Interesting tweet from @griff2742
My main thought on this is “proceed with caution”. There is certainly evidence supporting interleaving, but I’m not at all convinced that evidence supports re-designing a science curriculum in a way that breaks up topics more than we already do. However, I do think it indicates we should build in regular opportunities for revision and make use of links within, and sometimes across, topics.
Evidence that interleaving supports learning has been around for a while in education e.g. Dunlosky (2013), and before that I came across a similar idea in motor skill acquisition. The Learning Scientists have an excellent podcast summarising these ideas.
As people in science education may not be familiar with motor skill acquisition, this is a brief aside. Learning a skill that can be performed left- and right- handed (like shooting in basketball, or rolling a kayak) by practising on both sides at once is better than learning it just on one side first and then learning it again from scratch on the other side. It takes longer to reach initial competence because swapping sides slows down the acquisition of motor engrams (“muscle memory”) but it doesn’t take twice as long so overall is more efficient and it avoids ending up with a preferred side so the skill ends up stronger. I’m struggling to find articles on this but it was a big thing picked up by kayaking instructors in the ’90s.
However, whether learning to roll a kayak left- and right- handed together rather than right-handed first, or calculating the volume of four different geometrical shapes rather than one at a time (Rohrer and Taylor 2007) this is interleaving of very similar things. It perhaps works by introducing ‘desirable difficulties’ (Bjork 1994) or increasing ‘germane load’ (Sweller, van Merrienboer and Paas 1998) although these both seem to be somewhat slippery constructs. Alternatively, it may be that it helps to extract the key learning from the specific context somewhat, or helps expose the deep structure of problems, or highlights important differences between similar ideas, making the learning more flexible and therefore more useful. Willingham (2002) explains that new learning tends to be inflexible, initially. (I think this is a paper every teacher should read, by the way.) But I think it’s possibly just me making a connection between this idea and interleaving so I might be talking rubbish. Either way, my point is that these versions of interleaving, which have been shown to be helpful, are about interleaving similar things.
Interleaving during revision is about interleaving different things but this isn’t new learning. Plus, there is some evidence that if the revision content is too diverse, the interleaving doesn’t help (Hausman and Kornell 2014) (but note this is a single study learning very isolated items).
So. my conclusion is that interleaving is a good idea when learning very similar things for the first time, and when revising fairly similar things (like jumping around topics when revising GCSE Science) but I don’t think there is much support for shifting from a curriculum model that delivers new content in a series of linear topics, to a model that jumbles things up more than we are used to in secondary science. If you think about it the students already experience a massive amount of interleaving just from the design of the timetable. Even at A-Level they’ll have interleaved maths and chemistry, plus the rest of life, between every physics lesson. And across Key Stage 3 and 4, we re-visit most topics in some version of a spiral curriculum. In fact my inclination (although it’s just hunch with a bit of support from Mastery work) is to block the curriculum more rather than less. It’s only half thought-through but I wonder if spending longer on fewer topics at KS3 might mean fewer children experiencing science as a series of disjointed areas from which we move on before they’ve been properly grasped.
On the other hand, I’m definitely an advocate for a curriculum model that keeps coming back to previous topics in the form of regular revision. There are very few things I’d put money on as being demonstrably effective in science teaching but I am pretty sure that, with linear GCSEs and A-Levels, interleaving revision of earlier topics is a winner. I think Benney (2016) might have written the most apposite thing I’ve read about applying this principle in secondary science.
I also think, on the back of what I know about schema acquisition, that it’s a good idea to try to make links between ideas as a topic develops. Taylor (2019a) has explained this very nicely in relation to teaching chemistry of Groups 0, 1 and 7, and re-assuringly agrees with me about not breaking up topics (that’s his image I’ve used for this post). He has then written a subsequent post (Taylor 2019b) with some further examples and a tiny bit of quantitative data (nice, but obviously needs extensive replication). There may well be some other excellent subject-specific ideas in all the delicious #cogscisci goodness to be found on the Curriculum Links Part 2 page.
So that’s my thoughts, based on some relevant reading and my experience of teaching science. As always, I might be wrong.
Benney, D. (2016). Optimal Time For Spacing Gaps (?). mrbenney, [blog]. Available at < https://mrbenney.wordpress.com/2016/11/03/optimal-time-for-spacing-gaps/ > [Accessed 1 October 2020].
Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe and A. Shimamura (Eds.), Metacognition: Knowing about knowing (pp.185-205). Cambridge, MA: MIT Press.
Dunlosky, J. (2013). Strengthening the Student Toolbox: Study Strategies to Boost Learning. American Educator, Fall 2013 pp. 12-21.
Hausman, H. and Kornell, N. (2014). Mixing topics while studying does not enhance learning. Journal of Applied Research in Memory and Cognition, 3(3), pp. 153–160.
Rohrer, D., Taylor, K. (2007). The shuffling of mathematics problems improves learning. Instructional Science 35, pp. 481–498.
Sweller, J., van Merrienboer, J. and Paas, F. (1998). Cognitive Architecture and Instructional Design. Educational Psychology Review 10, 251–296.
Taylor, I. (2019a) This is Interleaving: A Concrete Example. Mr T’s Blog: Keeping it Simple, [blog]. Available at < https://mrtaylorsblog.home.blog/2019/03/03/this-is-interleaving-a-concrete-example/ > [Accessed 1 October 2020].
Taylor, I. (2019b) Teaching by Contrast. Mr T’s Blog: Keeping it Simple, [blog]. Available at < https://mrtaylorsblog.home.blog/2019/04/28/teaching-by-contrast/ > [Accessed 1 October 2020].
The Learning Scientists (2017). Episode 8 – Interleaving [podcast]. Available at < https://www.learningscientists.org/learning-scientists-podcast/2017/12/6/episode-8-interleaving?rq=interleaving > [Accessed 1 October 2020].
Willingham, D. (2002) Ask the Cognitive Scientist: Inflexible Knowledge: The First Step to Expertise. American Educator. Winter 2002.