Battling the Bandwidth of your Brain

Why some people think cognitive load theory might be the most important thing a teacher can understand.

Recently, there has been a surge of interest in cognitive load theory, perhaps aided by comments made by Dylan Wiliam on Twitter that it is ‘the single most important thing for teachers to know’ (Wiliam, 2017). So, what is cognitive load theory, how did it arise and what are the implications for teachers in the classroom?

The origins of cognitive load theory can be traced back to the results of an experiment published by John Sweller and his colleagues in the early 1980s (Sweller, 2016). In this experiment, students were asked to transform a given number into a goal number by using a sequence of two possible moves; they could multiply by 3 or subtract 29. Unknown to the students, the problems had been designed so that they could all be solved by simply alternating the two moves e.g. ×3, –29 or ×3, –29, ×3, –29.

The students who were given these problems were all undergraduates and they solved them relatively easily. However, very few of them figured out the pattern.

By that time, it had been established that people solve novel problems by the process of means-ends analysis: Problem-solvers work backwards, comparing their current state with the goal and looking for moves that will reduce this distance. Sweller wondered whether this process drew so heavily on the mind’s resources that there was nothing left to learn the pattern. In other words, solving problems induces a heavy ‘cognitive load’.

It has been known since the 1950s that our short-term memory is severely limited. In a classic 1956 psychology paper, George Miller argued that the maximum number of items that can be held in memory for a short period is about seven (Millar, 1956). However, an important question arises: what is an ‘item’? One of the tasks Millar examined was reciting a string of random digits, with each digit representing one item. Compare this with a string of digits such as, ‘SPIDERS’ – this is no longer seven items. Instead, it represents a single item because most people already possess a concept of what a spider is. An item is therefore the largest unit of meaning that we are dealing with and this will therefore depend upon what a person already knows. When we gain new knowledge – new meanings – we therefore reduce the number of items that we need to consider, a process known as ‘chunking’.

The concept of working memory is similar to that of short-term memory except that it doesn’t just store information, it also manipulates it. The limitations of working memory are what lead to cognitive overload.

We now know that different kinds of item impose different limits (Shriffin & Nosofsky, 1994). Words are generally more intensive than digits, cutting the short-term capacity further. Many cognitive scientists today accept a model of the mind that includes a ‘working memory’ (e.g., Baddeley, 1992). The concept of working memory is similar to that of short-term memory except that it doesn’t just store information, it also manipulates it. The limitations of working memory are what lead to cognitive overload.

Sweller’s initial experiments did not involve tasks that are educationally relevant and so a natural progression was to examine the kinds of problems that students are asked to solve in real academic courses. Working with Graham Cooper, Sweller tested whether school students and university students learned more by solving simple algebra problems or by studying worked examples. If Sweller’s hunch was correct, students may well be able to solve some of these problems, but the cognitive load imposed by this would lead them to learn little. Conversely, by imposing less cognitive load, the worked examples should lead to more learning. This was confirmed by the research (Sweller & Cooper, 1985) and this finding has now been replicated in many different situations involving a wide variety of subject matter (Sweller, 2016).

However, these results seemed counterintuitive and presented researchers with a conundrum. How is it possible for small children to pick up their mother tongue by simple immersion? Wouldn’t that lead to cognitive overload? If Sweller and colleagues were right, wouldn’t we need to give children worked examples of talking and listening in order for them to learn?

The answer to this problem may be found in the work of David Geary. His suggestion is that some forms of learning are ‘biologically primary’. Humans have presumably been speaking a kind of language for hundreds of thousands, perhaps millions, of years and this is long enough for evolution to have had an impact, equipping babies with a mental module for picking up language without conscious effort. In contrast, reading and writing (and all other academic subjects, for that matter) have been around for only a few thousand years and for much of that period, only a small elite engaged with them. They therefore cannot have been affected by evolution, rely on repurposing biologically primary mental modules and are therefore known as ‘biologically secondary’ (Geary, 1995).

Cognitive load theory suggests that all biologically secondary knowledge must pass through our limited working memories in order to be stored in long-term memory. For learning new, complex academic concepts such as algebra or grammar or the causes of the First World War – as opposed to learning simple lists – it is probably wise to try to minimise cognitive load by avoiding approaches that look like problem solving and to instead utilise those that provide clear and explicit, step-by-step guidance (Kirschner et al., 2006).

In the process of its development, cognitive load theory has also incorporated a number of learning effects that are related to the load that they impose. For instance, the ‘split-attention effect’ demonstrates that it is better to place labels directly on a diagram rather than provide an adjacent key because this avoids the need to cross-reference, which imposes unnecessary load. Similarly, the ‘redundancy effect’ shows that it is best to avoid adding unnecessary additional information for students to process. For example, if a diagram of the heart clearly shows the direction of blood flow then adding a label saying which way the blood flows is redundant (Sweller, 2016). This has clear implications for teaching – don’t provide lots of text on a PowerPoint slide and simultaneously explain the same concepts verbally. In general, it is best to minimise the number of different things that students have to pay attention to at any one time. Remove those fancy borders, animations and cartoons unless they are fundamental to what is being communicated.

And this is why cognitive load theory is so powerful. Unlike much of what we are told during training and professional development, cognitive load theory has real implications for teachers in the classroom that are based on sound evidence derived from robust research designs. Perhaps Dylan Wiliam is onto something. Perhaps cognitive load theory is an important thing for teachers to know.

Download a PDF version of this issue.


Baddeley, A (1992) ‘Working memory’, Science, 255 (5044) pp. 556–559.

Geary, D. C. (1995) ‘Reflections of evolution and culture in children’s cognition: implications for mathematical development and instruction’, American Psychologist, 50 (1) pp. 24–37.

Kirschner, P. A., Sweller, J., & Clark, R. E. (2006) ‘Why minimal guidance during instruction does not work: an analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching’, Educational Psychologist, 41 (2) pp. 75–86.

Miller, G. A. (1956) ‘The magical number seven, plus or minus two: some limits on our capacity for processing information’, Psychological Review, 63 (2) pp. 81–97.

Shiffrin, R. M. & Nosofsky, R. M. (1994) ‘Seven plus or minus two: a commentary on capacity limitations’, Psychological Review, 101 (2) pp. 357–361.

Sweller, J. (2016) ‘Story of a research program’, Education Review, 23.

Sweller, J. & Cooper, G. A. (1985. ‘The use of worked examples as a substitute for problem solving in learning algebra’, Cognition and Instruction, 2 (1) pp. 59–89.

Wiliam, D. (2017) ‘I’ve come to the conclusion Sweller’s Cognitive Load Theory is the single most important thing for teachers to know’ [Twitter], 26 January. Retrieved from