Best Books on Learning How to Learn, Ranked by Evidence

If you read one book on how to study, make it Make It Stick by Peter C. Brown, Henry Roediger and Mark McDaniel. Two of its three authors are research psychologists at Washington University in St. Louis, and the book rests on work by 11 cognitive scientists across six universities. After that, add one book that fits your goal. For math or science, that is A Mind for Numbers by Barbara Oakley. For the reason these methods work, it is Why Don't Students Like School by Daniel Willingham (2nd edition, 2021).

Each book below was weighed on three questions. Who wrote it: a researcher or a practitioner? What research does it rest on? Have its main claims held up since publication, according to published replications and independent reviewers? Where our sources say little about a book, this page says so and does not guess.

A note on prices: we have no verified October 2026 Amazon US prices for these titles, so none are quoted here. Check the listing before you buy.

Make It Stick (2014) has the strongest research base on this list. Henry L. Roediger III is the James S. McDonnell Distinguished University Professor of Psychology at Washington University in St. Louis, and Mark A. McDaniel directs the university's CIRCLE center. Peter C. Brown, a writer and former management consultant, turns their work into stories a general reader can follow, according to the official book site.

The central claim is that retrieval, which means pulling information out of your memory, teaches more than review does. Rereading feels productive but is, in the book's words, "labor in vain." The book also covers spacing, which means spreading study over days, and interleaving, which means mixing problem types instead of doing them in blocks.

The evidence is concrete. RetrievalPractice.org's summary reports that in the K-12 classroom experiments the book cites, retrieval practice raised grades from a C+ to an A-, and the gains lasted a full school year. The book also states plainly that no empirical evidence supports matching instruction to "learning styles." That still matters: a September 2026 study of 269 university faculty in Ecuador found 90.7% endorsed learning-styles matching. The sample was not representative, but the myth is clearly still around.

The drawback: it is narrative-heavy, and the core advice fits on a page. Even Scott Young, who ranks it fifth on his own list, admits he recommends it for Roediger's research reputation without having read it cover to cover.

Skip it if you already use flashcards with spacing and just want something new.

A Mind for Numbers (2014) is the practitioner's pick for technical subjects. Its main idea is the difference between "focused" thinking, which is concentrated work on a problem, and "diffuse" thinking, which is the looser mode your mind shifts into when you step away. Two independent lists place it second: Scott Young's ranking and a 2026 roundup from InterObservers, which calls it best for technical subjects.

The drawback: our sources do not include a replication record for the focused/diffuse framework comparable to the one behind retrieval practice. Read it for its study habits, and lean on Make It Stick for the evidence.

Skip it if your subject is a language or a reading-heavy exam.

Willingham's book appears on RetrievalPractice.org's curated reading list and fourth on Scott Young's, so both a researcher-run site and a self-learning writer rate it. Buy the 2021 second edition rather than the original.

The drawback: it is written with teachers in mind, so a student has to translate it. Skip it if you want steps, not explanations.

RetrievalPractice.org, founded by cognitive scientist Pooja Agarwal, lists this book next to Make It Stick and Willingham. That endorsement comes from researchers who work on these exact techniques, which counts for a lot.

The drawback: beyond that recommendation, our sources do not grade its specific claims or compare its formats. Treat it as a reputable second read, not a replacement for Make It Stick.

Scott Young ranks Dehaene's book third and files it under cognitive neuroscience, the study of how the brain physically carries out learning. Pick it if you want the mechanism, not a study plan. Skip it if your exam is weeks away. Don't confuse it with How We Learn by Benedict Carey, a different book; our sources don't grade Carey's.

Peak by Anders Ericsson and Robert Pool (2016) was written by the researcher behind "deliberate practice," which means effortful, feedback-driven practice aimed at a specific weakness. Its core rests on a 1993 study of violinists by Ericsson, Krampe and Tesch-Römer. That study relied on people recalling how many hours they had practiced years earlier, a method open to bias. A replication by Macnamara and Maitra found the broad pattern (more practice, more skill) but not the neat match between hours and skill level. A 2014 meta-analysis by Macnamara, Hambrick and Oswald found practice explains a modest share of performance overall: larger in games and music, smaller in education and professions.

In fairness, Peak says much of this itself. It limits the framework to fields with well-established rules, like chess, sports and music, and says it fits most modern jobs poorly. Ericsson also objected when the "10,000 hours" average from his violinists was turned into a universal rule. Read Peak for the idea of practicing at the edge of your ability, not for a number of hours.

Ultralearning by Scott Young (2019) is a practitioner's book with a nine-principle plan for self-directed learning projects. It is useful if you are learning a skill without a class. It is not on RetrievalPractice.org's list, which reflects a real divide between books grounded in classroom research and self-directed learning guides.

Moonwalking with Einstein by Joshua Foer makes the bottom of both Young's list (tenth) and the InterObservers list (sixth to tenth). It is a good read about memory, not a study manual.

Not yet graded: our sources don't evaluate Outsmart Your Brain by Daniel Willingham (2023) or Uncommon Sense Teaching by Barbara Oakley, Beth Rogowsky and Terrence Sejnowski, so we won't rank them on reputation alone.

Exam in six weeks. Make It Stick, then start retrieval and spacing right away. Six weeks is enough time to space your review sessions instead of saving it all for the end.

Teaching yourself math. A Mind for Numbers for habits, plus Make It Stick's chapter on interleaving.

A new language. None of our sources grades a language-specific book. Use Make It Stick's retrieval and spacing methods with a flashcard app.

The bar exam or a certification. Make It Stick. Be careful with Peak here: the 2014 meta-analysis found deliberate practice explains the least in education and professions.

High school. Make It Stick, since its best-known results come from K-12 classrooms. The week before a big test, cramming is the wrong bet.

Free first. RetrievalPractice.org offers free discussion questions, sketchnotes and research articles for Make It Stick, plus more than 100 resources tied to Powerful Teaching, a classroom companion by Pooja Agarwal and Patrice Bain.

• Make It Stick: everyone, first.
• A Mind for Numbers: math and science self-learners.
• Why Don't Students Like School (2nd ed., 2021): readers who want the reasons behind the methods.
• Understanding How We Learn: a researcher-endorsed second read.
• How We Learn by Dehaene: the curious, with time to spare.

If none of these fits, especially for a language, wait for published evaluations of the newer titles above. Whichever book you choose, try one technique from it this week, such as a ten-minute self-quiz instead of a reread.