Make It Stick - The Science of Successful Learning - Brown, Peter; Roediger, Henry; McDaniel, Mark (2014) (EN)

Résumé de lecture, chapitre par chapitre — rapporte ce que le livre dit.

Reading summary

🔬 Built from _dossiers/… — DOSSIER.md (150 candidates of 658) + — SONDES.md.

⚠️ 11 modules of kb/24_Sciences_apprentissage cite this book. Nothing here re-verifies them.

Identity

⚠️ A trade book written by a professional writer with two researchers, reporting their own field. It is a vulgarisation of a real literature, not the literature. Citing it establishes that the finding is reported by two of its principal investigators; the finding's weight belongs to the studies, which the book names.

Where the book bounds itself — and it is unusually good at it

On working-memory training, the claim this corpus most needs bounded:

Verified quotations

"Moreover, the study focused on only one training task, so it is unclear to what extent it might
apply to other working-memory training tasks, or whether the results are really about working
memory rather than some peculiarity of the particular training." (probed, 1 occurrence)

Verified quotations

"The brain is not a muscle, so strengthening one" (probed, 1 occurrence)

⛔ This directly contradicts Tuckman (2009), also in this corpus, which called Cogmed the one programme that seems to offer any real promise — ⚠️ italicised: that string is verbatim from Tuckman, not from this book. Same corpus, two sources, opposite readings — seven years apart, and the later one is the more cautious. ⚠️ The corpus should record the disagreement and note the direction of travel, not pick a side by preference.

On its own mechanism claims:

Verified quotations

"What's happening in this situation remains an open question for now" (probed, 1 occurrence — apostrophe courbe dans « What's » à la source)

The finding the book is actually for

Verified quotations

"after one week a study-only group showed the most forgetting of what they initially had been able
to recall, forgetting 52 percent, compared to a repeated-testing group, who forgot only 10 percent"
(probed, 1 occurrence)

Retrieval practice (the testing effect) — the book's core, and the best-supported thing in it. ⚠️ Still a reported study, with its own numbers; the corpus anchors it on the primary literature (module 40, Dunlosky et al. 2013), not here.

Good for / must not be used for

Good for: an accessible statement of retrieval practice, spacing and interleaving by researchers who ran the studies; its explicit cautions on cognitive training.

• ⛔ As primary evidence — it is a trade book reporting a literature.
• ⛔ Working-memory / brain training as effective; the book argues the opposite, with reasons.
• ⚠️ Its figures (52 % / 10 %) without the study behind them.

Provenance