Interleaving in Practice: insights and observations from a TSA research project

First posted on the Herts & Bucks TSA blog on 29 March 2020.

I hated times tables when I was young. I struggled and tended to find the memorisation of them an utterly miserable experience. Three years ago, I went into a maths class as part of my ‘SLT on tour’ duty and the well meaning maths teacher asked me what 6 X 7 is. I gave the wrong answer. Then panicked, thought about it, and gave another wrong answer. Everyone seemed embarrassed. 

Therefore, I am determined not to let my son experience this. We practice times tables regularly at home. This had been going badly until…

My mistake had been elementary; whereas my wife mixed the times tables up immediately after he had mastered each table in turn, I had been wasting time getting him to go over each times table in a rote fashion. Although my wife’s method is in no way groundbreaking – most parents will do this and it is similar to most peoples’ use of flash cards – it does demonstrate a very simple example of what is called ‘interleaving’. 

Interleaving is a method used to revisit learning, put spaced review into practice and improve memory recall. Moreover, interleaving emphases mixing pre-taught topics when planning spaced or distributed practice in order to facilitate deeper consideration of the knowledge being recalled. As Stephen Pan (2015) says, ‘Mixing it up boosts learning’ compared to more traditional methods of block learning where students master one topic before moving on to the next in preparation for exams. 

Interleaving improves the brain’s ability to tell apart, or discriminate, between concepts. The brain continuously focuses on searching for different solutions…which better enables you to execute the correct response.’ (Stephen C. Pan)

In terms of planning learning, interleaving is the opposite of blocked learning, which means pupils learn each topic in turn as opposed to ‘mixing up’ the subjects (see diagram). 

Screen Shot 2020-03-23 at 09.23.31
An interleaved curriculum (from @DavidDidau)

Essentially, interleaving builds upon research advocating distributed practice and retrieval. For example, purposely revisiting taught content and practicing relevant problems in activities that are distributed or spaced over time (i.e: A1….A2…..A3) results in better knowledge retention, especially in terms of long-term memory, than if the lessons or activities are blocked together or limited to a closed sequence of lessons where no other content gets revisited or reviewed (i.e: A1A2A3). 

Instructors at MIT suggest teaching content in an interleaved fashion so that a activity/problem related to one aspect of previously taught unit content is followed by a different activity/problem type related to a different aspect of taught unit content (i.e: A1B1C1B2C2A2C3A3B3). This has been found to have higher learning gains than simply practicing activities/problems grouped by types (i.e: A1A2A3B1B2B3C1C2C3). Moreover, they clearly advocate combining distributed/spaced practice with interleaved subject content:

‘Although spacing and interleaving can be separated into two different interventions, interleaving results in the spacing of the same problem types. Therefore, interleaving and spacing practice are linked and often used together (A1B1C1….B2C2A2….C3A3B3). Although spacing and interleaving can mean slower initial learning, they result in both increased retention and better ability to discriminate between problem types.’ (See here for more).

The comments cited above are gradually being backed up by plenty of empirical evidence. For example, Doug Rohrer of the University of Florida has demonstrated that whilst pupils using blocked practice were able to recall subject content more accurately in their immediate lesson (or learning), pupils who used interleaved practice were better at recalling subject content the next day.

Moreover, Rohrer found that interleaving helped pupils distinguish between information taught, especially when that information was similar; this avoided pupil confusion between concepts or terms when retrieving knowledge. For example, pupils often mistake words with similar spellings, such as ‘allusion’ instead of ‘illusion’, or confuse strategies when problem solving, including in mathematics.

Similar studies have reported the benefits of interleaving on the learning of:

  • geometric concepts (e.g., Taylor & Rohrer, 2010) and algebraic rules (e.g., Mayfield & Chase, 2002);
  • classify paintings of particular artists (Kornell & Bjork, 2008; Kang & Pashler, 2012); 
  • classifying different types of birds (Wahlheim, Dunlosky & Jacoby, 2011);
  • problem solving in engineering (Butler, Marsh, Slavinsky & Baraniuk, 2014);
  • medical knowledge and skills (Kao, 2016).

Why does interleaving work?

  • One prominent explanation why interleaving works is that it improves the brain’s ability to tell apart, or discriminate, between concepts (Pan, 2015).
  • A second explanation is that interleaving strengthens memory associations between these concepts (Pan, 2015).

How does interleaving boost cognitive skills?

It is clear from reading various blogs on interleaving that a number situations that occur in its use facilitate enhanced learning, particularly in pupils’ ability to recall subject knowledge. These include: 

  • Desirable difficulty – Interleaving is seen as aiding the recall of information as mixing topics up creates a desirable difficulty, which forces the brain to work harder and improve retrieval, especially in terms of long-term memory (Bjork & Bjork, 2011; see also Birnbaum, Kornell, Bjork & Bjork, 2012).
  • Contextual interference – This is another mechanism for improving the retrieval of information from the long-term memory; the learner has to identify between various concepts, especially those that might be similar (Rau, Alevena & Rummel, 2012)
  • Elaborative rehearsalThis is where learners focus on remembering the meaning of items that they are memorising, instead of focusing on simply repeating those items verbatim, which is more conducive if retrieved knowledge is interleaved as it involves greater mental effort to distinguish the correct answers (Shea & Morgan, 1979).

However, as will be discussed below, some suggest that interleaving could potentially involve re-jigging the actual taught curriculum or schemes of work as opposed to simply mixing up certain activities, such as homework or starter activities. This article will briefly question the practicalities of the former – even if the evidence for doing so is strong – and focus on the latter. 

As part of a wider Herts & Bucks TSA research project, I have been applying interleaved activities/practice in my Religious Education classes at GCSE. Although the overall impact is hard to measure – as I am using various interventions – my classes have met with relative success since interleaving has been applied. 

For example, I have been setting interleaved starter questions (different to the current content being taught) as well as interleaved homework (one question on the current topic and an additional question on a separate content that has been previously taught). These interleaved tasks are distributed/spaced out as per the table below:  

Screen Shot 2020-03-14 at 11.52.30
Year 10 RE GCSE interleaved starter and homework questions

Here, the starter questions are part of my Do Now expectations. The first three questions are on the current topic taught, the next two on a previous topic and the last on another previously taught topic. In order to give adequate coverage to all units taught, these are planned out as per the table. Pupils have 8 minutes to complete the starter, which looks like this: 

Screen Shot 2020-03-14 at 11.52.44
Interleaved starter: first 3 questions are on the current topic; questions 4 and 5 are from a previous unit and question 6 is from another previous unit.

In addition to the above, pupils are cyclically assessed on previous taught content. Within a six or seven week half term, all my GCSE classes will sit at least two timed assessments in exam conditions and – with the exception of formal mocks – will have two sections from the GCSE syllabus; these include the current topic taught and a previously taught unit. The latter is planned in a sequence to ensure pupils are assessed proportionally in terms of the course’s overall content. 

This combination of interleaved starters, homework and assessments seems to be working: my previous GCSE cohort attained a Progress 8 score of 1.72 (the best I have ever had) and my current Year 11s are currently averaging a Progress 8 score 0.92, which I hope would go up in actual exams (unfortunately cancelled this year). Of course, I have no direct evidence that interleaving has impacted on these scores – it could well be the simple use of retrieval or spaced practice as well as other interventions. Moreover, I am still to canvass my pupils views on this (part of the next phase of the TSA research) to ascertain their impressions of interleaving, which is explained and always practiced in my classroom. 

Interweaving not interleaving

However, not everyone may see my use of interleaving as anything more than spaced retrieval practice. Mark Enser, for instance, suggests that when we look at what many teachers are calling ‘interleaving’, what we often see is a form of spaced practice where material is introduced at one point and then returned to later, taking advantage of the testing effect; this is not a bad thing in itself, says Enser, but it is not necessarily interleaving. 

Enser goes on to say that pupils are rarely taught something in one block of massed practice and then left to never return to it again. Instead, instruction and practice is spaced out with pupils returning to the topic at regular intervals through tests, assessments and – probably more often than not – starters and homework assignments like the ones I use. According to Enser this practice is mistakenly labelled as ‘interleaving’ as opposed to spaced or distributed practice. 

Enser also questions whether breaking up a syllabus into units and interleaving lesson-by-lesson (i.e. a lesson on technics (one unit), followed by a lesson East Africa (another unit), followed by a lesson on the local landscape (another unit) is impactful if lessons are spaced out on a timetable, particularly a two week timetable, where pupils might simply become confused as the teacher jumps from topic to topic. This was something the TSA working group completely agreed with. 

Subsequently, Enser suggests that there is a more effective way to approach interleaving, spacing and retrieval in the curriculum. He calls this approach ‘interweaving’. In this approach, he identifies themes and concepts that recur throughout different topics and plans to teach them in the context of those topics. 

For example, interweaving could work like this in geography:

  • In the first unit, pupils study the mechanisms behind plate movement and how these form different types of volcanoes etc. 
  • In the second unit, they look at East Africa where – the tectonics thread re-emerges as they look at how the Great Rift Valley formed and how it affects people living there today. 
  • The third unit, pupils consider their local surroundings and landscape and the role of tectonics in that area; 
  • The last unit focuses on Haiti with the thread continuing, as pupils learn how earthquakes in the region are caused and how this impacts on the country’s development. 

Enser argues that this allows pupils ‘to re-encounter the material in a more meaningful way’. However, the TSA research group felt that this approach is simply good practise; it is synoptic and practised by many teachers with a curriculum/syllabus conducive to this type of teaching. None of us disagreed with Enser’s overall argument, but did question how unique this approach is and – considering Enser’s criticism of other peoples’ approaches to interleaving as mere ‘curriculum design’ – questioned whether his own view is really an example of good curriculum design and planning. Nonetheless, this may be unfair on the points Enser is making, so please read his TES blog here).

Complete overhaul of the curriculum

My own application of ‘interleaving’ is very much ad-hoc to my main lessons, which admittedly follows a traditional block learning approach. An alternative approach is to completely overall the curriculum or syllabus you teach. One excellent example developed by Dawn Cox is shown below (a traditional blocked approach comes first followed by Dawn Cox’s version). 

Screen Shot 2020-03-16 at 14.15.29
Traditional ‘blocked’ GCSE RE curriculum/syllabus
Screen Shot 2020-03-16 at 14.12.31
Dawn Cox’s interleaved GCSE RE curriculum/syllabus

This long term plan for a Religious Education GCSE course completely reworks the blocked units found in the text books and AQA specification. However, it does not lose sight of covering all the content in a logical manner. In some ways it not only interleaves key aspects of the GCSE syllabus, but takes Enser’s idea of interweaving one step further as key themes, and important abstract concepts, are mapped across the course to re-emerge in a thread like fashion in distributed/spaced intervals. This is further exemplified by the additional plan below that highlights how the topics related to Jesus Christ are interleaved (or knowledge of Jesus tested and recalled) at spaced intervals over the timeframe of the course (please see Cox’s ResearchEd presentation here).

Screen Shot 2020-03-16 at 14.12.51
This version of Cox’s plan shows where the life of Jesus will be taught and potentially assessed.

Although this impressed colleagues when we met to discuss interleaving, many felt that the change was too radical and that members of their departments might object. It was also suggested that many would not have the time to plan this. However, arguably any reworking need not be immediate, but gradually changed over time. In an email Dawn Cox also suggested that interleaving can take many forms; planning for as much retrieval as possible is essential here. 

Does this mean my own interleaving strategies are all wrong?

The ideas of both Mark Enser and Dawn Cox have made me question whether I am actually using interleaving strategies effectively or even at all. Nonetheless, Itamar Shatz, in a blog he calls Effectiviology, claims that research generally pinpoints a number of key ideas that should be followed if we are to label anything as ‘interleaving’. This has raised further questions over my approach to interleaving. Shatz’s points are outlined below with my initial responses in brackets and italics (you can read Shatz’s original blog here).   

  1. Mix things up in a way that makes sense. The criteria that you use when deciding how to mix the material should allow you to benefit from interleaving as much as possible. (OK, I think I do this in RE as all material is relevant – as we start with beliefs and practices in both Christianity and Islam everything that follows build upon these foundations; thus interleaved recapping makes sense).
  2. Avoid interleaving topics that are too different. If the two topics that you want to interleave are so widely different from one another, the mixed practice might end up hindering your learning, instead of facilitating it. (Oh, I was under the impression that I could mix up topics, which is exactly what I have been doing as part of my use of ‘interleaving’).
  3. Avoid interleaving topics that are too similar. If the two topics that you want to interleave are too similar to one another, it might feel more like blocked practice, at which point you won’t benefit from interleaving. (Oh, some of the beliefs in RE transgress topics; they are applicable in different situations).

However, on further reflection, and in discussion with the TSA research group, I feel that I am covering the points listed above in my more ad-hoc/half-way approach to interleaving. These are my responses to Shatz’s latter two points (in italics):

  • Avoid interleaving topics that are too different –  while interleaving different types of maths questions might be beneficial, interleaving maths questions with history lessons could end up being more confusing than helpful. (Fine, my interleaved activities will be in the same subject area. They will exhibit contextual interference by being interleaved). 
  • Avoid interleaving topics that are too similar – if your goal is to interleave different types of basketball throws, then simply switching between two forms of the same throw that are barely different from one another will likely not allow you to take advantage of the benefits of interleaving. (None of the interleaved questions or activities I use will be that similar). 

Defending my use of  interleaving!

  • Despite the idea that we should ‘interweave’ or break up our curriculum (and I would advocate these if feasible), a number of studies seem to suggest interleaving is still present if only limited to starter, homework and/or assessment activities. 
  • A study by cognitive psychologist Doug Rohrer and colleagues at the University of South Florida, published in the Journal of Educational Psychology in 2015, suggests the above. Interestingly, Pan suggests that Rohrer and his team are the first to research interleaving in actual classrooms; their studies on pupils learning about geometry and algebra involved homework and testing assignments as interleaved practice (see Rohrer, Dedrick & Stershic, 2015). 
  • Also, researchers suggest a way to ‘accomplish effective interleaving’ might be to ‘incorporate into current lessons and class activities brief reviews of previously-learned information’. These reviews could also be implemented as homework assignments, which may be particularly advantageous when class time is limited (see, for example, Carpenter, 2014). 

Conclusions to date

Although my own use of interleaving is part of a wider inter-school research project, I have reached some basic – perhaps self-evident – conclusions so far. 

  • Interleaving is a learning technique which involves mixing together different topics or forms of practice, in order to facilitate learning.
  • Interleaving is beneficial for several reasons, and most notably because it encourages in-depth processing of the material, by prompting people to use various learning strategies, due to the contextual interference that people experience when shifting between topics.
  • When implementing interleaving in your learning, you should make sure to interleave topics at a reasonable rate, and to avoid interleaving topics that are too different or too similar to one another.
  • Because different forms of interleaving will work better in different scenarios, there is no single method of interleaving that will work best for everyone, so you should experiment with different types of interleaving in order to figure out what works best for you.

References

Birnbaum, M., Kornell, N., Bjork, E. and Bjork, R. (2012). ‘Why interleaving enhances inductive learning: The roles of discrimination and retrieval’. Memory and Cognition, 41(3), pp.392-402.

Bjork, E. L., & Bjork, R. A. (2011). ‘Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning’ in M. A. Gernsbacher, R. W. Pew, L. M. Hough, & J. R. Pomerantz (Eds.), Psychology and the real world: Essays illustrating fundamental contributions to society (pp. 56-64). New York: Worth Publishers.

Butler, A. C., Marsh, E. J., Slavinsky, J. P., & Baraniuk, R. G. (2014). ‘Integrating cognitive science and technology improves learning in a STEM classroom.’ Educational Psychology Review, 26(2), 331–340.

Carpenter, S. K. (2014). ‘Spacing and interleaving of study and practice’ in V. A. Benassi, C. E. Overson, & C. M. Hakala (Eds.), Applying science of learning in education: Infusing psychological science into the curriculum (p. 131–141). Society for the Teaching of Psychology.

Kang, S. & Pashler, H. (2012). ‘Learning Painting Styles: Spacing is Advantageous when it Promotes Discriminative Contrast’. Applied Cognitive Psychology. 26. 97-103. 10.1002.

Kao R, 2016, ‘Inter-blocking: Integrating a Process of Inquiry Mindset into Medical Education’. MedEdPublish, 5, [1], 20.

Kornell, Nate & Bjork, Robert. (2008). ‘Learning Concepts and Categories Is Spacing the “Enemy of Induction”?’. Psychological science. 19. 585-92. 10.1111.

Mayfield, K. H., & Chase, P. N. (2002). ‘The effects of cumulative practice on mathematics problem solving’. Journal of Applied Behavior Analysis, 35(2), 105–123.

Pan, S. (2015). ‘Up Boosts Learning‘. Scientific American. August 4, 2017.

Pan, Steven & Pashler, Harold & Potter, Zachary & Rickard, Timothy. (2015). ‘Testing enhances learning across a range of episodic memory abilities.’ Journal of Memory and Language. 83. 10.1016.

Rohrer, D. (2012). Interleaving helps students distinguish among similar concepts. Educational Psychology Review, 24, 355-367.

Rohrer, D., Dedrick, R. F., & Stershic, S. (2015). Interleaved practice improves mathematics learning. Journal of Educational Psychology, 107(3), 900–908.

Rau, M.A., Aleven, V. & Rummel, N. (2010). Blocked versus Interleaved Practice with Multiple Representations in an Intelligent Tutoring System for Fractions in Aleven, J. Kay, & J. Mostow (Eds.): ITS 2010, Part I, LNCS 6094, pp. 413–422, 2010.

Shea, J. B., & Morgan, R. L. (1979). ‘Contextual interference effects on the acquisition, retention, and transfer of a motor skill’. Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179–187.

Taylor, Kelli & Rohrer, Doug. (2010). The Effects of Interleaved Practice. Applied Cognitive Psychology. 24. 837 – 848. 10.1002.

Wahlheim, Chris & Dunlosky, John & Jacoby, Larry. (2011). ‘Spacing enhances the learning of natural concepts: An investigation of mechanisms, metacognition, and ageing.’ Memory & Cognition. 39. 750-63. 10.3758.

Picture credit: Overlay Colours 1 by BA 1969 (used under a Creative Commons Licence)

 

 

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s