top of page
Search
  • Writer's pictureChristian Moore Anderson

Moving on from CLT fever: Perspectives from the classroom

Updated: Aug 26, 2023

I came to Twitter in 2012 in my first year teaching science, and it wasn't long before cognitive load theory (CLT) began its ascent in popularity on Edutwitter. But why did CLT become so popular in UK education?


In my perspective, and I'm sure there'll be those that disagree, it became popular not for its own merits per se, but as tool to defend against bad practice in schools. At the root of the problem was the lack of educational models permeating schools, which may have contributed to a rise in strange activities and games in the classroom.


I certainly don't ever remember having CPD on how students learn. It felt like it was just everyone else's anecdotes versus yours. CLT presented teachers with a simple model to replace the partial vacuum.


Crucially, CLT presented a model for learning which had a body of evidence behind it; more than those anecdotes. CLT came in to challenge the status quo, to help the profession move on from abusive observations, restrictions on 'teacher talk', and the pressure to provide fun lessons with games, rather than deep learning. That list could be longer.


CLT with its model of information processing certainly helped us move on from this, but the movement then became a behemoth as EduTwitterans took forward their interpretation of it.


CLT completed its remit, but got greedy, and began to swallow the whole debate on learning. No longer was it a model of learning, its ambition was to become the theory of education. All educational theories are partial, so there was only one way about it, by eating up everything else and arguing down anyone who saw the crumbs.


What did CLT ever do for me?

I read around CLT, and the blog posts by teachers on Twitter who put it into practice. And it improved my teaching at the level of the explanation, or the activity. I'm a big fan. Here's some examples of the practical results, which I think many other UK teachers will share with me:

  • I teach entirely by drawing on paper and a visualiser, while continually in dialogue with my class. Limiting cognitive load by building the picture as we go, retrieving and connecting to prior knowledge as we proceed.

  • I think carefully about the explanation I'll give, and plan the core of it beforehand on paper, including some questions.

  • I produce a set of core questions for my courses. This helps me plan, and gives the students a tool for self-quizzing.

  • When I plan tasks, I think about the cognitive load, depending on intrinsic and extrinsic factors, such as element interactivity, and support given.

  • For my science practicals, I now produce instructions using pictures, rather than those recipe style lists of text that suppose more cognitive load.


What has CLT not done for me?

I'm a better teacher as a result. But, there is a problem. None of this stuff has ever provided a model on how my students will make meaning of what they study. And this is surely the ultimate goal of our teaching.


Instead of searching for new models of learning to provide alternative views, multifaceted views, and understand the limits of each, EduTwitterans sought to push their subjective interpretations of CLT into domains where they did not belong. In a gross simplification, I see it a bit like this:


The expert-novice continuum and CLT were used by EduTwitterans as strong evidence that we should not have students imitating experts in the classroom. They had to walk before they could run. All well and good.


But then this is where EduTwitterans took CLT beyond its remit: we know from research that experts think differently to novices because they have a vast bank of subject specific knowledge while novices don't.


So, it was argued, we just have to teach them vast amounts of knowledge. We know that people forget things, but we're in luck. The cog sci work on retrieval practice, and the testing effect, means we have the tools to have students never forget all this vast quantity of knowledge.


What's missing? Understanding. It's not just that experts know more, they have made meaning of important core concepts and are able discern those core patterns easily in different situations.


The popular interpretation of CLT sheds no light on meaning making. Instead of searching for models that could help us with meaning making, EduTwitterans took to schemas, which are visually represented as lots of connected dots.


Cog sci tells us of schema and these images, but they show how understanding may be represented at an end point. It doesn't provide a mechanism for teachers to exploit.


CLT can't tell us what knowledge, or what connections, or how to help students make connections. Ultimately, CLT is a theory of constraints on learning, not a theory of learning itself. In CLT terms, it isn't a theory of how 'encoding' occurs. I tells us nothing of meaning-making, just ways in which the opportunities for meaning-making may be inhibited.


Nevertheless, the two were combined by EduTwitterans. Now, we're told, all we need is lots of knowledge (don't let them forget) and lots of connections. Understanding will emerge on its own.


The EduTwitter definition of learning—taken up by the UK government— "a change in long term memory" is based on EduTwitteran interpretations of CLT. Which allows learning to be anything from verbatim rote learning, to the learning of misconceptions, or even forgetting. How helpful is that to a teacher?


Marton and Booth (1997) have a definition I find useful for my teaching: Learning is a change in way of seeing. The world the student experiences changes. Oh the beauty of such a definition, which completely omits verbatim recall as learning. (That's not to say that there are instances where rote learning can be stepping stone to something better).


Some EduTwitteran CLT proponents would laugh at the lack of precision, I'm sure. But it's like that joke: A scientist is, at night, on the floor under a street lamp searching around. Another scientist comes along and asks what they're doing. They respond that they'd lost their keys on the other side of the road, but that they can see better under the street lamp.


This is 'learning is a change in the long term memory'. Easy to see, easy to measure, easy to teach, but not what matters. Ultimately, content knowledge fades with time if not used. We all know this. But a change in way of seeing, once seen, is never unseen. Once students make meaning of something, it's forever.


For example, I think the biology practical of testing for types of food is a waste of time. This is because there is nothing for students to discern, and therefore nothing that can change their way of seeing the world. It's just a recipe style practical experiment that they may learn verbatim to gain some marks on an exam.


Yet, if you only take to the current, CLT influenced, interpretation (in UK education) on learning, then you can tick the box of connections and claim you're building schema. After all, it's testing food, and food also appears in the topic of the digestive system. Here all connecting is equal. But we all know that it isn't.


On evidence

The irony is that CLT first rose to popularity because they brought 'evidence' to schools. All well and good, teaching should be an informed profession. But it was then subsequently, and subjectively, through conjecture, extrapolated to whatever a teacher wanted. And this is the problem, as it has been used to answer classroom questions that CLT tells us nothing about.


The novice-expert divide is a good case in point. Research that sought to understand the difference between real experts and real novices, has been used to justify the pedagogy of single lessons. First the behemoth gobbled up pure discovery learning, but then generalised across all learning. Students are novices and must therefore get all their learning from the teacher.


In a two recent blog posts (here, and here) I've challenged this, and got some of the kick back I had anticipated from those who only look to CLT and its associated theories. The novice-expert divide generalised to single lesson pedagogy tells us not to carry out a pedagogy of problem-solving first.


And yet, two recent meta-analyses have both found that for conceptual knowledge (not procedural knowledge) problem-solving before teaching is more beneficial than the reverse (Chen and Kalyuga 2019; Sinha and Kapur 2021).


With carefully planned pedagogy you can create the conditions for students to discern new patterns for themselves, which is highly fruitful and efficient for developing understanding (and do check the blog posts, this isn't about unguided discovery learning).


It works, yet we're told it won't. To attempt to make sense of this, EduTwitterans who only see learning through CLT will switch to say that the students were obviously experts. Yet they aren't, they are discerning patterns for the first time in my classroom, new patterns, they cannot transfer from previous experience.


If you only know CLT, this is impossible to understand.


Here something worked that wasn't supposed to, and yet I've also experienced failure where it was supposed to be a success. At the beginning of CLT fever I designed core questions and ran weekly quizzes with my older students, and fortnightly with younger students. But, the data never correlated well with larger assessments that I designed to test understanding.


What's more, I noticed a distinct switch in my students' conception of learning towards a more surface approach based on verbatim recall. Since removing this and switching to a new assessment framework I have designed, students have swung back to seeking meaning.


There's certainly more evidence that this happens in students (Trigwell and Prosser, 2020). Now core questions serve more as a safety net for students, and a planning tool.


The kick back will be that I didn't do it well, my questions weren't right. But no one can claim I'm one for half jobs. My blog posts and published work show how deeply I think about my practice. I even have a blog post on designing core questions for quizzes. If I'm not getting it right, who will?


The popular focus on knowledge, the novice-expert distinction, retrieval practice, and models of memory have all forced a focus on content to be learnt. If we asked 'what is to be learnt?' A CLT approach would make you list out content to know. A meaning making approach would have you ask what you want students to discern.


The former has led to 'just teach them and don't let them forget'. The latter leads to thinking about the necessary conditions (carefully planned pedagogy and explanations) for which students can discern for themselves the meaning.


A CLT focus has benefits that can add to multiple educational models, but alone it's blinding. Recently I've seen teachers discuss how blogging is dying out, and people say their own ideas are drying up. This is because there are limits to what we can get from CLT, and what it can give us is about basic classroom task creation or explanation. These have been explored and teachers have come up against the limits. Soon teachers will move forward to include other ideas.


Only through seeing differences can we find meaning. Indeed, what should they know of CLT who only CLT know? Without understanding other educational theories, how can you know where CLT models end and reality begins?


As a profession we need to explore other educational models and see their differences and limits. I predict this will happen soon, and when it does, we'll see a resurgence of teachers blogging about 'what I used to do, and what I do now.' If you've liked this then check out my book. Download chapter 1 here—English edition—edición española—or check out my other posts.


@CMooreAnderson (twitter)





References


Chen, O., and Kalyuga, S. 2019. “Exploring factors influencing the effectiveness of explicit instruction first and problem-solving first approaches.” European Journal of Psychology Education 35: 607–624.


Marton, F., and Booth, S. 1997. Learning and Awareness. New York: Routledge.


Sinha, T., and Kapur, M. 2021. “When Problem Solving Followed by Instruction Works: Evidence for Productive Failure.” Review of Educational Research 91 (5): 761–798.


Trigwell, K., and Prosser, M. 2020. Exploring University Teaching and Learning: Experience and Context. Switzerland: Springer Nature.







1,711 views
bottom of page