This study explores the effectiveness of an instructional strategy which displays appropriate knowledge structures versus a learning strategy that engages learners in defining knowledge structures by comparing the effects of providing graphical organizers in the form of completed frames with requiring students to complete frames as a study strategy prior to examinations. The learning variable being investigated is structural knowledge, i.e., the knowledge of how concepts within a domain are interrelated. The subjects were 56 students from a general psychology course at a large community college in Denver, Colorado, who were divided into two groups. Three subject-matter exams were designed to test recall of the text and lecture material, and subscales were developed to measure three aspects of structural knowledge: relationship proximity judgments, semantic relationships, and analogies. Students were also asked to complete a questionnaire assessing the extent of use and helpfulness of the instructor-provided and student-generated maps. Analyses of the data showed that both groups improved their relationship knowledge by focusing on the relationships when using the instructor-provided graphic organizers, and structural knowledge acquisition improved significantly. Between the second and third exams, however, an apparent task-by-group interaction occurred as Group 1 relationship scores appear to have benefited more from the learning strategy than Group 2 scores. Factors that may have influenced the results include the treatment, the content or test difficulties, and the difficulty of the mapping exercise. (12 references) (BBM)