What was the headline result of Roediger & Karpicke (2006) one WEEK later?

The tables flipped: the retrieval group recalled 61% versus 40% for the re-study group.

One week after studying, the group that had tested themselves decisively beat the group that had re-read: 61% vs 40%.

The lesson isn't subtle: re-study wins in the next five minutes; retrieval wins when it counts, at a delay. Because real exams and real life happen after a delay, "testing beats restudy at a delay" is the single most important finding in this whole field — and the reason spacing your self-tests out (see the next mission) matters so much.