Saturday, December 20, 2008

That Remains To Be Seen

Kia ora tātou – Hello Everyone
Learning retention with time

Learning, and one’s ability to retain it, depends on the distributed frequency of related study sessions over time.

Clark Quinn’s recent post, to do with the effectiveness of crammed learning, brought to mind a discussion I had with a colleague some years ago. We’d been inquiring about the rate of return of assignments from a distance learner who had crammed the equivalent of several months’ study into one day.

Does learning have a half-life?

Clark cited an article by Inga Kiderra outlining the research findings of Hal Pashler. What is learnt during a study session seems to decay. The rate of decay has a dependency based on the number of related study sessions in a series and its duration. A series of study sessions over a significant period of time has a cummulative effect and can lead to longer lasting retention.

Graph of a series of study sessions with time
Learning diminishes at a rate that relates inversely to the pace of distributed study sessions over time. It means that a series of crammed sessions, during the week before an examination, is unlikely to bring about learning that's
useful a year or so later.

Competence over time:

A simple example of the properties of learning over time is how the skill and knowledge
is remembered that’s needed to solve a quadratic equation in mathematics.

Though this is not an easy skill to obtain, it is one that can be acquired by a competent student of mathematics by cramming over a few days. To do this, competent learners have to grapple with new ideas, some concepts and some content. One piece of content that the experienced student needs to know is the formula for the solution:

Equation for roots of a quadratic equation

A learner who has acquired the skills and knowledge during a few days of crammed study may be hard pushed a year or so later to remember that such a formula
even exists, let alone how to apply it. If the practice of solving quadratic equations is not revisited during the interim period, there may be little remembered of the activity.

The learner who has gained skills and knowledge over several months of regular practice may not be able to remember the exact formula a year or so later either. However, recollection of the concept of solving a quadratic equation, as well as recalling that the formula exists, is more likely. It may be that the solution is only a Google search away.

What is really being assessed?

Every learner is different in the way they assimilate what is learnt. What one can gain usefully from a paced rate of learning may not be equivalent to that acquired by another, even if their end assessments are identical.

The ideas brought forward by Pashler’s research have implications for the results of tests that lead to qualifications, as in the New Zealand Qualifications Authority standards. One has to ask what is being assessed in these tests.

There is no doubt that a good result in a standard assessment shows that learning has occurred. This is a measure of the ability of the learner to learn and perhaps understand through study.

How do we test long-term retention?

Depending on what study has gone before, and the pattern of that over time, however, a grade in a standard test may not be a useful measure of learning that may be put to use in the future.

There are similar implications for the results obtained through online assessment. Study that’s performed online, by a learner who is able to access all the resources for a unit of learning, may not be carried out in the best way possible to enable long-term retention.

( 10 ) << - related posts - >> ( 8 ) ( 7 ) ( 6 ) ( 5 ) ( 4 ) ( 3 ) ( 2 ) ( 1 )

Haere rā – Farewell


Tony Karrer said...

Great graph showing the timing effect that Will talks about. I'm sure I'll have reason to use it at some point.

Tom Haskins said...

Ken: I draw a distinction between contrived and natural retention. Adults cannot remember how to fill out the insurance, expense reimbursement or vacation benefits claim forms when they taught in a class that covers precisely how to do it. But they never forget it once they really need to file the form and get the beneficial result from filling the forms out correctly. They even learn from feedback if it comes back as incomplete or in error -- which holds up getting sought after prize. This suggests just-in-time educational offerings that get retained naturally instead of preparing too far in advance which makes a contrivance of retention.

Adults also remember from their first day job orientation where the rest rooms are located, no problem. But they do not retain how to get into Room 417B which is accessed through the foyer in Room 422C -- when they will be working in Room 213A. The rest rooms appear immediately useful, practical and relevant. It's likely retained because the information gets immediately put to use. The "use it or lose it" rule applies here. This suggests providing performance support and job aids in lieu of classroom training.

Retention also increases when the information to be "crammed" is imbedded in a narrative context. Our minds seem to naturally organize unfamiliar material in terms of character conflicts, outcomes held in suspense and unfamiliar territories where we discover new things. When the information to be retained is framed as "held in contention", "getting abused by those in power" or "difficult to apply in situations without a thorough understanding", the information becomes "made to stick", memorable and naturally retained. This suggests the use of scenarios, role plays and immersive simulations to increase natural retention.

Blogger In Middle-earth said...

Tēnā kōrua Tony and Tom

This graph is really a blast from the past - I'm glad you're taken with it Tony, I knocked it up in Photoshop.

The first time I saw a graph like that was when my old Chemistry professor, Brian Gowenlock, drew one like it on the blackboard in 1969. He spoke of the power of regular study and the need for this over a sustained period for useful retention.

That was long before Will Thalheilmer was blogging, that's for sure. I wondered about mentioning him in this post - maybe I should have - but I could not find a post of his to link to that was relevant to the topic. I'm new to Will's blog as I only recently picked it up and RSSed it when I came across his neon elephant award.

Thanks Tom, I like the story about the rest room and retention of learning. There are all sorts of things that do enhance the moment of learning. Many of these have emotional associations, at least that's my experience. There are tricks that can be used in teaching to bring about a learning moment as it were, and there is no doubt that one-off instances of learning that's retained significantly do occur.

The main point here, of course, is to do with enhancing retention specifically through study patterns. This is a different effect - one that there now has been some recent research on.

Catchya later

Sarah Hanawald said...

Hi Ken,
Add me to the list of folks who like the graph! I really struggle with assessment that isn't performance based. The life students lead after they leave us is the real assessment. If everything we spend time on should be focused on that end goal, how should/does that change our classrooms?

Have you heard of Disrupting Class, a book by Clayton Christensen, with Curtis Johnson and Michael Horn? There are some edtechtalk podcasts of interviews with the authors I think you would like.

Merry Christmas!

Blogger In Middle-earth said...

Kia ora Sarah!

I agree with what you say about the life of the student being the true assessment. I'm glad you like the graph too!

I will look out for Disrupting Class by Clayton Christensen. It sounds interesting.

Thanks for dropping by.

Best wishes to you and your lovely family.