A study was conducted to demonstrate the value of a mnemonic strategy in remembering information from prose passages and to assess processing differences associated with three variations of the mnemonic strategy. The subjects were 220 eighth grade students who read four short fictional biographies and answered recall questions that were either randomly ordered or ordered to reflect the way the information was presented in the passages. The students participated as members of the following conditions: (1) two control conditions in which students used their own best methods of remembering the biographical information, (2) a keyword paired condition connecting keyword-stimulus and concrete referents at the sentence level, (3) a keyword-chained condition connecting the keyword and the concrete referents of two sentences, and (4) a keyword-integrated condition that combined the keyword with concrete referents throughout the prose passage. Students in the keyword groups recalled significantly more passage information than did students in the control groups. The likelihood of correctly recalling pieces of information that had been contiguously presented within a passage was found to vary as a function of instructional condition and question order. The experimental groups also differed in terms of the kinds of errors made. Overall, the results suggested the efficacy of using the keyword mnemonic strategy. (RL)