Peter Imeson explains how understanding cognitive load theory and dual coding
has enabled him to make his teaching more memorable.
The human brain’s ability to store information in long-term memory is almost unlimited; a study carried out by American scientists judged the brain’s storage capacity to be one petabyte which is the equivalent of 20m four drawer filing cabinets full of text or 4.7bn books. With such a huge storage capacity it is a wonder that anyone finds remembering facts difficult. However, one of the problems we face is that whilst the storage capacity of long-term memory is vast, working memory, which process information into long-term memory, is severely limited. This article considers ways in which teachers can take account of the limits of working memory to make learning more effective.
Although estimates differ, typically working memory is thought to hold between 4 and 7 pieces of information. Working memory capacity is heritable i.e. has a genetic basis and students with learning difficulties often have even more limited working memory capacity. Working memory can therefore be thought of as a bottleneck in the functioning of memory as shown in the diagram below:
Since working memory is so limited it is easy for it to become overloaded. When this happens, it can prevent students from learning as it can mean that they fail to recall verbal instructions, miss out on important steps in a set of instructions and fail to understand complex explanations. This is the essence of Sweller’s (1998) cognitive load theory that Sam Stones and Kathy Cameron wrote about in the Autumn 2020 edition of this journal. Despite the claims of many ‘brain training’ programmes, there is little evidence that anything can be done to increase the storage capacity of working memory (e.g. see Sala and Gobet 2017). However, by understanding how working memory works and designing instruction with its limitations in mind, it is possible to teach in a way that overcomes these limitations.
The diagram below is a model which represents the current understanding of working memory, though some parts, especially the ‘central executive’, are less well understood. The important aspect, for the purposes of this article, is the fact that the ‘phonological loop’ which stores words is a separate storage system to the ‘visuospatial sketchpad’ which processes visual images.
Knowledge of the different stores within working memory presents a useful insight into why working memory can get overloaded in a lesson and how it can be overcome. A common practice used in classrooms is to present written information on Powerpoint slides whilst explaining it verbally. The problem caused by doing this is that both the written information and the verbal explanation are processed by the phonological loop which is unable to cope with the quantity of information it is being presented with. An easy fix to overcome this problem is to present the information on the Powerpoint slide as visual images whilst giving a verbal explanation – this is the process of ‘dual-coding’. Doing this reduces the cognitive load as the images and the spoken words are being processed by different parts of working memory. The added benefit to this is that the images can perform a mnemonic function as images are remembered better than words.
Pre-existing knowledge and schema
An interesting set of experiments was carried out in the 1940s by the Dutch psychologist Adrian de Groot in which he tested the memory of chess grandmasters. De Groot wanted to understand what differentiated the grandmasters from other chess players and had a theory that it was due to their superior memory. In one experiment, he arranged twenty pieces on a chess board in a pattern that would be found during a game and asked players of varying abilities to glance at the board and then recreate the positions from memory. The grandmasters easily identified the exact positions whereas the amateurs struggled, which seemed to confirm his theory that the difference was the superior memories of the grandmasters. However, a second experiment was conducted in which the pieces were scattered randomly across the board rather than in a pattern consistent with gameplay. This time, the grandmasters performed no better than the amateurs. The conclusion de Groot drew was that the grandmasters used their pre-existing knowledge of gameplay to ‘chunk’ the information in a way that overcame the limitations of working memory. Rather than seeing the individual pieces, pawns, bishops, etc, they saw patterns – each pattern containing many pieces became a chunk in working memory allowing it to deal with much larger amounts of information. David Didau presents the following example in his book ‘What Every Teacher Needs to Know About Psychology’:
Try to remember the following string of letters:
XKD JSD PRL MXG LWK
Compare that to this:
BBC FBI ATM TLC CIA
Whilst the two tasks involve remembering the same number of letters, the first is very challenging as it is beyond the limits of working memory whereas the second is much more straightforward as the fifteen letters now become five chunks. Note though the importance of existing knowledge; someone who had not heard of the BBC or the FBI would not see the letters as chunks and therefore would find the second task no easier.
Psychologists refer to the pre-existing knowledge we have about a topic as ‘schema’. When we start to learn a subject, we have a schema which is under-develop; perhaps a few unconnected facts. As we learn more, the schema becomes larger and more interconnected, containing more knowledge and an understanding of how different elements of the knowledge relate to each other. When we learn new facts, they become integrated into the schema and the larger, richer and more detailed our schemas are, the easier it is to learn new information as there are more ways in which the new knowledge can become connected to our existing knowledge. It is very easy to forget this as teachers: we are experts on our subject and therefore have well developed schema; our students do not and therefore it can be difficult to remember how challenging it is for students to understand the concepts we are teaching them, the so-called ‘curse of knowledge’. This is especially true for those concepts that are abstract and far removed from students’ personal experiences. The important point is that the more we can use effective techniques to build up students’ knowledge, the better they will be able to learn the new information we teach.
Some teaching suggestions
Learning about cognitive load theory and the practice of dual coding has made me re-think how I deliver information to students. The main task I have given myself is changing my Powerpoints to remove the majority of text and substitute it with images. Below is an example of changes I have made for delivering the topic of sole traders and limited companies to my Y10 GCSE classes. Talking through the top two slides is very likely to lead to a cognitive overload for students as there is a large amount of information to take in where both means of delivering the information are using the same part of working memory: the phonological loop. In the new versions the visual information on the slides will be processed by the visuospatial sketchpad whilst my verbal explanation is processed by the phonological loop. Following the explanations, I would then get students to recall the information, thereby utilising the technique of ‘retrieval practice’ which research shows to be a very effective means of retaining information (e.g. Bjork and Bjork 2020). In subsequent lessons further opportunities for recall can be given, a technique known as ‘spaced-practice’, sometimes using the images e.g. asking students to write a paragraph explaining advantages and disadvantages of the different business types based on the slides. The use of images, retrieval practice and spaced practice will enable students to build up a stronger schema that will then make learning new related information, such as sources of finance, easier and more effective.
Using images can be particularly effective when explaining a complex abstract process. Above is an example of a slide I used to explain the process of short-selling when discussing the recent events involving Gamestop with my economics classes.
Here having the process represented as a diagram can help students to visualise a process which can be very difficult to understand as it is far removed from their personal experiences. I have used similar types of diagrams for topics such as the operation of quantitative easing and the transmission mechanism of monetary policy.
The information given above is based on solid research which is becoming better known and more widely disseminated among teachers; however, the examples given are my interpretations of how this knowledge of psychological processes can be utilised in a business and economics classroom setting. I am sure that other teachers will have applied the same research findings in different ways and I would encourage teachers who are interested in the psychology involved to do their own reading. The sources that I have found particularly useful include:
The Learning Scientists: an excellent blog which applies cognitive science to classroom settings: https://www.learningscientists.org/blog
The research of Robert Bjork who has pioneered much of the study of techniques such as retrieval practice and spaced practice: https://bjorklab.psych.ucla.edu/robert-a-bjork-publications/
David Didau’s book ‘What Every Teacher Needs to Know About Psychology’.
Peter Imeson is Head of Business Studies at Farmor’s School