Understanding cognitive load theory
The same year that the first Eurovision Festival was celebrated in 1956, a man named George A. Miller wrote one of the most cited articles in psychology. “The Magical Number Seven, Plus or Minus Two” deduces and explains that the capacity of our brain to process new information is limited..
“Average people can only hold 7±2 items in their working memory.” — Miller’s Law
Based on Miller’s work, John Sweller developed the Cognitive Load Theory in the late 1980s from his observations while studying problem-solving processes. Sweller continued to work on this theory and publish his advances during the 1990s, inspiring many other researchers to delve deeper into the subject. Although today the model has evolved, including more concepts and sophisticating the approach with new published research, the basic principles remain useful and are included within Psychology and related Cognitive Sciences (Howard’s Cognitive Hexagon is another article).
What the theory says
We know that the human brain works with two types of memory: long-term and short-term. Long-term memory preserves highly structured information for an indefinite time. However, short-term memory, or more correctly called working memory, receives unstructured information from the senses and has the task of processing it to give it a structure that allows its comprehension and storage in long-term memory, which constitutes the learning process.
This working memory is limited, and traditionally it is said that it can store up to seven pieces of information, although there are studies that defend other figures or even that it will depend on the type, complexity, and our familiarity with the elements.
When long-term memory establishes a structure for certain content, the processes associated with the information that fits with that structure can be automated, requiring less effort. As the individual manages to establish a variety of mental schemas, he acquires the ability to quickly structure the information he perceives through the senses, so that the understanding and analysis of this information is automated. At the moment this happens, even complex information, if structured, counts as a single element in working memory, relieving its limitation. We refer to this phenomenon as “gaining experience.”
We know that the process of learning, assimilating information or analyzing, can be facilitated if the information is dosed and grouped with an obvious sense that facilitates the creation of a new mental schema.
Another interesting detail is that working memory is independent for each sense, so it is also possible to relieve cognitive load if we distribute the information in various media (for example, visual and auditory).
In summary, to reduce interference in the learning or analysis process, it is necessary to present the information in such a way that attention and effort are devoted to the ultimate goal, not to precursor activities.
“The cognitive load theory has been designed to be a guide that assists in the presentation of information in a way that favors learning activities by optimizing intellectual performance.” — John Sweller
A quick search on Google Scholar returns countless articles to discover more about the cognitive load theory.
Disorders related to cognitive load
In some people, working memory and related processes may be affected by disorders that make it difficult to process information received through the senses. For example, we know that in dyslexia, ADHD, and SLD (specific language disorder), phonological or visual capacities are altered, making it difficult to process information.
In these cases, the effort required to assimilate new information received through the affected channels can be so high that it compromises the attention that the person can dedicate to a task. These are cases where adherence to good practices is not a matter of comfort but accessibility
Biological and evolutionary bases
A curious detail is that the process of analysis or learning has some exceptions. Several contexts have been identified in which the human brain has evolved to learn autonomously and unattended, such as psychomotor development, face recognition, or articulating sounds to speak. These cases are called biologically primary knowledge.
The opposite is true in the acquisition of other skills, for which, despite being relevant to our social, intellectual, and labor competencies, mechanisms that automate their learning have not evolved. These are called biologically secondary knowledge.
A hypothesis being investigated is to use processes associated with primary knowledge to facilitate the acquisition of secondary knowledge, such as promoting collaborative learning.
Implications NOT of this theory
Potentially confusing misinterpretations have arisen from the works of Miller and Skewer. Research on cognitive load focuses on understanding the process of learning and analysis, and its principles can be applied to design educational materials or structure complex information in a way that facilitates its analysis. However, it is entirely irrelevant in situations where the information remains available to the user
It’s absurd to limit the navigation of a website or a list in a PowerPoint to seven elements just because “Miller’s Law says so” when we are actually talking about data that is always in view and therefore does not need to be memorized. To establish an acceptable limit, it´s necessary to research user needs and usage contexts of each situation. In fact, navigating a menu whose grouping has been forced or artificial with the sole purpose of reducing the number of elements in each level generates confusion and disorientation due to the cognitive load involved in the task of finding meaning in the jumble.
To see this and other myths related to cognitive load dismantled in detail, I invite you to visit Andrea Cantú’s blog.
Best practices in digital product design
Avoiding the previous myths, it is true that by correctly understanding the theory of cognitive load, we will be better prepared to design more productive digital tools. Some points to remember are:
- Understand: we must research not only the challenge we will be designing for, but also what expectations future users have, and let them guide our design decisions.
- Structure: determining the most appropriate structure for presenting information according to its typology and the user’s familiarity with it. The goal is to facilitate the automation of learning and analysis processes, achieving that “experience acquisition” with the least friction as possible.
- Coherence: the information structure should be consistent over time and in all materials, so that exposure to this structure is maximized and assimilation is facilitated.
- Combining channels: distributing cognitive load using more than one communication channel.
- Dosage: partitioning information into small, coherent packages.
