Just what is working memory anyway? And what does it mean for language teachers?
By: Caroline Handley
Watching Peter Doolittle’s TED talk (an excellent introduction to working memory), two questions came instantly to mind, one based on prior reading about the brain, and the other more of a speculation about the implications for foreign language learning.
First, I was immediately reminded to question how working memory is defined. Just what do we mean when we talk about working memory capacity anyway? In Peter’s talk, and in the literature, the dividing line between working memory and attention seems very fuzzy, to the point where you have to wonder, does it really exist at all? Individual differences in working memory capacity are frequently associated with differences in the ability to control internal and external distractions (mind wandering and off-task sensory input). Similarly, mind wandering is associated with the default network, and a reduction in default network activity co-occurs with an increase in activation across the prefrontal cortex during successful completion of working memory tasks. So when we talk about students having high or low working memory capacity, are we really just talking about differences in the ability to focus effectively on the target task? Moreover, if working memory is the capacity to attend to what we are currently doing, how does this disassociate from the aspect of consciousness typically referred to as awareness? This implies that working memory could basically be understood as attention to tasks whose performance requires conscious cognitive processes (i.e., are not automatically carried out by subconscious neural processes). Another way of looking at this (as discussed in information theory) is the signal to noise ratio: People with high working memory capacity are good at focusing on the signal while ignoring the noise, whereas people with low working memory capacity don’t filter out so much noise and so process and store less of the signal.
What does this mean for educators? For me, there are three main implications for the classroom. One, this fuzziness is reassuring, as although I know I can’t do much to improve the working memory capacity of my students, encouraging them to stay on-task seems much more aligned to the pedagogical aim of creating engaging materials at the right level of challenge. Of course, this will still favour students with high working memory capacity, who, if I’ve been successful in my material creation, will now be super-focused and extract maximum information, and thus show greater learning gains (and, in fact, Peter himself has published research suggesting high working memory capacity is correlated with increased learning from multimedia sources).
Two, this view of working memory capacity suggests that exceeding capacity is not just about giving students too many things to remember at once, but also giving them low-priority activities at the same time as high-priority ones. Sometimes one signal needs as close as it can get to 100% attention. For example, if, as we are giving instructions, we also ask our students to pass around a handout, we are increasing the noise around the intended signal. Whereas if, as we give instructions, we also display them in writing (say via slides on a TV or projector), we reinforce the signal. This also increases the duration of this signal, allowing low working memory capacity students to pay repeated attention to it, as needed.
And three, the need to guide attention towards the signal and away from noise provides another reason why students benefit from knowing what the aim of the activity is before they start trying to perform it. If they know in advance which features of the activity they need to attend to, it must surely be easier for them to use and produce task-relevant information.
My second question is much more of a speculation—or an area for future research! Working memory is also associated with long-term memory. One way of viewing working memory is as the connection between attention and long-term memory, or prior knowledge. Prior knowledge can be divided into semantic and autobiographical memory, which are distinct yet interacting knowledge representations. The semantic component seems to help structure episodic memories by contributing gist knowledge (for example, a few days after teaching a class, you might not be able to remember whether an absent student was actually present or not, as although they were absent on that one occasion, your gist memory of them being part of the class on seven previous occasions in the last month interferes with the episodic memory of that day). Your extant knowledge guides what you attend to in your immediate environment, and this interaction determines learning.
But what I find relevant to teaching English as a foreign language is the suggestion that this extant knowledge is so efficient in guiding human behaviour because it is automatically “tagged” by language. The implication of this is that, in learning a foreign language, working memory (attention) is guided by previous knowledge, which is inherently tagged by the native language. This idea is supported by research showing that we remember the gist of what we read in a second or foreign language in our first language. In other words, the idea of immersion language teaching, in which only the L2 is used, is completely impossible to achieve, as all student learning is immersed in their prior, linguistically-tagged knowledge. But the big question for me is: How much does this inevitable intrusion of the L1 interfere with L2 learning? More specifically, what does it add to the working memory load associated with performing tasks and exercises in the L2? When we allow time for planning output or responding to input, do we allow sufficient time for the working memory load of using the L1 as well as the L2 to integrate the L2 input into our previous (long-term) knowledge? Do we have any idea what sufficient time is, for learners with high or low working memory capacity?
Overall, I guess the key reminder for me from Peter’s talk, and reflecting on working memory in the foreign language learning classroom, is that is highlights just how important it is that activities are engaging enough to keep distraction at bay as much as possible, yet also allow sufficient processing time so as not to overload learners. With such a delicate balancing act to achieve, and given it will vary according to individual differences between learners, maybe it shouldn’t be so surprising when a lesson plan works perfectly in one class yet falls to pieces in another!
Further Reading
Doolittle, P. E., & Altstaedter, L. L. (2009). The effect of working memory capacity on multimedia learning: Does attentional control result in improved performance? Journal of Research in Innovative Teaching, 2(1), 7-25.
Lunqvist, M., Rose, J., Herman, P, Brincat, S. L, Buschman, T.J., & Miller, E.K. (2016). Gamma and beta bursts underlie working memory. Neuron, 90, 152–164.
Ma, W. J., Husain, M., & Bays, P. M. (2014). Changing concepts of working memory. Nature Neuroscience, 17(3), 347–356.
Unsworth, N., Engle, R. W. (2007). The nature of individual differences in working memory capacity: Active maintenance in primary memory and controlled search from secondary memory. Psychological Review, 114(1), 104–132.
Caroline Handley is an Associate Lecturer at Seikei University and a PhD candidate at Swansea University in Wales. Her main research interests are in vocabulary and the mental lexicon, and the interaction between linguistic and conceptual knowledge.