Skip to main content
Education & Training

Desirable Difficulties: Why Making Learning Feel Harder Can Make It Actually Work Better

A specific category of learning conditions — reduced feedback frequency, varied practice, spaced review — feels less effective in the moment and reliably produces stronger, more durable long-term learning.

Key Takeaways
  • Desirable difficulties is an umbrella term for a category of learning conditions that increase difficulty and reduce immediate performance during practice, while improving long-term retention and transfer
  • Spacing, interleaving, reduced feedback frequency, and generation are all specific examples that fall under this broader umbrella concept
  • The common thread is that these conditions require more effortful, active cognitive processing than smoother, more immediately successful-feeling alternatives
  • Applying desirable difficulties well requires calibrating the difficulty to a learner's actual current level, since difficulty that's simply too great, disconnected from any relevant existing foundation, doesn't produce this same benefit

Several specific, seemingly unrelated findings in learning science — spacing practice over time rather than massing it, interleaving different types of problems rather than blocking them, reducing feedback frequency rather than providing it after every single attempt, requiring learners to generate answers rather than simply presenting them — share a common underlying pattern that researchers have organized under the umbrella term desirable difficulties: conditions that make learning feel harder and produce weaker immediate performance, while reliably producing stronger long-term retention and transfer.

What actually unifies this specific category of learning conditions

Each of these seemingly distinct techniques shares the property of requiring more effortful, active cognitive processing than a smoother, more immediately comfortable alternative — spacing requires effortful re-retrieval after some forgetting has occurred, interleaving requires effortful discrimination between problem types, reduced feedback requires the learner to actually attempt genuine self-evaluation rather than immediately being told the answer, and generation requires effortful production rather than passive recognition. This shared demand for effortful processing, rather than any single specific mechanism, is what the desirable difficulties framework identifies as the common thread producing stronger durable learning across all these otherwise distinct techniques.

Why these conditions feel counterproductive in the moment despite working better

Every one of these techniques produces weaker apparent performance and a less comfortable, less fluent subjective experience during the learning process itself compared to their smoother, easier-feeling alternatives (massed practice, blocked practice, frequent feedback, passive presentation) — which is exactly why they're called desirable difficulties rather than simply difficulties: the difficulty itself is precisely what's producing the durable learning benefit, not an unfortunate cost being paid despite the technique's effectiveness.

Why the difficulty specifically needs to be desirable, not simply present

Not every form of difficulty during learning produces this same benefit — difficulty stemming from genuinely confusing, poorly explained material, or difficulty that exceeds a learner's current foundation so substantially that no meaningful cognitive processing can actually occur, doesn't produce improved learning, it simply produces frustration and failure without the underlying productive cognitive engagement that desirable difficulties specifically depend on. The framework's name explicitly distinguishes this category from difficulty generally, since only difficulty of the right kind and right calibration relative to a learner's actual current level produces the documented benefit.

Why this framework matters for how instructional design decisions get made

Recognizing multiple, previously separate-seeming learning science findings as instances of one underlying pattern helps instructional designers apply the same underlying principle more broadly and more confidently than treating each specific technique as an isolated, unrelated tactic — understanding that the shared mechanism is effortful cognitive engagement, not any single technique's specific mechanics, supports appropriately extending the same underlying logic to other, not-yet-specifically-studied instructional choices that share this same effortful-processing structure.

What this means for designing courses and training programs

  • Deliberately incorporate spacing, interleaving, reduced feedback frequency, and generation activities into course design, expecting them to feel less immediately smooth or successful than easier alternatives
  • Calibrate the level of difficulty introduced to a learner's actual current foundation, since difficulty exceeding that foundation doesn't produce the same benefit
  • Communicate to learners explicitly that reduced comfort and increased effort during these specific activities is expected and consistent with the technique working as intended, not a sign something has gone wrong
  • Recognize the desirable difficulties framework as unifying several specific, well-documented techniques under one shared underlying principle, rather than treating each as an isolated, unrelated tactic

Desirable difficulties is a genuinely useful organizing concept precisely because it explains why so many of learning science's most counterintuitive, well-replicated findings point in the same underlying direction — the discomfort these techniques produce isn't an unfortunate side effect to minimize, it's the actual mechanism doing the productive cognitive work.

desirable difficultieslearning science principlesdurable retention researchcorporate trainerscounterintuitive study strategies