Desirable Difficulties

Conditions that make learning harder during practice but lead to better long-term retention and transfer — the paradox that effort during study is the price of durable knowledge.

---

What is it?

In the early 1990s, cognitive psychologists Robert and Elizabeth Bjork introduced a term that captures one of the most counterintuitive findings in learning science: desirable difficulties. The core idea is that certain conditions which slow down learning and increase errors during practice actually produce stronger, more durable, and more transferable knowledge in the long run.¹ The difficulty is “desirable” because it forces the brain to engage in deeper processing — the kind of effortful work that builds robust memory traces rather than fragile ones.

This runs directly counter to intuition. Students and teachers alike tend to assume that smooth, error-free practice is a sign of effective learning. If a student can read through their notes fluently, if practice problems feel easy, if the material seems familiar, then learning must be happening. But the Bjorks’ research reveals that this fluency is often an illusion of competence — a misleading signal that confuses recognition with recall and familiarity with understanding.² The conditions that make learning feel easy (massed practice, re-reading, blocked problem sets) produce rapid initial performance gains that fade quickly. The conditions that make learning feel hard (spacing, interleaving, retrieval practice) produce slower initial progress but dramatically better retention.

The framework rests on a distinction the Bjorks call the new theory of disuse: every memory has two independent strengths — storage strength and retrieval strength — and these two strengths follow different rules.³ Understanding this distinction is the key to understanding why desirable difficulties work and when they become undesirable.

The Bjorks identified five primary desirable difficulties that have strong experimental support: (1) spacing or distributing practice over time, (2) interleaving different topics or problem types, (3) testing or retrieval practice, (4) varying the conditions of practice, and (5) reducing the frequency of feedback.¹ Each of these makes learning feel harder in the moment but produces superior outcomes over time. Together, they form the theoretical backbone of the evidence-based-learning-strategies identified by The Learning Scientists.

In plain terms

Desirable difficulties are like training with a weighted vest. Running feels harder and your times look worse during practice. But when you take the vest off on race day, you are faster and stronger than someone who trained without resistance. The struggle during practice is not a sign of failure — it is the mechanism of improvement.

---

At a glance

The desirable difficulties paradox (click to expand)

graph LR
    A[Easy Practice] -->|Feels productive| B[High Fluency]
    B -->|But produces| C[Weak Retention]
    D[Difficult Practice] -->|Feels frustrating| E[Low Fluency]
    E -->|But produces| F[Strong Retention]
    style D fill:#4a9ede,color:#fff

Key: The paradox of desirable difficulties. Easy, fluent practice feels like learning but produces weak long-term retention. Difficult, effortful practice feels like struggling but produces strong, durable memory. The node highlighted in blue — difficult practice — is the path to lasting learning.

---

How does it work?

1. Storage strength vs. retrieval strength — the two-strength model

The foundation of desirable difficulties is the Bjorks’ new theory of disuse (1992), which proposes that every memory has two independent properties:³

• Storage strength reflects how well-learned something is — how deeply and richly it is encoded in long-term memory. Storage strength only increases; it never decreases. Once something is encoded, the memory trace persists even if you cannot access it.
• Retrieval strength reflects how easily you can access a memory right now — how likely you are to recall it at this moment. Retrieval strength fluctuates: it increases with recent use and decreases with time and interference.

The critical insight is that these two strengths are dissociated. A memory can have high storage strength but low retrieval strength — you “know” it deeply but cannot access it right now (the tip-of-the-tongue phenomenon). Conversely, a memory can have high retrieval strength but low storage strength — you can recall it easily right now (because you just read it) but it will not last.³

Think of it like...

A library with millions of books (storage strength) and a catalogue system (retrieval strength). The books do not disappear when the catalogue becomes outdated — they are still on the shelves. But if the catalogue is disorganised, you cannot find them. Re-reading is like checking the catalogue is working right now; retrieval practice is like rebuilding the catalogue pathways so you can find the books in the future.

Here is where desirable difficulties enter: the lower the current retrieval strength when you successfully retrieve a memory, the greater the boost to both storage strength and retrieval strength.³ In other words, the harder it is to remember something, the more benefit you get from successfully remembering it. Effortful retrieval is not just harder — it is more productive.

Example: two-strength model in action (click to expand)

Consider two students studying French vocabulary:

Student	Strategy	After study session	After 1 week
A	Re-reads word list 5 times	Retrieval strength: HIGH, Storage strength: LOW	Both have dropped — word is forgotten
B	Studies once, then tests herself 4 times with errors	Retrieval strength: MODERATE, Storage strength: HIGH	Storage remains high; retrieval rebuilt quickly with one review

Student A’s memory has high accessibility in the moment but shallow encoding. Student B’s memory required more effort but created deeper storage. After a week, Student B can recover the word with minimal review; Student A must start almost from scratch.

2. The five desirable difficulties

The Bjorks identified five specific conditions that exploit the two-strength model:¹

Spacing (distributing practice). Instead of studying a topic in one long session, distribute practice across multiple sessions separated by time. The gap between sessions allows retrieval strength to decay, so each subsequent session requires effortful retrieval — which boosts storage strength. The spacing effect is one of the most replicated findings in experimental psychology.⁴

Interleaving (mixing topics). Instead of practising one type of problem until mastery before moving to the next (blocked practice), mix different problem types within a single session. Interleaving forces the learner to discriminate between problem types and select the appropriate strategy — a more demanding task that produces better transfer to novel problems.⁵

Testing/retrieval practice. Instead of re-studying material, test yourself on it. The act of retrieval — whether successful or not — strengthens the memory trace more than additional study. This is called the testing effect, and it has been demonstrated across hundreds of experiments with diverse materials and populations.⁶

Varying practice conditions. Instead of practising in the same environment or with the same materials every time, introduce variation. Study in different locations, use different formats, approach the same concept from different angles. Variation creates multiple retrieval cues, making the knowledge more flexible and less context-dependent.¹

Reducing feedback frequency. Instead of providing immediate feedback after every attempt, delay or reduce feedback. Immediate, constant feedback can create dependency — the learner performs correctly only when guided. Reduced feedback forces the learner to develop their own error-detection skills, building metacognitive competence.¹

Think of it like...

Five different kinds of resistance training for memory. Spacing is like rest between sets (allowing partial fatigue before the next effort). Interleaving is like circuit training (switching exercises so no single muscle adapts to a predictable pattern). Testing is like lifting heavy (maximum effort produces maximum adaptation). Varying conditions is like training on different terrain (building versatility). Reducing feedback is like removing training wheels (forcing self-correction).

3. When difficulties become undesirable

Not all difficulties are desirable. The Bjorks are careful to specify the boundary condition: a difficulty is desirable only when the learner can respond to it successfully, even if with effort. When a difficulty exceeds the learner’s current ability to respond — when they cannot retrieve the answer even with struggle, when the material is too far beyond their existing knowledge — the difficulty becomes undesirable and simply produces failure, frustration, and no learning.¹

This connects to Vygotsky’s concept of the zone of proximal development (ZPD): the space between what a learner can do independently and what they cannot do even with help. Desirable difficulties operate within this zone — they are hard enough to require effort but achievable enough that the effort is productive.⁷ A difficulty that pushes beyond the ZPD is not desirable; it is simply overwhelming.

Example: desirable vs. undesirable difficulty (click to expand)

Desirable difficulty: A medical student who has studied the cardiovascular system attempts to draw the circulatory pathway from memory without notes. She struggles, makes errors, and has to reconstruct her understanding actively. The effort strengthens her mental model.

Undesirable difficulty: The same student, who has never studied neuroscience, is asked to diagram the neural pathways of the visual cortex from memory. She has no existing knowledge to retrieve. The struggle produces frustration and no learning because there is no memory trace to strengthen — only a void.

The difference: In the first case, the difficulty activates and strengthens existing knowledge. In the second, there is nothing to activate. Desirable difficulties require a foundation to build on.

Key distinction

Desirable difficulty = effort that activates and strengthens existing memory traces. Undesirable difficulty = effort with no existing trace to strengthen. The question to ask: “Does the learner have enough prior knowledge to make this struggle productive?“

4. The illusion of competence

One of the most important consequences of the desirable difficulties framework is its explanation of why students systematically misjudge their own learning. Koriat and Bjork (2005) demonstrated that students’ judgments of learning (JOLs) — their predictions about how well they will remember material — are heavily influenced by the fluency of their current experience, not the durability of their actual memory.²

When material feels easy to process (because you just read it, because the font is clear, because the text is familiar), you judge your learning as strong. When material feels hard to process (because you are struggling to recall it, because the practice is effortful), you judge your learning as weak. But the relationship between processing fluency and actual learning is inverted: the effortful condition produces better learning, and the fluent condition produces worse learning.²

This creates a vicious cycle: students use strategies that feel easy (re-reading, highlighting), judge their learning as strong because of the resulting fluency, perform poorly on exams, and then conclude they need to re-read even more next time. The illusion of competence is self-reinforcing.⁸

Think of it like...

Driving the same route to work every day. You arrive on autopilot and feel like you “know” the route. But if someone asks you to give turn-by-turn directions from memory, you struggle. The fluency of the drive (recognition) is not the same as the ability to reproduce the route (recall). Re-reading creates the driving-on-autopilot kind of “knowing.” Retrieval practice creates the give-directions-from-memory kind of knowing.

5. Practical applications

The desirable difficulties framework has direct implications for how students should study and how teachers should design instruction:⁴

Study scheduling. Space study sessions apart rather than cramming. Use a calendar to distribute practice across days and weeks. Embrace the forgetting that happens between sessions — it is what makes the next session productive.

Curriculum design. Interleave topics within lessons rather than teaching each topic in isolation. Return to previous material regularly, even when students (and teachers) feel it has been “covered.”

Self-assessment. Do not trust the feeling of fluency. After studying, close the book and try to recall the material. If recall is easy, you may need to wait longer before the next session (to increase difficulty). If recall is impossible, you may need more initial encoding first.

Feedback design. Consider delaying feedback in some practice contexts. Let learners attempt to detect their own errors before providing correction.

Concept to explore

See retrieval-practice for detailed guidance on designing effective retrieval-based study activities, and spaced-repetition for systems that automate optimal spacing intervals.

---

Why do we use it?

Key reasons

1. It explains the learning paradox. Without this framework, the fact that easier study produces worse outcomes makes no sense. Desirable difficulties provide the theoretical explanation: effort during retrieval is the mechanism that strengthens memory, and fluency during study is a misleading signal.¹

2. It unifies the six strategies. Rather than having six separate study tips with no connecting logic, desirable difficulties provides a single principle that explains why all six evidence-based strategies work: each one introduces productive difficulty that strengthens storage and retrieval.⁴

3. It corrects metacognitive errors. Understanding desirable difficulties helps students recognise that struggle during study is a feature, not a bug. This reframes the emotional experience of learning and helps students persist with strategies that feel less comfortable but produce better results.²

4. It defines the boundary. The framework does not say “harder is always better.” It specifies exactly when difficulty is productive (when existing knowledge can be activated) and when it is counterproductive (when it exceeds the learner’s ability to respond). This prevents misapplication.¹

---

When do we use it?

• When choosing study strategies and deciding between methods that feel easy and methods that feel effortful
• When designing practice schedules and determining how to space, interleave, and sequence material
• When evaluating your own learning and resisting the temptation to trust fluency as a signal of mastery
• When teaching others and choosing activities that maximise learning rather than student comfort
• When designing assessments and understanding why low-stakes testing is a learning tool, not just an evaluation tool
• When explaining to students why the study methods that feel worst often work best

Rule of thumb

If a study session feels smooth and effortless, you are probably not learning much. If it feels like a productive struggle — challenging but not impossible — you are in the zone where desirable difficulties do their work.

---

How can I think about it?

The vaccination analogy

A vaccine introduces a weakened version of a pathogen — just enough to trigger an immune response, but not enough to cause disease. The discomfort (sore arm, mild fever) is not a side effect to avoid; it is evidence that the immune system is learning to fight the real threat.

• The pathogen = the forgetting and difficulty that happens between study sessions
• The vaccine = a desirable difficulty (spacing, testing, interleaving) that introduces controlled challenge
• The immune response = the effortful retrieval that strengthens memory traces
• The side effects = the discomfort and frustration of struggling during practice
• Full immunity = durable, transferable knowledge that persists under pressure
• An overdose = undesirable difficulty — too much challenge overwhelms the system rather than training it

A doctor who avoids all discomfort produces a patient with no immunity. A teacher who avoids all difficulty produces a student with no durable knowledge.

The sword-forging analogy

A blacksmith forging a sword repeatedly heats the metal and then plunges it into cold water — a process called quenching. Each cycle of heating and rapid cooling (tempering) changes the metal’s internal crystal structure, making it harder and more resilient. The process looks brutal: the metal glows, hisses, and deforms. But each cycle produces a blade that is stronger than one that was simply left to cool slowly.

• Heating = the initial study session (encoding new information)
• Quenching = the forgetting gap between sessions (retrieval strength drops, forcing effortful recovery)
• The next heating = the spaced retrieval session (re-accessing the memory under difficulty)
• The crystal structure change = the strengthening of storage strength through effortful retrieval
• A blade left to cool slowly = massed practice (comfortable but produces a brittle result)
• The finished sword = knowledge that has been tempered through multiple cycles of difficulty

You cannot forge a strong blade without the stress of quenching. You cannot build durable knowledge without the stress of forgetting and retrieval.

---

Concepts to explore next

Concept	What it covers	Status
retrieval-practice	The testing effect — why retrieving information from memory strengthens it more than restudying	stub
spaced-repetition	The spacing effect and systems that automate optimal review intervals	stub
cognitive-load-theory	How working memory limitations interact with desirable difficulties — when load becomes overload	stub

Some cards don't exist yet

A broken link is a placeholder for future learning, not an error.

---

Check your understanding

Test yourself (click to expand)

1. Explain the core paradox of desirable difficulties in your own words. Why do conditions that make learning feel harder produce better long-term outcomes?
2. Distinguish between storage strength and retrieval strength. Give an example of a memory that has high storage strength but low retrieval strength.
3. Name the five desirable difficulties identified by the Bjorks and describe how each one exploits the two-strength model of memory.
4. Evaluate this scenario: a language teacher introduces 50 new vocabulary words in a single lesson and tests students on all 50 immediately. Students score an average of 15%. Is this a desirable difficulty or an undesirable one? Justify your answer using the framework.
5. Connect the illusion of competence to students’ study strategy preferences. Why do students consistently choose ineffective strategies, and how does the desirable difficulties framework explain this pattern?

---

Where this concept fits

Position in the knowledge graph

graph TD
    EBLS[Evidence-Based Learning Strategies] --> DD[Desirable Difficulties]
    EBLS --> RP[Retrieval Practice]
    EBLS --> SR[Spaced Repetition]
    DD -.-> RP
    DD -.-> SR
    style DD fill:#4a9ede,color:#fff

Related concepts:

• retrieval-practice — the most extensively studied desirable difficulty; the act of retrieving a memory strengthens it more than restudying
• spaced-repetition — applies the spacing desirable difficulty systematically using algorithms that optimise review intervals
• cognitive-load-theory — provides the complementary framework for understanding when difficulty becomes undesirable, because it exceeds working memory capacity

---

Sources

---

Further reading

Resources

• Making Things Hard on Yourself, But in a Good Way (Bjork & Bjork, 2011) — The most accessible introduction to desirable difficulties by the researchers who coined the term; written for a general audience
• Desirable Difficulties in Theory and Practice (Bjork & Bjork, 2020) — Updated review reflecting three decades of research, with attention to boundary conditions and practical implementation
• A New Theory of Disuse (Bjork & Bjork, 1992) — The original paper introducing the storage strength / retrieval strength distinction; technical but foundational
• Desirable Difficulties: Why Struggling Helps You Learn Better (DeepTerm, 2026) — A recent, student-friendly overview with practical study recommendations
• Robert Bjork: A Teacher’s Guide to Desirable Difficulties (Structural Learning) — Practical guide translating the framework into classroom strategies for teachers

Footnotes

1. Bjork, R. A. & Bjork, E. L. (2011). Making Things Hard on Yourself, But in a Good Way: Creating Desirable Difficulties to Enhance Learning. In M. A. Gernsbacher et al. (Eds.), Psychology and the Real World. Worth Publishers. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

2. Koriat, A. & Bjork, R. A. (2005). Illusions of Competence in Monitoring One’s Knowledge During Study. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31(2), 187-194. ↩ ↩² ↩³ ↩⁴

3. Bjork, R. A. & Bjork, E. L. (1992). A New Theory of Disuse and an Old Theory of Stimulus Fluctuation. In A. Healy, S. Kosslyn, & R. Shiffrin (Eds.), From Learning Processes to Cognitive Processes: Essays in Honor of William K. Estes (Vol. 2, pp. 35-67). Erlbaum. ↩ ↩² ↩³ ↩⁴

4. 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. ↩ ↩² ↩³

5. Rohrer, D. & Taylor, K. (2007). The Shuffling of Mathematics Problems Improves Learning. Instructional Science, 35(6), 481-498. ↩

6. Roediger, H. L. & Karpicke, J. D. (2006). Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention. Psychological Science, 17(3), 249-255. ↩

7. Metcalfe, J. & Bjork, R. A. (2014). Desirable Difficulties and Studying in the Region of Proximal Learning. In B. H. Ross (Ed.), The Psychology of Learning and Motivation (Vol. 60). Academic Press. ↩

8. Karpicke, J. D., Butler, A. C., & Roediger, H. L. (2009). Metacognitive Strategies in Student Learning: Do Students Practise Retrieval When They Study on Their Own?. Memory, 17(4), 471-479. ↩