Desirable Difficulties in Learning: Why Harder Study Methods Stick Better

Desirable Difficulties in Learning: Why Harder Study Methods Stick Better

There is a deeply uncomfortable truth sitting at the heart of learning science: the methods that feel most productive are often the least effective, and the methods that feel frustrating, slow, and effortful tend to produce the strongest, most durable memories. If you have ever highlighted an entire textbook chapter and felt genuinely accomplished, only to blank on the material two weeks later, you have experienced this mismatch firsthand.

Related: evidence-based teaching guide

The concept of desirable difficulties was introduced by psychologist Robert Bjork in the 1990s, and it has since accumulated one of the most robust empirical records in cognitive science. The core idea is deceptively simple: certain types of difficulties during learning — ones that slow you down, force errors, and demand more mental effort — actually strengthen the underlying memory traces. Not all struggle is useful, but the right kinds of struggle are not just tolerable. They are necessary.

For knowledge workers in their 20s, 30s, and 40s, this matters enormously. You are not sitting in a classroom with a single subject to master. You are juggling technical documentation, industry reports, new software systems, regulatory changes, and professional development courses, often simultaneously. Understanding which study strategies are genuinely building durable knowledge — versus which ones are just creating a comfortable illusion of competence — is one of the highest-leverage cognitive skills you can develop.

What Makes a Difficulty “Desirable”

Not every form of struggle improves learning. Trying to learn quantum mechanics with no foundation in basic physics is just confusion, not a desirable difficulty. The distinction matters. A difficulty is desirable when it challenges the learner in a way that can actually be resolved through effort, and when that resolution process strengthens encoding and retrieval pathways in long-term memory.

Bjork and Bjork (2011) describe desirable difficulties as conditions that “slow the rate of acquisition, reduce performance during training, or both, yet enhance long-term retention and transfer.” The key phrase there is during training. These methods hurt your performance while you are practicing, which is exactly why they feel unreliable. We conflate current performance with long-term learning, and they are not the same thing at all.

Think about re-reading, which is the single most common study strategy used by students and professionals alike. It is fast, it is easy, it produces a sensation of familiarity, and it does almost nothing for long-term retention. Familiarity is not memory. You can recognize something without being able to retrieve it under pressure, and in most professional contexts, retrieval under pressure is precisely what is required.

The Big Three: Testing, Spacing, and Interleaving

Retrieval Practice: The Testing Effect

If you take away only one principle from learning science, make it this one. Testing yourself on material — before you feel ready, before you are confident, while you are still struggling — is one of the most potent memory interventions known to researchers. Roediger and Karpicke (2006) conducted a landmark study in which participants studied prose passages either by re-reading them or by attempting to recall them from memory. One week later, the retrieval practice group outperformed the re-study group by approximately 50 percent on a final recall test. Fifty percent. From a simple strategy change.

The mechanism here involves something called retrieval-induced potentiation. Every time you successfully pull information out of memory, you strengthen the retrieval pathway. You are not just reviewing the information — you are actively rebuilding the mental route to it. Failed retrieval attempts also help, which is counterintuitive but well supported. Attempting to recall something you cannot quite remember, then checking the answer, produces stronger encoding than simply reading the answer passively (Kornell et al., 2009).

For practical application: close the document, close the slides, and write down everything you remember. Use flashcard systems like Anki that force active recall. After a meeting or a training session, spend five minutes writing a brain dump before you look at your notes. These habits feel inefficient. They are the opposite of inefficient.

Spaced Practice: Fighting the Forgetting Curve

Hermann Ebbinghaus mapped the forgetting curve in the 1880s, and what he found has been replicated so many times it is essentially bedrock: memory decays in a predictable, exponential fashion unless it is reinforced. Massed practice — what most people call cramming — compresses all your learning into a single session and produces sharp initial performance that dissolves quickly. Spaced practice distributes that same amount of study time across multiple sessions separated by intervals, and the retention advantage is dramatic.

Cepeda et al. (2006) conducted a large-scale meta-analysis of spacing research and found consistent, substantial benefits of distributed practice over massed practice across a wide range of materials and populations. The optimal gap between study sessions depends on when you need to remember the material, but a general principle holds: the gap should feel uncomfortably long. If you can still easily remember everything from your last session, you waited too long — or actually, you did not wait long enough.

Here is where this gets practically interesting for busy professionals. You do not need more total study time to implement spacing. You need to restructure when you study. Instead of one 90-minute session on a new framework, you could do three 30-minute sessions spread across a week and walk away with substantially better retention. The calendar adjustment is trivial. The cognitive payoff is not.

Interleaving: Mixing It Up Against Every Instinct

Interleaving is probably the most counterintuitive of the three core desirable difficulties. Conventional study wisdom says to master one topic completely before moving to the next. Practice all the problems of type A, then all the problems of type B, then all the problems of type C. This is called blocked practice, and it feels logical, organized, and productive.

Interleaved practice mixes problem types together — A, C, B, A, B, C — in an apparently random or varied sequence. During practice, interleaving performs worse than blocking. Students make more errors, feel more confused, and generally dislike it. Yet on delayed tests measuring actual learning, interleaving consistently outperforms blocking by meaningful margins (Taylor and Rohrer, 2010). The reason appears to be that interleaving forces learners to actively identify which type of problem they are facing before choosing a solution strategy, which is precisely the skill needed in real-world application where problems do not arrive neatly sorted by category.

If you are learning a new programming language, do not drill all the loops, then all the conditionals, then all the functions in separate blocks. Mix them. If you are studying for a professional certification, randomize practice questions across domains rather than working through one domain completely before starting the next. It will feel messier. The learning will be deeper.

Why We Resist These Methods (And Why That Resistance Is Itself a Signal)

Here is something worth sitting with: the reason most people default to re-reading, blocked practice, and massed studying is not laziness or ignorance. It is a reasonable response to false feedback. When you re-read a chapter, you recognize every sentence. That recognition feels like understanding. When you study in concentrated blocks, performance improves steadily within the session. That improvement feels like progress.

Desirable difficulty methods provide the opposite experience. You test yourself and fail to remember things you thought you knew. You space out your sessions and walk into the second one feeling like you have forgotten everything from the first. You interleave topics and feel lost without the structural scaffold of working through one thing at a time. Every signal your brain sends during these methods says: this is not working. But that signal is wrong, and the long-term data is unambiguous.

As someone with ADHD, I find this especially relevant. The methods that feel productive for my brain — re-reading with a highlighter while music plays, watching the same video lecture twice in a row — are precisely the ones that produce the least learning. My subjective sense of whether I have learned something is not a reliable guide. This is probably true for you as well, ADHD or not. Metacognitive accuracy about learning is surprisingly poor in almost everyone, which is why we need external frameworks rather than just trusting our intuitions about what is working.

Applying Desirable Difficulties in a Real Work Context

After Conferences and Training Sessions

Most professionals sit in a training session, take some notes, file those notes away, and never engage with the material again until they vaguely need to remember it months later. Instead, try this: immediately after the session, close your notes and write from memory everything you can recall. Note what you cannot recall as clearly. Then, two days later, open your notes and test yourself again on the sections that were fuzzy. One week after that, try to reconstruct the key frameworks from scratch without looking at anything. Three exposures, spaced out, with active retrieval each time. The time investment is modest. The retention difference is not.

Reading Technical Material

When you need to actually learn something from a report, paper, or technical document — not just skim it for a meeting, but genuinely internalize it — stop highlighting. Read a section, close the document, and write a short summary in your own words. Not the author’s words. Yours. This forces processing at a deeper level than passive reading. Then, crucially, return to the document and notice where your summary was incomplete or wrong. That comparison is high-value learning, not just a check on comprehension.

Building Skills in New Software or Tools

When your organization rolls out a new tool, most people follow the linear tutorial path, complete it once, and consider themselves trained. A more effective approach: go through the tutorial once for orientation, then close it and try to accomplish real tasks from memory. You will struggle. Look things up as needed, but try to retrieve first. Come back to the core workflows two days later and rebuild them from scratch. The frustration is the point. The frustration means the retrieval system is working.

The Role of Generation and Elaboration

Two additional desirable difficulties deserve mention. The generation effect refers to the finding that information you generate yourself is better remembered than information you passively receive. If you try to predict what a document will cover before reading it, the act of generating those predictions — even incorrect ones — primes the memory system and improves encoding of what actually follows. Similarly, generating an answer to a question before being told the correct answer improves subsequent retention, even when your initial answer is wrong.

Elaborative interrogation is related: asking yourself why something is true, rather than just accepting that it is, forces deeper processing and connects new information to existing knowledge structures. When you read that a certain business strategy failed, do not just accept the conclusion. Ask yourself why it failed, what conditions would have made it succeed, and what other situations are structurally similar. These questions cost cognitive effort. They produce the kind of rich, interconnected memory that transfers to novel situations.

This is the ultimate goal, really. Not just remembering information for a test or a presentation, but building knowledge structures flexible enough to apply in contexts you have never seen before. Desirable difficulties do not just improve retention scores on standardized tests. They improve the quality of thinking that is available to you when the problems are genuinely hard and the stakes are real.

The Meta-Skill: Learning How to Learn

There is a compounding effect that happens when you genuinely internalize the desirable difficulties framework. You stop evaluating study methods by how they feel and start evaluating them by what the evidence says about long-term outcomes. You become comfortable with the discomfort of not knowing, because you understand that struggling to retrieve something is doing useful cognitive work. You develop patience for the messy, non-linear feeling of interleaved practice, because you know the eventual payoff justifies the present confusion.

This shift in orientation — from comfort-seeking to evidence-based learning — is one of the most valuable cognitive habits a knowledge worker can develop. The information landscape is not getting simpler. The rate at which professionals need to acquire, integrate, and apply new knowledge is not slowing down. Given that reality, the people who understand how memory actually works, and who design their learning accordingly, are building a genuine and durable advantage.

The science on this is not new. Bjork has been publishing on desirable difficulties for over three decades. The testing effect was documented more than a century ago. What is surprising is how slowly this knowledge has diffused into actual practice. Most workplaces still organize training as passive information delivery. Most professionals still reach for the highlighter first. You do not have to. The harder path through the material is the one that sticks, and now you know why.

Last updated: 2026-03-31

References

• Bjork, R. A., & Bjork, E. L. (2020). Make It Stick: The Science of Successful Learning. Harvard University Press. Link
• Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. Metacognition: Knowing about Knowing. MIT Press. Link
• Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249-255. Link
• Kang, S. H. K. (2016). Spaced repetition promotes efficient and effective learning: Policy implications for instruction. Policy Insights from the Behavioral and Brain Sciences, 3(1), 12-19. Link
• Rohrer, D., & Taylor, K. (2007). The shuffling of mathematics problems improves learning. Instructional Science, 35(6), 481-498. Link
• Eich, T. S., et al. (2026). Why Desirable Difficulties ‘Work’: A Review of the Evidence From Cognitive Psychology and Health Professions Education. Medical Education. Link

Related Reading

• Restorative Practices in Schools [2026]
• How to Write Learning Objectives That Actually Guide Your Teaching
• Comparative Religion: Why Studying Multiple Faiths Makes