In this, the first of our Heresy of the Week mini-series, Dr Esther Saxey of LSE’s Teaching and Learning Centre discusses Elizabeth Bjork’s and Robert Bjork’s work on ‘desirable difficulties’ and explains how making things hard for students – if the right things are chosen  – can be good for learning.

I’m experiencing a pedagogic quandary. Often, in my work with academics, I emphasise the benefits of making things easier for students: making our expectations of students more visible; structuring programmes, courses and lectures so they unfold in a logical way; shaping the assessment methods to be a smoother fit for the subject.

Is it contradictory, then, that two of my favourite pedagogic researchers suggest making things harder?

Elizabeth Bjork and Robert Bjork argue that some kinds of difficulty help students to develop their understanding, and to better retain their learning in the long term. Their research, and that of others building on their work, has been to explore which difficulties in learning are useful and productive (as opposed to fruitless and frustrating).

A couple of their key proposals relate to varying what we ask of students, and when we ask it.

Interleaving the order in which subject matter is introduced.

Modern degrees tend towards ‘chunking’ subjects – introducing a subject, guiding students to a high level of understanding, and then assessing students, all within one academic year (or one term). Within courses, subjects are often ‘chunked’ further; one concept (or era, or region, or organisation) is studied per week, with minimal interconnection.

The Bjorks argue that interleaving subjects – studying a subject for a shorter period, but returning later – is much more likely to lead to a deep understanding by students, and improve their long-term retention of the knowledge and ability to use it.

Interleaving subjects may feel harder for students because it can be superficially more confusing. It also may reduce the satisfying sense of progress that comes from studying a ‘chunked’ course – a sense that can be illusory, if it leads to students overestimating how much they’ve grasped, or how much they will retain.

Varying the assessment requirements.

Bjork and Bjork give the example of a test in which participants aim at a specific target. Those who were sometimes asked to overshoot, and sometimes fall short, became more accurate than those only required to aim at the target.

Similarly, they argue, asking students to do one thing repeatedly (the same format of essay, the same type of problem) may be less effective than asking them to do a variety of related tasks. Making students demonstrate their understanding in different contexts promotes a more flexible, transferrable understanding (rather than knowledge that works only in one situation). It’s more work for the student, and again may reduce their sense of success, but it can be effective.

Bjork and Bjork’s proposals may seem to contradict other findings from educational research. Such research often calls for a smoother fit between our desired outcomes, how we teach, and how we assess (see the work of John Biggs on ‘constructive alignment’). Introducing more varied assessments may feel like a step backwards. Similarly, academic developers often prioritise a logical progression through course material – returning to topics feels counterintuitive.

However, in the end, there’s no paradox here. The drive to make study ‘easier’ (to simplify, clarify and make transparent) and Bjork and Bjork’s desire to make it ‘difficult’ (usefully challenging for the student) are two sides of the same coin.

As academics (and academic developers) we are looking for the optimal territory between ‘spoonfeeding’ students and throwing up needless obstacles. To avoid both extremes, we can consider:

  • Which of the things we ask of students – in their study habits, their assessment methods, and the activities they perform during contact hours – actually develop their understanding? Which are conventional but unhelpful?
  • Which ‘obstacles’ (gaps in the material we provide, or abilities that we require students to develop) encourage students to engage and learn? Which ones prevent students from being able to engage?

At root, the question remains the same – not ‘how can we make it easier’ or ‘should we make it harder’, but: what kinds of teaching and assessment actually help students to learn?


Bjork, E. L., & Bjork, R. A. (2014). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. In M. A. Gernsbacher and J. Pomerantz (Eds.), Psychology and the real world: Essays illustrating fundamental contributions to society (2nd edition). (pp. 59-68). New York: Worth.

Applying Cognitive Psychology to Enhance Educational Practice (an overview, with short films and further reading)


If considering Dr Saxey’s heresy has inspired you to discuss one of your own on the Education Blog, please email Natalie Paris ( with details of your heretical thoughts.

Print Friendly