That is: when students <span style="color: #000000;"><b>didn’t&nbsp;</b></span>review a particular set of facts, they could still reason with them — as long as they <strong>had practiced</strong> doing that kind of reasoning.

Although fact quizzes were beneficial for fact learning, they did not facilitate higher order learning, contrary to popular intuition based on Bloom’s taxonomy.

Contrary to popular intuition, building a foundation of factual knowledge via retrieval practice did not enhance students’ higher order learning. Instead, students’ final fact test and higher order test performance was greatest following retrieval practice that matched in cognitive complexity based on Bloom’s taxonomy: fact quizzes enhanced final fact test performance and higher order quizzes enhanced final higher order test performance. Retrieval practice increased learning by 20–30% under laboratory conditions with college students and also in an authentic K-12 classroom.

Why didn’t fact quizzes improve higher order learning in the present study, as many cognitive scientists and educators contend? First, students may have been unaware that information on fact quizzes was related to final higher order tests, thus they did not transfer their knowledge without explicit instructions to do so.

Mixed quizzes, comprising both fact and higher order questions, increased higher order test performance more than fact quizzes (in Experiment 2) and slightly more than higher order quizzes (in Experiment

If we want to reach the top of Bloom’s taxonomy, building a foundation of knowledge via fact-based retrieval practice may be less potent than engaging in higher order retrieval practice at the outset, a key finding for future research and classroom application.

Those teachers who keep a list of question prompts relating to the various levels of Bloom’s Taxonomy undoubtedly do a better job of encouraging higher-order thinking in their students than those who have no such tool. On the other hand, as anyone who has worked with a group of educators to classify a group of questions and learning activities according to the Taxonomy can attest, there is little consensus about what seemingly self-evident terms like “analysis,” or “evaluation” mean. In addition, so many worthwhile activities, such as authentic problems and projects, cannot be mapped to the Taxonomy, and trying to do that would diminish their potential as learning opportunities.

The categories or “levels” of Bloom’s taxonomy (knowledge, comprehension, application, analysis, synthesis, and evaluation) are not supported by any research on learning. The only distinction that is supported by research is the distinction between declarative/conceptual knowledge (which enables recall, comprehension, or understanding) and procedural knowledge (which enables application or task performance).