Written by Theresa Pham, MClSc/PhD program in Speech-Language Pathology at the University of Western Ontario and Lisa Archibald, PhD, Associate Professor, School of Communication Sciences & Disorders and Department of Psychology, The University of Western Ontario
Working memory is the ability to hold and manipulate information in mind over short periods of time. Complex thinking and learning draw on working memory. Working memory supports school learning across the curriculum, from following instructions, to learning to read, to solving mathematical and scientific problems. Given that much of classroom instruction depends on working memory skills, the academic environment may be particularly challenging for students with learning disabilities (LDs) who often have working memory deficits (Gathercole & Alloway, 2008; Dehn, 2008). Reduced working memory abilities might make it difficult for these students to process as much information or to process information as rapidly or automatically as their peers. Overall, students with working memory deficits will have to work much harder than their typically developing peers to learn and carry out classroom activities. Because we know that students with poor working memory will face substantial learning difficulties when task demands exceed available working memory resources, providing learning support in the classroom is important for overcoming poor working memory skills.
The components of working memory
Storage vs. Processing
Working memory is sometimes referred to as ‘the mental post-it note’ of our minds. Information we need for a task is held in working memory. We can think about working memory as a balance between the storage and processing demands of a task:
- Storage – the brief retention of items, such as list of numbers, words, or locations
- Processing – the cognitive component completed along with the task, such as re-ordering items in a list or following changes in location.
What ‘fits’ in working memory though will vary depending on the storage and processing demands of the task. For example, it is easier to remember 3 words than 8 words, but remembering 3 words becomes difficult when we add a processing task, such as thinking of a rhyming word.
Verbal vs. Visuospatial information
Beyond the demands of a task, the type of task also influences working memory functioning. Information can be verbal (spoken and written language) or visuospatial (forming objects in the space around us) in nature. Much of school learning is verbally mediated as teaching is largely delivered through language-based directions and explanations. For students with language-related LDs such as Developmental Language Disorder (DLD) or Dyslexia, the verbal working memory demands of the classroom will be very challenging.
Signs of working memory overload
It is important for teachers and parents to recognize working memory breakdowns as well as teach the child, themselves, to become aware of when their working memory is being overwhelmed. A student experiencing working memory overload may do any of the following:
- Give up or stop attending
- Complete only a part of the task
- Become frustrated by the learning challenges
- Act out or engage in other distracting behaviours
Once working memory breakdowns are recognized, strategies can be implemented to alleviate working memory overload.
Working Memory Training to Increase Capacity
Can working memory be improved through training? ‘Working memory training’ refers to interventions aimed at direct training of working memory skills to increase storage and processing capacity. Working memory training should lead to improvements in academic skills due to the relationship between working memory and cognitive performance. However, the evidence to date seems to show that working memory training doesn’t work this way. Let’s consider what this means.
The evidence from many studies suggests that working memory training is only effective for tasks you practice or very closely related tasks – this is known as “near transfer” (Melby-Lervag et al., 2013; Schwaighofer et al., 2015). In fact, more intensive training results in better performance on a trained task (Schwaighofer et al., 2015), but learning does not transfer beyond the trained tasks, that is, there is no evidence of “far transfer”. This means that training the ability to recall more numbers in a row does not necessarily influence educational attainment.
While the available evidence does not support recommendations for working memory training on academic skills, there are other lessons to be learned from these studies. First, we know that learning and training need to be highly related (near transfer). For example, to improve math skills, time and energy should be spent working directly on math skills. Secondly, sufficient frequency over longer durations will enhance and boost learning. Based on these two principles, the key to a successful working memory training program would be (1) tailored to the individual goal of the student and (2) implemented with high intensity. Given that working memory demands are inherent in classroom learning and students with poor working memory skills, in particular, are at an additional disadvantage, we need to focus on reliable ways for providing learning support.
Working Memory Strategies to Increase Efficiency
Although the evidence for working memory training is limited, studies examining strategy use provide a promising approach (Gonthier & Thomassin, 2015; Jarosza, Radena, & Wiley; 2019). The idea is that strategies can be used to support working memory by increasing efficiency and functioning even if working memory capacity does not change. The strategies, themselves, however, place different demands on working memory (Gonthier & Thomassin, 2015; Turley-Ames & Whitfield, 2003). Educators must evaluate and monitor the working memory burden or facilitative effects of these strategies for individual students. Drawing on lessons learned from examining memory and how the brain learns, we have identified strategies that have the potential to support working memory. The strategies are broadly organized into those that minimize processing demands to allow for better storage and those that minimize storage to allow for better processing.
Minimize processing demands when storage demands are high
Sometimes, the crucial component of a task is the need to remember the items as they were given. This is especially important when learning new information. The better the new information is encoded in working memory, the more durable the representation in long-term memory may become, which helps with learning of that information. The quality of a representation can be compromised when trying to store too many items or when trying to do additional processing tasks in working memory. By carefully monitoring working memory demands to optimize high-quality storage of new information, opportunities for retention in the long term are maximized. Strategies include repetition, rehearsal, phonological strategies, distributed practice, and multiple means of learning the information.
The need for repetition is so important, and yet so easily underestimated. In the classroom, a majority of students learn new information with only a few repetitions so the focus of teaching quickly shifts to completing additional tasks with that information. Students with working memory issues, however, require more repetition to retain information. Hearing a few repetitions will be insufficient for mastery leading to knowledge gaps when trying to apply the new information in other tasks. In fact, in a recent study (Storkel et al., 2017), children with DLD learned new words better after hearing it at least 36 times compared to 12 or 24 times in typically developing children. Teachers should make a conscious effort to use repetition in their teaching. Providing many opportunities for students with poor working memory to review and repeat new information will facilitate learning. In fact, repetition of task instructions, keywords, or comprehension checks can benefit all students. With repetition, the information will become so familiar that it will offset working memory demands once more complex tasks are introduced.
Rehearsal refers to repetition of to-be-remembered items for the purpose of retaining the items in memory, and it is a strategy that students, themselves, can adopt. Without rehearsal, to-be-remembered items decay within a few seconds (Baddeley, 1986). Importantly, parents and educators should not assume that a student with low working memory automatically knows how to use rehearsal or uses it effectively. Fortunately, the use of rehearsal can be taught explicitly (Miller, McCulloch, & Jarrold, 2015). When a student is presented with a list, they might be taught to restate the to-be-remembered information aloud immediately after hearing; first rehearsing aloud and then gradually shifting to rehearsing ‘in their head’ (not spoken aloud).
3. Phonological strategies
Several studies have reported results favouring the use of phonological over semantic strategies to support verbal memory in those with working memory difficulties (Gonthier & Thomassin, 2015; Turley-Ames & Whitfield, 2003). Phonological strategies may be more beneficial because they place lower demands on working memory, which is especially important during the initial stages of learning. Phonological awareness activities that emphasize the phonological structure of a word, such as counting syllables or identifying sounds, will improve the quality of the phonological representation, and thereby, facilitate encoding into long-term memory.
4. Distributed practice
Research has shown that memory is better when learning is spaced out rather than presented in one long session, even if the total time is the same. This is called distributed practice (e.g., Cepeda, Pashler, Vul, Wixted, & Rohrer, 2006). Not only does distributed practice help with remembering the new information, it can also promote generalization of learning (Vlach, Ankowski, & Sandhofer, 2012). This strategy is especially beneficial for students with poor working memory skills. Given that these students are already working hard to process information, they will become fatigued more easily during long learning sessions. To prevent cognitive fatigue and improve memory, learning of new material and ideas should be spread over time. This strategy also affords opportunities for answering questions and providing clarification without overloading the working memory demands for learning.
5. Multiple means of learning
As mentioned previously, most tasks require processing of verbal and visuospatial information. Multiple means of learning means presenting information both verbally and visually so that the information can be coded in different ways. Creating such high-quality representations of new information can facilitate learning by solidifying the ‘what’ of learning (Goldman, 2003; Eilam & Poyas, 2008). Although it may seem that focusing on two types of information increases working memory load and may impair learning, this is not the case. We know that verbal and visuospatial information are processed by unique networks in the brain (Nee et al., 2013), so presenting complementary verbal and visuospatial information actually reduces the working memory load compared to, for example, giving too much verbal or visuospatial information at once. By using both modalities, the working memory load is spread across the brain’s processors instead of being concentrated to any one network.
Presenting information in different ways will also make learning accessible to all students, as it is a principle of Universal Design for Learning (CAST, 2018). Students should have opportunities to interact with the material in different ways, such as talking about new information, viewing pictures of it, and manipulating objects related to it. Different representations will also give the student different ways of remembering the information later on.
Minimize storage demands when processing demands are high
Until now, we have discussed strategies aimed at supporting the storage (or memorization) of new knowledge. However, for many tasks, the most important part is working with what we know in novel ways. In short, this processing load is the task, and cannot be reduced (e.g., mental arithmetic). When the learning activity imposes a processing load on top of storage, then the bulk of information to-be-remembered should be reduced so that resources can be used for processing. We can reduce the amount of information that needs to be remembered by making the information more memorable. This includes making connections with long-term memory, using familiar information, using external aids, and self-testing.
6. Connections with long-term memory
One way to offset the demands of working memory is by relying on long-term memory, often described as ‘activating background knowledge’. Students can connect the new information with well-learned information stored in long-term memory to make it more meaningful, referred to as elaboration (Craik & Lockhart, 1972). When introducing a new topic, teachers can ask students what they know about the topic or about experiences related to the topic. Making connections between lesson plans and information held in long-term memory will reduce the reliance on working memory.
Another way students can use long-term memory to optimize storage is by chunking information (Cowan, 2001). This means grouping long, arbitrary sequences of information into meaningful chunks based on what you already know. For instance, this 6-item list, “I, F, A, I, B, C”, can be chunked into 2 common abbreviations, “FBI” and “CIA”. Finding associations between items, even weak ones, will reduce the number of “things” you need to remember. Interestingly, a new study found that chunking not only improved memory for chunked items, but also for non-chunked items, especially when earlier (vs. later items in the list) are chunked (Mirko, Alessandra, & Klaus, 2019).
Lastly, making connections with long-term memory is most effective when students themselves are making these associations, a process called self-generation. Students need to generate associations or chunk the information themselves, as opposed to reading or hearing it (Begg, Snider, Foley, & Goddard, 1989). Notably, making connections with long-term memory needs to be done cautiously as it may involve additional effort. If drawing on long-term memory itself is effortful and demanding, then fewer resources will be available for the working memory task. Nonetheless, additional efforts induced by making these connections can improve learning and memory.
7. Using highly familiar or automatized information
Another way to reduce storage demands is by ensuring that the task uses highly familiar or automatized information. When you recite the alphabet, you hardly need to think about it because you know the alphabet so well—it is automatized. This is why it is important to memorize (or automatize) sight words and math facts so they can become effortless building blocks for higher-level cognitive tasks (sight words, Yaw, 2012; math facts, Nelson et al. 2016). When students can easily retrieve and recall these basic concepts, they can use their working memory skills to hold more information or engage in another mental activity.
Familiar schemas and scripts can also be used to assist learning (Bransford & Johnson, 1972). For instance, try reading this passage,
“First you arrange items into different groups depending on their makeup. One pile might be sufficient depending on how much there is to do…”.
This passage is hard to comprehend at first, but if you think about laundry while reading it, it becomes much easier to process. Relatedly, students with poor working memory will be able to accomplish more complex tasks, if the tasks use information they already know. For example, if the student can use their script about playing a sport, then making an argument in a persuasive essay becomes easier. By drawing on what the student already knows, they will be able to activate and use their knowledge to anticipate or understand the activity more readily.
8. External aids
Don’t let your working memory do all the hard work! Encouraging students with poor memory to use external aids can alleviate working memory demands by providing a lasting record of the information to be remembered or processed. Having the information in front of them means that they do not have to use valuable resources to store information. Graphic organizers have been found to help students with LDs learn new academic vocabulary (Dexter, Park, Hughes, 2011) and increase reading comprehension (Kim, Vaughn, Wanzek, & Wei, 2004). In general, graphic organizers provide a visual outline that makes abstract concepts more meaningful by linking those concepts to existing knowledge. Combining verbal information with visual materials is also an example of multiple means of learning. We cannot just assume that students will spontaneously use or know how to use external aids. Teachers can explicitly teach students using an “I do, we do, you do” gradually release of responsibility strategy (Fisher & Frey, 2007): teachers will first provide explicit instructions and model an example (“I do”); then, teachers will work with students (“We do”); finally, students will work independently (“You do”).
Another form of the self-generation strategy is self-testing (Roediger & Karpicke, 2006). Self-testing is testing your memory through “free recall”, that is, challenging yourself to generate the newly learned information on your own. Self-generation is more effective at promoting meaningful learning than mere “recognition” (i.e., answering multiple-choice questions). Practice retrieval is one example that has received the most research attention. In brief, students read a text, set it aside, and then spend time freely recalling and writing down what they remember as opposed to re-reading. Students also become aware of what they do not know and can spend more time reviewing. Notably, the precursor to this strategy is to learn through repetition. But repetition is only good for storage and in fact, once information can be recalled, repetition may no longer be beneficial for learning (Karpicke & Roediger, 2008). Thus, once the student has achieved a level of mastery, then self-testing becomes a more powerful way to learn (Roediger & Karpicke, 2006).
We know that students with LDs are at an increased risk for having additional working memory deficits. We also know that many classroom activities draw on working memory, meaning that students with poor working memory skills will face many learning obstacles on a daily basis. So what can be done to improve their chances of success? We have described a variety of learning support strategies specific to working memory impairments. These strategies broadly fall into two categories: minimizing processing demands and minimizing storage demands. Although note that there will be great overlap during the practical applications of these strategies. This description is also not meant as an exhaustive list of strategies, nor is every strategy meant to be applied all at once. Instead, educators, parents, and the student will need to recognize when working memory difficulties arise and evaluate the working memory load to find strategies that work best. Providing learning support to each student will ensure that they are not missing out on rich classroom instructions or learning opportunities due to poor working memory skills.
About the Authors:
Theresa Pham is a graduate student in the combined MClSc/Ph.D. program in Speech-Language Pathology at the University of Western Ontario. She is supervised by Dr. Lisa Archibald. Her research focuses on understanding phonological (speech sounds) and semantic (meaning) contributions from different memory systems during language processing, and linking it with memory and language deficits in individuals with communication disorders. She completed her BSc in Psychology and Linguistics at the University of Toronto.
Lisa Archibald is an Associate Professor in the School of Communication Sciences and Disorders at the University of Western Ontario, Canada. Lisa studies links between memory and language processes in individuals with communication disorders. In particular, she is interested in working memory and language learning deficits in children. Recently, she has focused on SLP-educator school-based collaborations, and has been part of an international team of researchers and stakeholders considering terminology and profile for children with and unexplained, persistent language disorder now known as developmental language disorder (DLD).
Lisa is a member of the international organizing committee for Raising Awareness of Developmental Language Disorder (RADLD.org) and a founding member of DLDandMe.org. Prior to her research career, Lisa worked as a clinical Speech-Language Pathologist (SLP) for over 15 years providing services to children and adults, in schools, hospitals, and other facilities.
You can follow Lisa on twitter at @larchiba6.
Baddeley, A.D. (1986). Working memory. Oxford, UK: Oxford University Press.
Begg, I., Snider, A., Foley, F., & Goddard, R. (1989). The generation effect is no artifact: generating makes words distinctive. Journal of Experimental Psychology: Learning, Memory and Cognition, 15(5), 977–989. https://doi.org/10.1037/0278-73188.8.131.527
Bransford, J. & Johnson, M. (1972). Contextual prerequisites for understanding: Some investigations of comprehension and recall. Journal of Verbal Learning & Verbal Behavior, 11, 717–726.
CAST (2018). Universal Design for Learning Guidelines version 2.2. Retrieved from http://udlguidelines.cast.org
Cepeda, N., Pashler, H., Vul, E., Wixted, J., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: a review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. https://doi.org/10.1037/0033-2909.132.3.354
Cowan, N. (2001). The magical number 4 in short-term memory: a reconsideration of mental storage capacity. The Behavioral and Brain Sciences, 24(1), 87–114; discussion 114- 185
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning & Verbal Behavior, 11, 671–684. https://doi.org/10.1016/S0022-5371(72)80001-X
Dehn, M. J. (2008). Working memory and academic learning: Assessment and intervention. John Wiley & Sons Inc.
Dexter, D. D., Park, Y. J., & Hughes, C. A. (2011). A meta‐analytic review of graphic organizers and science instruction for adolescents with learning disabilities: Implications for the intermediate and secondary science classroom. Learning Disabilities Research & Practice, 26(4), 204–213. https://doi.org/10.1111/j.1540-5826.2011.00341.x
Eilam, B., & Poyas, Y. (2008). Learning with multiple representations: Extending multimedia learning beyond the lab. Learning and Instruction, 18(4), 368–378. https://doi.org/10.1016/j.learninstruc.2007.07.003
Fisher, D., & Frey, N. (2008). Homework and the Gradual Release of Responsibility: Making “Responsibility” Possible. The English Journal, 98(2), 40–45.
Gathercole, S.E., & Alloway, T.P. (2008). Working memory and learning: A practical guide. London: Sage
Gathercole, S.E., Hitch, G.J., Service, E., & Martin, A.J. (1997). Short-term memory and long-term learning in children. Developmental Psychology, 33, 966-979.
Goldman, S. R. (2003). Learning in complex domains: When and why do multiple representations help? Learning & Instruction, 13(2), 239-244.
Gonthier, C., & Thomassin, N. (2015). Strategy use fully mediates the relationship between working memory capacity and performance on Raven’s matrices. Journal of Experimental Psychology: General, 144(5), 916–924. https://doi.org/10.1037/xge0000101
Jarosz, A. F., Raden, M. J., & Wiley, J. (2019). Working memory capacity and strategy use on the RAPM. Intelligence, 77(August), 101387. https://doi.org/10.1016/j.intell.2019.101387
Karpicke, J., & Roediger, H. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966–968. https://doi.org/10.1126/science.1152408
Kim, A., Vaughn, S., Wanzek, J., & Shangjin Wei. (2004). Graphic Organizers and Their Effects on the Reading Comprehension of Students with LD: A Synthesis of Research. Journal of Learning Disabilities, 37(2), 105–118. https://doi.org/10.1177/00222194040370020201
Melby-Lervåg, M., & Hulme, C. (2013). Is working memory training effective? A meta-analytic review. Developmental Psychology, 49(2), 270–291. https://doi.org/10.1037/a0028228
Miller, S., Mcculloch, S., & Jarrold, C. (2015). The development of memory maintenance strategies: training cumulative rehearsal and interactive imagery in children aged between 5 and 9. Frontiers in Psychology, 6(MAY), 524. https://doi.org/10.3389/fpsyg.2015.00524
Nee, D., Brown, J., Askren, M., Berman, M., Demiralp, E., Krawitz, A., & Jonides, J. (2013). A Meta-analysis of Executive Components of Working Memory. Cerebral Cortex, 23(2), 264–282. https://doi.org/10.1093/cercor/bhs007
Nelson, P. M., Parker, D. C., & Zaslofsky, A. F. (2016). The relative value of growth in math fact skills across late elementary and middle school. Assessment for Effective Intervention, 41, 184–192. https://doi.org/10.1177/1534508416634613.
Roediger, H., & Karpicke, J. (2006). Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention. Psychological Science, 17(3), 249–255. https://doi.org/10.1111/j.1467-9280.2006.01693.x
Schwaighofer, M., Fischer, F., & Bühner, M. (2015). Does Working Memory Training Transfer? A Meta-Analysis Including Training Conditions as Moderators. Educational Psychologist, 50(2), 138–166. https://doi.org/10.1080/00461520.2015.1036274
Storkel, H., Voelmle, K., Fierro, V., Flake, K., Fleming, K., & Romine, R. (2017). Interactive book reading to accelerate word learning by kindergarten children with specific language impairment: identifying an adequate intensity and variation in treatment response. Language, Speech, & Hearing Services in Schools, 48(1), 16–30. https://doi.org/10.1044/2016_LSHSS-16-0014
Thalmann, M., Souza, A., & Oberauer, K. (2019). How Does Chunking Help Working Memory? Journal of Experimental Psychology: Learning, Memory, and Cognition, 45(1), 37–55. https://doi.org/10.1037/xlm0000578
Turley-Ames, K. J., & Whitfield, M. M. (2003). Strategy training and working memory task performance. Journal of Memory and Language, 49(4), 446–468. https://doi.org/10.1016/S0749-596X(03)00095-0
Vlach, H., Ankowski, A., & Sandhofer, C. (2012). At the same time or apart in time? The role of resentation timing and retrieval dynamics in generalization. Journal of Experimental Psychology: Learning, Memory and Cognition, 38(1), 246–254. https://doi.org/10.1037/a0025260
Yaw, J., Skinner, C. H., Orsega, M. C., Parkhurst, J., Booher, J., & Chambers, K. (2012). Evaluating a Computer-Based Sight-Word Reading Intervention in a Student with Intellectual Disabilities. Journal of Applied School Psychology, 28(4), 354-366.