Evidence for How learning works #

Every substantive claim on the How learning works page is checked against current research. Here is each claim, how well today’s evidence supports it, and the sources. The full, de-duplicated source list lives on the references page.

Supported · strong evidence — Study strategies that feel productive (rereading, highlighting, cramming) often produce weaker durable learning than slower, more effortful strategies; subjective fluency is a poor predictor of later retention.

A well-documented metacognitive illusion: learners conflate processing fluency with learning. Make It Stick (2014) synthesises decades of cognitive-psychology experiments showing massed rereading yields fluency without durability; this remains a consensus finding in 2026.

Sources: Brown, P. C., Roediger, H. L., & McDaniel, M. A. (2014), Make It Stick: The Science of Successful Learning · full reference ›

Supported · strong evidence — Working memory can hold only a few items at once, while long-term memory capacity is effectively unlimited; learning is the transfer of information from the limited workspace to durable storage.

The severe capacity limit of working memory (Cowan’s estimate of about 4 chunks, refining Miller’s 7+/-2) and the vast capacity of long-term memory are foundational, well-replicated findings in cognitive science.

Sources: Cowan, N. (2001), The magical number 4 in short-term memory: A reconsideration of mental storage capacity, Behavioral and Brain Sciences 24(1) · full reference ›

Supported · strong evidence — Reducing extraneous cognitive load — clutter, unnecessary complexity, doing too much at once — frees limited working-memory capacity for productive learning.

Cognitive load theory is one of the most extensively validated frameworks in instructional research; managing extraneous load to protect working memory has broad empirical and applied support through 2026.

Sources: Sweller, J. (1988), Cognitive load during problem solving: Effects on learning, Cognitive Science 12(2) · full reference ›

Supported · strong evidence — Presenting information through both verbal and visual channels together (dual coding) improves memory more than using a single channel alone.

Paivio’s dual coding theory and Mayer’s multimedia-learning research consistently show combined verbal-plus-visual encoding outperforms single-channel presentation; the multimedia principle is robustly replicated.

Sources: Paivio, A. (1986), Mental Representations: A Dual Coding Approach · Mayer, R. E. (2009), Multimedia Learning (2nd ed.) · full reference ›

Supported · strong evidence — Some difficulties introduced during study (desirable difficulties) feel harder and slow performance in the moment yet produce stronger, more durable long-term learning.

The desirable-difficulties framework (Bjork & Bjork) is well established: conditions that depress immediate performance but enhance retention and transfer are repeatedly demonstrated across testing, spacing and interleaving studies.

Sources: Bjork, E. L., & Bjork, R. A. (2011), Making things hard on yourself, but in a good way, in Psychology and the Real World · full reference ›

Supported · strong evidence — Actively retrieving information from memory (practice testing) produces stronger durable learning than rereading or reviewing the same material (the testing effect).

The testing/retrieval-practice effect is among the most robust findings in the science of learning, replicated across materials, ages and settings and confirmed in large meta-analyses by 2026.

Sources: Roediger, H. L., & Karpicke, J. D. (2006), Test-enhanced learning, Psychological Science 17(3) · Adesope, O. O., Trevisan, D. A., & Sundararajan, N. (2017), Rethinking the use of tests: A meta-analysis, Review of Educational Research 87(3) · full reference ›

Supported · strong evidence — Distributing study sessions across time (spacing) produces better long-term retention than massing the same total study time into one session.

The spacing effect is supported by a large meta-analysis (Cepeda et al. 2006, 254 studies) and over a century of research; distributed practice reliably beats massed practice for durable retention.

Sources: Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006), Distributed practice in verbal recall tasks: A review and quantitative synthesis, Psychological Bulletin 132(3) · full reference ›

Supported · moderate evidence — Interleaving different problem types or topics, rather than blocking one to mastery before the next, feels harder but improves discrimination and long-term learning.

Interleaving benefits are well demonstrated for category learning and mathematics practice, though effects are more task-dependent than spacing or testing; the core finding holds in 2026 with noted boundary conditions.

Sources: Rohrer, D., & Taylor, K. (2007), The shuffling of mathematics problems improves learning, Instructional Science 35 · Brunmair, M., & Richter, T. (2019), Similarity matters: A meta-analysis of interleaved learning, Psychological Bulletin 145(11) · full reference ›

Supported · strong evidence — A review of ten common study techniques rated practice testing and distributed practice as having high utility, while highlighting, rereading and summarising rated low.

Dunlosky et al. (2013) in Psychological Science in the Public Interest is the canonical evidence-graded review; its top-vs-bottom rankings of techniques remain the standard reference and align with subsequent syntheses.

Sources: Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013), Improving students’ learning with effective learning techniques, Psychological Science in the Public Interest 14(1) · full reference ›

Supported · moderate evidence — Transfer — applying learning from one context to a new situation — is harder to achieve than people assume, so practice under conditions resembling the target task helps.

That far transfer is difficult and not automatic is a long-standing, well-supported conclusion (Barnett & Ceci’s taxonomy); the practical emphasis on context-similar practice is sound, though the size of transfer effects varies by domain.

Sources: Barnett, S. M., & Ceci, S. J. (2002), When and where do we apply what we learn? A taxonomy for far transfer, Psychological Bulletin 128(4) · full reference ›