a daily or weekly basis is not an efficient use of time or energy.  Long-term, a review period of once every two months seems to maximize retention for life. (2) But how do we get from a new piece of information to a retrieval practice once every two months, especially when we consider that we want our students learning a LOT of new pieces of information every day and retaining them?

First, use an effective study method: Teach students to self-test, rather than review material passively.  Second, revisit material on an expanding review schedule, as described in the next section.

Application in the Homeschool

There are three possibilities:

  • Informal test-enhanced learning
  • Scheduled test-enhanced learning with a paper system
  • Scheduled test-enhanced learning with a computer system

You may find a combination of all three, based on subject matter and grade level.

Informal Testing

When you are working closely with a student during each lesson, you may find you have an instinctive feel for what topics need to be revisited.  You can perform a parent-guided retrieval practice in written or oral form, such as in these examples:

  1. Prior to a math lesson, write out ten warm-up problems which include topics from the last six months.  Concentrate on newer concepts or concepts your student is struggling with.  In our home, this work is done while I glance at the teacher’s book for the day’s lesson to get my footing.
  2. Before starting a new chapter in a read-aloud, ask your students to help you recall what has already happened, and what they think might happen next.
  3. Take a hint from public schools, and give your child “homework.”  Go over a lesson and do several example problems.  Then, rather than having the student immediately attack their independent work (spelling lists, math drills, verb conjugations…), set it aside and  move to a new subject.  When your student is ready to complete their homework, it will serve as a first review, rather than as a long lesson.

Scheduled Testing with a Paper System

If your curriculum includes a lot of memorization, you may find it is too hard to keep track of which Latin root, which insect order, and which historic event each of your children is currently struggling to remember.  You can create a file card system that is based on spaced repetition, and use it daily for retrieval practice.  Simply Charlotte Mason has created a scripture memory system that is easily adapted for any fact memorization.  Watch this video for details.

Scheduled Testing with a Computer System

Some of the best work in spaced repetition systems was done by pioneering software designers, who tried various algorithms of testing frequency until they had found the most effective.  We use Anki, but there are many other available- see the list on Wikipedia.  All of these programs are intelligent flash card programs.  They are incredibly powerful tools, and best suited to subject areas with large amounts of memorization: foreign language, specialized vocabulary (i.e. medical, legal), etc.  However, they can also replace a paper system for family memory work, especially when mom or dad uses the computer to prompt retrieval practice across a broad subject range as part of a morning circle time or other general review time.  I will go into much more detail about the power of Anki when I review Fluent Forever.

What if a student misses the ideal window, and is unable to remember the desired information at the next study session?

Step 1:  Remaster the material immediately.  Self-test at 1-2 minute intervals during a study session until you have reliably relearned the material.  It does not take nearly as long to relearn a forgotten piece of information as it does to learn a new piece, so take heart!

  • Step 2:

    2000px-Leitner_system_alternative.svg
    As correctly retrieved cards are moved to the right, they are studied less and less frequently. When a student fails to retrieve the desired information from memory, the card moves all the way back to the beginning. Image from Wikipedia
  • If you need to have the material solidly memorized in the near term (1-2 weeks or months, for example when studying for an exam),  then move that piece of information back to the very first slot in your card review system.  If you are using a computer based program, this will be done automatically!
  • If you are looking for retention over a much longer time period (years, or a lifetime),  leave the card where it is, with a spaced retrieval planned for no more than two months away. (2)  Over the long term (five year study) this works.

consensusA Meeting of the Minds

Charlotte Mason may not have heard of the spacing effect, but she certainly knew how to use it.  She suggested reading books at a snail’s pace: often one chapter per week, with perhaps 10-20 books being read in this incremental fashion during the same time period.  Presenting information in this way is a great example of the spacing effect.  A book read over four months is much more likely to be remembered than one read over four hours.

Mason was also a proponent of short lessons, regularly given.  We know from research that there is cognitive science backing her up.  Two thirty minute lessons are far more memorable than a single one hour lesson.

Next in the Retention Toolkit, I will be looking at research supporting a move form a skill-centered approach to a content-centered approach to education.

 

(1) Sisti, H. M., Glass, A. L., & Shors, T. J. (2007). Neurogenesis and the spacing effect: learning over time enhances memory and the survival of new neurons. Learning & memory, 14(5), 368-375.

(2) Bahrick, H. P., Bahrick, L. E., Bahrick, A. S., & Bahrick, P. E. (1993). Maintenance of foreign language vocabulary and the spacing effect. Psychological Science, 4(5), 316-321.

In addition to these sources, this post contains a summary of the spacing effect information contained in this article:

Roediger, Henry L. et al. The critical role of retrieval practice in long-term retention. Trends in Cognitive Sciences, Volume 15, Issue 1, 20-27. 2010

12 thoughts on “The Retention Toolkit #2: Spacing Effect”

  1. I like Memrise. They even have a little Swiss German. 🙂

    Read one book over four months? [shudder] I will grant that my husband, who reads books slowly, retains them better than I do. I have excellent retention of some books, to the point of being able to quote from them, or find passages based on their position in the book, but that comes from rereading over time (naturally taking advantage of the spacing effect). I’d rather read many books quickly, then revisit ones I find especially valuable.

    I imagine it takes extreme self-control to read a book over four months. The only way I could possibly do it would be to read, say, one chapter; then the next time read that and next chapter; then the first three chapters, etc. Granted, by the end of the book I’d pretty much have it memorized. But the torture might turn me off from reading altogether.

    1. I would never be able to read a single book over four months either… but I am an avid re-reader like you. For school books with the kids, I find this a bit easier. For example, we are reading one chapter a week from A Little History of the World, one chapter a week from Paddington, and one chapter a day from The Princess and the Goblin. We also do one poem a day, though that’s more related to time management than spacing!

      As an aside, one of the things I miss about reading on a kindle is knowing instinctively where a passage is in a book. In a paper book, I can generally remember what quadrant a particular passage was in, making skimming very easy to re-find something. No such effect on the kindle.

      I will be talking about Flow and Personal Connection later in this series, which will touch on why we can sometimes remember with incredible clarity after only one exposure… and sometimes not. If there is a big emotional reaction, our brains are wired to hang on tight to the information!

  2. Wow. Great stuff. I’ve noticed with my kids how I can pack a power lesson at the right time (i.e. when they ask) and the retention and growth is amazing! I’ve often thought about how I could teach my kids more in the time commuting to some class, than the class could teach, for much the same reason you outline here. Of course with commuting time, most classes aren’t less than an hour, but if the kid checks out even after 30 minutes, the whole exercise is a waste of time and money compaired with being more efficient at home.

    Thanks for these great summaries! I hope your blog is successful, because it sure saves me time and gives good inspiration in a powerful punch.

  3. I’m loving this series! This article reinforced to me what a great program Math On The Level is. After you learn a concept, you put in a review rotation (first every day, then every other, then every week, up until you’re only reviewing it every 4 weeks), to help move it to long term memory.

    I love to read books quickly, but I often end up reading them slowly, just because I only manage to get in one chapter before I fall asleep at night! 🙂

    1. I have not used that math program before, but I’m happy to hear that they are taking spaced repetition into account in a serious way! Hopefully more homeschool resources will do so, but if not, most can be modified in some way to make them more memorable. Unfortunately, reading tends to make we wired, and I read too late at night. Podcasts, on the other hand, I have to listen to in 10 minute chunks or while moving, otherwise they knock me out, even when I’m really interested! 😀

  4. Yes, Monica! I have that Kindle problem, too. With print books I have a physical link with where things are: where, generally, in the book, and even where on the page. When I remember a quote, I often remember its position on the page. When I was younger, I never bothered with bookmarks, as I could always find my place easily. That changed when I became a mother, and subject to frequent, random interruptions.

    I love Princess and the Goblin, by the way — and in fact am currently reading it (on the Kindle) for the umpteenth time, as part of my “read everything by George MacDonald in the order he wrote it” project.

    1. Ambitious project! This is my third time through The Princess and the Goblin, each time with my oldest. I tried to read Phantastes by MacDonald (having never read anything by him before) after reading CS Lewis’s praise, but I got bogged down by his 30+ word sentences. I should give it another shot now that I’ve got my foot in the door with Irene and Curdie.

  5. I LOVED Phantastes, but I think my mom is the only one else I know who loved it as well. I love building a mental picture of descriptive scenes, so those long passages never bothered me . . . but I am a slow reader.

  6. I have a question about the video. I’d seen it before, but always thought it seemed very tedious and overdone. After a while, you have LOTS of stuff that you have to do each day, even though they claim it only takes 5-10min. 10 minutes of boring recitation can feel like a long time, and as your first chart shows, we don’t seem to need all that many review sessions before a two-month review suffices.

    Is it possible to overdo the review, so that we turn our kids (and ourselves) off of what we learned because we have to recall it too often? I did a quick calculation, and the system in the video has you review nearly twice as offen as deemed necessary in your graph. Multiply that by each thing you’ve learned, and you’re talking a lot of extra time, bordom and recordkeeping!

    What do you think?

    1. The SCM memory box is a simplification of the original Leitner box, which is totally impractical. I have attempted to use the SCM index card box as shown in the video for memorization across the curriculum, and I found in lacking for the same reasons you mentioned- tedious and cumbersome. I think it would be an excellent system if you were only memorizing in one domain, for example Latin word roots, or if you were attempting a screen-free learning environment. But I ended up moving our family memory work (inc. roots, Baltimore Cat., grammar definitions in French and English, math terms, etc) to Anki. I interact with the computer and askt he kids the questions. This is working better. I can set how many cards to show so that we don’t go over ten minutes or so. In our family, my kids seem to like recitation. I know, they’re weird! The only other thing we use Anki for is German. For most subjects, I use informal testing as part of our warm-up, relying on my instincts for how often to revisit a topic.

      There has to be some balance between efficiency and joy in learning. We all know that learning a foreign language is rather boring until you acquire the skill to engage in basic conversation- so why not speed up that process as much as possible rather than dragging it out? Once you are using a working vocabulary on a regular basis, you can stop your SRS altogether or else teach yourself specialized vocabulary in your interest areas. But if recital is making kids hate learning, then all the efficiency in the world won’t do you any good. SRS are all about getting it solid as fast as possible.

      Personally, I think the very robust spaced review systems should be used with moderation, in subjects where their usefulness outweighs their tediousness- foreign vocabulary acquisition specifically comes to mind. Spelling is perhaps another.

      1. I’ll have to check out the electronic helps. I know kids love repetition much of the time, but on their own terms. Forced repetition seems like another story, but I haven’t tested the limits yet, so I just need to try.

        Since I use a Tickler system daily anyway, I’ve put a few things we’ve learned on the schedule suggested by your graph, simple by writing the spacing dates on the item itself. When an item comes up, I’ll put it on my desk for review during the day, then check it off, and look how many days later I need to put it in the Tickler for. This won’t work if we get too many things going, but it will be an interesting test of the method, and if it works, I can maybe build a seprate Tickler for things memorized and build a specific time to check and review it.

  7. You might try Lilith before Phantastes; I found it helped prepare me, mentally. In a different way, it’s not unlike reading G.K. Chesterton, who tends to make my mind spin. Well worth it, however, in both cases. Of MacDonald’s “regular” novels, I find Sir Gibbie a good starting point. Some of them are ‘way too 19th century romantic for me, but even those I love for his diversions into the nature of God, faith, sin, and redemption. (Diversions from the story, but really the most valuable parts of the books.)

    Sorry; I seem to have gotten a tad off topic….

Leave a Reply

Your email address will not be published. Required fields are marked *