Evidence for Exercise #

Every substantive claim on the Exercise 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 — Forming creative mental associations (vivid imagery) lets you memorize a 15-item word list far more accurately than rote study/rehearsal — the exercise ‘proves the power of association.’

The most current (2025) systematic review/meta-analysis of an imagery-association mnemonic finds a large advantage over plain rehearsal for serial list recall — exactly the kind of task the exercise uses. The author flags high study-level bias, so treat the size as directional rather than precise.

Sources: Ondřej, J. (2025). The method of loci in the context of psychological research: A systematic review and meta-analysis. British Journal of Psychology, 116(4), 930-986. https://pmc.ncbi.nlm.nih.gov/articles/PMC12514325/ · Reggev, N. et al. effect-size conventions: d=0.88 = large per Cohen.

Supported · moderate evidence — Linking the items into a single connected story (narrative chain) is an effective way to memorize a list in order with great accuracy.

This is the foundational and frequently-cited demonstration of the story method and it directly validates the page’s core technique; the ~93% vs ~13% figure is striking but comes from a single small 1969 study, hence moderate rather than strong strength.

Sources: Bower, G. H., & Clark, M. C. (1969). Narrative stories as mediators for serial learning. Psychonomic Science, 14(4), 181-182. https://doi.org/10.3758/BF03332778 · Ondřej, J. (2025). British Journal of Psychology, 116(4) — confirms imagery-mediator mnemonics outperform rehearsal on serial recall.

Supported · moderate evidence — Engaging multiple senses in the imagery — not just seeing but hearing, smelling, even tasting each scenario — strengthens the memory.

Mechanistic and behavioural work supports the idea that multimodal encoding creates extra, independently-cueable retrieval pathways (dual-coding). The neural engram study is in Drosophila so generalisation to human imagery is partial; behavioural multisensory-memory studies in humans concur.

Sources: Zeng, J., et al. (2023). Multisensory learning binds neurons into a cross-modal memory engram. Nature. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10208976/ · Order of multisensory associative sequences reinstated during recognition (2025), Scientific Reports. https://www.nature.com/articles/s41598-025-02553-3 · Paivio dual-coding theory — multimodal traces recalled separately (review).

Mixed · moderate evidence — Making the images bizarre/‘crazy’ (the exercise’s explicit instruction, ‘Be creative!’, and naming it the ‘Crazy List’) is what makes the items stick.

The bizarreness benefit is real and replicable, but the mechanism is distinctiveness, not the vivid imagery the page leans on — so ‘be crazy’ helps for a different reason than implied.

Sources: Besken, M., & Mulligan, N. W. (2022). JEP:LMC. https://pubmed.ncbi.nlm.nih.gov/34351199/ · McDaniel & Einstein distinctiveness account, as reviewed in bizarreness-effect literature.

Mixed · strong evidence — Making the images bizarre/‘crazy’ (the exercise’s explicit instruction, ‘Be creative!’, and naming it the ‘Crazy List’) is what makes the items stick.

ADVERSARIAL: the exercise’s whole list is uniformly bizarre — a ‘pure list’ — which is precisely the condition where the bizarreness boost washes out (and can even hurt serial order). The exercise’s effectiveness is therefore driven by imagery + story chaining + the items being distinctive against ordinary memory, NOT by bizarreness per se; the page over-attributes the gain to ‘being crazy.’

Sources: DeLosh, McDaniel et al. (2000). Order information and retrieval distinctiveness. https://pubmed.ncbi.nlm.nih.gov/10946377/ · McDaniel & Einstein (1986/1995) distinctiveness framework; ’the bizarreness effect: it’s not surprising, it’s complex’ https://pubmed.ncbi.nlm.nih.gov/7738508/ · Besken & Mulligan (2022) — distinctiveness-based retrieval, not imagery.