Make It Stick — Peter Brown, Henry Roediger, Mark McDaniel
That’s the central finding of Make It Stick: The Science of Successful Learning, published in 2014 by cognitive scientist Henry Roediger III and psychologist Mark McDaniel, both of Washington University in St. Louis, with science writer Peter Brown. The book synthesizes decades of research on human learning and memory into a set of practical principles that challenge most of what students, teachers, and training designers believe about how learning works. The principles carry evidence unusually strong for social science — replicated across populations, domains, learning contexts — and they’re almost universally ignored by educational practice.
Understanding why learning feels the way it does, and why that feeling is a poor guide to whether it’s working, is the first step toward becoming the kind of deliberate, effective learner who builds durable knowledge rather than fleeting familiarity. Not a minor optimization. For anyone whose success depends on continuously building expertise — most people operating in complex environments — potentially the highest-use improvement available.
The Illusion of Knowing
The most dangerous enemy of real learning is fluency — the feeling, after reading something or hearing it explained, of understanding it. Fluency is not understanding. It’s recognition. And recognition is far easier to produce and far more perishable than the genuine encoding of knowledge in long-term memory.
The authors call this the “fluency illusion” — the gap between how well something feels known and how well it’s actually known. The illusion comes from familiarity: reading a passage that seems clear and comprehensible, the brain registers this as knowing, even when the comprehension is entirely a product of the immediate context and won’t survive removal of that context by more than an hour or two.
The practical consequence: the most common study strategies are fundamentally miscalibrated. Students reread their notes, highlight key passages, reread chapters — and feel productive doing it, because the material grows increasingly familiar and fluency increases with each pass. But the research shows consistently that this increased familiarity doesn’t translate into durable memory or the ability to apply the material in new contexts. Students who study by rereading perform modestly better on a test given immediately after studying. They perform no better — often worse — on a test given a week later.
The mechanism behind this failure is well understood in cognitive science. Rereading is a recognition exercise. It tells the brain “yes, seen this before” — a low-threshold, metabolically cheap operation. What it doesn’t do is build the retrieval structures that allow actual recall and use of the material when needed. Building those structures requires retrieval — actively pulling information out of memory rather than passively recognizing it on the page — and retrieval is difficult, effortful, and feels nothing like the smooth familiarity of rereading. Retrieval feels like work. Rereading feels like learning. Which is exactly why most people do more rereading and less retrieval than they should.
Retrieval Practice: The Single Most Powerful Tool
If there’s one principle in Make It Stick that deserves immediate, consistent implementation, it’s retrieval practice — actively recalling information from memory rather than reviewing it passively. The research on retrieval practice is among the strongest in all of learning science: testing yourself on material, formally or informally, produces dramatically superior retention compared to equivalent time spent reviewing the material.
The effect is called the “testing effect” or “retrieval practice effect,” demonstrated in hundreds of studies across subjects ranging from elementary school vocabulary to medical school pharmacology to pilot training. The basic finding holds consistently: students who study by testing themselves retain more information over longer periods than students who study by rereading, even when total study time is identical. The superiority of retrieval practice over rereading isn’t marginal. It’s a reliable, replicable, large difference in outcomes.
Why does retrieval work so much better than review? The authors offer a mechanistic explanation: every successful retrieval of a piece of information from memory strengthens the neural pathways supporting that retrieval. The attempt also activates the surrounding network of associations — connections to related concepts, contextual cues linking the information to particular situations and applications — in ways passive review doesn’t. The effort of retrieval is itself the learning. The difficulty isn’t a sign the method is too hard. It’s the mechanism by which the method works.
The practical implications are straightforward. After reading a chapter, close the book and write down everything recallable — not what just got highlighted, not the main points as they appear on the page, what can actually be recalled without the book in front of you. The gaps in recall aren’t frustrating evidence of not learning anything. They’re the map of exactly where learning needs to go next. Make flashcards and use them actively, testing before looking at the answer. Take practice tests before feeling ready, not after. Quiz on yesterday’s material before starting today’s. Not clever tricks. The application of the most well-established finding in learning science.
Spaced Practice: The Spacing Effect
The second major principle in Make It Stick is spaced practice — distributing learning over time rather than massing it in a single session. The spacing effect is one of the oldest findings in memory research, first documented by Hermann Ebbinghaus in the 1880s, and remains one of the most reliable. Learning distributed across multiple sessions separated by time intervals produces better retention than learning massed in a single session of equivalent total duration.
The mechanism relates to retrieval. Returning to material after a time gap, some forgetting has occurred, and recalling it — bringing it back from a state of partial forgetting — is more effortful and more beneficial than reviewing material still fully fresh. The slight struggle to retrieve fading material is itself a retrieval practice event, strengthening the memory more effectively than reviewing the same material while still easily accessible.
Spaced practice feels less productive than massed practice. After a single four-hour study session, saturation with the material sets in — highly familiar, retrieval easy, performance on an immediate test good. After four one-hour sessions spread across a week, the material feels less fresh at each return, retrieval feels more effortful, performance on an immediate test runs modestly lower. But performance on a test two weeks later is dramatically higher for the spaced learners.
The short-term cost in fluency is an investment in long-term retention.
Practical implementation requires scheduling — a decision to return to material after specific time intervals rather than whenever it feels like a good idea. The optimal spacing interval depends on the time horizon: material needed for a month, spacing over days works. Material needed for a year, spacing over weeks works better. Spaced repetition software — programs like Anki that automatically schedule flashcard review based on previous performance — automates this optimization, presenting cards the moment a slight degree of forgetting has occurred and retrieval will therefore be maximally beneficial.
Interleaving: The Counterintuitive Scramble
The third principle — perhaps the most counterintuitive — is interleaving: mixing different types of problems or subjects within a single study session, rather than completing all problems of one type before moving to the next. Studying algebra, interleaving means mixing linear equations, quadratic equations, and word problems in the same session rather than finishing all linear equations, then all quadratics, then all word problems.
Why would mixing things up beat focusing on one thing at a time? Again, retrieval. Interleaving problem types means the immediately preceding problem can’t tell you which approach to use for the current one. The problem type has to be identified and the appropriate approach selected independently, from stored knowledge, every time. That identification step — which blocked practice eliminates — is itself a form of retrieval practice, building the discriminative ability to recognize which approach applies in which context. That discriminative ability is exactly what real-world application requires, where problems never arrive pre-labeled with their category.
The research on interleaving is particularly striking in domains involving categorization and pattern recognition. Medical students who studied diagnoses with interleaved examples — cases from multiple diagnostic categories presented in random order, rather than all cases of one type, then all of another — showed substantially superior diagnostic accuracy testing on novel cases. The interleaved students worked harder during study. They performed worse on immediate practice tests. But they showed superior transfer of learning to new, unlabeled cases — the only kind of case real patients present.
The discomfort of interleaving is real and worth acknowledging. Studying by interleaving, the material feels less clear, sessions feel less organized, performance on immediate tests runs lower. Not signs the method is failing. The characteristic signature of deep learning, always messier and more effortful in process than the clean fluency of massed practice. The goal isn’t feeling like learning is happening. It’s actually retaining and being able to use what got studied.
Elaborative Interrogation and Self-Explanation
Two additional strategies deserve particular attention: elaborative interrogation and self-explanation. Both involve generating explanations for material rather than simply receiving explanations, and both produce substantially better retention than passive reception of information.
Elaborative interrogation means asking “why” questions about material while studying it. Not “what” — not simply rehearsing facts — but “why is this true? What mechanism produces this outcome? How does this connect to what’s already known?” Generating an explanation — even an imperfect one, even a wrong one later corrected — activates a richer network of associations than simply receiving an explanation does. Cognitive work is happening, and the cognitive work is the learning.
Self-explanation is similar: pausing during problem-solving or text reading to explain, in your own words, what’s happening and why. This is the learning mechanism behind the “rubber duck” method software developers use — explaining code to an inanimate object often reveals misunderstandings and bugs, because the act of explanation makes implicit knowledge explicit and exposes gaps the fluency illusion had concealed. Students who pause during problem-solving to explain their reasoning make fewer errors and retain the material better than students who just work through problems without self-explanation.
The mechanism behind both strategies is the generation effect: information generated by oneself is more memorable than information received passively. A strong finding, appearing across domains — students remember vocabulary words better generating a sentence using them than reading a sentence using them, even when the read sentence is more elegant and informative. The act of generation, regardless of quality, drives encoding in ways passive reception doesn’t.
Calibration: Knowing What You Don’t Know
One of the book’s most important but least immediately practical contributions is its treatment of metacognition — the ability to accurately assess one’s own knowledge state. The fluency illusion, described earlier, systematically distorts this assessment. But even students who understand the fluency illusion intellectually tend to overestimate how well they know recently studied material, and underestimate how quickly that knowledge will decay without reinforcement through retrieval practice.
Calibration — the alignment between confidence in knowledge and actual ability to retrieve and use it — is itself a learnable skill. The authors describe research showing students who practice retrieval regularly become better calibrated over time: their assessments of what they know and don’t know grow more accurate, allowing more efficient allocation of study time. Students who study by rereading, by contrast, tend to stay poorly calibrated — consistently overestimating readiness for tests and consistently surprised when performance falls short of expectations.
The practical implication: retrieval practice isn’t only a learning strategy, it’s a diagnostic tool. Testing on material and discovering it can’t be recalled reveals something important — that fluency with the material was illusory and the material needs more work. Uncomfortable discovery — it disrupts the sense of progress rereading provides — but accurate, and accuracy about learning state is the precondition for effective allocation of study effort.
Desirable Difficulties
A unifying concept running through Make It Stick is “desirable difficulties” — a term coined by cognitive psychologist Robert Bjork for the class of learning conditions that feel harder in the moment but produce superior long-term outcomes. Retrieval practice is a desirable difficulty. Spaced practice is a desirable difficulty. Interleaving is a desirable difficulty. The common thread: all introduce a degree of cognitive effort passive review eliminates — and it’s precisely this effort that drives encoding.
Not all difficulties are desirable, of course. Difficulties arising from poor instruction, inadequate prerequisite knowledge, or distracting environmental conditions are simply obstacles to learning rather than drivers of it. The distinction: is the difficulty related to the process of encoding the target material — likely desirable — or to factors irrelevant to that encoding — simply a nuisance.
The concept of desirable difficulties has implications well beyond academic study. Skill acquisition in any domain runs on the same principles: the practice conditions that feel most comfortable — drilling skills in isolation, blocking similar tasks together, practicing in conditions closely matching the training environment rather than the deployment environment — are often less effective than conditions introducing variability, uncertainty, and the requirement to adapt. Varied practice — solving the same type of problem in multiple different contexts, with multiple different surface features — produces stronger transfer than blocked practice, even though it feels less clean and produces worse performance on immediate tests of the specific practiced form.
Mnemonics, Interleaving, and the Art of Making It Stick
The authors also address mnemonic devices — memory aids, memory palaces, acronyms, and other encoding strategies. Mnemonics can be powerful for specific types of material, particularly material lacking inherent structure or meaning — lists of unrelated items, arbitrary associations, sequences with no logical order. But the authors carefully distinguish mnemonics as an encoding strategy from deep learning as a comprehension and application strategy.
Mnemonic encoding can produce impressive on-demand recall of specific material, but it doesn’t automatically produce the kind of understanding that allows flexible application. A medical student who memorized the bones of the wrist using a mnemonic can recite them in order. That doesn’t mean they understand the functional anatomy making certain injuries possible and certain interventions effective. The mnemonic serves recall; understanding serves application. Both have their place, and neither substitutes for the other.
The book closes with recommendations for learners, teachers, and training designers, deceptively simple given the complexity of the evidence behind them: practice retrieving, not reviewing. Space your practice. Mix different types of material and problems. Generate explanations rather than consuming them. Test yourself before feeling ready. Accept the feeling of difficulty as a sign of learning rather than a sign of inadequacy.
Why Most Education Gets This Wrong
The gap between what learning science recommends and what actually happens in classrooms, training rooms, and self-directed study is enormous, and the authors address it directly. The gap exists for several reasons worth understanding.
First, the techniques producing the best learning aren’t the ones that feel best in the moment. Students prefer rereading to self-testing because rereading feels like learning while self-testing feels like evaluation. Teachers prefer blocked practice to interleaved practice because student performance in class looks better with blocked practice. Training designers prefer massed instruction followed by comprehensive testing over spaced practice with periodic retrieval tests because the former is easier to schedule and produces better immediate performance.
Second, the feedback loops in education are misaligned. Students get rewarded for immediate performance — on quizzes, assignments, exams — rather than long-term retention. The study strategies maximizing immediate performance (massed review right before a test) aren’t the strategies maximizing long-term retention (spaced practice over weeks). Students rationally optimize for the incentives they face, which are the wrong incentives.
Third, institutional inertia slows the diffusion of evidence-based teaching practices. Teachers teach the way they were taught. Curriculum designers design curricula around familiar templates. The research on retrieval practice, spacing, and interleaving has been available for decades and has barely penetrated mainstream educational practice. The gap between research and practice in education runs as large as the gap in any other domain where the people implementing the practice have limited exposure to the research and limited incentive to update their methods.
The Learner Who Takes This Seriously
The practical implications of Make It Stick are available to any individual learner regardless of what happens in their formal educational environment. No need for a teacher to change methods. No need for a training program to get redesigned. Only a change in how material gets personally engaged with — replacing passive review with active retrieval, spacing practice rather than massing it, interleaving rather than blocking, generating rather than receiving.
The learner implementing these principles isn’t working harder than the learner who rereads. They’re working differently — in a way that feels harder in the moment but produces dramatically superior outcomes over time. The discomfort is temporary and productive. The frustration of not being able to recall something isn’t evidence of failure. It’s the signal that the retrieval attempt is working exactly the memory it needs to work.
Over a career of continuous learning — and almost every serious professional career requires continuous learning — the compounding advantage of superior learning strategies is enormous. The person who encodes new material durably, who retains it under stress and time pressure, who can apply it flexibly in novel contexts, who knows what they know and what they don’t — that person carries a systematic advantage over the person who studies more hours but uses strategies producing only the illusion of knowledge.
The science is clear. The strategies are accessible. The only requirement is willingness to replace comfortable habits with effective ones, to trust evidence over intuition, to invest in the long term over the short. A small ask for a very large return.
Transfer: The Ultimate Goal of Learning
One dimension of learning science Make It Stick addresses most compellingly is transfer — applying learned knowledge or skill to contexts differing from the learning context. Transfer is the ultimate purpose of education and training. Chemistry doesn’t get learned to perform well on chemistry exams; it gets learned to understand chemical phenomena in the real world. Business case studies don’t get learned to analyze those specific cases; they get learned to think more clearly about business situations that will actually get encountered. The gap between what’s learned and where it must be applied is always present, and transfer is the bridge.
The research on transfer is sobering. Direct transfer — applying a skill in a context nearly identical to the learning context — is relatively strong. Near transfer — application in contexts moderately different from the learning context — is less reliable. Far transfer — applying principles learned in one domain to problems in a very different domain — is difficult and uncertain. The conditions promoting transfer are, not coincidentally, the conditions effective learning strategies produce: deep encoding through retrieval practice, varied practice across multiple contexts, understanding of underlying principles rather than surface features, and interleaving that requires discrimination among problem types.
The learner who studies by rereading and massed practice builds knowledge strongly associated with the specific context of the learning material — the textbook, the classroom, the specific framing of the problem. When the same knowledge is required in a different context, the retrieval cues differ, and the knowledge may not activate. The learner who practices retrieval in multiple contexts, with varied problem framings, builds knowledge linked to more diverse retrieval cues and therefore more likely to activate across a broader range of situations. The same amount of information, differently encoded, produces dramatically different transfer performance.
Interleaving in Professional Development
The principle of interleaving has applications far beyond academic study the book gestures toward but doesn’t fully develop. Consider professional development in any knowledge-intensive field: the doctor who reads journal articles on a single topic for a month, the lawyer who works through all cases in a single area of law sequentially, the engineer who focuses on one type of problem until it’s mastered before moving to the next. Each is doing blocked practice — accumulating fluency in a single context before moving to the next. The research suggests this is systematically inferior to interleaved practice, which in professional contexts might mean deliberately rotating across different types of problems or cases within each work period.
More practically, interleaving suggests the most effective professional development programs would mix different skill areas within each learning session rather than organizing by topic. A management training program alternating between communication skills, decision-making frameworks, financial analysis, and people management within each session — rather than a week on each sequentially — should produce better transfer to the actual managerial role than the sequential approach, even though the interleaved approach feels more disorganized and produces worse performance in the immediate evaluation at the end of each section.
Organizations that understand this and design their development programs accordingly will produce better performers. The problem is that organizations understanding it are rare, because the evidence base for interleaving — like the evidence base for retrieval practice and spaced repetition — isn’t part of the standard knowledge base of HR professionals and training designers. The gap between what the research recommends and what organizations actually do is one of the largest unfilled arbitrage opportunities in professional development.
Building a Personal Learning System
The practical synthesis of Make It Stick‘s principles for any serious learner points toward designing a personal learning system — a set of practices and habits consistently implementing the evidence-based strategies rather than relying on intuition about what feels productive. Such a system has several components.
The first component is a retrieval practice habit: after every significant learning event — a chapter read, a lecture attended, a document studied — taking ten minutes to write down, from memory, the most important points. No reviewing the material first. Just retrieval. The gaps in recall become the learning agenda for the next engagement with the material. Writing it down rather than just thinking it through adds the additional benefit of requiring more explicit formulation, which deepens encoding.
The second component is a spaced review schedule: a system for returning to previously learned material at increasing intervals. The simplest version is a calendar with recurring review reminders — return to last week’s material this week, last month’s material this month, last year’s material this year. More sophisticated versions use the spaced repetition algorithms built into flashcard software, optimizing the review schedule based on performance on each item. Either approach substantially beats the default, which is no planned review at all, and therefore rapid decay of most of what got learned.
The third component is a self-testing habit before important performances — not a final check on readiness, the primary study method. Before a presentation, write out the key points from memory and compare to notes. Before an important client meeting, recall the relevant background without looking at preparation materials. Before an exam, take practice tests under realistic conditions — timed, closed book, no notes. The self-testing reveals gaps that reviewing would leave hidden, and it produces better performance on the actual performance than equivalent time spent reviewing would produce.
The Broader Implication
The final implication of Make It Stick is one the authors state explicitly and that deserves to be taken seriously: the education system most people experience is systematically miscalibrated in its learning methods, and the miscalibration is largely invisible because the incentives driving educational practice don’t reward long-term retention. Not a minor inefficiency. An enormous waste of human potential — years of schooling producing knowledge that decays within weeks of the final exam, leaving graduates with credentials but not the durable knowledge the credentials are supposed to represent.
The individual response is becoming a deliberate learner — taking personal responsibility for implementing the strategies the educational system fails to implement, learning not just the content education provides but the learning skills education neglects. The person who graduates with both the content knowledge of their field and the learning skills to continue acquiring content knowledge throughout their career carries an enormous advantage over the person who graduates with only the former. Content knowledge depreciates. Learning skill compounds.
The strategies in Make It Stick aren’t complicated. No expensive equipment, no specialized training, no unusual intelligence required. Only the willingness to do learning that feels harder than what feels natural, to trust the evidence over the intuition, to invest in the long term over the short. A small ask for a very large return.
How Teachers Can Apply These Findings
While much of Make It Stick addresses individual learners, its implications for teaching practice are substantial and deserve attention. The authors address teachers directly in several chapters, arguing that the gap between evidence-based teaching practice and what happens in most classrooms isn’t primarily about teacher skill or dedication — it’s about teacher knowledge, specifically knowledge of what the research on learning actually shows. Most teachers teach the way they were taught, using methods learned through their own educational experience, shaped by the same misconceptions that make student studying so ineffective.
The evidence-based teacher uses low-stakes quizzing regularly — not as assessment, as a learning strategy. Rather than ending a class session with a summary of what got covered, they end it with a brief retrieval exercise: students write down the key points they remember without consulting notes, or answer questions about the material without reviewing it first. This practice, taking ten minutes and requiring no grading, produces better retention of the day’s material than any summary or review could. It’s also an honest diagnostic for both student and teacher: the gaps in students’ retrieval reveal exactly which parts of the lesson need reinforcement and which have been successfully encoded.
The evidence-based teacher also plans the curriculum with spaced repetition in mind — returning to important concepts across multiple class sessions separated by time, rather than covering them once comprehensively and moving on. Difficult within curriculum frameworks organized around sequential topic coverage, but the teacher who finds ways to weave earlier material into later lessons — who treats review not as wasted time but as essential consolidation — produces better long-term retention in students than the teacher maximizing fresh content coverage. The curriculum is not the learning; the learning is what sticks, and what sticks is what’s been retrieved multiple times across time.
The Argument Against Matching Learning Styles
One of the book’s most welcome myth-busting sections addresses the popular belief in “learning styles” — that different students learn best through different sensory channels (visual, auditory, kinesthetic), and instruction should be tailored to match each student’s preferred style. The learning styles concept is one of the most persistent myths in education, embedded in teacher training programs and school policies despite the consistent failure of research to find evidence it works.
The authors review the research on learning styles directly and find it unambiguous: no credible evidence exists that instruction matched to a student’s self-reported learning style produces better outcomes than instruction not matched to it. The studies seeming to support learning styles are consistently methodologically weak; the well-designed studies consistently fail to find the predicted advantage. The concept survives not because evidence supports it but because it’s intuitive, because it’s taught in teacher training programs, and because it provides a flattering account of student diversity that avoids the less comfortable conclusion that some students simply need more practice than others.
The evidence-based alternative to learning style matching is presenting material in multiple modalities — visual, verbal, kinesthetic where the content allows — not because each modality serves a different type of learner, but because multiple representations of the same concept produce richer encoding for all learners than a single representation does. The visual representation and the verbal explanation and the hands-on demonstration each add encoding pathways that reinforce the others. The right conclusion from a reasonable observation (some students seem to engage better with visual material), but it follows from principles of encoding richness rather than the learning styles myth.
A Call to Become a Deliberate Learner
The final synthesis of Make It Stick is a call to become the kind of learner who takes personal responsibility for the effectiveness of their learning rather than leaving it to the design of external systems. In a world where the formal educational system runs largely misaligned with what the research recommends, the individual who understands the evidence and applies it holds a systematic and substantial advantage. That advantage compounds over decades of continuous learning — producing a person whose knowledge is durable, whose skills are flexible, and whose capacity to acquire new knowledge is itself a practiced and refined skill.
The barriers to becoming that kind of learner aren’t intellectual. The principles are simple and accessible. The barriers are psychological: the discomfort of learning that feels harder than rereading, the ego cost of testing yourself and discovering less knowledge than assumed, the investment of designing a learning system rather than defaulting to patterns that feel natural. Real barriers, but surmountable ones, and the reward for surmounting them is a fundamentally different relationship to the acquisition of knowledge — more effortful in the moment, dramatically more productive over time.
Make it stick. Not by reviewing more but by retrieving more. Not by studying longer but by studying smarter. Not by seeking the comfort of fluency but by accepting the productive discomfort of genuine learning. The difference between the two paths, compounded across a lifetime, is the difference between a person who is always learning and a person who is always forgetting. The choice is available. The science is clear. What happens next is up to the reader.
Key Lessons from Make It Stick
- Retrieval practice is the most powerful single learning strategy available. Testing yourself on material produces better long-term retention than any amount of re-reading, re-listening, or re-watching — by large margins, consistently, across all subjects and ages.
- Spaced repetition compounds learning the way compound interest compounds money. Returning to material at increasing intervals after initial learning is far more efficient than mass practice, because it exploits the memory system’s tendency to consolidate information that is retrieved after partial forgetting.
- Desirable difficulty is the signature of genuine learning. The study strategies that feel most productive (rereading, highlighting, massed practice) are systematically less effective than strategies that feel harder (self-testing, spaced review, interleaving). If studying feels too easy, you’re probably not learning.
- Learning styles are a myth. The research is unambiguous: there is no credible evidence that instruction matched to a student’s self-reported learning style produces better outcomes. Multiple representations improve learning for all learners, not because they serve different types but because they create richer encoding.
- Learning skill is itself a learnable skill. The people who understand how learning actually works have a systematic, compounding advantage over those who rely on what feels natural. This advantage is available to anyone willing to apply the evidence over their intuition.
Real Talk on Make It Stick
Make It Stick is the most practical book on learning science available to general readers. Not the most comprehensive — academic texts go deeper — but the most accessible, the most directly applicable, and the most honest about the gap between what intuition says works and what the research demonstrates. Anyone who studies, teaches, trains, or develops people in any context should have these principles in their operating manual. The gap between knowing them and implementing them is the main obstacle to benefit; the book supplies everything needed to close that gap except the decision to actually do it.
Books Similar to Make It Stick
These books provide complementary frameworks and deeper context. Ultralearning by Scott Young provides a more ambitious and adventurous approach to the same learning science principles. A Mind for Numbers by Barbara Oakley provides the neuroscience foundation for focused/diffuse learning and is particularly valuable for quantitative subjects. Peak by Anders Ericsson provides the deliberate practice framework explaining how expert performance is built on the foundations Make It Stick describes. Thinking in Bets by Annie Duke covers the decision-making and calibration skills that are the natural complement to learning efficiency. Mindset by Carol Dweck provides the growth mindset foundation that makes the learning strategies in Make It Stick feel worth implementing.
Who Should Read Make It Stick
Students at any level who want to study less while learning more — specifically by switching from the ineffective strategies that feel productive to the effective strategies that feel harder. Teachers and trainers who want their instruction to produce durable learning rather than temporary performance. Professionals in any knowledge-intensive field who want to acquire and retain new information more efficiently. Parents who want to help their children study effectively. Coaches and athletic trainers who want to apply learning science to skill acquisition. Essentially: anyone who learns, teaches, or trains, which is everyone.
Practical Protocol
- Replace rereading with retrieval practice immediately. This week, after every significant learning event — chapter read, lecture attended, document studied — close the source and write down everything remembered. Don’t look back until everything producible from memory is written. The gaps become the learning agenda. This single change will improve long-term retention more than any other modification to learning practice.
- Build a spaced review system. Calendar: return to last week’s key material this week. Return to last month’s material this month. Return to last year’s material this year. The simplest possible implementation is a weekly fifteen-minute review of the previous week’s notes, converted to questions answered from memory. More sophisticated is spaced repetition software (Anki is free). Either substantially outperforms no system.
- Use self-testing as the primary preparation method before important performances. Before presentations, important meetings, exams, or any performance where preparation quality matters: write out the key content from memory without looking at notes. Compare to the original. Repeat with the gaps. This produces better performance than any amount of rereading, by concentrating practice on exactly what isn’t yet known.
- Embrace the difficulty of interleaved practice. Developing a complex skill with multiple components, resist the temptation to master each component in isolation before combining them. Practice the components interleaved — rotating between them within each practice session — and accept this will feel less organized and produce lower session-level performance. The long-term retention and transfer run significantly better. The discomfort is a sign it’s working.
Make Stick Summary Q&A
If retrieval practice is so powerful, why don’t schools use it? Because educational systems are optimized for several things other than long-term learning: covering curriculum, managing large groups efficiently, providing administratively convenient assessments, and minimizing student distress. Frequent low-stakes testing is more effective for learning but requires more implementation effort and generates more resistance from students who associate testing with high stakes and judgment. The gap between what research recommends and what schools do is one of the largest in any applied field.
How does retrieval practice work for procedural skills, not just factual knowledge? The exact same principles apply. For procedural skills, retrieval practice means performing the procedure from the performance side rather than studying descriptions of it — doing the calculation, not reading about how calculations are done; performing the surgical knot, not watching demonstrations of it. The specificity and difficulty of the practice should match the specificity and difficulty of the actual performance. Mental rehearsal of procedures also activates the retrieval mechanism and produces retention benefits, though physical practice is more effective for motor skills.
What is the best free tool for implementing spaced repetition? Anki is the most widely used and most research-aligned free spaced repetition tool available. It uses an algorithm based on the research to schedule flashcard reviews at the optimal interval for each card based on performance. It has a steep learning curve and requires the time investment of creating or importing card decks. The investment pays substantial compound dividends for any domain requiring maintenance of large amounts of factual or conceptual knowledge over time.
How much harder does retrieval practice feel compared to rereading? Much harder, which is both the sign it’s working and the main obstacle to implementation. Rereading feels productive because the material feels familiar — that’s fluency, which feels like learning but is actually just activation of existing memory. Retrieval practice feels difficult because information is getting produced against resistance — the difficulty is the learning mechanism. The discomfort is the signal that the memory system is doing the work that consolidates the learning. Trust the discomfort.
Can these principles help with skill acquisition as well as knowledge acquisition? Yes. The retrieval practice principle applies directly to procedural skills: practicing the skill from performance is more effective than studying descriptions of it. The spaced practice principle applies directly to physical skills: distributed practice sessions with rest periods outperform massed practice of the same total duration. The interleaving principle applies to skill development: mixing different related skills in practice produces better long-term performance than blocked practice on individual skills. The desirable difficulty principle is perhaps most visible in physical training, where the most effective training protocols are systematically those that feel hardest in the session.
Related: Five Dysfunctions of a Team Summary
Related: Dune Summary
FROM THE LIBRARY ›
References
Editorial StandardsCorrectionsMedical DisclaimerAbout Our ContentAffiliate DisclosureSite Map