Demystifying desirable difficulties 2: What they’re NOT

Mirjam Neelen & Paul A. Kirschner

In the first blog, we attempted to demystify what Robert and Elizabeth Bjork meant by desirable difficulties (and later also many others!), hoping that we can stick to their definition and call the ‘other things’… other things. Today, in blog 2, we discuss two ‘other things’ that are often mistaken for or confused with desirable difficulties.

There are two general misconceptions (maybe more, but these are the ones that we’ve run into repeatedly) where people refer to desirable difficulties incorrectly. First, when they talk about Vygotsky’s zone of proximal development and second, when they discuss the idea that errors support learning.

Desirable difficulties are NOT challenges and efforts caused by the difficulty of the task

Desirable difficulties are often explained in the context of Vygotsky’s zone of proximal development[1]. This ‘zone’ refers to the difference between what a child can do without the help of an adult or a more knowledgeable other and what they can achieve with this help. Wikipedia tells us: “Vygotsky saw natural, spontaneous development as important, but not all-important. He believed that children would not advance very far if they were left to discover everything on their own. It’s crucial for a child’s development that they’re able to interact with more knowledgeable others”.

When thinking about ‘difficulties’ in this context, the idea is that as learning designers, trainers, or teachers, we need to present the learner with tasks that are just beyond their current ability level of the learner or trainee, yet not so difficult that they will fail when trying to complete the task. This is all true as in ‘it is in line with Vygotsky’s ideas’ and yes, we should design in a way that tasks are challenging and/or effortful (they shouldn’t be easy or trivial for the learner) and at the same time we still need to make sure that learners can successfully complete them (with guidance when they’re a novice in the context of the task, without guidance when they’ve achieved a certain level of proficiency). However, the difficulty of the task has nothing to do with Bjork & Bjork’s desirable difficulties[2].

The difference? Bjork and Bjork’s desirable difficulties are about learning-strategies that slow down the learning process to increase learning. The ‘challenge’ and ‘effort’ in that context refer to the strategies that stimulate or even force the learner to remember and retrieve (i.e., expend more mental effort during study) and not the learning task itself. This is different from the learner being challenged and having to put in more effort because the task is (too) difficult for them. In other words, while it is desirable to increase the difficulty of the tasks a learner must carry out, this isn’t Bjorks’ desirable difficulties!

Desirable difficulties are NOT about teaching or designing in a way that encourages the learner to make errors or ‘fail’ to complete a task successfully

We also often hear the term ‘desirable difficulties’ when people talk about the idea that errors support learning. In this context, people use ‘desirable difficulties’ to express that a learning task needs to elicit errors. Without errors, there’s no learning; that’s the (silly) idea.

In the 1950s and 1960s B.F. Skinner introduced the idea that errors would be counterproductive for learning and were the result of poor instruction. Based on this, teachers started to introduce instructional techniques in which they provided ‘bite-sized’ learning and then quizzed students immediately afterwards (Brown et al, 2014). You can probably see how this relates to Bjork and Bjork’s desirable difficulties in the sense that they distinguish between performance and learning. When you quiz learners so quickly after instruction, they’re retrieving things when they’re still fresh in the mind and hence, the quizzes are testing performance, not learning.

Skinner’s idea quickly went out of fashion as research moved on to figure out the role of errors in instruction and learning. There are various ways to think about this. For example, Lorenzet, Salas, and Tannenbaum (2005) describe four design approaches for error occurrence.

  1. Avoidance – We should design training in a way in which we try to prevent learners from making errors. The goal is to increase learning and motivation, while reducing anxiety. The ‘Skinner way’, so to speak.
  2. Allowed – We should design training in a way that doesn’t necessarily prevent errors, yet there is also no intentional design to elicit errors.
  3. Induced – We should make intentional design decisions to evoke errors, for example increasing task difficulty or withholding certain support, guidance, or information (also see Kapur’s research on productive failure). We’d like to add that in this category, we could also consider designing for specific types of errors to support deeper understanding (e.g., to demonstrate the importance of place value to calculate with units, tens, and hundreds)
  4. Guided – We should use a design approach in which learners are intentionally instructed or guided to make errors in a way that shows them that errors are expected as part of the learning process.

Though errors in learning might be useful (the results are mixed), they’re different from desirable difficulties as the Bjorks defined them. Again, their desirable difficulties are about learning strategies that appear to slow down the learning process. In contrast, Avoidance and Guided are both about increasing motivation and reducing anxiety; a different topic altogether. Allowed and Induced are based on the assumption that errors serve some kind of informational function (e.g., they provide feedback that helps learners to developbetter schema accuracy that then hopefully helps to improve performance, in particular accuracy and speed). They’re also about improving self-efficacy (the feeling that you can do something successfully) because learners learn to better distinguish between accurate and inaccurate. In other words, here too we see that it might be more difficult for the learner and that this might be desirable (the jury’s out on some of these) but they aren’t Bjorks’ desirable difficulties!

Back to the original problem: Robert and Elizabeth Bjork’s desirable difficulties refer to study strategies that require extra mental effort by the learner to support both storage and retrieval of information such as contextual interference, interleaving, spaced practice, reducing feedback, and retrieval practice. Vygotsky’s Zone of Proximal Development is about scaffolding difficult tasks and the role of errors in learning are yet another ball game, focused on increasing motivation, reducing anxiety, increasing schema accuracy, and improving self-efficacy.

Demystifying concepts and theories clear up misty business, which hopefully helps us as practitioners to make clear design decisions, based on the actual evidence.


Battig, W. F. (1966). Facilitation and interference. In: E.A. Bilodeau (ed.), Acquisition of skill. Academic Press, 215-244.

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). MIT Press.

Bjork, R. A., & Bjork, E. L. (2020). Desirable difficulties in theory and practice. Journal of Applied Research in Memory and Cognition, 9(4), 475-479.

Bjork, E. L., & Bjork, R. A. (2011). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. Psychology and the real world: Essays illustrating fundamental contributions to society, 2, 59-68.

Brown, P. C., Roediger III, H. L., & McDaniel, M. A. (2014). Make it stick: The science of successful learning. Harvard University Press.

Clark, D. (2022). Learning experience design: How to create effective learning that works. Kogan Page Publishers.

Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363–406.

Kirschner, P. A., Hendrick, C., & Heal, J. (2022). How teaching happens: Seminal works in teaching and teacher effectiveness and what they mean in practice. Routledge.

Lorenzet, S. J., Salas, E., & Tannenbaum, S. I. (2005). Benefiting from mistakes: The impact of guided errors on learning, performance, and self‐efficacy. Human Resource Development Quarterly, 16(3), 301-322.

[1] ZPD isn’t a learning theory; it’s a developmental psychological theory that has implications for learning.

[2] Also note that the difficulty and complexity of a task are two different things. A non-complex (simple) task in quantum physics can still be very difficult for someone who knows next to nothing about quantum physics. However, a complex task in quantum physics can still be relatively easy for an expert in quantum physics.