Continuing from part 1. There is some logic in this world!
Like all good learning programmes let’s start with a little revision.
The brain has 2 memory systems which interact.
The time and size limited working memory which processes, does the thing we call thinking, and can call on stored schema in the long term memory.
The working memory can only act on items it has in its own, limited storage – probably a maximum of 8 items and for anything slightly complex it can “think about” 2 or 3 of them.
But, working memory can call into its storage schema, however large and complex, and each schema only counts as one item.
Learning is the process of putting memories into long term storage. Working memory does that, often by attaching a new item to an existing schema, although working memory can create new schema. Think of schema like parts of a spider’s web. These can be big or small and probably will be connected to other “spider’s webs”. Think of a colour, say, green. When you do that you may also think of things that are green. Grass, tree leaves, bogies. Perhaps green also triggers thoughts of the Green Party. You get the idea. Schema are linked in all sorts of ways. The more links there are the more you are able to access a memory, because you can get at the memory from a greater number of places. We are completely unconscious of our stored memories until they get brought into working memory. Dreams are a special case. Odours can trigger memories as can sights and sounds or touches. You will need to Google if you want to know how senses fit into this.
Let’s get back to the limitations of working memory. CLT suggests that there are three loads, unfortunately not well defined or understood exactly what a load means as a brain function within CLT, that working memory has to deal with. The theory says these loads add and, I guess, we can think of these loads as totalling to a number either within the capacity of working memory or totalling and overloading working memory. This is where the, common sense to some extent, practice of Direct Instruction comes in.
The loads are named Intrinsic, extrinsic (my name as it fits better with intrinsic – Google if you don’t like it.) and germane.
Intrinsic load is the difficulty of the material to be learned. So 2 + 2 will present a lower intrinsic load than, find x where x=(245+93-18)/3.
Extrinsic load is the elements of distraction that are in the environment of the learning. Reducing this distraction is one of the features that CLT and DI folk get so worked up about. If you follow Twitter you will know of the minor ruffling that Mr Men caused. There are lots of bits to that argument but the one that impacts on CLT is that Mr Men is an additional piece of learning that is not directly connected to the core learning, which in the Mr Men case, was about the rise of the Weimar Republic. The Mr Men are seen by the DI folk to be an unnecessary distraction. And one can see from CLT and the limited space in working memory. The opposite camp will have reasons why the use of Mr Men is good. Probably something to do with engagement. I know it is a little more complex than that in the Mr Men case but I can’t go into all the possible issues. They don’t matter for CLT theory.
Germane load is the cognitive load put on working memory in creating or modifying schema, that you know are held in long term memory. The process of learning.
Now, CLT seems very sensible and one of the things I impress on teachers I work with is to make things as simple as possible.
But there are a number of matters that CLT enthusiasts have to deal with if they want to have a theory supporting their claims that DI is a better way to get children, and adults, to learn.
What exactly is a cognitive load? How is it measured?
What causes the limitations of working memory? Is it related to neurons or some other physical structure?
How do we include motivation/or engagement or whatever you choose to call it? Why do some challenging tasks lead to better learning than simpler ones? (Difficult fonts make learners work hard in reading the material but there is some evidence that learning is better than for more readable fonts.)
Given CLT theory and DI are directly accessing brain structures for learning and working within their limits, why is DI, at best, only around 10% more effective in securing learning?
Happy to have any comments, and especially corrections. I am no expert. I just read stuff and think a bit.