www.lesswrong.com/posts/33KewgYhNSxFpbpXg/scientific-self-help-the-state-of-our-...

The post cites "Karpicke & Roediger 2003" in support of retrieval practice, but the referenced paper is The Critical Importance of Retrieval for Learning in Science, volume 319, pages 966–968. That article was published on February 15, 2008.

This is easy to verify from the publication metadata:

• Washington University’s research profile for the paper lists it as Karpicke, J. D., & Roediger, H. L. (2008), Science, 319(5865), 966–968.
• The Science DOI record also lists the article as published in 2008.

So the year in the post is not just debatable formatting; it is factually incorrect bibliographic information. The reference list at the bottom of the post repeats the same mistake, which further suggests this is an error rather than an alternate citation style.

The problem here is the categorical wording: "Cramming doesn't work."

The evidence does support a weaker and more accurate claim: cramming (massed practice) is usually inferior to spaced practice for durable, long-term learning. But that is not the same as saying it doesn't work at all.

Two lines of evidence contradict the blanket statement:

1. A peer-reviewed review in Advances in Physiology Education states that repeated study without spacing—i.e. massed practice or “cramming”—can produce good performance on an immediate test.
2. A study on second-language vocabulary learning found that students using massed practice performed virtually identically on the immediate test compared with distributed-practice students; the advantage for spacing emerged on the later delayed test.

So the strong claim in the post is inaccurate. A fair summary of the literature would be something like: cramming can help short-term performance, but it is much less effective than spaced practice for long-term retention.