One hundred nine college students participated in an experiment that involved reading a passage and responding to a 20-item multiple choice test. The students were randomly assigned to one of four groups. The four conditions involved reading and copying, reading and highlighting, reading and taking notes, and reading and generating questions. Students who generated questions performed better than two of the three other groups on the test. The authors attributed this improved performance to generative learning. (Contains 2 tables and 2 figures.)