Book Title: Make It Stick: The Science of Successful Learning
Authors: Peter C. Brown, Henry L. Roediger III, and Mark A. McDaniel.
Published: 2014
Genre: Learning Science / Education
Table of Contents
- 1. Book Basics
- 2. The Big Idea
- 3. The Core Strategies: Evidence-Based Learning
- 4. What I Liked
- 5. What I Questioned
- 6. One Image That Stuck
- 7. Key Insights
- 8. Action Steps
- 9. One Line to Remember
- 10. Who This Book Is For
- 11. Final Verdict
- 12. Deep Dive: The Science of Memory, How Encoding Actually Works
- 13. Deep Dive: Practical Application Across Learning Contexts
- 14. Deep Dive: Common Mistakes in Applying the Framework
- 15. Deep Dive: Comparison to Related Frameworks
- Final Reflection: The Technical Foundation of the Series
1. Book Basics
Why This Book Exists
Make It Stick: The Science of Successful Learning was published in April 2014 by Harvard University Press. It was written by Peter Brown, a narrative writer, in collaboration with two of the most distinguished memory researchers in the world: Henry Roediger III, a cognitive psychologist at Washington University in St. Louis who has spent four decades studying how memory works, and Mark McDaniel, also at Washington University, whose research focuses on prospective memory and the science of encoding. The collaboration is unusual, a skilled writer and two working scientists, and it produces a book that is both empirically rigorous and genuinely readable.
The book’s premise is a paradox: the learning strategies that feel most effective, including re-reading, massed practice, highlighting, and reviewing lecture notes, are among the least effective methods ever studied. And the strategies that produce the most durable, transferable learning, including retrieval practice, spaced repetition, interleaving, and desirable difficulty, feel ineffective because they are harder, slower, and produce more errors during the learning process itself. The subjective experience of learning is a poor guide to actual learning, and the mismatch between how learning feels and how learning works has produced a global epidemic of inefficient study.
Roediger and McDaniel bring to the book decades of research on the testing effect (the finding that retrieving information from memory strengthens it more than re-reading it), spaced practice (the finding that distributed practice sessions produce far better long-term retention than the same amount of time spent in a single session), interleaving (the finding that mixing up different types of problems during practice produces better long-term performance than practising each type in isolation), and elaborative interrogation (the finding that asking “why?” and connecting new information to prior knowledge produces better encoding than passive review). The book synthesises this research into a unified account of how human memory actually works and what follows from that account for how to learn.
The intended audience is broad and explicitly stated: students, teachers, trainers, coaches, and anyone who wants to learn more effectively. The book is structured around a set of core principles, illustrated throughout with vivid case studies, including a military pilot surviving a crash by applying mental simulation, a neurosurgeon who rebuilt his skills after a stroke, and a Chicago teacher whose students dramatically outperformed peers using retrieval practice, that make the abstract findings concrete and memorable.
2. The Big Idea
The central claim of Make It Stick is that the most effective learning strategies are the ones that feel least effective in the moment, and the most popular learning strategies are the ones that feel most effective but produce the least durable retention. This double inversion, effective feels bad and ineffective feels good, is the cognitive trap that the book exists to name and dismantle.
The mechanism behind the trap is what the authors call the fluency illusion: when you re-read a chapter, the text becomes familiar, and familiarity feels like knowledge. But familiarity is not the same as the ability to retrieve and use knowledge in a different context. Re-reading creates the subjective experience of having learned while producing relatively little actual encoding. Retrieval practice, being tested on material before you feel ready, generating answers rather than recognising them, is uncomfortable, error-prone, and slow. It also produces encoding that is three to four times more durable than re-reading, according to the research.
The book’s second foundational insight is about the structure of difficulty. Not all difficulty is desirable, but certain kinds of difficulty, those that require the learner to reconstruct knowledge rather than simply recognise it, produce stronger encoding. The difficulty is desirable because the act of reconstruction strengthens the memory trace in ways that passive review cannot. Struggling to retrieve an answer before being shown it produces better retention than being shown the answer directly. Making errors during practice and then correcting them produces better retention than error-free performance. The difficulty is not an obstacle to learning. It is the mechanism of learning.
What Changes
The primary change in a reader who has absorbed Make It Stick is in how they study, practise, and train, both for themselves and for others they teach or manage. The re-reader becomes a self-tester. The blocked-practice advocate becomes an interleaving advocate. The single-session crammer becomes a spaced-repetition practitioner. These changes are not incremental improvements. The research suggests they produce qualitative differences in retention and transfer.
The secondary change is metacognitive: readers develop a more accurate model of their own learning. The ability to distinguish between the subjective experience of learning (fluency, familiarity, ease) and objective evidence of learning (the ability to retrieve, apply, and transfer knowledge to new contexts) is itself a durable cognitive skill that improves all subsequent learning. Once you know the fluency illusion exists, you cannot be entirely fooled by it again.
3. The Core Strategies: Evidence-Based Learning
Retrieval Practice (The Testing Effect). Testing yourself on material through flashcards, practice problems, free recall, or low-stakes quizzes, rather than re-reading or reviewing notes. Retrieving information from memory strengthens the memory trace more than passive review. The act of reconstruction, not recognition, is what produces durable encoding. Roediger’s research shows retrieval practice produces two to four times better long-term retention than re-reading.
Spaced Practice (Distributed Learning). Distributing practice sessions over time rather than massing them into a single extended session. Returning to material after a gap, when some forgetting has occurred, before the next practice session. Forgetting is a feature, not a bug. When you retrieve information after a gap that has produced some forgetting, the retrieval effort is greater and the resulting encoding is stronger. The spacing effect is one of the most robust findings in cognitive psychology.
Interleaving (Mixed Practice). Mixing different types of problems or subjects within a single practice session rather than completing all problems of one type before moving to the next. Interleaving forces the learner to identify which strategy or concept applies to each problem, not just execute a known procedure. This discrimination practice produces better long-term performance even though it feels slower and more error-prone during learning.
Elaborative Interrogation. Connecting new information to prior knowledge by asking “why does this work?”, “how does this relate to what I already know?”, and generating explanations rather than consuming them. Connections to existing knowledge create more retrieval routes to new information. The more a new concept is woven into an existing web of understanding, the more accessible it becomes from multiple starting points. Explanation is encoding.
Concrete Examples. Grounding abstract concepts in specific, vivid instances before or after encountering the abstraction. Abstract principles are harder to retrieve than concrete instances. Concrete examples create multiple retrieval cues for abstract concepts and make transfer to new contexts more likely.
Dual Coding. Combining verbal and visual representations of the same concept, including diagrams alongside text, charts alongside prose, and mental imagery alongside verbal descriptions. Verbal and visual information are encoded in different memory systems. Encoding the same information through both systems creates redundant retrieval routes and significantly improves recall.
Generation Effect. Attempting to produce an answer, solve a problem, or summarise material before receiving instruction or feedback, being wrong first, then corrected. Generating an incorrect answer before receiving the correct one produces better retention than being shown the correct answer directly. The error creates a distinctive retrieval cue that the subsequent correction attaches to.
Reflection. After a learning session, taking time to ask what went well, what could be improved, what connections exist to prior knowledge, and what remains unclear. Reflection combines retrieval practice with elaborative interrogation. It consolidates encoding by forcing reconstruction of what was learned and connecting it to the learner’s existing knowledge and experience.
The Illusions of Knowing: What Does Not Work
Re-reading is the most common study strategy and one of the least effective for long-term retention. It creates familiarity, which is mistaken for learning, while producing minimal encoding benefit relative to time invested. Massed practice (cramming) produces rapid short-term improvement, which is why it feels effective, but minimal long-term retention. Highlighting and underlining produce the illusion of active engagement while requiring no retrieval or reconstruction. Blocked practice, practising one type of problem repeatedly until mastered before moving to another, feels efficient because performance improves rapidly within the session, but the improvement reflects procedural familiarity rather than durable encoding. And the feeling of knowing, the subjective sense that you could retrieve something if asked, is unreliable as a measure of actual knowledge: familiarity with a concept creates the feeling of knowing without the ability to retrieve and use it.
4. What I Liked
The empirical credibility is unimpeachable. Unlike most learning-advice books, which are based on intuition, interviews, or loosely cited research, Make It Stick draws on decades of rigorously controlled experiments by two of the world’s leading memory researchers. The testing effect, the spacing effect, and the interleaving effect are not speculative frameworks. They are among the most replicated findings in cognitive psychology. Readers can trust what the book recommends.
The case studies are the book’s pedagogical genius. The authors practise what they preach by illustrating every abstract principle with a vivid, concrete story: Matt Brown the pilot who survived a crash through mental simulation, Nathaniel Fuller the actor who used retrieval practice to memorise roles, the Chicago schoolteacher whose class dramatically outperformed peers through daily low-stakes quizzing. These stories function as elaborative examples, creating concrete retrieval routes to the abstract principles, exactly as the book prescribes.
The fluency illusion is the most practically important single concept. The gap between “this feels familiar” and “I can retrieve and use this” is where most learning strategy errors live. Naming the illusion and explaining its mechanism gives readers a diagnostic that immediately upgrades their metacognitive accuracy. Once you understand the fluency illusion, you cannot blithely re-read a chapter and mistake familiarity for learning.
The desirable difficulty framework resolves a genuine confusion. Many learners and teachers associate struggle during practice with failure to learn. Make It Stick inverts this: the right kinds of difficulty, including retrieval effort, spacing gaps, and interleaved problems, are the mechanism of learning, not evidence of its absence. This inversion has immediate practical consequences for how learners evaluate their own practice and how teachers interpret student struggle.
The sections on teaching and training are unusually specific. Most learning science books stop at the individual learner. Make It Stick has substantive, actionable guidance for teachers, coaches, and corporate trainers on how to redesign instruction around retrieval practice, spaced testing, and interleaving. The examples from medical education, military training, and corporate learning are particularly valuable.
The chapter on the illusions of knowing is the most intellectually honest section. The authors spend significant time documenting not just what does not work but why it feels like it does work, the mechanisms by which re-reading, massed practice, and highlighting generate false confidence. This metacognitive section is more useful than a simple list of bad strategies would be, because it explains the trap rather than just identifying it.
5. What I Questioned
The motivational dimension is underaddressed. The strategies the book recommends, including retrieval practice, spaced repetition, and interleaving, are harder, slower, and more error-prone than the strategies they replace. They are also less enjoyable, particularly for learners who are already anxious about performance. The book acknowledges that desirable difficulty produces discomfort but does not spend adequate time on how to sustain motivation through strategies that consistently feel worse than the alternatives. Pink’s Drive is the necessary companion text.
The context specificity of the research is not always explicit. Most of the studies cited were conducted in laboratory settings or academic classroom contexts, primarily with university students. The book presents its findings as broadly applicable, and many are, but the transfer of specific recommendations including optimal spacing intervals, interleaving ratios, and testing frequency to very different learning contexts including professional skills development, motor learning, and creative practice is not always as seamless as the text implies.
The role of prior knowledge is underdeveloped. Elaborative interrogation and the testing effect both work differently depending on the learner’s existing knowledge base. A learner with rich prior knowledge in a domain encodes new information more efficiently because they have more existing nodes to connect it to. A learner with minimal prior knowledge has fewer connections available and may need different sequencing. The book discusses elaborative interrogation but does not fully work through the implications of knowledge asymmetry.
The learning styles critique is correct but briefly made. The book dismisses learning styles correctly, as the research does not support matching instruction to style, but the argument is made quickly and may not be persuasive to readers who hold the belief strongly. A more thorough engagement with why the learning styles idea is appealing and why the research consistently fails to support it would serve the argument better.
The spacing recommendations are underpractical. The book establishes convincingly that spaced repetition is superior to massed practice. It is less helpful on the practical question of how to actually implement optimal spacing in a busy life or institutional schedule. The gap between “spacing is important” and “here is a specific, implementable spacing system” is wider than the book acknowledges.
6. One Image That Stuck
The Quizzing Teacher in a Chicago Elementary School
A Chicago public school teacher, working with students who came disproportionately from low-income households and were performing below grade level in reading and science, began using a simple daily practice: at the start of each class, she gave a short, low-stakes quiz on material from previous sessions. Not a graded exam, not a high-stakes test. Just a few questions, answered quickly, checked immediately. The quizzes were short enough to feel routine. They were frequent enough to require regular retrieval of prior learning.
By the end of the year, her students had dramatically outperformed comparable students in other classrooms, including classrooms with more experienced teachers and students with higher baseline performance. The gap was not marginal. It was large enough to be noticed, investigated, and reported. The cause was not a curriculum innovation or a technology intervention or a change in teaching philosophy. It was the daily practice of retrieval: the students were regularly required to pull information from memory rather than simply receive it.
What makes this image so durable is its democratic implication. Retrieval practice does not require expensive technology, specialised training, or advantaged students. It requires a teacher who understands that the act of remembering is not a test of learning. It is a form of learning. The students who were asked to retrieve information regularly were not being assessed more; they were being taught more effectively. The quiz was not an evaluation instrument. It was a learning instrument.
The image also corrects a deeply embedded assumption about what practice is for. Most practice in educational settings is conceived as rehearsal for later evaluation: we practise so that we can be tested. Make It Stick inverts this: testing is practice. Retrieval is encoding. The quiz at the start of class is not preparation for the real learning. It is the real learning. And for students who had been underserved by the traditional model, this inversion produced results that the traditional model could not.
7. Key Insights
1. Learning that feels easy is often shallow; learning that feels hard is often deep. The subjective experience of learning is inversely correlated with its effectiveness for a wide range of common strategies. Fluency feels like knowledge. Difficulty feels like failure to learn. Both feelings are unreliable guides. The most durable encoding comes from effortful retrieval, spaced practice, and interleaved problems, all of which feel slower and harder than re-reading, massed practice, and blocked practice.
2. Retrieval practice is not a test of learning; it is a form of learning. The act of retrieving information from memory strengthens the memory trace more than any form of passive review. This is the testing effect, and it is among the most replicated findings in cognitive psychology. The implication is direct: students should spend more time testing themselves than reviewing notes, and teachers should use low-stakes quizzing as a primary instructional tool, not a summative assessment.
3. Spacing works because forgetting is not the enemy of learning; it is part of the mechanism. When you retrieve information after a gap that has allowed some forgetting to occur, the retrieval effort is greater and the resulting encoding is stronger. The spacing effect exploits the forgetting curve rather than fighting it. Returning to material at intervals that require genuine retrieval effort produces far better long-term retention than returning to it while it is still fresh.
4. Interleaving produces discrimination learning that blocked practice cannot. When you practise different problem types in a mixed sequence, you must identify which concept applies to each problem, not just execute a known procedure. This identification process is the learning that transfer requires. Blocked practice produces rapid within-session improvement and poor long-term performance. Interleaved practice produces slow within-session improvement and far better long-term performance. The immediate feeling of progress is misleading.
5. Generation before instruction improves retention even when the generation is wrong. Attempting to answer a question or solve a problem before receiving instruction produces better retention than receiving instruction directly. The generation of an incorrect answer creates a distinctive memory trace that the subsequent correct answer attaches to. Being wrong first, then corrected, is a more effective learning sequence than being correct immediately.
6. Elaboration creates a denser web of retrieval routes. Every connection between new information and prior knowledge creates an additional retrieval route to that information. The more elaborately you connect a new concept to what you already know, through explanation, analogy, application, and questioning, the more retrieval routes you create. Well-encoded knowledge is not stored more firmly; it is stored more multiply. Retrieval can begin from any of its connecting nodes.
7. The fluency illusion is the root cause of most learning strategy errors. Fluency, the ease with which information can be processed, is mistaken for knowledge because it feels similar. Re-reading creates fluency but minimal encoding. Familiarity with a concept feels like knowing it. The feeling of knowing is unreliable as a predictor of actual retrieval ability. Calibrating metacognitive accuracy, developing a more honest model of what you actually know, is itself a learnable skill and a prerequisite for effective self-directed study.
8. Mental simulation and reflection are retrieval practice in disguise. After a performance, mentally replaying what happened, what went well, what went wrong, and what you would do differently, is a form of retrieval practice that strengthens encoding and improves future performance. The pilot who mentally simulates emergency procedures before a flight is doing the same thing as the student who closes the textbook and tries to recall the main points. Both are retrieving rather than reviewing.
9. Transfer requires variation in practice, not just repetition. The ability to apply knowledge to new contexts requires that practice has included variation in context, problem type, and application. Practising in exactly one format, in one sequence, in one context produces performance in that specific format and poor transfer. Varied practice, interleaving, and diverse examples are what build the flexible, transferable knowledge that real-world performance requires.
10. Knowing your own learning style matters less than using evidence-based strategies. The learning styles hypothesis, that people learn better when instruction matches their preferred style (visual, auditory, kinaesthetic), has been tested extensively and is not supported by the research. What does matter is whether strategies produce active encoding (retrieval, elaboration, generation) or passive processing (re-reading, listening, watching). The distinction that matters is not style but effort.
8. Action Steps
START: Replace Re-reading with Retrieval Practice
Use when: You are studying anything you want to remember and be able to use, not just recognise on a test.
The Practice:
After reading a section, chapter, or attending a lecture, close the material and write down everything you can remember: main ideas, key arguments, specific details, examples. Do not look. Do not check. Simply retrieve.
After you have retrieved everything you can, check your notes or the material. Note what you missed, what you got wrong, and what connections you did not make. Correct the errors explicitly, writing the correct information next to the error, not over it.
One to two days later, retrieve again without looking. The gap is important: some forgetting should have occurred. The retrieval effort at this point is more valuable than retrieval while the material is still fresh.
For technical or factual material, use flashcards or practice problems rather than free recall. The principle is the same: generate the answer before seeing it. If using digital flashcards such as Anki, use the spacing algorithm built into the software, as it is implementing the spacing effect automatically.
Why it works: Retrieval practice forces reconstruction rather than recognition. The reconstruction effort strengthens the memory trace in ways that re-reading cannot. The errors you make and correct are encoded more distinctively than correct responses that cost no effort. Checking what you missed provides calibration feedback that re-reading systematically withholds.
STOP: Massed Practice and Last-Minute Cramming
Use when: You are tempted to spend a long, uninterrupted session studying material you need to retain beyond the next 48 hours.
The Practice:
Identify the material you need to learn and the date by which you need it. Work backwards from the date to create a spaced study schedule: multiple shorter sessions with gaps between them, rather than one or two long sessions close to the deadline.
In each session, begin with retrieval of material from the previous session before introducing new material. The retrieval at the start of each session is not review. It is practice. It should require effort.
When you catch yourself re-reading material you have already read, stop. Ask yourself: what do I actually remember from this? Can I explain it without looking? If not, close the material and try to retrieve before returning to it.
Use the discomfort of not knowing as information rather than anxiety. When retrieval is difficult, you are learning. When re-reading feels comfortable, you may not be.
Why it works: Massed practice produces rapid within-session improvement that feels like learning but evaporates quickly. Spaced practice produces slower within-session improvement that feels less satisfying but produces durable retention. The uncomfortable truth is that learning strategies optimised for immediate performance are suboptimal for long-term retention, and the strategy that feels best in the moment is often the one that serves you least.
TRY FOR 30 DAYS: Build a Spaced Retrieval System
Use when: You are learning anything you want to retain and be able to use over months or years, including a language, a professional field, a technical skill, or a body of knowledge.
The Practice:
Week 1. Set up the system: Choose one subject or skill area. After each study session, create a set of retrieval prompts: questions, problem stems, or blank summaries that require you to reconstruct key concepts rather than simply recognise them. Handwritten index cards, Anki digital flashcards, or a simple notebook all work.
Week 2. Begin spacing: Return to your retrieval prompts from week one, but not the same day you created them. Leave a gap of at least two to three days. Attempt to answer each prompt without looking. Note which ones you retrieve confidently, which ones are effortful, and which ones you cannot retrieve at all. Review only the ones you missed or were uncertain about.
Week 3. Introduce interleaving: In your practice sessions, mix prompts from different topics or time periods rather than working through them in the order they were created. The mixing is uncomfortable and your performance within the session will be worse than if you practised each topic separately. This discomfort is evidence that the interleaving is working.
Week 4. Calibrate and continue: Review your pattern of errors. What are you consistently missing? What have you solidly encoded? For solid material, extend the spacing interval, returning to it less frequently. For consistently missed material, increase the retrieval frequency until it is solid, then space it out.
Why it works: The spaced retrieval system implements three evidence-based strategies simultaneously: retrieval practice (you are always generating answers, not reviewing notes), spaced repetition (the gaps between sessions exploit the spacing effect), and interleaving (mixed practice produces discrimination learning). The cumulative effect after thirty days is not just better retention of the specific material. It is a fundamentally more accurate model of what you actually know versus what merely feels familiar. That metacognitive upgrade applies to all subsequent learning.
What you will notice by day 30: The gap between what you thought you knew and what you can actually retrieve will have become visible and then narrowed. The discomfort of early retrieval attempts, the effortful reaching for something that is not quite there, will have become recognisable as the feeling of learning rather than the feeling of failure. Material retrieved multiple times across spaced intervals will feel differently encoded from material you have only read: more accessible from multiple starting points, more connected to other things you know, more genuinely yours.
9. One Line to Remember
“Learning is deeper and more durable when it’s effortful. Learning that’s easy is like writing in sand, here today and gone tomorrow.”
“Trying to recall something you’ve studied is often more productive than studying it again. Practice at retrieving new knowledge or skills from memory is a potent tool for learning and durable retention.”
“The testing effect is one of the most robust and replicable effects in all of cognitive science. It’s also one of the most ignored by students and teachers.”
10. Who This Book Is For
Students at any level. The most direct and immediately useful book available for improving how you study. Replacing re-reading with retrieval practice alone, the simplest takeaway, will produce measurable improvement in retention and exam performance.
Teachers and instructors. The book provides a research-grounded argument and practical toolkit for redesigning instruction around retrieval, spacing, and interleaving, with specific classroom applications that have been tested and shown to work.
Corporate trainers and learning and development professionals. The evidence-based critique of conventional training design, including lecture, review, and single-session delivery, and the specific alternatives, including retrieval quizzing, spaced follow-up, and interleaved practice, are directly applicable to professional learning programmes.
Coaches and sports trainers. The chapters on motor learning, mental simulation, and varied practice are specifically relevant to skill acquisition in athletic and performance contexts.
Anyone learning a new skill or language. The practical sections on building a spaced retrieval system are immediately applicable to any self-directed learning project.
11. Final Verdict
Make It Stick is the single most practically important book in this series for anyone whose goal is to learn things and keep them. It is not the most motivationally sophisticated account of learning: Drive (Book 23) and Mastery (Book 22) cover motivation and skill development in more depth. It is not the most philosophically rich account of what learning is for: Four Thousand Weeks and Big Magic address that territory. But it is the most empirically rigorous and immediately actionable guide to the mechanics of encoding, to how human memory actually works and what follows from that account for how to study, practise, and train.
Its greatest strength is the combination of scientific credibility and narrative accessibility. Roediger and McDaniel are world-class researchers presenting their own field’s most robust findings. Brown is a skilled writer who ensures those findings arrive with stories rather than statistics. The result is a book that is simultaneously trustworthy and readable, a combination that is rarer in learning science than it should be.
Its greatest limitation is the motivational gap. The strategies Make It Stick recommends are better than the strategies they replace by every objective measure, and they are reliably less enjoyable in the short term. The book acknowledges this honestly but does not adequately address the question of how to sustain practice strategies that consistently feel worse than alternatives. For readers who already have intrinsic motivation for their subject, this gap is manageable. For readers who are struggling with engagement, it is a real obstacle.
In the context of this series, Make It Stick is the technical complement to Mastery (Book 22), Drive (Book 23), and Peak (Book 14). Peak establishes that deliberate practice is the mechanism of expertise. Drive explains why people persist in deliberate practice. Mastery describes the educational architecture that should deliver both. Make It Stick provides the specific cognitive science of encoding that tells you exactly what to do during the practice sessions those frameworks describe. Together, the four books constitute a complete account of how human beings learn: why they should, how they are motivated to, what the institutional conditions should look like, and what the specific mechanics of effective practice are.
The gap between how learning feels and how learning works is where most wasted study time lives. Close that gap and you do not need to work harder. You need to work differently.
12. Deep Dive: The Science of Memory, How Encoding Actually Works
The Encoding-Storage-Retrieval Framework
Human memory is often described as a storage system, a filing cabinet or a hard drive in which information is deposited and later retrieved. This metaphor is misleading in a way that has practical consequences. Storage implies that information is either present or absent, and that the goal of learning is to get information in. The retrieval-practice research suggests a different model: memory traces are not fixed deposits but dynamic structures that are strengthened each time they are successfully retrieved and weakened by disuse.
Encoding is not a single event but a process of connecting new information to existing knowledge. The richness of the encoding, how many existing concepts the new information is connected to, determines how retrievable it will be. This is why elaboration works: every connection to prior knowledge creates an additional retrieval route. A fact stored in isolation is accessible only from its own starting point. A fact woven into a network of related concepts is accessible from any of its connected nodes.
The Role of Sleep and Consolidation
The consolidation of learning, the process by which newly encoded information is stabilised and integrated into long-term memory, occurs primarily during sleep. Research by Matthew Walker and others has shown that specific stages of sleep are associated with the consolidation of different types of memory: REM sleep is associated with the integration of new information into existing knowledge networks, while slow-wave sleep is associated with the strengthening of factual and procedural memories. The practical implication is that sleeping after learning is not just restoring energy. It is completing the encoding process. Studying late at night and then sleeping is more effective than studying and then staying awake.
Why Making Errors Helps
The generation effect, the finding that attempting to produce an answer before receiving instruction improves retention even when the attempt is wrong, has an intuitive explanation in terms of prediction error. When you generate an incorrect answer and then receive the correct one, your brain registers a prediction error: the world did not match your expectation. Prediction errors are strong memory signals. The brain tags them as important and encodes the corrective information more durably than it would encode information that confirmed an existing expectation. Being wrong, then corrected, is not wasted effort. The error is the hook on which the correct answer hangs.
13. Deep Dive: Practical Application Across Learning Contexts
For Students
The single most impactful change a student can make is replacing re-reading with retrieval practice. Close the textbook after reading a section and write down everything you can remember before checking. Use the errors and gaps to guide the next retrieval attempt, not the next reading. Create a schedule that distributes study across multiple sessions with gaps between them, rather than concentrating study in single long sessions. If preparing for exams, begin retrieval practice from the first session, not as a way to assess readiness, but as the primary learning activity.
For Teachers
The research on the testing effect has direct classroom implications that require minimal resources: low-stakes quizzes at the start of each class, covering material from previous sessions, produce significant retention gains over re-teaching or review. The quizzes need not be graded. They need not be anxiety-producing. Their function is retrieval practice, not evaluation. Teachers who are concerned about testing culture misunderstand the mechanism: the issue is not whether quizzes are high-stakes or low-stakes, but whether students are regularly required to retrieve rather than merely review.
For Professional and Corporate Learning
The dominant model of corporate training, a day or a week of intensive instruction followed by a return to regular work with no structured follow-up, is almost perfectly designed to produce the forgetting curve. The evidence-based alternative has three components: spaced follow-up sessions (brief retrieval exercises returned to the learner at intervals of days, then weeks, then months), application assignments (learners required to apply new concepts in their actual work context between sessions), and peer teaching (explaining new concepts to a colleague produces elaboration and retrieval simultaneously). These components do not require more total training time. They require the same time redistributed differently.
14. Deep Dive: Common Mistakes in Applying the Framework
Treating retrieval practice as self-testing rather than primary learning. Many students who adopt retrieval practice use it as a check at the end of a study session, to see what they know after studying. The more powerful application is at the beginning: retrieve before reviewing. Use retrieval to identify gaps, not to confirm what is already solid. The gap is where the learning is.
Spacing without retrieval. Distributing study sessions across time without incorporating retrieval practice in each session produces some benefit but misses most of the spacing effect. The benefit of spacing comes from the combination of spaced returns and effortful retrieval. Spacing without retrieval is just divided re-reading.
Interpreting interleaving difficulty as evidence of poor learning. The most common reason learners abandon interleaved practice is that their performance within a session is worse than it would be with blocked practice. This is evidence that interleaving is working: it is forcing discrimination learning rather than procedural repetition. If interleaving feels too easy, it probably is not mixed enough.
Applying retrieval practice only to factual material. Retrieval practice works for conceptual understanding and problem-solving, not just facts. Practising explaining a concept in your own words, solving novel problems from a given domain, or applying a principle to a new situation are all forms of retrieval practice. Reducing it to flashcard-style fact recall misses most of its potential.
Using the absence of errors as evidence of solid learning. Error-free practice produces the same fluency illusion as re-reading. If retrieval practice is consistently producing correct answers with no effort, the spacing interval is too short or the prompts are too easy. Learning that feels easy is a warning sign, not a reassurance.
15. Deep Dive: Comparison to Related Frameworks
Peak by Ericsson and Pool (Book 14) provides the macro-level framework; Make It Stick provides the specific cognitive science of what happens during effective practice sessions. Peak tells you why to practise at the edge of your ability; Make It Stick tells you the specific mechanics that make practice encode durably.
Mastery by Wagner and Christensen (Book 22) describes the educational architecture; Make It Stick provides the evidence base for why that architecture’s components work. Learner agency supports encoding through elaboration. Portfolio assessment is retrieval practice. The mastery loop is a spaced retrieval system.
Drive by Pink (Book 23) explains why people persist in hard practice; Make It Stick explains what makes hard practice effective. The two books address the motivational and cognitive dimensions of the same activity: deliberate, effortful, spaced practice.
The Power of Habit by Duhigg (Book 9) provides the behavioural architecture; Make It Stick provides the memory science that explains why habits encode. The cue triggers retrieval of the routine, and the reward strengthens the encoding.
Flow by Csikszentmihalyi (Book 12) describes the optimal challenge zone from an experiential and motivational perspective; Make It Stick describes the same cognitive state from the memory science side. Both books are describing desirable difficulty from different angles.
A Mind for Numbers by Oakley is the natural companion for STEM learners specifically. Both are grounded in cognitive science and make the same core recommendations. Make It Stick is the broader synthesis; A Mind for Numbers is the domain-specific application.
Final Reflection: The Technical Foundation of the Series
Twenty-three books into this series, Make It Stick arrives as something like the user manual that should have come with the human brain. The series has been, in aggregate, an account of how to live and work from the inside out, from values to habits to creativity to motivation to mastery. Make It Stick is the cognitive science that explains how any of that learning actually sticks.
Every strategy that previous books in the series have recommended involves learning. Duhigg’s habit formation is an encoding process. Ericsson’s deliberate practice is retrieval and correction at the edge of current ability. Gilbert’s creative practice is elaborative interrogation applied to one’s own imagination. Cameron’s morning pages are reflection and retrieval combined. Goggins’ mental accounting of hard things done is spaced retrieval of evidence of capability. Every one of these practices is, through the lens Make It Stick provides, a form of effortful, desirable-difficulty encoding.
The book’s most durable contribution to the series is the correction it provides to the fluency illusion. The series has been asking readers to change how they work, how they relate, how they create, how they manage attention and energy. All of those changes require learning. And the single most reliable predictor of whether that learning will stick is whether the reader is retrieving and applying it: practising with the book closed, returning to the ideas after gaps, explaining the concepts to others, applying the frameworks to their own situations. The reader who treats this series as reading rather than practice has acquired fluency. The reader who has been doing the work, retrieving, applying, spacing, interleaving, has acquired something that will last.
Learning is not something that happens to you when you read. It is something you do when you retrieve, apply, and connect. The book is just the starting point. The real encoding begins when you close it.
