The finding that reverses the intuition
Roediger and Karpicke had students study prose passages and then either restudy them or take a recall test. Restudying produced better performance when students were tested five minutes later, which is why the technique feels effective. At a one-week delay the pattern reversed sharply: the students who had tested themselves retained substantially more. The students themselves predicted the opposite, expecting rereading to serve them better. That gap between what feels effective and what is effective is the central practical problem in studying, and it points in a consistent direction — the more comfortable technique is usually the weaker one.
It beats the techniques that look more sophisticated
Karpicke and Blunt compared retrieval practice against concept mapping, an elaborative technique that looks far more like real studying: building a structured diagram of how ideas connect. Retrieval practice produced more learning on a later test, including on questions requiring inference rather than recall. This matters because the intuitive objection to self-testing is that it only drills facts. The evidence does not support that objection. Retrieving information appears to strengthen the material in a way that supports using it, not merely repeating it.
What this means for how you use a practice test
Take the test before you feel ready rather than after. An early attempt is a study session with a diagnostic attached: the questions you get wrong are marked for attention, and the act of attempting them helps even when you attempt them badly. Waiting until you feel prepared converts the test from a learning tool into a confidence check, which is the less valuable of the two. Our practice tests score by topic for exactly this reason, and running one first tells you which parts of the material deserve your remaining hours.