I was lucky enough to be involved recently with a conference in Virchester that brought together some great clinical educators from across the world. During the conference, there was a lot of talk about cognitive load and cognitive offloading to free up the brain to think, but nothing on the theory itself (it wasn’t that type of conference) so I thought I would redress that in this forum.
Ed – you can read more about educational theories that alter your practice on the St Emlyn’s site here.
Before I start I need to make clear I am not a cognitive psychologist nor an expert in this field, but as a medical educationalist I have looked into this a fair bit, most recently in a fabulous book, Understanding how we learn: a visual guide , as I think it gives us some important things to think about when it comes to designing teaching sessions.
There are a few different theories on cognitive load but the main one for education is based on the work done by Sweller . Cognitive load is the load created by all the information that comes into our working memory through our senses, that needs processing.
Each one of us has a cognitive capacity, the amount of stuff we can store and process in our working memory, and whilst there is variation between individuals, it is fixed in size; you can’t increase cognitive capacity. Our cognitive load can never exceed our cognitive capacity otherwise cognitive overload takes place. You may have experienced this yourself when you are driving a car and get lost, or have to do a complex parking manoeuvre. Suddenly you have to think more about your driving, increasing your cognitive load and in order to cope, you turn the radio down.
How learning takes place
Before looking at the theory it’s worth revisiting how learning takes place. The cycle of learning begins with attention since if something doesn’t come to our attention we can never learn it. Attention is based on how interested we are in the subject matter (individual interest) and how interestingly the subject matter is taught or presented (situational interest) . We are also prone to mind-wandering. Your goal may be to read this blog (in cog-psych speak it’s called ‘task set’) but as you read, your mind may stray to other things like, ‘what’s shall I have for lunch?’ (This is unimaginatively called ‘loss of task-set). It’s been estimated that learners spend about 50% of their time during a teaching session mind-wandering  – just think of that for a second; half of every teaching session you do is wasted because no one is paying attention!
The information that has come to our attention then enters our working memory. Limitations to our cognitive capacity means we can only attend to between 5 and 9 items at once, for between 15-30 seconds. Now that doesn’t sound like a lot of information but our brains are very good at chunking information together to reduce this cognitive load (more of that in a second).
Our working memory now needs to process or encode the information. The most simple way of doing that is through repetition, such as when someone gives us a phone number and we repeat it over and over, whilst running around like a lunatic looking for a pen and paper. As you will no doubt have experienced, this is a poor way of processing and rarely works. The other, and more lasting, way of encoding this information is to attach it to previously learned information from our long-term memory to create new learning. For example, think about how we might learn the following 22 letters: FB CUR EAC REA TNA KGL UCL FT
Now we might try to go about remembering them by repeating all 22 over and over but more likely we will try to do it by chunking bits together; FB for Facebook (the 2 items ‘F’ & ‘B’ become 1 item ‘Facebook’), ‘CUR EAC REA’ could become ‘cure’ ‘acre’ ‘A’ (9 items become 3).
In fact, if we rearrange the spaces, those 22 letters become FBC UREA CREAT NA K GLUC LFT, so 22 items now become 7. You have used previously held information in your long-term memory to chunk and simplify the information, reducing your cognitive load. To a non-medic these would still be 22 random letters. So one individual’s cognitive load will be different from another’s depending on their previous knowledge.
Once this information is encoded it is passed on to the storage unit of the long-term memory where many more things happen, the details of which are beyond the scope of this blog, but essentially involves sleep consolidating the learning of the information and then the recalling of the information before it is lost. So let’s now look at the theory itself.
Cognitive load theory.
Cognitive load is made up of 3 separate loads; the intrinsic load, the extraneous load, and the germane load (there are different names for these depending on who you read but they amount to roughly the same thing).
The intrinsic load is the load created by the more factual information such as A > B and B > C. The more complex the information, the higher our intrinsic load. The less we know about a subject before learning it, the higher the intrinsic load will be and vice versa – so if you know a bit about CLT before you started this blog your intrinsic load is less than someone who knew nothing.
The extraneous load is the load created by the learning method. The more complex the method, the higher the extraneous load as we need to process it. So something like simulation will cause us a very high extraneous load as we work out what’s happening within the scenario and deal with the distractions, whilst a chalk ‘n’ talk workshop will cause us a lower extraneous load. Similarly, a busy slide set will have a very high extraneous load as will a poorly designed chart. Even my use of the abbreviation CLT in the paragraph above increased your extraneous load, as you had to work out what I meant (how many times do we see that in academic papers?).
Finally the germane load. This is the good load. This is the load created when we get the information and go “so that means if A > B and B > C then A must be > C“. This is where we make the connections and schemas, make deductions and apply what we already know to create new knowledge… This is learning at a higher level and it’s what we want.
Because both the intrinsic and extraneous loads are relatively fixed (by the complexity of the information, our previous knowledge and the learning method), and our cognitive capacity is fixed this means the space available for this type of higher learning will be determined by them.
Space for germane load = cognitive capacity – (intrinsic load + extraneous load)
Thus the higher the intrinsic and/or extraneous loads, the less space there is for the germane load and therefore higher level learning.
What does this mean in practice?
When designing a teaching session we should be aware of our learners’ cognitive loads so that we can maximise learning – a teaching session isn’t a mass data dump from educator to learner.
Think about learners’ pre-existing knowledge on the subject – what is it likely to be at their stage of training so we can build on that knowledge. We can bring people up to similar levels of knowledge for a session by using a flipped classroom model – read an article, a blog, or watch a (short) video before attending. There is also evidence that if the intrinsic load is either too low (too easy) or too high (too complex) learners will rapidly lose attention 
This is the biggy that we need to reduce. Concentrate on their learning needs (not our teaching wants), and deliver education in an interesting, thought-provoking and bite-sized way. How does the educational material look – it is easy on the eye or jarring, can it be followed simply or is a degree of thought needed to work out what is going on (“Sorry for this busy slide…”) .
Avoid unnecessary distractions where possible – teaching in a clinical environment is particularly challenging, as there is so much going on that the brain still needs to process (“Can you just sign this for me please…”) – can we educate at the bedside and then move somewhere else for discussion.
If we do need to teach a lot of complex information then we need to choose a learning method that is low in extraneous load, so simulation may be a poor choice for this however much you want to have a go (perhaps a chalk ‘n’ talk before and then using a simulation to apply it after a period of consolidation).
We need to maximise this and by using some of the above ideas we can free up space in our learners’ working memory for generative thought. Other ways we can help is to get them thinking – asking questions such as “what does this mean?”, “what’s your experience on this?” and “why does this happen?”. Allowing opportunities for discussion as opposed to teaching from the front makes our learners think and create the connections they need in order for higher level learning to occur.
If we want our learners to learn we need to think about the cognitive load we are placing on them with our teaching – as with everything in life, balance is important.
 Weinstein, Y., Sumeracki, M. and Caviglioli, O., 2018. Understanding How We Learn: A Visual Guide. Routledge.
 Sweller, J., Ayres, P. and Kalyuga, S, 2011. Cognitive load theory. Springer. New York.
 Hidi, S. and Harackiewicz, J.M., 2000. Motivating the academically unmotivated: A critical issue for the 21st century. Review of educational research, 70(2), pp.151-179.
 Smallwood, J., Fishman, D.J. and Schooler, J.W., 2007. Counting the cost of an absent mind: Mind wandering as an underrecognized influence on educational performance. Psychonomic bulletin & review, 14(2), pp.230-236.