Desirable difficulties: A not so difficult approach

First published by SecEd Magazine on 7 December 2022.

Creating desirable difficulty in the classroom is a key part of lesson planning and delivery, but we must be cautious about clumsy interpretations of what this means.

Although the concept has been around for sometime, over the last couple of years I have heard more and more colleagues referencing “desirable difficulties” in blogs and on #edutwitter.

The idea of desirable difficulties is worth continually pursuing as we plan, devise and teach lessons. However I cannot help but think that it may be initially daunting for busy teachers to explore this concept without having the time to fully unpack what it entails.

Moreover, a haphazard approach to implementing these difficulties could make them redundant or even “undesirable”.

What are ‘desirable difficulties’?

A desirable difficulty is a learning task that is not particularly easy and requires a desirable amount of effort to complete or solve. This effort, it has been found, can improve long-term memory and performance (Bjork, 1994; Bjork & Bjork, 2011).

However, compared with other teaching methods or strategies, desirable difficulties can initially slow learning down and seem counterproductive.

Bjork and Bjork (2011) suggest desirable difficulties can include: varying the conditions of learning; spacing practice sessions or study; interleaving topics as opposed to teaching them in blocks; using the generation effect; and using tests.

In a similar vein, Jeffrey Bye (2011) in his list of effective desirable difficulties suggests: getting pupils to generate the target material through a puzzle or similar task or process; making the target material less clearly organised for pupils who already have an appropriate amount of background knowledge; and using fonts that are slightly harder to read.

Caution is needed

I am sure that some colleagues reading this, particularly those working with pupils with SEN, English as an additional language (EAL), or even apathetic pupils, will have some concerns that desirable difficulties, if implemented rashly or clumsily, could cause more damage than good; that they will become, essentially, undesirable difficulties (see Beale, 2020 for a good discussion on this). This risk is also acknowledged by the cognitive scientists proposing the idea.

Therefore, any application of desirable difficulties needs to be adapted and pitched according to pupils’ abilities. Teachers should avoid a gung-ho approach that bypasses this somewhat obvious point.

Moreover, simply telling teachers that this evidence-based approach is better than their current practice could lead to mistakes in how some of the “difficulties” listed above are applied in lessons. It is important that collaborative planning or peer-to-peer feedback is undertaken when considering incorporating these “difficulties” into teaching.

I would add that a number of blogs, articles and tweets championing “desirable difficulties” should explicitly warn that poor formulation could have a detrimental impact on learning.

Using ‘desirable difficulties’ in the classroom: Simple, not difficult

While delivering INSETs, I have had colleagues baulk at the idea of using the generation effect and spacing and interleaving, but this need not be the case. Some of the ideas outlined by advocates of desirable difficulties can be implemented in more simple ways than first presumed, for instance…

The generation effect

If you search online for the “generation effect” you are likely to come across explanations like this: “The generation effect is a robust memory phenomenon in which actively producing material during encoding acts to improve later memory performance.” (Rosner et al, 2013).

However, fear not, the generation effect is relatively simple and many of us do it already. It is the idea that pupils generate the target material from cues devised by the teacher as opposed to merely reading it. Essentially, pupils need to recall and/or generate words, terms and ideas in order to deepen learning. So long as they are not designed to be too easy or straightforward, generation effect activities could include:

  • Letter stems or word stem completion.
  • Word fragment completion.
  • Word fills.
  • Anagrams.
  • Cryptic crosswords.
  • The generation of synonyms or antonyms (or opposing ideas).
  • Verbal scaffolding, even call and response.

Spacing and interleaving

Spacing suggests allowing pupils time to forget aspects of previous learning as the gap between recall tasks makes learning harder but more effective in the long run (Benjamin & Tullis, 2010). Interleaving adds to this by suggesting topics are not taught in one go, but “mixed up” (Pan, 2015).

This can panic some teachers. Many of the suggestions for interleaving suggest completely overhauling the way you deliver your curriculum or syllabus, in that you should avoid teaching in blocks.

This could mean hours of replanning, cause havoc with data tracking (if aligned to a syllabus) and ignore the fact that exam boards block their syllabuses, which could create confusion.

It could also cause tension in bigger departments where there is no consensus on the need for incorporating the concept, regardless of the “evidence”. Of course, if poorly, but well intentionally, put together, desirable difficulties could have a detrimental impact on pupils, especially the cohort experiencing the changes as “guinea pigs”.

That said, I am not suggesting it cannot be done – it can (see, for instance: Fawcett, 2014; Didau, 2015; Cox, 2016 ).

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 separate content that has been previously taught). I also always double up unit assessments with a spaced rotation of previous unit tests. Doing these three simple things allows the main lesson to continue with the blocked components of the syllabus while intertwining previous learning.

Varying the conditions of practice

This is the hardest for me. Research suggests that practising retrieval or application of knowledge, as well as subject-specific skills, has more impact if conditions are varied, particularly in terms of the learning environment or physical parameters in which tasks are completed (Bjork, 1994).

In some curriculum subjects, such as PE, conditions can vary both in location or in weather conditions etc. However, doing this in a humanities class is far harder considering the finite space of the school and timetabling.

Nonetheless, Bjork does suggest that “increasing the variety, types and range of exercises” has an effect on learning (2022). This, then, is quite simply a case of choosing, adapting and applying different learning activities that focus on the retrieval, retention or application of subject knowledge. It could include: some of the ideas above on generating prior learning, especially in terms of recall activities; varying the wording of questions; or varying the timings of written exercises to make the completion of the task more challenging.


Basically: “Interference is a counterintuitive strategy where learning is interfered with in such a way that while the interference is unwelcome, it is beneficial for learning.” (Brown et al, 2014 – cited in Beale, 2020).

In my own classroom, I have not used different fonts, as suggested by Jeffery Bye, but I have used more challenging texts with rarely used vocabulary and grammatical styles. In RE, I have used the King James Bible, as opposed to the more modern versions favoured in textbooks. In order to make sense of the text, pupils need to decontextualise the 17th century language and rephrase or explain verses in a more contemporary style.

I have also experimented with disorganising materials and asking pupils to rearrange the chronological order of an event, especially if aspects of the subject have been taught in a certain order.

On a small scale, card sorts and incorrect timelines are ways to facilitate “interference” in a way that makes “learning sense” to pupils.


There are plenty of ways subject specialists can adapt the above for their own lessons using variation on the themes suggested.

I would again urge caution, however. While “desirable difficulties” is a fine conceptual idea, we need to ensure that we take time in choosing, planning, and implementing challenging activities that are “desirable” and not “undesirable”. They should also be adapted and pitched according to our pupils’ needs.

Further information & references

  • BealeDesirable difficulties, The Tony Little Centre, December 2020:
  • Benjamin& Tullis: What makes distributed practice effective?Cognitive Psychology, November 2010:
  • Bjork: Memory and metamemory considerations in the training of human beings. In Metacognition: Knowing about knowing, Metcalfe & Shimamura (eds), MIT Press, 1994:
  • Bjork & Bjork: Making things hard on yourself, but in a good wayCreating desirable difficulties to enhance learning, Psychology and the Real World: Essays illustrating fundamental contributions to society, January 2011:
  • BjorkApplying cognitive psychology to enhance educational practice, UCLA Bjork Learning and Forgetting Lab (accessed 2022):
  • Brown, Roediger, & McDaniel: Make it StickThe Belknap Press2014.
  • Bye: Desirable difficulties in the classroom, Psychology Today, May 2011
  • Cox: The new GCSE & planning for learning:A practical guide, Missdcoxblog, March 2016
  • Didau: Using Threshold concepts to design a KS4 English curriculum, Learning Spy Blog, March 2015:
  • Fawcett: Can I be that little bit better at… designing a better GCSE curriculum? My Learning Journey Blog, April 2014:
  • Pan: The Interleaving Effect, Scientific American, August 2015:
  • Rosner, Elman, & Shimamura: The generation effect: Activating broad neural circuits during memory encoding, Cortex, 2013:

Photo credit: Amen Clinics (used under a Creative Commons Licence)

Leave a Reply

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

You are commenting using your 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