I gave a talk at Chicago State University yesterday, the inaugural Scheinbuks Lecture (honoring a new endowment that provides scholarships to outstanding students).  In the post-talk Q&A, I got a question about how to study better so that a student would have better performance on the upcoming test.  As a memory guy, I should really be more prepared to answer this question as it comes up regularly.  But since we focus mostly on implicit learning, most of our direct research has little to say — I can give advice on skill learning, even cognitive skill learning (practice!), but not much on effective memorization.

At the time, I fell back on the suggestion to use various tricks: mnemonics like acronyms, songs, memory palace (spatial) strategies.  It feels lame and I think that opinion was shared by the questioner.  On the way home after, I decided I should come up with a standard answer and I think I have an idea that better connects to our work. I will now recommend 2 ideas:

1. Treat memory use like a skill.  When we study, we often practice encoding by repeatedly reading over the material to be memorized.  We should also practice retrieving the information (a la Karpicke/Roediger — this is just a different account of their finding) to train ourselves to more quickly and easily pull up the information at test.
2. Take advantage of statistical learning in memory through spacing effects.  Per standard advice, study a little bit every day to create long-lasting durable memories (good sleep will help, too).  And then cram like crazy right before the exam.  The other side of the spacing effect is that massed study is perfect for immediate retrieval, suggesting cramming probably works pretty well.

Both these answers tap into ideas about how cognitive processes are shaped by practice and the statistics of use, which are what we study (but not usually in the context of explicit memory).  And I think both parts of the advice probably work pretty well in actual practice.