Cognitive load theory (CLT) is the concept that information should be presented at a pace and level of difficulty that corresponds to how the human brain processes information. When an instructor takes CLT into consideration, he or she purposely seeks to lower demands on the learner's processing capabilities in order to build comprehension and bring about more effective transfer of information into long-term memory (LTM). Cognitive load theory plays an important role in artificial intelligence (AI) research and is applicable to a broad range of teaching strategies for both machine learning (ML) and human cognition.
John Sweller, the educational psychologist who popularized CLT in the late 1980s, was interested in the idea that working memory is limited in the number of elements it can process at once. Although researchers do not agree on the exact number, several studies have shown that asking people to process more than seven elements simultaneously causes cognitive overload and some researchers insist the number is only four.
In his research, Sweller examined the concept of cognitive overload and identified three types of cognitive load that affect working memory:
- Intrinsic load - those elements that must be processed simultaneously.
- Extraneous load - those elements that require additional mental processing but do not add to the learning experience.
- Germane load - those elements that help the learner transfer information from short-term memory into long-term memory and vice versa.
When working memory is taxed, it can slow the learning process, cause confusion and lead to frustration. In fact, cognitive overload is one reason why some older people find learning a new technology to be so frustrating. Aging is known to decrease the efficiency of working memory, and if you impose that variable upon a person who has no frame of reference to help with germane load, the result is likely to be cognitive overload. Similarly, very young children are vulnerable to cognitive overload because they have very little information stored in long-term memory to provide new information with a frame of reference.
Intrinsic load is not something an instructor can control, but extraneous load can -- and should -- be minimized to free up working memory. In contrast, instructional techniques that facilitate germane load should be maximized to free up working memory. Recommendations include the use of concise introductory explanations, reinforcing new information both verbally and visually (graphically) and chunking content into segments to facilitate germane load.