Cognitive Load Theory – every teacher should know this!

Cognitive Load Theory

It was around 18 months ago that I first came across Cognitive Load Theory (CLT). Recently CLT has gained a lot of traction on social media; helped by the fact that Dylan Wiliam cited it as the most important thing for teachers to know earlier this year:


Oliver Caviglioli also recently created one of his fantastic illustrative summaries on Sweller’s book and this reignited the CLT flame for me. A few weeks ago I posted about CLT on the Society for Education and Training’s Blog, in an attempt to further promote what I and many others consider to be an essential learning theory. I thought I’d share it on my blog in an attempt to reach a few more practitioners, so here it is:


What is the one learning theory that I feel all teachers should be made aware of?

Cognitive Load Theory (CLT) – Coined in 1988 by John Sweller, this theory posits that our working memory is only able to hold a small amount of information at any one time and that instructional methods should avoid overloading it in order to maximise learning (Sweller, 1988).


Why have I chosen this theory?

We’ve all been in learning sessions where the teacher has whizzed through the content, leaving us with little to remember. We’ve also been in those sessions where the content is so complex that we leave more confused than when we entered. CLT goes some way to explaining why this happens and what we, as teachers, can do to maximise the learning of individuals within our classrooms.
Building on the work of Baddeley and Hitch (1974), CLT views human cognitive architecture as the working memory and long term memory. Put simply, the working memory has a limited capacity and consists of multiple components that are responsible for directing attention and coordinating cognitive processes. Long term memory on the other hand, has an endless capacity for storage and works with working memory to retrieve information (Baddeley, 2003).

What can teachers do to reduce cognitive load?

  • Activate prior knowledge before sharing new information with students – Our long term memory is said to have a number of organised patterns of knowledge (known as schema). Each schema acts as a single item in working memory, so can be handled easier than having lots of new, isolated information. Through retrieving information from the long term memory via quizzes, visual aids and discussions, students can bring crucial information to working memory (see image 1) and assimilate new information to build upon what they already know (Baddeley, 2003). Activating prior knowledge is also supported in the work of Marzano, Gaddy and Dean (2000), who found a substantial improvement in achievement (0.59ES). Furthermore, retrieval practice has shown to strengthen our retention of the information (Wenger, Thompson and Bartling, 1980) – a win win!

  • Use visual and verbal information to present information to students – This has nothing to do with the infamous ‘learning styles’, rather empirical research suggests that our working memory has two points of entry (Chandler and Sweller, 1992). One accepts auditory information, whilst the other visual. If the auditory and visual information correspond to one another, then the burden on working memory is far less than using one pathway alone. Image 1 shows the effect of using one pathway to working memory, whereas image 2 shows the use of both. However, please note that if the text and visual information are not clearly integrated, then it could have adverse effects on learning (Chandler and Sweller, 1992).

  • Use worked examples and models to support learning – There are a wealth of studies that have shown the positive impact of using worked examples to enhance learning (Chandler and Sweller, 1991). According to Clark, Nguyen and Sweller (2006, p.190), ‘a worked example is a step-by-step demonstration of how to perform a task or how to solve a problem’. These steps provide learners with direction and support to create mental models of how to tackle a problem/task, or what ‘good’ looks like. Discovery or problem-based learning on the other hand can be burdensome to working memory due to learners having insufficient prior knowledge to draw upon to support their learning. Moreover, the vast amount of information they have to consider in completing work independently can result in a struggle to direct their attention. As learners develop a greater understanding of the topic, elements of the worked or modelled examples can be ‘faded’ (removed) to foster greater independence.

Worked Example


In summary, regardless of one’s philosophical predisposition, I argue that all teachers need to have an awareness of the potential benefits and limitations of the ways in which they present learning opportunities for learners. CLT and the associated empirical research provides us with an understanding of how we process, organise and store information most effectively and for this reason, all teachers should acquire a basic understanding of the premise.

Blog by @FurtherEdagogy

Baddeley, A.D. (2003). Working memory: looking back and looking forward. Nature Reviews Neuroscience, 4, p.829-839.

Baddeley, A.D. and Hitch, G. (1974). Working Memory. Psychology of Learning and Motivation, 8, p.47-89.

Chandler, P. and Sweller, J. (1991). Cognitive Load Theory and the Format of Instruction. Cognition and Instruction, 8 (4), p. 293-332.

Chandler, P. and Sweller, J. (1992). The split-attention effect as a factor in the design of instruction. British Journal of Educational Psychology, 62 (2), p.233–246.

Clark, R.C., Nguyen, F. and Sweller, J. (2006). Efficiency in learning: evidence-based guidelines to manage cognitive load. San Francisco: Pfeiffer.

Marzano, R.J., Gaddy, B.B. and Dean, C. (2000). What works in classroom instruction. Aurora, CO: Mid-continent Research for Education and Learning.

Sweller, J. (1988). Cognitive Load during Problem Solving: Effects on Learning. Cognitive Science, 12, p.257-285.

Wenger, S.K., Thompson, P. and Bartling, C.A. (1980). Recall facilitates subsequent recognition. Journal of Experimental Psychology: Human Learning and Memory, 6 (2), p.135-144.

Are these the 7 pillars of classroom practice?

A really useful summary of research from the TES via Nick Rose

12th May 2017
Not all research is perfect – or even worth reading, argues Nick Rose. But here, he presents a septet of reliable reviews that time-pressed teachers can bet their house on being useful in the classroom

There are two main problems with the idea that teachers should be reading research. The first is that they don’t have time to do it (see Studies by the Education Endowment Foundation (EEF) have examined methods of encouraging teachers to engage with academic research but report that time pressures – and a lack of opportunities for teachers to work together – appear to undermine the initiatives.

So, if we genuinely want teaching to be a research-informed profession, we should reduce the number of hours we expect teachers to work – and ultimately set aside some time for teachers to read and apply research to their practice.

The second problem is actually finding research related to the questions we have as teachers. Searching through academic journals can be a frustrating and often fruitless investment of time and effort. Too often, even when something looks promising, it’s rare to find clear guidance on how it might be successfully implemented in our classrooms. So instead we become rather passive consumers of the latest “big idea”, which is sometimes presented to us like a magic bullet for all our woes.

One example would be the much misunderstood concept of growth mindset. Following the bestselling book and supportive studies on the subject, many schools enthusiastically began delivering growth mindset-inspired lessons and assemblies. But in contrast to the sometimes heavy-handed approach seen in schools (“growth mindset, good; fixed mindset, bad”), the successful interventions in the studies involved carefully targeted, often imperceptibly subtle and deliberately short-lived messages. This makes them extremely difficult to replicate at scale, though you wouldn’t guess that from the extraordinary (and unjustified) influence the “mindset meme” has had in schools. Even Carol Dweck, the author of so much of the research, has been critical of the way many schools have attempted to apply her ideas (see

It’s not just schools that get taken for a ride on the “latest research” bandwagon. A recent publication in Harvard Business Review purported to claim that only one type of headteacher can turn around schools and that they tended to have a history or economics background. This story was picked up by a number of news outlets, despite the fact that the HBR piece contained no details regarding the methodology. Analysis by Education Datalab tried to replicate the findings using publicly available data and found no relationship between subject background and lasting improvements to schools (see

This might seem an odd introduction to an article encouraging teachers to engage with research. But despite my concerns, there are some research reviews that I think all teachers should know about. None of them represents magic bullets, but each of the seven examples on the following pages have been written for teachers and have implications for school leadership and classroom practice.

In a field of research plagued with the problems I have outlined and more, these are the seven research reviews you could bet your house on being useful and reliable.

Coe R, et al (2014) What Makes Great Teaching? Review of the underpinning research, Centre for Evaluation and The Sutton Trust

Early in my career, I became frustrated by things I was expected to plan into my teaching that had no evidence base indicating that they actually helped my students to learn. Aspiring to become a more evidence-informed practitioner doesn’t mean giving up professional judgement – quite the opposite, it means challenging and cultivating that judgement so that even more of our students can experience success in school.

As well as reviewing a range of evidence-supported approaches to teaching and professional feedback, the CEM/Sutton Trust report What makes great teaching? also identifies a number of ineffective practices – things teachers and school leaders almost certainly should stop doing.

Some teachers might think this sounds a bit negative, but I suspect most will see the advantage of identifying things that might be adding to workload and stress, while not helping students.

Some of these ineffective practices will likely be familiar to readers of Tes: things such as planning different resources and activities for different learning styles ( or grading lesson observations ( But others – such as trying to address issues of confidence before you teach content; the limitations of group work; or the mistake of thinking active learning must involve physical or social activity, rather than reading or listening – might surprise some readers.

Nisbett RE (2013) “Schooling makes you smarter: what teachers need to know about IQ”, American Educator

Carol Dweck’s research into mindsets is intimately tied to people’s understanding of intelligence. Perhaps one of the reasons why schools struggle to implement it effectively is because many teachers might not have a clear understanding of the complex relationships between genetics, IQ and education ( Science rejects the deterministic view that intelligence is fixed at birth, but equally doesn’t support the naïvely optimistic “you can do anything if you just try hard enough” rhetoric that Dweck herself has criticised.


In “Schooling makes you smarter”, Richard Nisbett provides a teacher-friendly summary of decades of research into intelligence. For example, many teachers might not be aware that two types of intelligence are measured by IQ tests: fluid intelligence, which is related to working memory capacity (see point 4, page 40) and involves things such as being able to solve problems in the absence of much prior knowledge; and crystallised intelligence, which consists of the store of knowledge (facts, concepts, vocabulary) we possess. Both types can be increased (

Another counter-intuitive point is that we can use the heritability of intelligence (the extent to which variation in intelligence is influenced by genetic differences) as a measure of social equity.

A high heritability means genetic rather than social and economic differences (eg, quality of diet or education) are the main influence in the differences between people’s intelligence. Where heritability is low, social and economic differences are likely the strongest influence. Schools and teachers cannot tackle all of these societal inequalities on their own, of course, but great schools and great teachers play a vital part in helping children reach their full intellectual potential.

Dunlosky J (2013) “Strengthening the student toolbox: study strategies to boost learning”, American Educator

Since Hermann Ebbinghaus’ pioneering studies in the 19th century, psychologists have recognised forgetting as an important part of learning. Students will forget much of even a brilliant lesson, but research has identified that forgotten material is not truly lost – students re-learn it quicker and retain it for longer each time they return to it.

More recent research by Robert Bjork identifies an important difference between the retrieval strength of a memory and the storage strength ( Put simply, the ease with which a memory can be accessed or retrieved (usually with the assistance of contextual cues or recency of learning) is different from the more permanent, entrenched trace in long-term memory. This often undermines a student’s “judgement of learning”. For example, a student re-reading some revision notes might mistake the feeling of familiarity with the material with their ability to genuinely recall the material without the cues and context provided by the notes.

John Dunlosky presents a straightforward summary of the research into study skills and independent learning. A key finding is that, rather than re-reading or using techniques such as highlighting, testing your memory tends to help us learn new material better. Frequent but low-stakes quizzes and tests are a great way to model the techniques that students can use in order to revise more effectively.

Clark RE et al (2012) “Putting students on the path to learning: the case for fully guided instruction”, American Educator

A foundational model for learning involves the relationship between long-term memory, which has a vast capacity for storing facts and concepts organised into what psychologists call “schemas”, and working memory, which has a very small capacity and relies upon these schemas to organise information into meaningful chunks in order to process new learning effectively. This tried-and-tested model underpins the modern science of how we learn and has a number of important implications for teaching.

We’ve evolved, and are motivated, to learn some things readily through play ( But the hard-won cultural and scientific advances of the last few thousand years didn’t (and don’t) come easily. For example, while our first spoken language is naturally learned without effort or instruction, reading is a relatively recent cultural advance that, though building on the core ability of spoken language, benefits from explicitly being taught.

“Putting students on the path to learning” explores how easily working memory can become overloaded when learning new, ideas. Carefully structured sequences of learning allow the foundations of a subject or topic to be consolidated in long-term memory; this helps students to later go on to tackle complex and novel problems.

Gadsby B (2017) Impossible? Social mobility and the seemingly unbreakable class ceiling, Teach First

Decades of research into human motivation provides lots of examples showing that tackling attitudes doesn’t necessarily change behaviours. For example, attitudes towards organ transplants don’t appear to affect willingness to join a pool of potential donors; and attempts at encouraging smoking cessation through direct persuasion rarely changes people’s smoking behaviour. This means that the way some schools approach mindsets will likely have little influence on pupil outcomes.

Changing behaviour is really hard. Think about occasions when you’ve tried eating more healthily, taking more exercise or using your car less. You may readily agree that it’s important and you really want to do it, but it doesn’t necessarily mean you’ll successfully change your behaviour without support.


Another example of this is the idea that students, especially those from low-income backgrounds, have low aspirations that schools need to raise if they are to succeed. The EEF summarises the impact of aspiration interventions as “+0 months”, indicating that, on average, interventions that aim to raise aspirations fail to produce a positive effect on student outcomes.

The evidence suggests most young people actually have high aspirations, and that underachievement stems not from low aspiration but from a host of social and institutional barriers that hold them back (see page 50).

The recent Teach First campaign Challenge the Impossible examines the available evidence regarding what it calls the “class ceiling” in an accessible way and explores the range of ways that this affects the life chances of young people. Good grades are necessary, but not sufficient, it appears; students from poorer backgrounds face hurdles at every phase of education – from early years to apprenticeships or postgraduate study.

Rather than thinking of the attainment gap as being caused by some deficit of character, aspirations or mindset, it might be more helpful to consider the range of barriers that undermine the aspirations and attainment of young people from low-income backgrounds (

Rosenshine B (2012) “Principles of instruction: research-based strategies that all teachers should know”, American Educator

One of the problems with the latest research is that the conclusions are necessarily tentative and there’s a good chance that the next researcher might identify something that contradicts it. This leaves teachers with a problem when trying to identify evidence-informed approaches to developing their teaching: is it worth embarking on something involving lots of time and effort, only to discover that researchers change their minds in a year’s time?

One way around this problem is to look for findings that have been triangulated. For example, if we find an outcome in well-controlled, but quite artificial, laboratory experiments and we find the same result in authentic classroom settings, despite all the noisy variables involved, then that likely makes it a good bet to try to implement it in your classroom.

In this paper, Barak Rosenshine reviews different bodies of research, including cognitive science and classroom studies, to identify where the science and practice appear to tell us the same thing about how we might take a research-informed approach to improving our teaching. If you have time to read only one research summary this year, I would recommend this one.

Deans for Impact (2015) The Science of Learning

In contrast to claims by some educationalists, advancements in information technology haven’t rendered “knowing” obsolete: children learn by connecting new learning to what they already know. But “knowing stuff”, while a necessary foundation, does not mark the limit of our ambitions for student learning; we also want children to be able to apply their learning to novel situations and problems.

Since Edward Thorndike’s experiments in the 1920s, psychologists interested in education have considered the issue of transfer. The problem involves how individuals learning in one context often fail to transfer learning to another, similar context. For example, as a science teacher I might find that students struggle to mark out an appropriate scale on the axis of a graph, but discover that the same students appear to do this task quite well in their maths lessons.

The issue is that they know this stuff in maths, but don’t readily or automatically transfer this knowledge to apply it in the different context of my lessons.

The question of how to help students transfer their learning to new situations, whether inside or outside the classroom, is one of the areas summarised in the Deans for Impact Science of Learning review. If you encounter the same issue as I’ve had when trying to encourage students to apply and transfer their knowledge and skills, the practical suggestions in this review are worth experimenting with.

Nick Rose is a former leading practitioner for psychology and research and now works as a researcher for Teach First. He tweets @Nick_J_Rose

The 5 worst education arguments by graphics

Well worth a 5 minute read!

Below are the graphics and images that we see trotted out time and again to support various arguments in education. And all of those arguments, if based on these graphics as evidence, are most likely bad arguments. I’m not the first to point out how these have been debunked, and I’m sure I won’t be the last.

1. This cartoon as an argument against an academic curriculum:5232012052424iwsmt

In an age where almost everyone in education is against the looming government policy of expanding the grammar school programme, there exists a cognitive dissonance whereby many of the same anti-selection advocates also embrace the idea that pupils should be assessed differently/be given a vocational education up to the age of 16 because ‘not every child is academic’. Michael Fordham skewers the argument for introducing a vocational education earlier far better than I can here, but suffice to say, this seems like just another form of academic selection to me. One of the other troubling things about this image is that it equates children with different classes of animal. I think that children have more in common with each other than they do differences, and certainly not differences of such extremes represented by this image. Of course, there will always be a minority of pupils who do have more extreme physical or educational needs, and we should meet those needs, but this doesn’t involve changing the approach to education for the majority of pupils and holding these pupils to different standards by shutting them off from an education they might later wish they’d had the opportunity to acquire.

And if you see the related quotation, ‘Everyone is a genius. But if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is stupid.’ attributed to Albert Einstein, you might want to point out that they are misattributing it.

A good counter for this argument is – and this is something Einstein did actually say – a speech given to the State University of New York in 1931, ‘On Education’:

“I want to oppose the idea that the school has to teach directly that special knowledge and those accomplishments which one has to use later directly in life. The demands of life are much too manifold to let such a specialized training in school appear possible […] The school should always have as its aim that the young man leave it as a harmonious personality, not as a specialist. This in my opinion is true in a certain sense even for technical schools, whose students will devote themselves to a quite definite profession. The development of general ability for independent thinking and judgement should always be placed foremost, not the acquisition of special knowledge. If a person masters the fundamentals of his subject and has learned to think and work independently, he will surely find his way and besides will better be able to adapt himself to progress and changes than the person whose training principally consists in the acquiring the detailed knowledge.”

Oh, and all this is compounded by the fact that, if it weren’t for fish getting out of the sea and climbing trees, we probably wouldn’t have evolved to be here now, arguing about a silly cartoon. Good job fish! A* in that exam!

2. This graphic as an argument against teaching systematic synthetic phonics:blog

In fact, this is a good argument for teaching phonics. It shows that there are clear correspondences between graphemes and phonemes and it highlights that there are rules for these correspondences – rules which the graphic ignores. One of the rules, for example, is that gh is never pronounced ‘f’ at the beginning of a word. Likewise, we’d never pronounce the grapheme ti as ‘sh’ at the end of a word. So if you see someone using this as an argument that English isn’t a language that can be taught through phonics, tell them that it proves just the opposite. You just have to know the rules.

(And if they attribute the ghoti/fish idea to George Bernard Shaw, you can tell them that they are doubly wrong.)

3. This graphic as an argument against direct instruction:2882821_orig

Pyramids are like catnip to teachers. Present anything in a pyramid and we are sold. If you don’t believe me, put a Toblerone and a bar of Dairy Milk in the staffroom on Monday and see which gets finished first. This particular pyramid – often referred to as Dale’s cone – is made even more enticing as it has lovely numbers and abstract concepts spewed all over it. And it often has a nice citation of its source at the bottom: National Training Laboratories. Mmmm, laboratories. It’s all just so… sciencey. But it really isn’t.

Before I get on to provenance, the first thing one should notice is the numbers. Such perfectly rounded numbers going up in such neat increments. When was the last time you saw some research produce numbers like this?

Yes, Dale’s cone is pretty much made up. It began life as a simple idea from the American educator Edgar Dale in 1946, as an intuitive model of how different media effect us. This original model didn’t include any numbers, and Dale himself even warned people not to take the cone too seriously.

The numbers were later added by an employee of the Mobil oil company, in a (non-academic) article he had written.

And as for the National Training Laboratories? Well, when they were asked about its provenance, the reply came back that “we no any longer have – nor can we find – the original research that supports the numbers.” Harumph.

Don’t listen to anyone using this as evidence for anything. Well, anything other than evidence for how much we all love pyramids. And even then, I’d always go with a tray piled high with Ferrero Rocher instead.

4. This comparison as an argument for education reform:

This is often used to point out that the structures and rules of schools are wrong. The problem is that all institutions have structures and rules. Hospitals have dress codes and, the last time I was in one, I was glad the doctors didn’t defer any decision making to me. I trusted them with my health. And I was glad of the silence and order too. I was glad that they had set times for visiting hours and people didn’t wander in and out as they pleased. You know, like a prison.

You see, we could easily draw attention to similarities between all sorts of institutions based on these structures and rules. If you worked at the Magic Kingdom in Disney World, you’d find much of those lists above structuring the way the place is run. Are we suggesting there should be reform there? Unless you want chaos in the park, a lack of safety on the rides and the guy in the Mickey Mouse suit turning up to work drunk, then you are going to need those rules and structures strictly adhered to. Is Magic Kingdom like a prison? No. And neither are schools. Schools have many other things that prisons don’t have, such as gates that kids can walk out of once their relatively short day stuffed full of learning and wonder is over.

This is a daft argument and is not the basis for school reform. I’m happy to listen to arguments for school reform, but this isn’t it. This is an argument that the Texas sharpshooter fallacy is alive and well.

5. Contextless EEG images of brain activity to claim positive effects of an intervention or activity:

I don’t know much about neuroscience, and I’d never claim to. The thing is, neither do neuroscientists. Of course, they actually know staggering amounts about the brain, but what I really mean is that neuroscience is such a developing science that we (them, not me) are only just beginning to understand about the brain.

Unless these sorts of images above are being used by a neuroscientist, I’d be extremely skeptical of what that person is saying with them. As suggested on the excellent Neurobollocks blog here, these sorts of images need more than just a picture of two brains with different colours on them. EEGs can measure lots of different types of brain activity and unless they tell you what it is measuring, the image is pretty useless. What’s more, you’ll also need to know what the colours represent to understand what it is suggesting. I think these images largely play on our instinct that ‘more red is good’, whether it is or not. This is not to say that the original EEG didn’t have meaning. But if it is presented without this important information, that meaning is lost.

These sorts of images are mainly used to tell us that something (the thing someone is ‘selling’ us) makes our synapses ‘fire’. The problem here is that everything causes synaptic activity. Your synapses are firing reading this. But that doesn’t stop even top academics misunderstanding or misusing such images.

I’d steer clear of laypeople using these sorts of images to present an argument. Or at the very least, ask them a couple of questions. Ask them exactly what the images are representing: what particular brain activity is being measured and what the colours represent. And for further help, you could draw it to the attention of public skeptics of pseudo-neuroscience such as @neurobollocks.

By the way, the brain on the right in those images is actually an EEG of your brain when you read my blogs. I took the image through your webcam just now. Honest. Why would I lie to you?

Interleaving and spaced practice..

Lots of this on Twitter and blogs recently, certainly worth looking at our schemes of work to incorporate. I have pout a few examples below:

Spread sheet for teachers –