Richard P. Feynman already called modern pedagogics a pseudo-science. Now Karpicke demonstrates the fruitlessness of highfalutin learning techniques and proves the old experienced elementary teachers right, who have always stressed the importance of repetition and testing.
Miller sure won’t be read by epidemiologists. They profit far too much from the ever changing weekly panics and scandals.
Zwane is a clear warning. However carefully you may choose a random sample, by the end of the investigation it will have become unrepresentative, at least for longitudinal studies, and invalidate your results.
Shea asks the right questions about the current debate. See also the commentary by O’Connell.
Eren confirms an impression you get from the literature anyway. There is consensus among knappers that blade technology is comparatively easy to learn and Levallois far more demanding and requiring longer practice. It seems to me as if the proud European artisans hand crafted top quality Rolls-Royces until the blade makers swamped the market with mass produced and standardized Model-Ts. We too have a cultural leaning towards quality tools by I have to acknowledge that from the point of view of most efficient goal achievement the cheap China Exports (CE) from the building centre often are the better choice.
I ought to be ashamed for only now having read the much cited classic by Tversky and Kahnemann.
Crimmins and Kahmen belong together in a certain way. Crimmins shows how neither temperature nor precipitation determine climate but rather a combination of both, while Kahmen offers a palaeometer for just that compound.
Ghosh describes an independent thermometre for cases, where oxygen 18 can’t discriminate between surface temperature and ice volume. To use it we need unburnt cellulose without any extraneous oxygen, though.
Here’s the link to this week’s complete list.