Patrik Hultberg, Coordinator of Educational Effectiveness, Associate Professor of Economics, Economics and Business, Kalamazoo College
April 3, 2017
All teachers want their students to learn. Although there are many different definitions of learning, most would agree that learning implies that a student can use concepts and skills taught long after leaving the classrooms and in situations different from the classroom environment (Soderstrom and Bjork 2013), what is often called “transfer.”
We might also agree that the content of our fields, the concepts and skills needed to “really” master our field, is far greater than what can possibly be “covered” in the limited time available to us in a single course, or even several courses. This is important because it implies that teachers have to make choices in terms of what to teach. All of us also recognize that there are many different ways to present material to the students. Whether we like it or not, we are constantly exposed to new and different teaching strategies: lecture, discussion, inquiry-based learning, problem-based learning, team-based learning, case-based teaching, experiential learning, flipped, active, etc. All these choices imply that teachers have to make choices in how to teach.
The choices we make, in terms of both content and approach, are made explicit in the way we design courses. Teaching, in that sense, is actually a design problem, and being able to effectively design a course is an important, if complex, skill to develop. It requires effort, research, reflection, experimentation, as well as a willingness to relinquish control, and perhaps even to fail. It is always easier to do what we have “always done.” The choices and possibilities are overwhelming, while the time to research, think, and plan is scarce. But then again, we remember that we want our students to learn; to learn as much as possible, to retain it for as long as possible, and to apply what they have learned in future situations that we cannot even predict.
Although teachers have always faced the instructional design problem, the task seems more critical now for three reasons: our knowledge of teaching and learning has grown dramatically, the range of alternative approaches has expanded due to technology and research, and, importantly, colleges and universities are today becoming more diverse in terms of backgrounds, experiences, and levels of preparation, which increases the importance and consequences of our choices of content and approach. It also makes the teacher’s task more complex, and thus requires more effort in terms planning and designing our courses.
This paper, sorry to say, will not provide the solution to the instructional design problem; nor is there a single solution. It will not tell you what you should do in your classroom to maximize learning that is both durable and transferable. It will instead offer suggestions and a structure to allow teachers to frame and inform the choices that each one of us makes for our own specific courses and classrooms.
Inside the Brain
I begin by arguing that since learning is represented by a change in the student’s brain, it is quite important, as we design instruction for learning, to have some understanding of the how the brain actually works. This does not mean that all teachers need to be cognitive scientists. Nevertheless, having a basic understanding of how the brain works can inform our choices of both content and teaching approaches and help us design courses for more effective learning.
The cognitive architecture of the brain can be described as the necessary memory structures – in particular referencing working memory and long-term memory – that are required to input, process, store, and retrieve information which we need in order to learn, think and solve problems. To learn our brains must process information in working memory before permanent storage can occur in long-term memory. We are only conscious of the contents of our working memory. All other cognitive functions are hidden in long-term memory until retrieved into working memory, where they can once again be processed (Sweller et al. 1998). Thus in order for a student to think about and learn information presented in our courses, processing must take place in the brain’s working memory.
So here’s the issue: working memory is severely limited in both capacity and duration. When thinking of students we need to understand that their working memories are only capable of holding about seven elements of information at a time, and for less than 30 seconds, without rehearsal (Miller 1956; Sweller 2004). These limits hold for both novice and expert learners; that is, for students who have no experience in our fields, as well as for students who are well-prepared, having completed previous coursework. However, there is a big difference between students of different backgrounds, experiences, and levels of preparation in that “experts” are able to “chunk” many pieces of information into a single element thus allowing them to process more information in their working memory. By chunking information a student can go beyond simple memorization of facts and begin to understand more complex materials.
These limitations on working memory are important to take into consideration when we are designing courses. If we present too much new material too quickly or in an inefficient way, a student’s working memory will “overload” and needed processing (and hence learning) cannot occur. The material will not be stored in the student’s long-term memory and therefore cannot be retrieved and used in the future. However, by properly sequencing material we can use a student’s long-term memory to circumvent the limitations of working memory.
Schemas and Automaticity
Information (facts, concepts, and skills) in long-term memory is stored in schemas of varying degrees of complexity and automation. A schema is defined as a set of elements of information categorized according to their potential future application. A single schema can combine many pieces of information that are then processed as a single element by the more limited working memory. Thus by chunking material and building schemas the cognitive load on working memory can be reduced. Generating increasing numbers of ever more complex schemas by combining elements of lower-level schemas into higher-level schemas is what creates expertise and skilled performance.
Another way to reduce the demand on working memory is by practice and repetition. By practicing, a student’s use of concepts and skills become “automatic”; their schemas become automated. Automated schemas allow for automatic performance on familiar tasks and therefore require less conscious processing, which frees up working memory space by reducing the load on working memory (Ericsson 2008). With sufficient practice, a procedure (use of information) can be carried out with minimal conscious effort; i.e. with minimal working memory cognitive load. We should have both processes in mind as we plan our instructional designs.
By creating ever more complex schemas, and making such schemas ever more automated, students are able to increasingly process information in their (more) limited working memory and thus avoid overload (Van Merriënboer and Kester 2005). Despite the severe limits on the number of elements that can be processed in working memory, constructing complex schemas eliminates all limits on the amount of information that can be processed. These constructed schemas are stored in long-term memory, but retrieval and processing of this stored knowledge require the use of working memory. Thus the practice of retrieving and using learned material is also limited by working memory’s capacity constraints. It is worth emphasizing that schemas are stored in long-term memory. The usefulness of sophisticated and automated schemas thus depends on a student’s ability to store and, importantly, retrieve these schemas on demand in the future. That is, creating complex schemas is important, but in order to make learning both durable and transferable – that is, useful for the student in the future – the information must be available for retrieval on demand in novel situations in the future. Hence, this should also be an instructional design goal.
If learning basically consists of storing automated schemas in long-term memory (Sweller 1994), then the task of instructional design is one of carefully managing the cognitive load of the student’s working memory in order to allow for the required processing that leads to the construction of schemas and their automation. That is, instructional design must promote construction and automation while ensuring that students’ cognitive limit is not saturated, as well as promoting retrieval today and in the future.
Cognitive Load Theory
These are the goals that cognitive load theory attempts to address. Cognitive load theory was developed by John Sweller in the late 1980s out of the study of problem solving. Sweller (1988) argued that instructional design may be used to reduce cognitive load in learners. According to the theory, and as argued above, during learning, information must be held in the limited working memory until it has been sufficiently processed and transferred into long-term memory. Cognitive load theory argues that for individuals to learn effectively, it is important that their cognitive architecture and the learning environment be aligned, and such alignment is the task of instructional design. The ability of working memory to effectively process information is clearly the main concern of cognitive load theory and there are three types of working memory in which considerations of cognitive load should be considered: intrinsic, extraneous, and germane.
Intrinsic cognitive load is determined by the complexity of the material and the level of interactivity between elements relative to the learner’s level of expertise. An element is anything that has been, or needs to be, learned, e.g., a Spanish vocabulary word, an economic concept, or a mathematical rule. Element interactivity is a measure of the number of elements that must be processed simultaneously in the working memory in order to learn the specific concept (i.e., construct a schema). Low-element interactivity tasks require low cognitive load, e.g. learning basic facts. These elements can be learned serially, rather than simultaneously. High-element interactivity tasks must be processed simultaneously in order to be learned and understood. Any subject matter that requires multiple steps for understanding tends to have a high intrinsic load.
The intrinsic cognitive load of information cannot be altered by instructional design per se. However, the teacher is able to control the intrinsic load by presenting content and material in certain ways; e.g., simple to complex or part-whole sequencing. The intrinsic load can also be reduced by giving students time to process new material (remove time pressure), as well as encouraging students to write down intermediate results in order to “offload” their working memory (de Jong 2010). These are important aspects to consider in the instructional design process.
A student’s level of expertise will also affect the degree of intrinsic cognitive load of particular materials. A large number of interactive elements for a novice learner may be a single element (a schema) for someone with more expertise, which implies that the cognitive load of the novice will be higher for the same exact material than for an expert. As suggested by Schnotz and Kurschner (2007), instructional design should identify the zone of proximal development (Vygotsky 1987) that aligns intrinsic load to the level of expertise of the learner in order to optimize working memory. This implies that an important aspect of choosing what to teach comes from knowing the students in our classes. In order to properly match the material to students’ level of preparation, we should consider getting to know our students, perhaps by conducting a form of knowledge analysis.
The difficulty facing a teacher in an (experientially) diverse classroom is obvious. It will not be easy to design a course that presents content that matches each learner’s level of experience without overloading anyone’s working memory. This problem is probably unsolvable if the only teaching strategy available is the lecture, but other teaching approaches may be better able to accommodate diverse learners; e.g., discussion- or inquiry-based learning approaches might better accommodate students’ varied experiences. Designing a course that is able to accommodate a diverse set of students is a challenge, but worth our careful consideration.
Extraneous cognitive load is related to the way content or tasks are presented to students, and represents things that we do that not only don’t directly contribute to learning but may actually take away from it. Purposeful instructional decisions and interventions can help reduce extraneous cognitive load, which will allow students to either process more elements in working memory or elements of higher interactivity. For example, presenting irrelevant and peripheral information or using a poor layout may cause overloads in the working memory and negatively affect students’ storage of information. The main focus of cognitive load theory has traditionally been to suggest how to reduce the extraneous load while best presenting the material to be learned. So, for example: avoid redundancy (i.e. providing multiple sources containing the same information); avoid split-attention (i.e. asking students to mentally integrating disparate sources of information); and ensure that students possess the required schemas to solve problems. Again, these considerations should be part of instructional design.
The most important implication of achieving low extraneous load is the possibility of introducing instructional designs that lead to greater schema construction or automation (Artino 2008). That is, approaches that support an increase in germane cognitive load. Germane cognitive load is cognitive load that facilitates the construction and automation of schemas. Increasing the germane load of our courses is perhaps the main task of instructional design as it answers the question of what exactly we should ask our students to do inside and outside of our classrooms. These choices are difficult since a clear distinction between different types of cognitive load cannot always be made, especially between intrinsic and germane loads (Debue and van de Leemput 2014). In addition, approaches that can be categorized as germane load for an expert learner may be extraneous load for the novice learner (de Jong 2010). Nevertheless, the distinction is helpful and instructional design choices should aim to decrease extraneous cognitive load, increase germane cognitive load, while maintaining total cognitive load at manageable levels (avoid cognitive overload).
Designing courses for effective learning must address this issue carefully, and while the topic deserves separate treatment, cognitive psychology offers several evidence-based learning strategies that can increase the germane load, and hence durable and transferable learning. Five well-known approaches are:
- Distributed practice: spacing out study sessions over time to interrupt forgetting and reinforce durable learning (Benjamin and Tullis 2010).
- Interleaving: mixing related, but distinct material, during study sessions to force students to discriminate between ideas and problem types that reinforce learning and its transferability (Rohrer 2012).
- Retrieval practice: asking students to recreate something learned in the past from memory to reinforce learning and make information more easily retrievable in the future (Roediger et al. 2011).
- Elaboration (or reflection): asking students to explain and describe ideas with many details, and making connections to own experiences, to promote the acquisition of more complex schemas (Brown et al 2014).
- Concrete examples: linking abstract concepts with concrete examples to help students understand new material by combining new concepts to existing schemas.
We began by arguing that in order to effectively promote student learning, teachers must carefully choose both what teach and how to teach. These choices are made more difficult with classrooms that are becoming ever more diverse in terms of our students’ preparation and skill levels. I have argued is that cognitive load theory offers a framework for how to approach these difficult choices and thus design our courses to match the complexity of the material to students’ diverse backgrounds and levels of preparation (intrinsic load), and how to present content in order to minimize irrelevant information and poor layouts (extraneous load), as well as how to promote design practices that are known to facilitate learning (germane load). The instructional design goal is therefore to purposefully choose strategies that control intrinsic load, reduce extraneous load, and increase germane load. There are many ways of achieving this goal, and all of us can do it, but it may require a willingness to reflect on our current practices and perhaps explore new and different approaches.
Artino, A.R. 2008. Cognitive Load Theory and the Role of Learner Experience: An Abbreviated Review for Educational Practitioners. AACE Journal, 16(4), pp. 425-439.
Benjamin, A.S. and Tullis, J. 2010. What makes distributed practice effective? Cognitive Psychology, 61, pp. 228-247.
Debue, N., and van de Leemput, C. 2014. What does germane load mean? An empirical contribution to the cognitive load theory. Frontiers in Psychology, 5, 1099. http://doi.org/10.3389/fpsyg.2014.01099
de Jong, T. 2010. Cognitive load theory, educational research, and instructional design: some food for thought. Instructional science, 38(2), pp. 105-134.
Ericsson, K.A. 2008. Deliberate practice and acquisition of expert performance: a general overview. Academic emergency medicine, 15(11), pp. 988-994.
Kirschner, P., Kirschner, F. and Paas F. 2009. (December 23). Cognitive Load Theory. Retrieved from https://www.education.com/reference/article/cognitive-load-theory/
Miller, G.A. 1956. The Magical Number Seven, Plus or Minus Two: Some Limitations on Our Capacity for Processing Information. The Psychological Review, 63: pp. 81-97.
Roediger, H. L., Putnam, A. L., and Smith, M. A. 2011. Ten benefits of testing and their applications to educational practice. In J. Mestre & B. Ross (Eds.), Psychology of learning and motivation: Cognition in education, (pp. 1-36). Oxford: Elsevier.
Rohrer, D. 2012. Interleaving helps students distinguish among similar concepts. Educational Psychology Review, 24, pp. 355-367.
Schnotz, W. and Kürschner, C. 2007. A Reconsideration of Cognitive Load Theory. Educational Psychology Review, 19(4), pp. 469-508.
Soderstrom, N. C., and Bjork, R. A. 2013. Learning versus performance. In D. S. Dunn (Ed.), Oxford bibliographies online: Psychology. New York: Oxford University Press.
Sweller, J. 2004. Instructional design consequences of an analogy between evolution by natural selection and human cognitive architecture. Instructional science, 32(1-2), pp. 9-31.
Sweller, J. 1994. Cognitive Load Theory, Learning Difficulty, and Instructional Design. Learning and Instruction, 4: 295-312.
Sweller, J. 1988. Cognitive Load During Problem Solving: Effects on Learning. Cognitive Science, 12, 257-285.
Van Merriënboer, J.J. and Kester, L. 2005. The four-component instructional design model: Multimedia principles in environments for complex learning. The Cambridge handbook of multimedia learning, pp. 71-93.
Vygotsky, L. 1987. Zone of proximal development. Mind in society: The development of higher psychological processes, 5291, p. 157.