Defining Cognitive Load Theory (CLT)

Cognitive load refers to consumed amount of working memory resources; it is one of the key areas of cognitive psychology mainly dealt with learning and teaching.

According to Sweller. J [1-2], Cognitive load theory is an instructional strategy which deals with information processing of learning structures consisting long term memory and working memory. The former one stores all the knowledge and skills whereas latter involves in intellectual tasks related to consciousness. 

Feinberg and Murphy [3] opined that the instructional module must be the instruction itself as the basic assumption in CLT by designing web-based instruction that enhance working memory and reduce cognitive load.  As the process of acquiring and skills are greatly impacted when the working memory is overloaded. Ultimately, long term memory for storing knowledge and skills can be strengthened by presenting learning effectively to our working memory.

Pass et al. [4] viewed Cognitive Load Theory as a functionality of schemas of unlimited long-term memory in interaction with visual and auditory information in a limited working memory. It primarily focuses on the design of instructional strategies which help people to apply acquired knowledge and skills to new situations within their limited cognitive processing. 

In the words of De Jong [5] ‘cognitive load theory asserts that learning is hampered when working memory capacity is exceeded in a learning task.

Historical Accounts of CLT

During 1950s, glimpses of CLT began to appear with the earlier beginnings of cognitive science as an emerging field. Miller, George A. [6] in his classic paper titled ‘The magical number seven, plus or minus two: some limits on our capacity for processing information’ suggested inherent limits of our working memory that is 7 plus or minus 2 units of information could be hold in short-term memory by citing experimental evidences. So, Miller, G.A is perhaps said to be the first to have experimented in the direction of Cognitive load. 

Later, in early 1970s Psychologists such as Chase, William G. [7], Simon, Herbert A. wrote a paper on “Perception in chess” used the term ‘chunk’ for the first time to refer how we organize information in short-term memory. In schema construction process also, the chunking of memory elements were explained exclusively.

In the late 1980s Sweller, John [1] who is known to be the founder of Cognitive Load Theory worked on Cognitive Load during Problem Solving: Effects on Learning. In a number of studies, he and his associates found that the learners used a strategy called ‘means-ends analysis’ for solving given problems. He also came up with a view that means-ends analysis does not help in schema construction, but it demands huge amount of cognitive processing capacity. Sweller advised instructional materials which do not target problem solving should be developed in such a way to reduce unnecessary cognitive load by instructional designers. The worked-examples and goal-free problems could be used as alternative instructional materials.

Applications of cognitive load theory were widely used in number of contexts in 1990s. The empirical results from these studies led to the demonstration of several learning effects: the completion-problem effect; Paas, Fred G. [8] in his work on ‘Training strategies for attaining transfer of problem-solving skill in statistics: A cognitive-load approach’, He developed three computer-based training strategies named conventional strategy, worked-out and completion strategy. Interestingly, completion strategy and the worked strategy were found to be more fruitful than the conventional strategy in learning transfer.

Types of Cognitive load

In order to understand the form of cognitive load, experts in the domain explained the following three types, Intrinsic, Germane and Extraneous cognitive load:

• Intrinsic Cognitive Load: It occurs when the difficulty of learning material itself is inherent that may influence previous knowledge of the topic. It is built into the content to be learnt. For eg, some topics are easier to learn whereas others are tough like addition and subtraction are easy to learn than linear equations.  We really feel helpless in reducing intrinsic cognitive load, but there are ways and means to minimize it by providing training programs for teachers regarding online teaching

• Germane Cognitive Load: In order to construct schemas, some components are needed in processing the information those trigger germane cognitive load. The teachers who are undergoing training programs could really get benefited by gaining knowledge; developing skills with the help germane cognitive load.  Sometimes, it gives counter measure of teachers who leave the class in the middle of a training that leads to germane cognitive load. So, we need to have evidence based training practices such as stating learning outcomes, chunks of training materials, implanting spaced learning and feedback driven

• Extraneous Cognitive Load: It is the worst form of cognitive load occurring in the classroom to make a learner to process all the stuff actively which many of them do not offer learning. For instance, taking huge amount of calories but no nutritious value in that. Activities such as making confused user interface for LMS or E-Learning, irrelevant topic that does not fulfill training objectives, huge sentences, unfamiliar words, narrating to read word on scree in an e-learning program are all come under extraneous cognitive load. As an instructional designer, the teacher has to take necessary steps to eliminate this kind of load as it is in the hands of him/her

Possible Effects of Overload on Cognition

Spatial contiguity and modality effect: Moreno, Roxana; Mayer, Richard E. [9] experimented on Cognitive principles of multimedia learning: The role of modality and contiguity relative contributions of spatial contiguity and modality to multimedia learning. Learning was impaired due to separation of onscreen text from visual information spatially which is identical to results of limited resources in visual working memory in text by Baddeley's and Mousavi et al., [10-11] Most of the visual-text presentation needed mixed modality presentations for better learning as earlier reported by Paivio's [12] theory. He concluded that auditory working memory holds words whereas visual working memory does pictures, both are providing attention resources for linking connections between them altogether. Mixed presentation mode was advocated by Mousavi, Seyed Yaghoub; Low, Renae; Sweller, John  in their review on Reducing cognitive load by mixing auditory and visual presentation modes. Visual and auditory are two independent processors in working memory suggested by presentation-modality effect, according to them. They concluded that mixed presentation mode is helpful than unitary mode in enhancing working memory adding split-attention effect might only be seen in students with multiple sources of information as they split their attention.

Split-attention effect: The most common effect was noticed by Chandler, Paul; Sweller, John in their experiments on the split-attention effect as a factor in the design of instruction between conventional group and modified group. The results of these experiments have gone favored to modified group in each and every item of testing than conventional group.  For a realistic testing, the split-attention effect was reported between illustrations and text by these experiments. To confirm, other laboratory studies and field studies by Sweller et ai. [13] and Chandler and Sweller have also been conducted. The same results were again appeared that the integrated formats are found to be more helpful in learning rather than conventional designs. Instructional time was said to be the indirect measure of cognitive load in these laboratory experiments. Despite of having less time for integrated group, they had done better than the conventional group. In a series experiments conducted by Chandler and Sweller and Sweller e tal, irrespective of partition and sequential arrangement of textual information the integrated group outperformed the conventional group. It was further supported by in another experiment integrated group performed comparatively well than the conventional group though the latter was given partitioned and well-structured information to be learnt. Therefore, split-attention effect tested again and confirmed in the learning process.

Worked-example effect: Cooper, Graham; Sweller, John [14] have conducted number of experiments on the effects of schema acquisition and rule automation on mathematical problem-solving transfer. The worked-example group with less time has shown better performance on the transfer problem in comparison to conventional group in acquisition. Majority of the participants in the worked-example group (9 out 0f 12) solved the problem, only one subject made two or more errors. In contrast, 9 out of 12 subjects in the conventional group committed 2 or more mathematical errors on the transfer problem. It clearly indicated that the worked-example group was far ahead than the conventional group that explains us worked-example effect experimentally.  Sweller, John; Cooper, Graham A.  have again reported the same results in another experiments regarding the use of worked examples as a substitute for problem solving in learning Algebra. They assumed that algebra problem-solving skill requires number of cognitive structures as in other domains. They also added conventional problem solving techniques do not enhance in developing those cognitive structures. In the first experiment, the university students who have much experience shown better algebraic cognitive representation in comparison to the students with less experience in the areas of recalling equations, and distinguish between closely similar equations. Second experiment, not proved worked examples as better facilitators for problem solving and acquisition of knowledge but also took less processing time to conventional examples. In addition to that, mathematical errors and solution time were also found to be low when the problems were presented identical to previous ones. 

Expertise-reversal effect: Kalyuga, Slava et al. [15] reviewed number of research reports for the last 20 years to explore the expertise reversal effect. They observed findings reported by McNamara et al. [16] who viewed  original instructional text increased learning in low-knowledge readers  whereas minimally coherent text helped in for high-knowledge readers rather than highly enhanced text material. Kalyuga et al. [17] and Yeung et al. also concluded that the eliminating redundant material was beneficial for experienced learners as it minimized cognitive load. Healy, A.F., and McNamara, D.S. [16] reasoned expertise reversal effect behind the results, but not active engagement of processing of text redundant information. 

• Redundancy effect: External cognitive load is increased when we present the same information many more times. Kalyuga et al. [18] reported learning inhibition in experienced learners due to over repetition of auditory information. In a series of experiments conducted by them, reported effectiveness of diagram-only conditioning was increased and advantages of audio text disappeared. It further strengthened redundancy effect as most of the subjects felt higher cognitive load with loaded and repeated auditory information during training

• Element interactivity effect: Often, negative effects of split attention effect and redundancy effects are seen as the element interactivity increases. Because, some learning elements demand high cognitive load intrinsically which could not be processed separately stated by Sweller [2] Sweller and Chandler. The learners should be in a position to process huge number of interacting elements side by side in their working memory in order to understand structured complex instructional information.  To understand the instruction, number of elements to be attended should be used as indicators for calculating the degree of element interactivity which is of two types low-element interactivity material and high-element interactivity material. For instance, translating a word into second language considered as low-interactivity element material whereas correct word order that needs to be learnt simultaneously said to be high-element interactivity material. The former type is learned individually, so it can not lead to working memory load. On the other hand, the latter one is done simultaneously rather than individual elements and resulting in heavy load on working memory.  There can be learning failure due to heavy load on the capacity of working memory as a result of high element interactivity for students.   However, learner develops new schema when she or he all the interactions between learning elements. Whenever the students face similar situations, this schema works as an unitary element. Element interactivity drastically goes down at this level turns into experienced learners and helping them to put schemas together for at least some of the elements for incoming information.  In the process, cognitive load on the working memory be minimized by expert learners through keeping some of the elements united and incorporating into schemas. Pollock, Chandler, and Sweller [19] opined that the presentation of isolated elements may find a way to avoid working memory limitations for novice learners gradually leading them to deal high element interactivity elements. But it is clear that the causal relationships are cut down artificially in presenting isolated elements as they do not target understanding.  Without acquiring schemas to put interacting elements together, understanding of very high element interactivity would be impossible for novice learners. In the next stage of instruction, better understanding is provided to compensate prior learning without understanding. In Pollock et al.’s [19] reported better learning in novice learners through a mixed instructional method rather than conventional method.  Mayer and Chandler also supported presentation of animated isolated elements allowed inexperience learner to control speed and yielded better ac compared to students who had high-element interactivity information

• Imagination effect: Better outcomes are obtained from interaction and mental simulation of elements. Cooper, Tindall-Ford, Chandler, and Sweller [14] observed imagination effect when students are imagining the content of instruction in comparison to simply studying the content. In Cooper et al.’s [14] two alternative instructional strategies named worked examples and imagining procedures and relations adapted on using a spreadsheet: described in instruction. Imagining procedures and relations were found to be highly advantageous for more knowledgeable students as compared to studying with worked examples. But negative effect was seen with imagining procedures and relations when compared with going worked example and termed it as expertise reversal effect. In the working memory, mental representations are constructed and lead to mental imagination. Inexperienced learners like to fail in mental imagining as they do not have prior appropriate schemas ultimately turns task impossible due to limitations of working memory. However, learners who follow imagination instructions may learn little, but experienced learners learn better by practicing imagining task procedures though they lost redundant instructional guidance.  Experienced learners improve performance by easily following imagination instructions with no working load. But novice learner use worked examples for constructing schemas of interacting elements rather than imagination procedures. Redundancy is developed in expert learners when they were asked to study the material as they had already developed those schemas

Mitigation of Cognitive load 

Some of the following strategies could be used to eliminate cognitive load of the learners:

• Cognitive tools: In order to diminish extraneous cognitive load, teacher may use number of cognitive tools such as concept maps, checklists for completing learning tasks, nomenclature related quick reference guide, computer based scaffolding etc. These tools could automatically offload the forces on working memory and facilitate cognitive processes

• Collaborative learning: Kirschner, Paas, and Kirschner [20-21] found high cognitive load was easily dealt by collaborative learning though learning elements are to be re-integrated. The learning material is broken down into small parts and done by one of the members and combing also be done when and where needed. Effective collaborative learning strategies include word webs, talking chips, test-taking teams, video call meetings, think-pair-share etc

• Metacognitive tools: Regulation of cognition, planning, monitoring, controlling and evaluation are said to be the best tools of thinking the teacher has to possess in online classes. Accordingly, the teacher has to map out his or her learning situation so as to reduce cognitive workload up to some extent. In fact, metacognition adds some more load to cognition. However, teacher may use support mechanisms as per the demands. So, the teacher has to be very careful in selecting appropriate processing strategies and assure should not be high unnecessarily. It should be noted that some instructional models do not target to eliminate intrinsic and germane cognitive load. For instance, learners are to be given high cognitive load situations in project centered learning designs

Shortcomings of CLT

• Most of the thoughts were driven by old theories of cognitive processes that did not focus on present context

• Lack of strong empirical evidences as the recommendations were just given in the form of guidelines

• Most of theories of working memory confirm limited amount of information only maintained in short memory(Miller's magic number 7 plus or minus two, Baddaley and Hitch early model of working memory,  dual coding theory by Clark and Paivio) 

• Advanced CLT theories ignore most recent finding on attention and working memory

• Constraints regarding memory functioning is visible in CLT

• Finally, applications of cognitive psychology with ill-informed statements were often observed in the CLT

Sum Up

To conclude, CLT is widely accepted when it comes to web based instruction though it has few loopholes. By the nature, many teachers at different levels of education often feel being provided cognitive load to the learners. This theory may provide clear understanding about how working memory of the learner moves on. It also explains the various effects and consequences of cognitive load in the learners. In the present era, the cutting edge technologies came into education for fulfilling the basic purposes of instruction in online classes. Teachers should not be the slaves of technology, rather the asters of technology. Keeping all the pros and cons in the mind, the instructional design should be made and applied in online classes to make learners active for better learning by reducing cognitive load as much as possible.

1. Sweller J. “Cognitive load during problem solving: Effects on learning.” Cognitive Science, vol. 12, no. 2, 1988, pp. 257–285.

2. Sweller J. “Cognitive load theory, learning difficulty, and instructional design.” Learning & Instruction, vol. 4, no. 4, 1994, pp. 295–312.

3. Feinberg S. and M. Murphy. “Applying cognitive load theory to the design of web-based instruction.” Proceedings of IEEE Professional Communication Society International Professional Communication Conference and Proceedings of the 18th Annual ACM International Conference on Computer Documentation: Technology & Teamwork, Cambridge, Massachusetts, September 24–27, 2000, IEEE Educational Activities Department, pp. 353–360.

4. Paas F. et al. “Cognitive load measurement as a means to advance cognitive load theory.” Educational Psychologist, vol. 38, no. 1, 2003, pp. 63–71.

5. De Jong T. “Cognitive load theory, educational research, and instructional design: Some food for thought.” Instructional Science, vol. 38, no. 2, 2010, pp. 105–134. https://doi.org/10.1007/s11251-009-9110-0.

6. Miller G.A. “The magical number seven, plus or minus two: Some limits on our capacity for processing information.” Psychological Review, vol. 63, no. 2, 1956, pp. 81–97. https://doi.org/10.1037/h0043158.

7. Chase W.G. and H.A. Simon. “Perception in chess.” Cognitive Psychology, vol. 4, 1973, pp. 55–81.

8. Paas F.G.W.C. “Training strategies for attaining transfer of problem-solving skill in statistics: A cognitive load approach.” Journal of Educational Psychology, vol. 84, 1992, pp. 429–434.

9. Moreno R. and R.E. Mayer. “Cognitive principles of multimedia learning: The role of modality and contiguity.” Journal of Educational Psychology, vol. 91, no. 2, 1999, pp. 358–368.

10. Baddeley A. and G. Hitch. “Working memory - psychology unlocked.” January 2017. Archived from the original on January 2020. Retrieved January 2017.

11. Mousavi S.Y. et al. “Reducing cognitive load by mixing auditory and visual presentation modes.” Journal of Educational Psychology, vol. 87, 1995, pp. 319–334.

12. Paivio A. Imagery and verbal processes. Holt Rinehart & Winston, 1971.

13. Sweller J. “On the limited evidence for the effectiveness of teaching general problem-solving strategies.” Journal for Research in Mathematics Education, vol. 21, no. 5, 1990, pp. 411–415.

14. Cooper G. et al. “Learning by imagining.” Journal of Experimental Psychology: Applied, vol. 7, 2001, pp. 68–82.

15. Kalyuga S. et al. “The expertise reversal effect.” Educational Psychologist, vol. 38, no. 1, 2003, pp. 23–31.

16. Healy A.F. and D.S. McNamara. “Verbal learning and memory: Does the modal model still work?” Annual Review of Psychology, vol. 47, 1996, pp. 143–172.

17. Kalyuga S. et al. “Levels of expertise and instructional design.” Human Factors, vol. 40, no. 1, 1998, pp. 1–17.

18. Kalyuga S. et al. “Incorporating learner experience into the design of multimedia instruction.” Journal of Educational Psychology, vol. 92, no. 1, 2000, pp. 126–136.

19. Pollock E. et al. “Assimilating complex information.” Learning & Instruction, vol. 12, no. 1, 2002, pp. 61–86.

20. Kirschner F. et al. “A cognitive load approach to collaborative learning: United brains for complex tasks.” Educational Psychology Review, vol. 21, 2009a, pp. 31–42. doi:10.1007/s10648-008-9095-2.

21. Kirschner F. et al. “Individual and group-based learning from complex cognitive tasks: Effects on retention and transfer efficiency.” Computers in Human Behavior, vol. 25, 2009b, pp. 306–314. doi:10.1016/j.chb.2008.12.008.