Best Study Techniques Backed By Cognitive Science (Ranked)

Most Study Advice Is Wrong

The overwhelming majority of study advice that students receive from teachers, parents, peers, and even well-intentioned study blogs has never been tested in controlled experiments. Surveys conducted across multiple universities consistently find that students' most frequently used study strategies include rereading notes, highlighting textbook passages, and cramming the night before an exam. These strategies feel productive because they are easy, fluent, and create an illusion of competence. Unfortunately, cognitive science research reveals that they are among the least effective approaches to long-term learning.

The landmark study that established this disconnect is Dunlosky, Rawson, Marsh, Nathan, and Willingham's (2013) comprehensive review, "Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology," published in Psychological Science in the Public Interest. This team evaluated ten of the most commonly recommended study techniques, rating each one on a scale from low to high utility based on the quality and consistency of the available research evidence. Their findings were striking: several of the most popular techniques received the lowest possible ratings, while lesser-known strategies that demand more cognitive effort consistently outperformed them across dozens of experiments.

The core problem is what psychologists call fluency illusions. When you reread a highlighted passage, the information feels familiar and accessible. Your brain interprets that ease of processing as evidence of mastery, even though you would struggle to recall the same information without the textbook in front of you. This mismatch between subjective feeling and objective performance is one of the most robust findings in cognitive psychology. Robert Bjork at UCLA has described this as "desirable difficulties" -- learning strategies that feel harder during study actually produce stronger, more durable memories. The techniques that feel effortless during practice tend to produce weak, fragile memories that fade quickly.

Since Dunlosky's review, subsequent meta-analyses and replication studies have largely confirmed these rankings. The gap between what students believe works and what science demonstrates works remains wide. This guide closes that gap by ranking the ten study techniques with the strongest evidence base, explaining why each one works at a cognitive level, and providing practical guidance for implementation. The techniques below are ordered by their combined rating across three criteria: effect size in experimental studies, reproducibility across different populations and contexts, and practical applicability for real students studying real material.

Ranked: The 10 Most Effective Study Techniques

Not all evidence-based study techniques are created equal. Some produce large, consistent improvements across multiple experiments and diverse subject areas, while others show promising but variable results that depend heavily on implementation quality. The ranking below considers three primary criteria derived from the published research literature.

Effect size refers to the magnitude of learning improvement that a technique produces in controlled experiments. In cognitive science, effect sizes are typically measured using Cohen's d or similar metrics. A technique with a large effect size might improve test scores by a full standard deviation compared to a control condition, which translates to a substantial real-world difference in exam performance. Techniques with small effect sizes may still be statistically significant but produce improvements so modest that they are unlikely to meaningfully change a student's grade.

Reproducibility captures whether a technique's benefits have been replicated across multiple independent research groups, different populations of students, and various subject domains. A technique that works in one lab with college students studying word lists but fails to replicate in high school classrooms studying chemistry is less valuable than one that shows consistent benefits across diverse settings. The reproducibility criterion also accounts for the quality of the underlying evidence, favouring techniques supported by large-scale meta-analyses and randomised controlled trials over those supported only by small-scale or underpowered studies.

Practical applicability considers how easily a real student can adopt and sustain the technique in everyday study sessions. Some techniques require specialised software, extensive training, or impractically large time investments. Others are simple enough to implement immediately with no equipment beyond pen and paper. Practical applicability does not override evidence quality, but it does affect ranking when two techniques have similar research support. The techniques below are ordered from highest to lowest across these combined criteria, giving you a clear hierarchy of where to invest your limited study time for maximum returns.

#1 Practice Testing (Active Recall)

Practice testing received the highest utility rating in Dunlosky et al.'s (2013) landmark review, and subsequent research has only strengthened its position as the single most effective study technique available. Also known as active recall or retrieval practice, this technique involves deliberately retrieving information from memory without looking at the source material. The mechanism is straightforward: every time you successfully retrieve a memory trace, you strengthen the neural pathways that encode that information, making future retrieval faster and more reliable. Conversely, failed retrieval attempts still provide valuable learning benefits by identifying gaps in knowledge and signalling to your brain that the information needs to be re-encoded more robustly.

The evidence base is extensive. Roediger and Karpicke (2006) demonstrated in a series of experiments that students who practised retrieval retained significantly more information after a one-week delay than students who simply reread the material, even when the rereading group spent more total time with the material. A meta-analysis by Rawson and Dunlosky (2011) found that practice testing produced effect sizes ranging from 0.45 to 0.80 across various conditions, placing it firmly in the medium-to-large range by psychological standards. Importantly, these benefits extend beyond simple factual recall to complex transfer tasks, including problem-solving and application of concepts to novel situations.

The format of practice testing matters, but not as much as you might expect. Free recall (writing everything you remember about a topic), cued recall (using flashcards with prompts), and short-answer questions all produce strong benefits. Multiple-choice testing also works, though it is slightly less effective because the provided answer options reduce the retrieval demand. The key principle is that testing must involve active generation of the answer from memory, not passive recognition of the correct option among alternatives. For a complete breakdown of implementation strategies, see our full active recall guide.

#2 Distributed Practice (Spaced Repetition)

Distributed practice, more commonly known as spaced repetition, earned the second-highest utility rating in Dunlosky's review and is widely regarded as one of the most robust findings in all of cognitive psychology. The principle is simple: spreading study sessions across multiple days produces dramatically better long-term retention than concentrating the same amount of study time into a single session. This is not a minor effect. The spacing effect, first documented by Hermann Ebbinghaus in 1885 and replicated hundreds of times since, routinely produces retention improvements of 200% to 400% compared to massed practice.

The underlying mechanism is rooted in the forgetting curve. When you first learn something, the memory trace begins to decay almost immediately. Each time you review the material before it has decayed completely, you reset the forgetting curve and strengthen the underlying neural connections. Critically, the optimal time to review is just before you are about to forget the information -- a point that algorithms in spaced repetition software estimate based on your performance history. Cepeda et al. (2006) conducted a comprehensive meta-analysis of 839 assessments from 184 studies and confirmed that spacing effects are robust across different materials, different age groups, and different retention intervals. They also found that longer spacing intervals tend to produce better long-term retention, though very long intervals risk allowing the memory to decay beyond recovery.

Practical implementation ranges from simple manual approaches to sophisticated algorithmic systems. The most basic approach is to review material at increasing intervals -- for example, one day after initial study, then three days later, then one week later, then two weeks later. For more precise scheduling, spaced repetition software such as Anki uses the SuperMemo SM-2 algorithm to calculate optimal review times based on your individual performance data. Research by Rawson and Dunlosky (2011) suggests that even modest spacing -- reviewing material across three sessions rather than one -- captures a substantial portion of the benefit. The critical factor is not the precise interval but the act of distributing practice across time rather than concentrating it. For detailed implementation guidance, see our full spaced repetition guide.

#3 Interleaved Practice

Interleaved practice refers to the strategy of mixing different topics or problem types within a single study session, rather than studying each topic in a concentrated block. This technique received moderate utility in Dunlosky's review, but subsequent research by Rohrer, Dedrick, and Stershic (2015) has elevated its standing considerably. The key finding is that interleaving consistently outperforms blocking for tasks that require discrimination -- the ability to distinguish between similar concepts and select the appropriate strategy for each problem.

Rohrer and Taylor (2007) conducted a landmark experiment with college students learning to calculate the volume of different geometric solids. One group practiced each solid type in blocked sessions (all spheres, then all cones, then all cylinders), while the other group interleaved the problem types randomly. On a final test one week later, the interleaved group scored 43% correct compared to just 20% for the blocked group -- more than double the performance. The advantage persisted even though the interleaved group performed worse during practice itself, an illustration of Bjork's desirable difficulties principle in action. Blocked practice creates a temporary fluency because students can apply the same formula repeatedly without having to identify which formula is appropriate. Interleaved practice forces students to constantly discriminate between problem types, building a more flexible and robust understanding.

The practical benefits are strongest in mathematics, science, and any domain where students must learn to categorise problems and select appropriate solution methods. Rau, Aleven, and Rummel (2010) demonstrated similar benefits in physics education, where interleaving different problem types improved students' ability to transfer concepts to novel situations. Implementation is straightforward: instead of completing all problems of one type before moving to the next, shuffle problems from multiple types together and work through them in random order. The initial difficulty is a reliable signal that interleaving is working -- it should feel harder than blocked practice, and that difficulty is precisely what produces the learning gains.

#4 Elaborative Interrogation

Elaborative interrogation is a technique that involves asking yourself "why is this true?" for each fact or concept you encounter during study. Rather than passively accepting information, you actively generate causal explanations that connect new information to existing knowledge structures. This process of generating explanations creates deeper semantic encoding, which means the information is stored in a richer, more interconnected network of associations within long-term memory.

The research evidence, while somewhat variable across studies, consistently shows moderate benefits. Dunlosky et al. (2013) rated elaborative interrogation as having moderate utility, with a typical effect size around d = 0.40 to 0.60 in controlled experiments. Seifert (1993) found that students who generated explanations for facts retained significantly more information than students who simply repeated the facts, even when the explanation group spent less total time studying. The key mechanism is that generating explanations requires activating relevant prior knowledge, making inferences, and constructing logical connections -- all of which create additional retrieval pathways that support later recall.

The technique is most effective when students have some prior knowledge of the domain. If you know absolutely nothing about a topic, generating meaningful explanations is extremely difficult, and the benefits diminish. However, for most university-level study where students have at least introductory knowledge, elaborative interrogation is highly practical. When reading a textbook, pause after each paragraph or key statement and ask yourself why the statement is true. Try to articulate the causal mechanism, connect it to something you already understand, or identify what would have to change for the statement to become false. Writing these explanations down is more effective than merely thinking them through, because the act of writing imposes additional cognitive demands that strengthen encoding.

#5 Self-Explanation

Self-explanation shares conceptual ground with elaborative interrogation but operates through a slightly different mechanism. Rather than asking "why is this true?" in general, self-explanation involves articulating the reasoning behind each step in a problem-solving process or explaining how a concept connects to related concepts. The foundational research comes from Chi, Bassok, Lewis, Reimann, and Glaser (1989), who observed that the best-performing students in physics problem-solving were those who spontaneously explained each step of their reasoning aloud, even when no one was listening.

Subsequent experimental research has confirmed these observational findings. Rittle-Johnson (2006) demonstrated that students who provided self-explanations while studying mathematics examples showed significantly greater procedural understanding and flexibility than students who simply studied the examples silently. The effect sizes are moderate but consistent, typically ranging from d = 0.30 to d = 0.55 across studies. The critical distinction from elaborative interrogation is scope: elaborative interrogation targets factual knowledge and asks for causal justifications, while self-explanation targets procedural knowledge and asks for step-by-step reasoning articulation.

Self-explanation is particularly valuable for mathematics and science subjects where problem-solving requires chaining multiple logical steps. When working through a practice problem, pause after each step and explain aloud why that step follows from the previous one, what principle it relies on, and how it moves you closer to the solution. Research by Renkl (1997) found that self-explaining worked examples was more effective than solving equivalent problems without explanation, because the explanation process builds a mental model of the solution strategy rather than a fragile sequence of rote steps. Students who struggle with a concept often benefit most from self-explanation precisely because it reveals the specific point where their understanding breaks down, allowing targeted remediation.

#6 The Feynman Technique

The Feynman Technique, named after the Nobel Prize-winning physicist Richard Feynman, is a learning strategy that leverages the generation effect combined with retrieval practice in a powerful pedagogical framework. The core procedure involves four steps: first, choose a concept you want to learn; second, explain it as if teaching it to someone with no background knowledge; third, identify the gaps in your explanation where you struggle to find simple, clear language; and fourth, return to the source material to fill those gaps. The technique draws on multiple well-established cognitive science principles simultaneously.

The act of teaching or explaining a concept engages deep semantic processing, similar to elaborative interrogation and self-explanation. However, the Feynman Technique adds an additional constraint that makes it uniquely powerful: the requirement to use simple, jargon-free language. When you force yourself to explain a concept without relying on technical vocabulary, you are compelled to understand it at a fundamental, structural level rather than merely memorising definitions. If you can explain quantum tunnelling without using the phrase "quantum tunnelling," you genuinely understand it. If you cannot, you have discovered a specific gap in your knowledge that targeted study can address.

The cognitive science behind this technique is substantial. The generation effect, documented by Slamecka and Graf (1978), demonstrates that self-generated information is remembered significantly better than externally provided information. The Feynman Technique maximises generation because you must construct the entire explanation from scratch, not merely reproduce a pre-existing explanation from memory. Additionally, the technique incorporates self-testing: when you reach a point in your explanation where your knowledge breaks down, you receive immediate, accurate feedback about what you do not know. This metacognitive monitoring is essential for effective self-regulated learning. For practical applications of this and related memory strategies, see our guide to exam memory tricks.

#7 Concept Mapping

Concept mapping is a visual learning technique that involves creating structured diagrams showing the relationships between concepts within a domain. Typically, concepts are represented as nodes (boxes or circles) connected by labelled arrows that specify the nature of the relationship between them. The approach is rooted in Ausubel's (1968) assimilation theory, which posits that meaningful learning occurs when new information is consciously connected to existing knowledge structures. Concept mapping makes these connections explicit and visible, supporting the integration of new concepts into the learner's existing cognitive framework.

The research foundation comes primarily from the work of Novak and Canas (2008), who developed concept mapping as an educational tool over several decades of research. However, the evidence for concept mapping is more nuanced than many educators realise. Dunlosky et al. (2013) rated concept mapping as having low utility overall, because the benefits depend heavily on implementation quality. When students create concept maps by comparing them to a provided expert map, the technique shows moderate benefits. When students create concept maps without any reference standard, the benefits are inconsistent because students may construct inaccurate maps that reinforce misconceptions rather than correcting them.

The strongest evidence suggests that concept mapping is most valuable as a knowledge integration tool rather than an initial encoding tool. It works best after you have already developed a basic understanding of individual concepts and need to organise and connect them into a coherent framework. Using concept mapping too early, before you have sufficient domain knowledge to draw accurate relationships, can actually impair learning by creating premature and potentially incorrect associations. When used correctly as an advanced organisational strategy, concept mapping supports the development of structural knowledge -- understanding how concepts relate to each other within a domain -- which is critical for transfer and application in novel situations.

#8 Visual Imagery and Dual Coding

Allan Paivio's dual coding theory (1971) proposes that the human cognitive system maintains two separate but interconnected memory systems: one for verbal-linguistic information and one for visual-spatial information. According to this model, information that is encoded in both systems simultaneously is more robustly stored and more readily retrieved than information encoded in only one system. This theory provides the theoretical foundation for visual imagery as a study technique: by creating vivid mental images to accompany verbal information, you engage both coding systems and create redundant memory traces that support retrieval.

The experimental evidence supports a moderate benefit for dual coding, particularly for concrete information that naturally lends itself to visualisation. Dunlosky et al. (2013) rated imagery as having moderate utility, with effect sizes typically in the d = 0.30 to d = 0.50 range. The technique is straightforward: when studying a concept, deliberately create a detailed mental image that represents the information. For example, when learning about the mitochondria as the powerhouse of the cell, you might visualise a miniature factory inside each cell, complete with turbines and conveyor belts. The more vivid and elaborate the mental image, the stronger the dual coding benefit.

However, dual coding has important limitations. It is most effective for concrete, easily imageable information and substantially less effective for abstract concepts that resist visualisation. Trying to force visual images onto highly abstract material (such as philosophical arguments or mathematical proofs) can consume cognitive resources without producing proportionate learning gains. Additionally, some students are naturally more inclined toward visual processing than others, and individual differences in spatial ability moderate the technique's effectiveness. For students with strong spatial reasoning skills, dual coding is a powerful addition to any study routine. For those who struggle with mental imagery, the technique should be used selectively rather than applied universally.

#9 The Memory Palace (Method of Loci)

The Method of Loci, commonly known as the memory palace, is one of the oldest documented mnemonic techniques, with origins tracing back to ancient Greek and Roman rhetorical traditions. The method involves mentally placing items to be remembered along a familiar spatial pathway, such as the rooms of your house or the route you walk to work. During recall, you mentally retrace the pathway and retrieve each item from its imagined location. Despite its ancient origins, modern neuroscience has validated the technique's effectiveness and begun to reveal its neural mechanisms.

Functional MRI studies by Maguire, Valentine, Wilding, and Kapur (2003) compared the brain activation patterns of competitive memory athletes with those of matched controls. They found that memory champions showed significantly greater activation in brain regions associated with spatial memory and navigation, including the right posterior hippocampus, the retrosplenial cortex, and the lateral parietal cortex. Remarkably, a follow-up study by Dresler et al. (2017) demonstrated that training naive participants in the Method of Loci over six weeks produced measurable increases in connectivity between the medial prefrontal cortex and the right hippocampus, suggesting that the technique physically alters brain network architecture to support superior memory performance.

The practical effectiveness of the memory palace has been demonstrated in controlled experiments. Ericsson, Delaney, and Weaver (2017) showed that participants trained in the Method of Loci improved their recall of word lists by approximately threefold compared to untrained controls. The technique works best for ordered information that can be associated with specific spatial locations, such as lists of items, sequences of steps, or structured information with a natural progression. It is less suited to abstract conceptual understanding or information that requires flexible application rather than verbatim recall. For specific applications and variations, see our guide on how to memorise faster. The memory palace is a specialised but genuinely powerful tool that is supported by both millennia of practical use and cutting-edge neuroimaging evidence.

#10 Mnemonic Devices

Mnemonic devices encompass a broad family of memory aids designed to enhance encoding and retrieval by creating systematic associations between to-be-remembered information and more easily retrievable cues. The two most well-studied categories are the keyword method and first-letter mnemonics. The keyword method involves creating a vivid mental image linking a familiar keyword (that sounds like or shares features with the target word) to the target's meaning. For example, to remember that the Spanish word "caballo" means horse, you might imagine a horse riding in a cab. First-letter mnemonics, such as "ROY G. BIV" for the colours of the rainbow, create memorable phrases or acronyms from the initial letters of the target items.

The research evidence for mnemonics shows moderate but highly consistent effect sizes, typically in the d = 0.30 to d = 0.60 range across dozens of experiments. Dunlosky et al. (2013) noted that keyword mnemonics are particularly effective for foreign language vocabulary acquisition, where the technique has been validated across multiple languages and age groups. First-letter mnemonics are effective for ordered lists where the target items have a fixed sequence, though they are less useful for information that requires flexible application or deep conceptual understanding.

The key limitation of mnemonic devices is their specificity. They are highly effective for the specific information they are designed to encode, but they do not generalise broadly to other types of learning. A keyword mnemonic helps you remember a specific word's translation, but it does not help you understand grammatical structures or produce spontaneous speech. First-letter mnemonics help you recall an ordered list, but they do not support the ability to explain why the items appear in that order or how they relate to each other. For these reasons, mnemonic devices rank tenth on this list: they are genuine and reliable tools, but their benefits are narrower in scope than the higher-ranked techniques. They are best used as supplements to broader strategies like active recall and spaced repetition rather than as primary study methods.

Techniques That Don't Work (Or Barely Work)

Understanding which techniques to avoid is just as important as knowing which ones to adopt. The most commonly used study strategies are also among the least effective, a paradox that perpetuates poor academic outcomes for millions of students worldwide. Dunlosky et al. (2013) assigned the lowest possible utility rating to two of the most popular techniques: highlighting and underlining, and rereading.

Highlighting and underlining received low utility because the research consistently shows minimal benefits unless the technique is paired with additional strategies. The core problem is that highlighting is a passive selection activity that creates a false sense of engagement without demanding the cognitive processing necessary for robust encoding. Students who highlight profusely often perform no better on tests than students who do not highlight at all, and in some studies, heavy highlighting actually impaired comprehension by disrupting the natural flow of reading and encouraging students to focus on isolated fragments rather than integrated understanding. See our guide on remembering textbooks for alternatives to passive highlighting.

Rereading is similarly ineffective. When students reread textbook chapters or lecture notes, they experience processing fluency -- the material feels familiar and accessible -- but this fluency does not translate into improved test performance. Rawson, Karpicke, and Dunlosky (2011) found that rereading produced minimal benefits beyond the first exposure, with effect sizes approaching zero for long-term retention. The slight improvement that rereading produces is almost entirely attributable to the additional time spent with the material, not to any specific benefit of the rereading strategy itself.

Summarisation received a moderate utility rating, but with important caveats. Writing summaries can be effective when students are skilled at identifying key ideas and generating integrative syntheses. However, for many students, especially those with limited domain knowledge, summarisation devolves into copying key sentences with minor rewording, which produces negligible learning benefits. Cramming, the practice of concentrating all study into a single prolonged session immediately before an exam, is actively harmful. While cramming may produce adequate performance on an exam the next day, the resulting memories decay extremely rapidly and are essentially useless for cumulative or final examinations. Cramming also fails to support the transfer of knowledge to new contexts, which is the hallmark of genuine understanding rather than superficial memorisation.

How To Combine These Techniques

No single study technique is sufficient on its own. The most effective learners combine multiple evidence-based strategies in complementary ways, creating a study system that leverages the unique strengths of each approach while compensating for individual weaknesses. Below is a practical framework for combining the techniques ranked above into a coherent weekly study system.

Foundation layer: active recall and spaced repetition. These two techniques should form the backbone of every study session. Begin each session with a retrieval practice warm-up by testing yourself on material from previous sessions without looking at your notes. This provides immediate feedback on what you actually know and reactivates relevant prior knowledge, which facilitates the encoding of new information. Schedule reviews using spaced repetition intervals, either manually or with software like Anki, so that every piece of critical information is revisited at the optimal moment before it is forgotten. Together, active recall and spaced repetition address the two fundamental challenges of learning: initial storage and long-term retention.

Problem-solving layer: interleaving and self-explanation. When studying quantitative subjects or any domain that involves problem-solving, interleave different problem types and explain each step of your reasoning as you work. This combination builds both procedural fluency and conceptual understanding simultaneously. The interleaving ensures that you learn to discriminate between problem types and select appropriate strategies, while self-explanation ensures you understand why each strategy works and when to apply it.

Understanding layer: Feynman Technique and elaborative interrogation. For conceptual material that requires deep understanding rather than procedural skill, allocate time to explain concepts in your own words as if teaching someone else. After working through a challenging section, close your notes and attempt to explain the entire section from memory, noting where your explanation breaks down. Return to the source material to fill those gaps, then try again. This iterative cycle of explanation, identification of gaps, and targeted re-study is the essence of efficient, self-regulated learning.

Supplementary layer: mnemonics, dual coding, and concept mapping. Use mnemonic devices for specific items that resist other encoding strategies, such as foreign language vocabulary, ordered lists, or detailed factual information. Apply dual coding selectively for concrete concepts that benefit from visual representation. Use concept mapping after developing basic familiarity with a topic to organise and integrate your knowledge into a coherent structural framework. These supplementary techniques are not replacements for the foundation and problem-solving layers but rather precision tools for specific learning challenges.

A sample weekly schedule might allocate Monday and Wednesday to active recall sessions with spaced repetition review, Tuesday to interleaved problem-solving with self-explanation, Thursday to Feynman Technique explanations of challenging concepts, and Friday to concept mapping and review. Weekend sessions can consolidate the week's learning through comprehensive retrieval practice that tests all material together, simulating exam conditions. The specific schedule matters less than the principle: consistent, distributed practice using multiple complementary techniques will always outperform sporadic, single-strategy study sessions regardless of the total hours invested.