I found a set of five experiments similar to the one I described. Getting the wrong answer first appears to be a good method to get to the right answer.

Specifically, we evaluated the benefits of testing novel science instructional content before learning. Thus, the likelihood of failed tests was high, but we were able to extend our theory of testing to better understand whether trying and failing on test questions actually improved learners’ longer term retention of subsequently presented information...in the current study, the prequestions required participants to produce nouns or descriptive statements that they were unlikely to be able to answer on the basis of prior knowledge (e.g., “What is total colorblindness caused by brain damage called?”). This allowed us to isolate and examine the effects of unsuccessful retrieval attempts.

...

Previous research has demonstrated the memory benefits of successfully answering test questions. The five experiments reported herein provide evidence for the power of tests as learning events even when the tests are unsuccessful. Participants benefited from being tested before studying a passage—a pretesting effect—although they did not answer the test questions correctly on the initial test, as compared with being allowed additional study time. Furthermore, the benefits of pretests persisted after a 1-week delay.

...

The current research suggests that tests can be valuable learning events, even if learners cannot answer test questions correctly, as long as the tested material has educational value and is followed by instruction that provides answers to the tested questions.