This study was conducted to test the effect of practice or adjunct questions on learning in a real-world training environment. Subjects were 289 males enrolled in a self-study Naval communications course. Subjects were assigned to one of three experimental groups which received workbooks, or to a control group which did not receive workbooks. The workbooks varied as to the amount of questions that were identical to those in lesson tests or in the final test. In the first experimental group, all the workbook questions were identical; in the second group, half of the questions were identical; and in the third group, none of the questions were identical. These groups were subsequently referred to as the All, Half, and None Groups, respectively. At the end of the course, groups were compared on test and subtest scores, time required, and number of tries on lesson tests. In all cases, the performance of the All Group was superior. The Half and None Groups performed no better on the final test than the control group. Comparisons on subtests showed that practice questions that were not related to test questions could adversely affect both performance and study time. (Author/RD)