Skip to content

The science of learning

Desirable difficulty: why harder study lasts longer.

A desirable difficulty is a condition that makes study slower and harder now and makes the memory last longer. Spacing, mixing, recall and generation are the classic examples, and a difficulty that does not make you do useful mental work is not one of them.

By the Axo team · Last updated November 10, 2026

What the term means

The psychologist Robert Bjork introduced the term in 1994. He saw the same pattern in many experiments: a change that made practice harder and slower gave better results on a later test.

The word desirable is the important one. Not all difficulty helps. A difficulty is desirable when it makes you do the mental work that builds the memory, and when you are able to get through it.

Performance is not learning

In 1992 Richard Schmidt and Robert Bjork compared research on movement skills and on verbal learning. They found the same thing in both. Methods that give the best performance during training can be bad for the long term. Methods that slow you down during training can be good for it.

This is why the term is necessary. Performance is what you can do now, during practice. Learning is what you can still do later, and in a new situation. Teachers and learners see performance, so they choose the methods that make it go up quickly. Those are often the wrong methods.

The classic difficulties

DifficultyWhat you doWhat a study found
SpacingStudy in parts with gaps between them, not in one sitting.A 2006 review found that study which is spread out gives better recall later.
MixingPractice different kinds of items in one session.In a 2008 study, people learned the styles of painters better when the painters were mixed, and most of them believed the opposite.
RecallAnswer from memory before you look.In a 2006 study, recall lost to rereading on a test five minutes later and won on tests two days and one week later.
GenerationProduce the answer yourself, even from a hint.In a 1978 study, people remembered words that they had to complete better than words that they only read.

A wrong answer can help too. In a 2009 study, people who tried to answer a question and failed, and then saw the answer, learned it better than people who only studied the answer.

There is more on three of these in the forgetting curve, interleaving and retrieval practice.

When a difficulty is not desirable

A difficulty helps only if it makes you think about the meaning of the material. A good test of this idea is the font experiment.

In 2011 a study reported that students remembered more when their material was printed in a font that was hard to read. The idea spread quickly. In 2018 a meta-analysis put the experiments on this effect together and found no benefit for learning from text.

A font that is hard to read makes you slow, but the extra work goes to the letters and not to the ideas. The same is true of a noisy room, a confusing textbook, or material that is too far above your level. If you do not have the knowledge to get through a difficulty, it is only a barrier.

Why it feels wrong

Easy study gives quick, visible progress. Hard study gives slow progress and many errors. Both are signals, and both point the wrong way.

In the 2008 study of painters, people did better after mixed practice, and most of them still said that practice in blocks had worked better. They had the result in front of them and did not believe it.

How to use it

  1. Do not trust the feeling of ease. If a session feels smooth, ask if you could do the same tomorrow with the book closed.
  2. Answer first, then look. A guess that is wrong and then corrected is better than no guess.
  3. Leave a gap. Come back when the material has started to fade.
  4. Mix topics that are easy to confuse.
  5. Keep the difficulty in the thinking. Do not make the page, the room or the tool harder to use.
  6. Lower the level when you cannot start. A difficulty that you cannot get through teaches nothing.

How Axo uses it

Every Bite ends in questions, and you answer before you see the result. Keep Sharp brings an idea back after a gap, mixes ideas from all your Paths, and asks the question another way than you first met it. On review you type the answer from memory.

Axo also keeps the other side of the rule. Every question has “Help me learn this”, and a Path has a reading level that you can change. The difficulty stays in the recall, not in the reading.

Questions people ask.

What is a desirable difficulty?

It is a study condition that makes learning slower and harder now and makes the memory last longer. Spacing, mixing, recall and generation are the classic examples.

Who came up with desirable difficulties?

The psychologist Robert Bjork introduced the term in 1994. Much of the later work is by Robert Bjork and Elizabeth Bjork.

Is harder study always better?

No. A difficulty helps only when it makes you think about the meaning of the material and when you are able to get through it. A font that is hard to read, for example, did not help in a 2018 meta-analysis.

Why does easy study feel more effective?

Easy study gives quick progress that you can see during the session. That is performance, not learning. Learning shows only later, on a test after a delay.

Sources.

Every claim on this page comes from one of these. Each link opens the paper.

  1. Bjork, R. A., and Bjork, E. L. (2020). Desirable difficulties in theory and practice. Journal of Applied Research in Memory and Cognition.
  2. Schmidt, R. A., and Bjork, R. A. (1992). New Conceptualizations of Practice: Common Principles in Three Paradigms Suggest New Concepts for Training. Psychological Science.
  3. Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., and Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin.
  4. Kornell, N., and Bjork, R. A. (2008). Learning Concepts and Categories: Is Spacing the “Enemy of Induction”? Psychological Science.
  5. Roediger, H. L., and Karpicke, J. D. (2006). Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention. Psychological Science.
  6. Slamecka, N. J., and Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory.
  7. Kornell, N., Hays, M. J., and Bjork, R. A. (2009). Unsuccessful retrieval attempts enhance subsequent learning. Journal of Experimental Psychology: Learning, Memory, and Cognition.
  8. Diemand-Yauman, C., Oppenheimer, D. M., and Vaughan, E. B. (2011). Fortune favors the bold (and the italicized): Effects of disfluency on educational outcomes. Cognition.
  9. Xie, H., Zhou, Z., and Liu, Q. (2018). Null Effects of Perceptual Disfluency on Learning Outcomes in a Text-Based Educational Context: a Meta-analysis. Educational Psychology Review.