CLT and desirable difficulties
Cognitive Load Theory (CLT) and the idea of desirable difficulties might initially seem to present a dichotomy. A basic reading of the literature around CLT might give the impression that simplification of learning is the key to improved learning. The Bjorks discussion of desirable difficulties, on the other hand, paints a picture of there being a need to make learning a little more challenging to be effective. How can we rectify these ideas?
TLDR: We want ensure that what the students are doing and practicing is challenging, but our presentation and explanations are clear and simple (without being dumbed down).
Adam Boxer has a brilliant blog on his simplification of CLT, and I’ll use a combination of his language (cognitive load=task demand/available resources) and the more usual language of CLT (cognitive load=intrinsic load+extraneous load+germane load).
Let’s imagine that my aim is to teach a class of interested adults about physics. I’ve got options as to how I present material to a class. I might pick to use the textbook: ‘The Feynman Lectures on Physics’ as the basis of my teaching as it’s got the content that I want and I know that the style of writing is balanced between academic and readable. I notice online that I could buy a class set of the textbooks (in English) for an obscene amount of money, but there’s a translation of the book into Mandarin (because this is the image I could find on google) for an absolute bargain price. The content is identical right? So the intrinsic cognitive load (or the task demand in Adam’s language) is the same? And I do love saving money. (Worth mentioning that there’s plenty to criticise about Feynman outside of physics – see this excellent blog)
The issue is the presentation of the material always provides an extraneous cognitive load – I view this as any cognitive demand placed on my students that isn’t the intended focus of my lesson. Clearly, presenting the material to my non-Mandarin speaking students in a completely unfamiliar language with an unfamiliar alphabet is a completely unnecessary and illogical way to present the material. In Adam Boxer’s language, I’ve reduced the available resources on hand to deal with the material by removing reading comprehension as an internal tool that my students can use.
So, to me, CLT suggests that I should be aiming minimise that extraneous cognitive load and make the material (the intrinsic cognitive load) as understandable as possible. Notice, I’m not making the material itself easier, just its presentation. So I should be considering what the simplest way is to get the information across (it may be a carefully labelled diagram or flow chart, live modelling of a concept, some carefully considered concrete examples, but it could just as easily be a list, or a piece of prose or a simple demonstration). The medium I use needs to be the most appropriate to make the message understandable without simplifying the message.
I could feign ‘desirable difficulties’ in the presentation of an idea by making simple material appear more difficult purely due to the presentation. Simply the act of translating my bargain-price textbook would have my class hunched over the book (for a brief window of time at least), seemingly engrossed in this difficult task. But this is not at all the kind of difficulty that is desirable. I should not present an activity that means my students struggle with the mechanics of the task rather than the underlying material.
So I have a rethink and decide against the Mandarin version of The Feynman Lectures. Instead, I consider whether to stick with the expensive, English version of The Feynman Lectures or to use a title from the Baby University series of books (which, incidentally, are brilliant for a much wider audience than babies).
Newtonian Physics for babies seems to cover the same topics as the first few chapters of The Feynman Lectures, and it has the advantage of being a cardboard book so I can chew it. The diagrams are also crystal-clear throughout; they’re actually some of the best diagrams I’ve come across to explain ideas in Physics simply. They reduce the extraneous cognitive load brilliantly.
But, and this is a big but, Newtonian Physics for babies reduces the intrinsic load too (it reduces the task demand in Adam’s language). By trying to reduce the extraneous cognitive load, it is very easy to make choices that actually ‘dumb-down’ the material so that it’s no longer quite the content that you’d intended. The adults in my class would already be familiar with the majority of the material and might even be so complacent about it that they entirely miss some of the material that is actually new to them. The Bjork’s refer to this idea in a Mr Barton Maths podcast that “perceptual fluency does not equal increased performance”. It’s nice to read about simple content, and students often prefer this, but it’s not an effective way to learn.
We can view the progression from the idea of simplification/clarity of presentation that CLT suggests to the challenge of desirable difficulties like the levels of a computer game. Students only carry on playing a game if the levels get more difficult because that’s where the enjoyment lies. But if a game is made too difficult to begin with then nobody learns how to actually play and then they soon stop altogether.
So we start with clear and simple explanations (CLT) and then just keep cranking up the challenge of the material (desirable difficulties)? It’s certainly one option and this feeds into our discussions of curriculum. How ambitious should we be? How much further than the syllabus can we go?
Are there other ways to increase the ‘difficulty’ of the subject without just making it more challenging? Desirable difficulties can also be achieved through retrieval practice and the linking of ideas across topics. We all understand the inherent challenge of retrieving old information – and we know how vital this process of recall is to learning. This can be another tool in ensuring that material never stays too simple.
I use CLT as the lens through which to view my explanations and the presentation of material (including the layout of question sets) to my students.
I use desirable difficulties as the lens through which to view which content to choose in the first place, and how questions ramp up in challenge (due to the Qs being more inherently difficult within the topic or by including the opportunity for retrieval of older content).
